Project Hearoid.

Let me be the first to say thanks for posting this thread. I really like that model robot and you got a nice one! He has a really cool looking face. I am looking forward to seeing what ideas you come up with and how you impliment them. Good luck, have fun and we will be watching!

I love this style robot and the rad robots too. If your going to run him for long periods then a larger battery is in order. If you stay with 6 volts power sonic and other brands make a 6 volt battery that you can easily fit two of in the bottom of your robots torso. Two 6 volt 12ah batteries means 24ah of capacity.given the original battery powered the omnibot for almost an hour you could have 2 to 4 hours constant run time between charges. Also I've seen great success from other members telling the robot to follow a glyph and getting it to dock like a roomba face first.

The battery is what I'm currently looking in to although it's the last item on the shopping list - for now my ez-b is powered by a mains transformer for testing everything. Once I've got the robot back together with everything else in there I'll have a better idea of the space available for the battery/batteries but it looks like there is going to be more than enough space for them.

The batteries and charging station will be the last thing I look at though. I'm more interested in getting the robot back together, looking like new and moving around again.

My biggest problem at the moment is the face. The blue tinted visor is no good if I want to use colour tracking as the camera will be behind that. I need to either find a Robbie Sr. head (if it's the same size which I believe it is?) and use the visor from that, see how it looks without the visor or attempt to mould my own which may be beyond me.

Besides, the blue visor is scratched and washing it has showed those scratches up even more. I may try to polish them out if it's possible but I have limited experience with that.

I've ordered a mini tilt/pan bracket, it uses mini servos but hopefully they will be powerful enough to move the head, if not then I'll find another way of doing it.

Have rewired the camera from the EZ-B kit to be powered and switched by the EZ-B, although my soldering iron was playing up and only getting hot enough to just about melt the solder so I'll have to redo it. It works but I'm not happy with the soldering. I think it's time I invest in a better soldering iron (mine is just a cheap gas powered thing).

Half of the robot is now clean, the other half is in the dishwasher being cleaned now (small dishwasher).

And I've got a bunch of servos, mini servos, cables etc. on order now too so plenty to play with.

Welcome to EZ-B! I love the Hearoid, but could not find one so I did the omnibot. Here is a link to my build - it may give you some ideas.
Bob was completed and now I've gone back into him to make Gunslinger Bob. Good luck with your build!

@Bret your project is going to be a great help to me as are all of the Omni projects, the body is the exact same so other than my pan/tilt head modification which I think I have figured out.

The Hearoid wasn't my first choice, the Omni was. But when I bought my Omni (after over 25 years of wanting one) he was in such good condition I didn't want to take him apart. Looked for another Omni to modify and come across the Hearoid... I love the head, it just makes him look that much more human like, which is why I really want to figure out a way to make the blue tinted visor clear (well, either make a clear one or find a Robbie with a good condition one).

I love your idea of the EZ-B in the cassette drawer, that's an idea I may be looking at "borrowing"...

I'm currently trying to reuse the original drive but replace the motors for modified servos... I'm thinking opting for a HBridge would be the easier option... But I haven't given up yet...

Rich email me real quick kandyred@gmail.com

@Josh - head swap?

Maybe so , trades are great

Right then, so after messing with the original Hearoid/Omnibot drive trying to figure a way to run it with modified servos I gave up, it'll be easier to use the existing motors as they work perfectly fine.

But, how would you guys suggest going about this? Is the TB6612FNG H-Bridge the best option?

That little controller looks OK to me, a little more hands on with soldering (not a bad thing for some) but a good price and newer, more efficient tech then the L298N based 2.5A. I would slap a heat sink on it even if it seems already rated higher without then the L298N with.

If you are OK with modding the face dome, I would consider drilling a small hole front and center for the camera lens to peek out of (also allows easy focus adjustment if ever necessary). Remove the cameras shell and set up the rest of the head space for the EZ-B and whatnot. Their visible circuitry, LED's blinking and such behind the blue tint would look like the brain it is @Josh did a great mod making a mounting deck under the dome on his Squeegee bot that might work for you as well.

Soldering is not an issue for me, I prefer not to have to do it but I can do it very well - if my iron works like it should unlike yesterday (but that's because the tip is dirty and oxidised and I have no more cleaner left).

To be honest, I'd rather not drill through the dome. Ideally, although perhaps not realistically, I'd like everything like the camera, servos etc. to appear totally hidden on a quick glance but we will see what happens (the elbow servos will be the biggest challenge there). The possibility of swapping the blue for a clear dome is one which is very appealing but that all depends on availability, cost etc. - I've already spent far too much this weekend and I've barely started...

If the clear dome doesn't work out then he still looks pretty cool without the dome fitted, in fact it shows his face off a lot better without the dome and I'm starting to like that look a lot.

Heh, the expense never ends does it I like the idea of your stealth approach... and probably would have realized your goal if I had red your posts over again before offering an idea blush Actually I think a good stealth method shows a much higher degree of skill and planning anyhow... almost an art

A quick update while I am on here...

Currently I'm battling to get the head to pan and tilt. My pan/tilt bracket I bought wasn't up to the job of lifting his head at all, it wouldn't even hold his head. So I guess mini servos are out of the question. Just ordered a heavy duty bracket and servos but it's a bit bigger which will pose an issue when it comes to hiding it (I have ideas though but any others would be appreciated).

Still waiting for my servo extension cables to turn up, I have 2 lots on order from 2 different places and still none have shown up which means I can't wire the H-Bridge and get the base all wired up and ready to go.

So at the moment, it's still in bits and looks like I have made no progress at all, which kinda kills the motivation. Although I have planned everything, cleaned it, currently waiting for the fibre glass to set where holes have been filled... the boring prep stuff - I just want something that moves now though!..

ust a few images of his head now it's got the new "nose" camera and been relamped... Tempted to leave the bubble mask off now (thoughts?)

The mouth light may change. If nothing else it will be activated when the robot speaks but tempted to change it to a chasing strip of red leds (like KITT's scanner)

Eyes are bright white, reason being my house is dark at the best of times so he needs to light up where he is looking with the camera, this seems to do it very well.

I kinda like that look without the bubble.

Finally got the batteries decided upon for this project now, 2 x 6V 12Ah sealed lead acids should do the trick and hopefully fit nicely in the base.

Also have the head all fixed up (short of bubble mask, which I'm almost certain wont be fitted) and the bottom of it fibre glassed and filled to make a good base for the pan/tilt bracket I'm waiting on.

The base of the neck is also all fibre glassed up waiting for the bracket, it's going to be a challenge fitting a strong enough pan/tilt bracket & servos in as I don't want it to have a huge neck but hopefully the bracket can be shortened if needed (or more likely half of it shoved up inside the head).

Was working on the arms last night but didn't have the guts to start cutting out for the servos yet. In teh process of making some mock ups for the arm joints out of thick card, EVA foam, PVA & fibre glass, so far they look like I could use them as the finished part but we will see once it's dried and fitted etc. I assume the standard servos in the EZ-Kit are strong enough to lift an Omnibot arm?

Also put the hands back together... I didn't note how they come apart so that was one hell of a puzzle, thank god for the old robots site and the Omnibot Companion guide!..

Still waiting on a bunch of parts and still trying to find a cheap enough Omnibot that I can use the battery panel cover from as the Hearoid has the Omni 2000 version with the extension ports which I've removed.

Getting there slowly, slower than I had hoped to be honest but it'll get there.

Quick (and short) updated before I go to bed...

Hearoid's neck is now fitted, although it looks too long to me so have a shorter C bracket on order which should arrive tomorrow or Friday. If that doesn't help much then I guess I will have to fix the bracket up into the head rather than fix to the bottom.

It's very top heavy at the moment without the base fitted or any weight at the bottom, which is why there are no videos of it moving just yet (I tried but he just wanted to tumble over causing me to throw the camera to the floor to save the robot.

Eventually, once it's the right length it'll be covered with some flexible duct.

I'll do more images, videos etc. when I get more time.

I have to say I've got so used to no bubble on the head that it just looks right now and with the bubble it looks wrong, which is great news

if you dont need the bubble ,i do refinish them to look almost new

You can make them nice and clear? Can you do that do a robie jr dome?

@rich I keep my solder iron tip clean with 400 grit sand paper. Works great cleans oxidation and carbon right off.

yes,bought a plastic refinish kit and 4000 grit sandpaper and 1000

It may not be much to you guys but it's a great motivation booster getting the HBridge wired up and working.

I'm still struggling with the head, well, not struggling I know what I need I just cant source it in the UK... Basically, the pan/tilt bracket is too long and it means the neck is as long as the head is high, which looks wrong even when the flexi duct is attached... That's tomorrow's project... then on to playing with tracking

can easy cut the bracket and make it shorter,one cut and splice it together or cut it to length and drill holes for the mounting of tilt servo ,can easy cut the end off and use that as a template to drill the holes

That's the plan. I guess I should have just fabricated my own bracket from the start, it would have been quicker and cheaper... At least I know for next time.

Well I've sorted out the neck although I'm not 100% pleased with it but it'll do for now... It's still longer than I had hoped and after having to cut down the C Bracket and in the end making it an L bracket due to friction etc. it's very springy... It's not that bad but will be looking for ways to support it better soon... I may even scrap the pan/tilt bracket and try it another way with the tilt fixed to the bot and the pan to the tilt mech to save space (I literally have about an inch of space to use before the neck looks too long.

The ultrasonic testing didn't quite go to plan. It seems there are issues with black furniture when it comes to ultrasonic and typically, most of my furniture is black. So an IR sensor is also being used (although has it's own issues one being unable to control the Movement Panel but that's a minor setback).

Now, as far as the electronics are concerned it's just a case of building the switching circuits for the eye and mouth lights and building the battery monitoring circuit... once I figure out which batteries to use (thinking some 7.2v LiPo batteries at the moment but that's changed a few times already).

I still need to build the arms and attach them. Stupidly I managed to trim off too much of the original mount so the servo horn doesn't reach the servo when fixed to the arm (as DJs method in his Omnibot), that's something I need to fix but it's nothing that can't be built back up or fixed with spacers etc.

I also need to work out how, if at all, to move the elbow joints without having any servos or linkage on show... or at the very least, servos made to look like they are supposed to be there not just bolted on the side.

It's mainly been playing around with ARC and EZ-Scripts lately, writing scripts for various things like testing the head movement up, down, left & right. Moving at various speeds with PWM. Building dance routines (I was bored, sue me)... just generally getting used to what commands do what, working out how far a Forward(255,1000) command will move the bot, how many ms it takes to turn 90, 180, 270 and 360 degrees, how to make the head nod or shake... Just the basics...

The good news is, in 2.5 days I will finish work for a nice long 2 week break to be spend mainly on the bot, so come mid January I hope to have him (or her, undecided on sex yet, it depends how awkward it becomes and how much it costs) ready to be painted up all nice and pretty.

I'll also be posting more photos, sketches, calculations etc. soon... we don't see enough calculations and sketches on here IMO

Here's his head moving a little... controlled by the camera but really it was just picking up phantom faces (I assume due to low light levels in my house).

Like I said, the neck looks too long to me... It should be half that length.

I had a post about using sonars and IR,I guess you missed it Best way to use them and material that they cant detect,And on sonars it wont detect black material and more. Also ways to reduce the cone slope detection on sonars to pick up small objects in path,as thin as a pencil,one main item is table legs

I also work with metals ,plastics and some gears alot at work,i design at home my own servo gear boxes some better then you can buy (high torgue and low current)

RICH here is a idea i am using on my arms i hate plastics,not very strong,so on the inside of the plastic i cover it with a thin sheet of aluminium and then mount my servo's to it,makes the plastic much stonger and makes it easy for servo repairs nobody really thinks about it,just glue the servo hope it holds I do the same for my whole body of my omnibot 2000 project. Dont need to mount anything to my plastic body,no sensors or anything else check out my project my omnibot 2000 project

I agree rich , i love drawing the ideas for robots body and stuff. Yup his neck does have a giraffe thing going on. On my neck I just used an lynx motion bracket. 10 bucks. Not too long. I lost your email with your address rich email me agian so I can quote you on the clear dome/mask kandyred@gmail.com

I not great in drawings and not good at it,put schematics i am super good at it.only it takes a lot of time I also forgot i have 2 HEAROIDS like your too ,havent made plans on them yet,one most likely restore for my big robot collection BUT the other one will have EZB in it,cant wait for APRIL a very big month no more work ever only build robots

On drawings may need JOSH on that,hope one day to visit him,but so much work,and then travel for work plus my big vacation with my girlfriend for 3 weeks in jamaica I never talk to much about jamaica ,but been there 53 times so far and best place ever to visit ALSO MY SEXY GIRLFRIEND IS JAMAICAN

RICH i see you are in UK,I went there 4 times so far great place many castles to see and people very nice,not like france or some parts of Europe what part of UK are you in

It uses the LynxMotion bracket, it had the long C on it which I changed for the short C and still too long, cut down a cheaper bracket for this but again, giraffe neck. The next plan is to reverse the servos so tilt is fixed to the body and pan fixed to the tilt servo (that makes sense if I had the drawing on me to upload)... If that still looks too long it's going to need the neck cut out and bracket fitted inside the body or head more... It's all about trying different things to find what works and looks right.

@robotmaker, I'm just playing with a bunch of sensors at the moment finding the right ones. I have no thin legs on my furniture so that's not a problem. It's all big and flat.

I'm on the outskirts of the Cotswolds so a few castles around here and a lot of nice scenery. Not too far from Stratford-Upon-Avon (Shakespears birthplace).

@jstarne1, I'm going without the bubble now so his face is exposed.

But, if anyone wants/needs it my email is rich[at]richpyke[dot]net

yes been there nice place,LEEDS castle my favorate, i have lots of chairs in my place with thin legs (bar stools) but the sonar reduce sid slopes not just for thin legs,i guess i didnt add more info on it if sonar has smaller detect zone (side slopes or cone) has a much higher accuracy navigation most problems is the corner of walls ,very hard to detect it,and door ways too pretty easy to make,also you dont use on every sonar only left and right sonars at a angle if using a radar not great 100% percent because first no pinpoint type of sonar ,like reducing side slopes ,second not a great idea to use it on radar because doesnt have a wide angle slope needed so you need both,just some ideas to improve navigation most really good robot builders will tell you for very good navigation you need all types of sensor.IR.sonar,LIDAR ,camera,and thermal and compass ,plus real good codes and a map every made sensor has its good points and bad points i love my maxsonars worlds best sonar for accuracy and super low current ,plus ultrasonic sensor it use is both to send and receive,so takes up less room current is about 3 ma on srf-04 types 35ma,cost it a little high,but i got very lucky instead of the $29.95 i got it for $12 new from a company that going out of business,sorry bought every one they had 30 of them dont thing the ones from china ,same as EZ sells i may not need them will sell them i know my grammar is bad,just no time to fix it

I already have a floormap of my house drawn in AutoCAD which is accurate to a few mm, eventually it may become a complete 3d model furniture and all but I've not had the time to add to it since first drawing it due to my workload. By using that along with calculating the distance travelled in a certain time at specific PWM values for the HBridge that'll be better than any collision detection using sensors (although it will need collision detection to avoid other robots (most likely my "whiskers" robot vacuum which isn't intelligent in the slightest), humans and anything I move or leave lying around... But that's all for future consideration.

This project is to get to grips with the board, with the sensors, with the programming, ARC, EZ-Script and the SDK. It's not a challenging build in the slightest, it's basically one of the run of the mill Omnibot conversions.

In other words, perfect collision detection and object avoidance is not high priority for this build. Future builds may be a different story, my ultimate project I am working up to (life size biped android) will be a different story but that is way in the future and needs a lot of time planning even before the build, calculations to figure out how to balance, how to move lifelike... I could go on but the list of challenges to overcome is already 3 pages and it's only a brainstorming activity at the moment... But I digress, more will come on that bot in a few months/years time.

thats been tried before doesnt work well at all,but does help ,a 3 d floor map it the best for navigation sometimes furniture moves alot in a house also.but using only pwm values is not enough,need compass and sensors to detect any changes in the house

that a design i am working on life size biped android with very real face emotions one reason i bought a very good lathe and milling machine cost was high but worth it on the robots here is this forum is too easy more for beginners a real good robot design i found out takes a lot of planning,like my omnibot 2000 project may take another 6 months or more,for calculations like torque ,making is very strong,balance ,big idea i care about super low wattage,too many dont care about saving current on a robot,i guess they dont care if it last a hour or 2 mine i get it to last 12 hours without going to homebase to charge and lot more on other stuff added to it

robotmaker, If EZB never has what is needed for all your hord of robots then why do keep comming back here? Seems you would stay with the platforms that seem to give you everything you require. Why do you insist on coming here and chat up other platforms over EZB and point out your conceived downfalls over and over again? mad

I GUESS you missed the point the other software lacks hardware and sensors and other stuff that EZB has ,that why its my favorate of platforms most all of my robot designs WILL USE EZB like the AI SOFTWARE,the board they use only have motor and feedback control,7 sonars only and 6 IR plus digital port,doesnt have any near what EZB HAS and any other board and a little slower processor,i think is 32 bit at 35 mhz and at the time (3 years ago) i bought 6 boards at $250 each hope someday to use both EZB AND THE OTHER BOARD together ,so not to waste good money ROBOREALM is working on a interface for EZB ,IT add KINECT,navigation and so many many video filters i know DJ is trying to get better AI,he is getting very close, and so far not many video filter for camera yet so your point is wrong (you would stay with the platforms that seem to give you everything you require) IT DOESNT not right to be mad at someone with that red avatar you put up ,its not very nice

will say it again one last time i love EZB a lot more then you or i would not buy 15 of them i thing is one of the best designs made,but like in lif everything as good and bad points including electronics been building robots for over 20 years and so far i see EZB might be the best

all roborealm is mostly for video filters about a 100 or more ,but many other interfaces roborealm list of interfaces and more get added each month,it will a very good add-on for EZB I heard also someone in this forum said that the EZB is almost max out i never get mad at all ever

Not mad, just tired of the constant negative and limiting comments. One or two are OK but you never quite. I get the point. Maybe I should have used tired

none of my points are negative or limiting at all,just the things i found using both board each has it own faults,but i think you think there is any bad faults at all faults dont make a bad board I LOVE MY EZB way more then you do and in the 9 robot clkubs i pass rthat info about EZB HOW GOOD IT IS, but only a few guys from those clubs like it,thomas and mel are 2 i told it about i dont know why they dont like it ,they saw the site and what it can do one club is one of the biggest robotics club and has club meeting at a college,may i go to the meeting its in calif i dont care for that state,since i travel almost all over the world there is a big robotics convention in china this time,hope my work will let me see it just one day going in march there for work

Anyway...

Finally got around to focussing the camera and getting camera tracking working... Not very well for some things due to the lighting conditions of my house (LED warm white lamps aren't the greatest for light output but haven't finished the living room yet).

Tested in the kitchen where there is much better light and it's pretty good at face tracking, colour tracking and movement tracking. Especially now I've setup the config correctly!..

Front panel for the sensors is now all designed too, ready to be made out of some nice Sintra, cut to fit the upper opening in the body's front (where the digital display is on the Omnibot). I wont be rotating the sensors on a servo for the radar, it will be fixed to face forward only.

Have also decided to go for LiPo 7.2v batteries, to start with a single 5000ah and will see how long it lasts before needing to be recharged. Not ordered anything yet though, no point this close to Christmas, it wont be dispatched until after Christmas anyway, so plenty of time for me to convince myself otherwise (again!.. Gone from SLA to LiPo to NiCd to god knows what else so many times so far)...

I am having a few issues with the bluetooth disconnecting though, it may be brownouts or it may be the bluetooth dongle's range, it's something I need to look at. If it's due to the load of the servos and hbridge I'll just knock up a small board which supplies the servos their Vcc without going through the board. The HBridge power is direct from the battery so I know it's not the DC motors causing it. Could be the battery I guess, it's the replacement 6v 4.5ah SLA I got for the Omnibot which I have read is good for power so chances are low.

And I've finally triggered my electronics knowledge that's been stored in my memory for 15 years but not used in as long, I can't believe I struggled to figure out how to build a voltage divider but that's the human brain for you. The good news is, now I have a new soldering iron which actually gets hot enough to melt the solder, I haven't lost my soldering skills... Perfect solder joints all shiney and rounded (unlike the dull flat dry joints my gas iron was giving).

Finally, I'm tempted to keep the neck as long as it is. I'm not 100% sure if the last youtube video I put up was with the long C, short C or modified C bracket (I can't tell...) but once the arms are attached and it's all painted up in it's new colours it should make the neck look a bit shorter... but we will see... that's an easy change to make if it doesn't look right after then.

P.S. Couple of sketches to explain the neck pan/tilt option I tried explaining earlier (apologies for poor quality, taken with camera as scanner is upstairs and I'm lazy).

on the bluetooth problem i have the same problem DJ SAID it was RFI and i agree with 100 % my setup i have almost all my computers 9 of them on WIFI with 2 on 1GIG SETUP Second is i have a X10 MODULES on everything in my house,inc pool and spa,a/c (i love my X10) 100 % FULL AUTOMATION control of everything including my kitchen stuff second i have one of the strongest WIFI SETUP with special antenna so i dont know what is causing it,one reason using direct connection or WIFI to EZB on my pan and tilt using a custom design 360 pan and full tilt foward ,back and side to side .plus at the same height as orginal has

I kinda agree with dshulpius about robotmaker. We always seem to hear both extremes within the same posting,robotmaker, and while its good to hear of shortfalls(few) of EZ-Board/Builder its does get tiresome . I think your pretty amazing robotmaker with all your experiance but try and make your comments positive and not keep bring up the same negative ones! Thanks for your positive postings and understanding IMO

And, for the love of all things robotic, keep it to your own posts!..

My Hearoid WILL be using the EZB, will be programmed by me using ARC and EZ-Script, with all due respect, I don't care about anything else, at least not in this forum topic. This project is intended for me to learn about the EZ-B and the basics, nothing more. No fancy AI, no linking to billions of other bots.

So, please, use the topics you have already started regarding other boards, software etc.

i disagee with you on some of it,only problem i might keep saying it alot but it nothing really bad about it and other reason person who started the post didnt hear of the faults ,so it mostly for his post only

same if i bring it up again it will be for that persons post too,to give ideas nobody should say my faults are wrong unless that tried that software,so how you you know is good or bad its also good to bring up negative ones or they never will get fixed Thats the main reason i bring it up also not every board made is perfect

ok i look at my posts in this tread it ,i really didnt say much about the EZB BOARD on how bad it is,i might of missed it

some of the points or idea's giving to RICH was first about sonars ,since i know more about them then most people do and other sensors samne about navigation since i tried a few and see some professional designs that only uses sensors,lidar or sonar or compass and so many more and there the best designers in robotis,witch in calfornia i had a good offer there working in the robotics lab

same with the pan and tilt and the arm problem he had,gave idea i use alot because plastic is cheap and breaks alot,plus easy to remove servo's and there are other points too also i did mention about going on my trips (i guess thats not good for any posts) but i see other does it too so thought it was ok i did bring up i buy professional robots ,no matter the costs only to look at it and take it apart and try reverse engineering to use the ideas on my robots

RICH i thought you wanted to here other persons idea ,only trying to help you

I welcome and am thankful for any suggestions and any constructive criticism on my project and how to achieve my goals using EZB and ARC/EZ-Script.

But, your posts about how your AI software is great and ARC isn't are spilling all over every topic, as are many of your other complaints.

doesnt seem from your other post that you like any of mine

update i remove everything not related to this project ,if i missed something let me know like i said only trying to help you since i MAY no more about robots then you,i dont know you or how long you working on robots or working with sensors or software and yes i will keep all my stuff to my projects only ,i sometimes get carried away mostly about my work, Only trying to give you ideas on AI software or sensors or brackets or anything else for you project dont have to use them only a a idea thats all you say you really dont care i think about other ideas since you are doing the program adding sensor by your self,people here might have ideas on script to help you are anything else also never check the software too,so nobody can say its bad or good except me worst thing about it is NO NO EZB HOOK UP.but EZB I AM GLAD is almost there,i hope soon it is

Useless post removed

i would do the same on my projects also if anyone has a good suggestions on how to achieve my goals too

My TIP122s turned up the other day along with the 1k resistors needed to make the switching circuits for the eyes and mouth lights so immediately these were quickly soldered together and fitted to the bot. It's the usual TIP120 circuit so I wont post any pictures or schematics (unless asked for them).

Then quickly adjusted the "head test" script I wrote while playing with EZ-Scripts so it would flash the eyes and mouth 5 times quickly before and after the movement test...

I also finally got around to taking a look at how to mount the servo to move the elbows on the robot today. My biggest challenge is trying to make it all look like I haven't touched it so where possible hiding the servos inside the robot and inside the arms. The other challenge seems to be getting everything in without using hot glue or epoxy as it all needs to come apart again for when he receives his new colours (thinking a slight metallic white for the main body with a nice metallic candy blue for the face, hands and anything else that was originally blue on him, but that might change)

To do this would mean grinding down the original mounts;

Offering up a mini servo seemed to fit perfectly;

Trimming away at the first half of the lower arm so it can rotate, all the red needed to go but avoiding the original mount for the screw;

This does mean the elbow can't rotate a full 180 degrees but can yours?

And the other half also needed the original mounts grinding down (everything with a red x on) for it to all fit back together;

It's all ground down now and ready to be fitted back together (tomorrow's job, it's 2:30am here!) The photos of the parts ground down ready to go are still on the camera so I will post those up tomorrow.

It should all fit back together nicely. A servo horn is screwed to the piece where the join for the upper and lower arm is which should be hidden nicely when all on the robot. Hopefully the mini servos will have enough guts to lift the lower arms - they seemed pretty strong when used on the pan/tilt bracket (but not strong enough to lift the heavy head).

And because I've been grinding and cutting the arm parts and I've not exactly too confident with that just yet I kinda wimped out a little and won another Omnibot on ebay today just in case... the good thing is, the new one is complete in original box too, so depending on it's condition he will either be kept nice and original in his box or swapped out for the one I bought a few months back and the other used for parts if I need any (if I don't then I have my next EZB project)

It feels so much better now it's finally coming along, things are moving and starting to take shape.

Looking really good. I love the movement of the head.

Your doing a great job! Don't worry about the range of motion. Most people only have about 100 degree of motion. My. Shoulders and elbows are 220 to 240 degrees range of motion because I surface mounted the standard servos and connected them with push rods. Its more than anyone needs anyways lol. Can you post exactly how you did your circuit for your lights to turn on and off? Could be a great wiki submission. Thanks bud! - Josh S

I'm undecided on the length of the neck at the moment. It was longer than that too but changed the pan/tilt bracket from using a lynxmotion long c to a short c slightly modified. I may, when I get chance, try just fixing the two servos together using the universal bracket and cutting a slot for the pan servo with the tilt servo fitted to the base...

Apologies for the poor sketch and even worse photo of a sketch.

But then that would restrict the already restricted range (currently 50-70). I guess I will see how bored I get over my 2 week Christmas break.

@Josh, I'll do a video when I do the third one (want the eyes to be switched individually).

In the mean time, the schematic

For ease I used a servo extension cable, pulled off the surround of the female end to expose the three pins. Cut the white wire, soldered it to the 1k resistor, the other leg of the resistor went to the base of the TIP122 Cut the black wire about half way, the half connected to the male connector (EZB side) was soldered to the emitter. The other half of the black wire went to the collector. The red wire was left alone. Using the existing plug on the omnibot wiring I plugged the female end of the extension in to the plug. All soldered joints were covered with heatshrink to protect against shorts.

If using LED lamps the eye circuit needs rewiring due to Tomy's weird way of wiring it up. Also pay attention to the polarity, the red wire needs to go to the positive of the lamp.

You may want to re-write that if it doesn't make sense, it's 4:30am and I've not yet been to bed...

Nice job! Excelent crafstsmenship. You must be giving your drummel a work out. I like the way you used a cable extension and used the end as a socket to slip over the resistor. Useing the remaining wires to complete the circuit was very neat.

Thanks for the schematic. Hope you get some sleep now.

Well I still haven't been to bed... had too much to do for Christmas so just stayed up all night (I'll pay for it tomorrow).

Just spent the past half hour playing with the arm again and unfortunately it seems the mini servos aren't man enough to lift the existing plastic hand (never realised how heavy they are). So I will need to revisit that after Christmas and see if I can find a way to make it work (claws are an option but wanted to try and keep it as original looking as possible).

I may try the possibility of fitting springs in the arms so that there is always a force puling the arm up to aid the servo on the up movement and hoping gravity plus the servo will be enough to pull the arm down. It's all ideas at the moment (so any others would be very much appreciated)

Yes my "dremmel" (cheap mini grinder set which was a quarter of the price of a dremme) is getting a lot of use and my kitchen is getting covered in a lot of small bits of plastic. My own fault for not wanting to go outside in the cold and wet to do it and as I live alone I have nobody to complain at me about it.

A few more photos as I had the chance to pull them off the camera;

And some photos of one of the TIP circuits. Note, the resistor is covered by the heatshrink but it is on the white wire fixed to the base of the TIP122.

And the end of the extension cable which plugs directly in to the existing wiring of the Omnibot/Hearoid head for the eyes/mouth;

Unfortunately it doesn't lock in to anything so a slight tug and it would come apart. That is something I need to look in to, be it a dab of glue, a piece of tape or soldering that end to the internal wiring.

A small update on the arms for the bot. Typically I didn't do any calculations for the servo torque needed to move the elbow/lower arm and the micro servo fitted wasn't man enough for the job so it needed upgrading.

I found the highest torque micro servos I could get which were around the same size, although slightly taller which causes a slight issue but nothing that can't be solved.

I also found I cut the existing mount on the arm where it fits to the torso too short so please excuse the crude card covered fibreglass extension (which is surprisingly strong!).

Anyway, photos...

Tower Pro MG90S fitted which will (hopefully) have enough torque to move the arm.

Currently just superglued in position as it'll come out when it get's painted. I'll probably permanently fix it with glue, fibreglass or possibly just milliput it in, dependant on space available.

Cleaned up the inside of this piece to allow for a smoother movement.

Because the original shaft from the hand clashed with the movement I decided to cut it off and fix it with one of the micro servos originally intended for the elbow. This may not work with the claw hand as the original mechanism goes in to the shaft but I'll cross that bridge when I get to it.

Again, the servo is just super glued in position but will be fitted better once painted.

The elbow is mounted to the servo like so. The horn broke but it's not an issue as I need to space the horn away from the plastic by about a mm or two anyway, and then it'll be covered in plastic to form a nice hump that looks like it has always been there.

Inside the hand the servo horn is fitted with hot glue. I know I have said I want to avoid it where possible but I had little choice when trying to get the horn in the exact position. I may trim the hot glue down a bit at some stage and even may replace it all.

The servo cable for the hand rotation pops out here. Enough of a gap is there to avoid any rubbing from the elbow servo.

Then a nice slot cut in the shoulder for the hand and elbow servo cables to come through, again with enough space to avoid any rubbing from movement.

The elbow doesn't quite have 180 degree sweep but I didn't want 180 degrees anyway, it's pretty much exactly as I wanted.

I'm yet to test it on the bot but have tested it off the bot and it can just about lift the hand from the elbow servo. The shoulder servo in the torso may need upgrading though as the arm is pretty heavy now, that's the next job but need some milliput and plastic card to build up the inside of the torso a little to support the servo and make it hidden from outside the bot.

He is getting there now though, and the software side is looking good too with a bunch of scripts written, servos set up, digital ports planned etc.

Another hint on your light switching circuit,you should add a cap,so the light slowly comes on the light will last longer and no burn-outs

It's not my circuit, it's the one that's been floating around the forum for a while. I did think about adding a cap for smoothing out the voltage (didn't realise it would bring it on slow - by slow how slow?) and a diode (which is sometimes shown) but for the sake of this build, and because I don't have any, I decided against it.

The head is complete now too, I decided against two switches for the eyes as I am getting low on available digital ports. I only have 3 spare and that's without the MP3 board which I may put in there.

It takes about .6 volts to about .7 volts to turn on a transistor,when a cap going to ground as it charges it slowly ramps up the voltage,mostly a mosfet is a little better,but a transistor as voltage is apply to it ,acts like a resistor and slowly turns it on DIODE is not really needed ,only needed on relays for back emf spikes very rare that transistors are used any more for switching,mosfets are much better,lower RDS ON and more

Looking Really Good! I am anxious to see your arm tests.

Me too Bret! I just hope they work as I can't get a stronger micro servo than I fitted and really want to avoid having it on the outside of the arm. I calculated it roughly and it's pretty much on the limit of the servo, and I think that's when horizontal so lifting the hand above horizontal may be a struggle... We will see...

If everything goes as it should (i.e. if my milliput and plastic card turn up tomorrow) I should have one of the arms fitted some time tomorrow for a test.

The arm is fitted and connected up, and I just did the first arm test (see video - it's pretty much unedited so excuse any long gaps of nothing happening while I was changing settings in ARC)

It needs stronger servos for the upper arm joints but I think that's a pretty good result. The hand rotation is perfect and smooth. The elbow servo is borderline OK but it's as good as it's going to be.

The hardest part, fitting the arm! The servo horn for the upper arm is inaccessible with the arm put together but needs to be accessible for fitting to the robot. The arm needs to be together before it goes on the robot. What a nightmare it was feeding the cables through the tiny holes made just about big enough...

So anyway, need to order some heavy duty servos now, the one used was the standard servo from the EZ Kit, so I know I can get much stronger ones.

In the mean time I can firm up the arm, which will add more weight but not that much more so it should still be OK once done.

Are you running 6 volts to the metal gear MG90 servo? It can take 6 Volts and that will be the best torque you can get from it. All of mine are that same servo and run on 6 Volts. A separate power source than from the EZ-B - this will be the optimal for those little servos. Just tie the power to your 6 volt battery and only run the signal wire to the EZ-B

No, until about 2 hours ago I didn't realise I could give them their own supply, it's now on the to-do list for at least the elbow servos (if not all servos) to give them a bit more torque and reduce the chances of brownouts. Although I'm using a 7.4v lipo now so I guess a voltage regulator will be needed but that's no problem.

i would use a LDO (low dropout) regulator,i use them in every robot design,and we use them for all the test equipment we sell at work. there is DC-DC converters too,but they waste more power. 7.4 volt battery is near the border line of a circuit using 6 voltswnen near low battery voltage 7.4 VOLTS mostly good for 5 volt circuits, IT will work only you will have the battery drain fast from servo's

Battery drain isn't a major concern to be honest, if it drains too fast once it's done I'll change the battery for a bigger/better one, or throw more in there (plenty of space and they are very light compared to the old SLA which was in there).

I don't know if something is out there but what I picture using is some kind of servo bus, a supply in of 6-12v, regulated to 6v and then a strip of connectors with the Vcc and Ground... Like the EZB but without the processor etc. I haven't looked yet but that's, if they exist, what I'll be looking to use. Otherwise I guess I'll have to make my own (I may do that anyway)

I guess the SLA I used to test the arm with is going flat (I've never charged it since I bought it and played with the Omnibot I bought it for for a few hours so I am not surprised) as I just hooked the LiPo up to the EZB for another quick test and it moved much better, although will still be upgrading the servo on the top of the arm and running a separate 6v supply to the arm servos.

Now to figure out how to fix the servo horn to the top of the arm as hot glue wasn't man enough for the job. Also need to cut out the old battery compartment in the back as it's not being used now, make and fit some brackets in the robot for the hbridge board, ezb and something to keep the servo wires away from the wheels, cut out for the sensors on the front... So plenty to get on with while I wait for the heavier duty servos to turn up. I may even take another look at the neck and see if I can shorten it (although it's grown on me despite it not looking 100% correct)

sounds like good progress

A small update;

The battery compartment has now been removed (what a horrible job that was to remove whilst keeping the back door usable) and the front is filled with some Jervis plastic card ready to be cut out for the LCD display, volt meters (yes plural, one for each cell), IR and Ultra Sonic sensors.

The arms are still being worked on, a lot of milliput has been used in small doses to build a nice surround for the servo to keep it in there nicely without having to have screws and bolts holding it there from the outside. Although I am having issues connecting the arm to the servo still as the existing support with the horn milliputted to the end is long and the tiny screw seems stuck in there at a funny angle and wont budge (but it will do when I get the drill in there again!)

I've also taken the advice of using 6v direct to the servos for the arms, elbows and hands. I ordered a lipo regulator (Etronix LiPo Regulator 6.0V 5A) which is supposed to be used to connect a lipo to a 6v receiver but as long as 7.4v goes in and 6v comes out I don't care what it's supposed to do. but there is no indication of which way round it goes in the circuit (i.e. no label for supply and load) - anyone have any ideas? It has JST connections (male one end, female the other)? Or am I going to have to try it one way and see if it works and turn it round if it doesn't?

It's all getting there but has slowed down with me being back at work as of this week and having a lot of my own work to do when I get home too (both self employed and employed full time to fund the robot habit but it loses me a lot of time to play!)

Oh, and I almost forgot one of the best things I have done, which is to have ARC interact with my home automation via EventGhost... We have all seen DJs Omnibot watching TV I expect, mine can choose what he wants to watch or listen to (on XBMC)... I'll go in to more detail when I have more time as it's not a simple set up at all (uses a Windows PC, a Ubuntu PC, web server, PHP scripts, EventGhost, XBMC and probably more that I've forgotten about)... But basically, I could have my robot be doing his own thing, exploring or whatever and when he "walks" (or rolls) in to the living room he could stop what I'm watching, throw on his favourite song and start dancing Now that's something you can't do with just any old controller... I have an idea for a video now too but the robot needs finishing first...

My door on my omnibot comes of easy with the right tool I really like EVENTGHOST MY whole house uses automation,every outlet,every light,plus A/C control.pool and spa control ,alarm and more mostly use X10 modules,

Another item you may look into if you dont have already is USB-UIRT it uses EVENTGHOST to control everything IR (stereo's and tv's and more) and with a IR REPEATER system only need one IR SENDER and then emitters on each item to be controlled

ALSO one of my robot uses USB-UIRT module controlled by a AI software

THESE are ideas you may like or dont like

ON the LI-PO regulator, the way i think it works its a step-down to a step-up dc converter or step-up to regulated 6 volts output.

I opened up the regulator but couldn't follow the circuit board, it'll work for any source so I'll throw it on a 9v battery to test it, it'll save setting the lipo on fire or anything stupid like that. Basically in there is a coil, an IC, some resistors and a capacitor - nothing too complicated.

I looked in to IR through EventGhost but most of my stuff is RF controlled anyway (via EventGhost) or has an API for network control and what is IR controlled never gets touched anyway other than the TV power. It's tempting to throw in an IR repeater for those IR controlled devices (basically TVs and amplifier) but it's not worth the cost (both financial cost and the cost of building it all in to the fabric of the house and making it invisible - there's no way in hell I'm chasing out brick walls and replastering again!) but that's on the digital house (not home automation, it's specifically not completely automatic) side of things not the robot (I need to remember this robot is just a pet and nothing more).

here is info on USB-UIRT with EVENTGHOST

USB-UIRT THERE are wireless IR repeaters Thats funny a robot PET mine is my helper,my friend and more working on a design (almost done) to fetch a beer,but since not EZB related cant post it

I agree robotmaker"mine is my helper,my friend and more" since watching bicentennial man (1999) for the first time.

I'll have to watch that film then Always looking for new robot films to watch...

This is my first project of many and really a test of what I can and can't achieve with the EZB and ARC, and testing my construction skills - especially with the stealth servos, not a single servo, cable or extra screw (to original Hearoid screws) externally but servo powered hands, arms and neck (so far so good).

I have plans for helper bots in the future, one being a life size android version of Iron Man, War Machine or possibly C-3PO built totally from scratch using Papakura, card, eva foam, fibre glass etc. (cosplay costume maker methods) for the outer shell and an aluminium or carbon fibre inner frame - that will be quite some build I'm sure but a long way off.

Besides, I already have a helper bot named JARVIS who runs my house (yes, JARVIS from Iron Man was the inspiration for it, mine is a dumbed down, more domestic version) but it doesn't use EZB as he is not a physical robot but a more virtual one, running inside the wiring of the house, inside all PCs through a combination of bespoke software and specialist software - all of which can be interacted with via ARC and the SDK through the HTTPGet function, TCP and Telnet

Lol and im making the physical version of Jarvis

My main robot project is JOHNNY FIVE from short circuit movie got some parts so far,thinking total build is near $500 ,like the R2-D2 project like many others working on and made them already i bought while back the wooden prop from the first short circuit movie AND WEBSITE input-inc.com almost has the plans ready for full detail and full size johnny five

I am more of a robot NERD collecting every type of robots and movies

Just tested my LiPo charging circuit and it works great, if not slowly, but that's fine for this application... I guess charging at 0.2c (or 1/5c) isn't the fastest way to charge but it gives a very balanced charge with next to no heat given off (so little heat that I'm not at all worried about the fire hazard of charging a LiPo and will use it unattended without any fire containment precautions).

Basically what I've done is used the Turnigy 2S/3S charger which feeds, via a 2S balance plug extension, the 5000mAh battery. I cut the balance plug extension and soldered in a servo extension, 1 wire from each of the balance plug wires, this then feeds the cheapo LiPo voltage alarm/tester to save having to change over from charger to tester manually. Eventually this will change to feed two 3 digit 7 segment volt meters, one for each cell, which will be fitted in the instrument cluster on the robots front (where the tape deck once was).

All of this will be fitted inside the robot once it's ready and an extension wired from the 12v barrel jack of the charger up to the original Omnibot/Hearoid charging barrel jack so the 12V 1A power supply plugs in like the original charger would have (and eventually the robot will back on to this when it triggers the low power alert in ARC).

It's late now so it'll be tomorrow (if not later) than I get some photos of this put up.

The next challenge is to move the two LEDs from the charger to the front of the robot so they are in his instrument cluster. And also figure out a way for the EZB to know that the charger has completed (green LED turns off when done, thinking a simple transistor circuit, LED +ve to base, so when it turns off it stops current flow through collector/emitter andt he EZB detects a 0v on an ADC port)

I've taken the advice of 6v to the servos for the arms and it works great now.

It's a bit fast now and needs setting up correctly in ARC but there is enough torque in the elbow servos now.

The top of the arm servos are still going to be changed for some higher torque, metal gear ones too once they turn up but this means I can get on with the other arm and that'll pretty much be him done other than the LCD and volt meters in his chest, and paint of course.

I'm torn on whether to put 6v to the drive motors or go for the full 7v-8.4v of the LiPo, 6v sounds safer as they are fine at 6v, speed is good, noise isn't too bad and I know the motors can safely take 6v.

Thanks for sharing Rich,robotmaker, and Josh, your projects/visions are great! Since just receiving my EZ-Board I hope to reveal FRED as parts come together.....in the Project Showcase(thanks Josh)

You name your robot FRED,thats my name

I've just put him all back together, hidden everything inside and given it a good test and, well, I'm not happy with both how the arm looks and the constant servo buzz coming from one (or more) of the servos.

It may be another story once I firm up the mountings for the servos etc. I've not permanently fixed anything yet as it's all coming apart again when he gets painted so that explains some of the flimsiness that is happening. At least I have spare arms from an omnibot if I need them (although I'll probably just fit the connecting rods on and disconnect the elbow servos as it would be a shame to strip down a perfectly good Omnibot just to take it's arms).

I'm probably also going to do away with the volt meters from his chest area and opt for the simpler and cleaner approach of using the ADC ports to monitor each cell of the battery, which I have to do anyway for the auto charging function I plan to do. He has an LCD display coming anyway which can display those values.

On the plus side, I've now changed the pan/tilt bracket for his neck for a sturdier bracket so it doesn't flop from side to side now.

Anyway, time for bed, I've stayed up far too late working on him again!..

Well today was fun....

Took time out of building to play around with the battery monitor for the auto charge function. The LiPo battery I am using has a balance plug which is there to give an equal "balanced" charge to all cells of a LiPo battery.

My battery being a 2S LiPo battery has 2 cells wired in series and the balance plug is a 3 pin plug which connects to the + of cell 1, - of cell 1/+ of cell 2 and - of cell 2. I thought it would be ideal to run a cable from it to the ADC ports to read the voltage, what I forgot is ground, being - of cell 2, and cell 1's connection give the voltage of both cells, 8.4v, which is too much for the ADC port.

The LiPo battery is wired like this;

So I now have the first cell wired in for monitoring by tapping off the ground and +v from cell 1. I need to do cell 2 which would be between - and + of cell 2 but the question there is, is it safe to connect the ground from an ADC port to between the two cells?

I also spent some time writing the EZ-Script for monitoring the ADC ports, converting the ADC value back to a voltage value, checking for errors with cells reporting 0v and ADC of 255 (i.e. not connected), and also added in a multiplier for voltage division if required after some discussion here.

I also made the Battery Monitor EZ-Script available on the EZ-Cloud if anyone needs it (or wants to give me any constructive criticism.

Now there is a mass of wiring to be tidied up in the torso. Some photos

After a lot of googling for my LiPo voltage monitor I finally found the solution, although it's not what I wanted but it'll do.

Seems like the only way to read the 2nd cell's voltage (and 3rd, 4th etc. if more cells if anyone else comes across this and uses a 3S1P LiPo) is to use a voltage divider, measure, multiply in EZ-Script, minus cell 1 voltage and you're left with Cell 2.

So, now to build the simple voltage divider circuit;

10k resistors are shown as I have those here, is it worth upping them to 100k? Any other additional safety precautions that could be put in place? These simple dividers are all over the net as the easiest way to do it so I assume they are good to go?

And adjust the script, which wont be hard at all.

Nice

As long as the 10k resistors drop the volatge to below 5 volts and using 1/4 watt you should be good to go. Of course a VOM should confirm the voltage before terminating to the EZ-B.....but try and get confirmation from others

I just jumped in and went for it, it works fine It's reporting back pretty much the same as the lipo voltage meter I have so it's all good, although the next task is to figure out how to limit the decimal places, currently they are showing in ARC to too many decimal places, I want it to 2, but that's tomorrow's job.

The total the battery can get to is 8.4v (4.2v per cell) so just halving the voltage did the job. When I get chance I might pick up some bigger resistors to reduce the current draw, for now the 2 10ks wil do though.

better idea is to use a divider for each cell and have analog switch using digital output pins like the 4053 series called multiplexer/demultiplexer chip

with your circuit idea where you say minus 1 cell,each cell has a different voltage as it charged or drain in a robot,so it wont give you a good reading of the second cell

Rich! good to hear(oid) about your bold leap working out! I am thinking the issue of "too many decimal places" is task for ....DJ...unless there is way in your script to strip the extra places.

It gives a spot on reading of the second cell, or at least it did for the 20 minutes I spent testing it.

ADC0 reads the voltage of cell 1 ACD1 reads the voltage of cell 1 + cell 2 So ADC1 - ACD0 gives me the voltage of cell 2

What advantages would the multiplexer/demultiplexer chip give over the resistor divider?

@irobot58 I'm sure DJ has implemented functions in EZ-Script to round off the values to 2 decimal places, I just need to read the manual

you still need voltage dividers,but only one is needed ,way i do it is on your circuit i would use 3 resistors and using only one analog port and would be able to read all 3 voltages using only one porT.

I guess most guys only look at 3 1/2 digits accuracy ,with me i look at about 8 1/2 digits just some thing i i am use to doing. SINCE i have made test equipment of to 24 bits and more robots i guess dont need it since only the A-D is 10 bits ,

BUT how i design my robots ,voltage or current or temperature and more

on resistors for the divider you dont want to go too high,mostly noise ,but 200k total may work

ALSO if you notice i look at noise in the system ,it can cause problems like random resets and more so i use whats called SPG (single point ground ) or wire i try to keep short and twisted oon keeping wire short you save some current ,longer the wire on power you add more resistance where if you do ohms law and wire gauge resistance you will see power loss might not be good at programming but more of a expert in this field

I don't see the benefits of having the voltage monitored over one ADC port, I have 8 ports and only use 1 of them (possibly 2) leaving 7 (or 6) for battery monitor.

There is a reason why I will be rounding it to 2 decimal places, so it looks right on the LCD display. It's an 8x2 and 2 decimal places like "C1 0.00v" takes up 1 row.

Now stocked up on resistors of all kinds of values as I know I will need them on future projects. So I will be trying the divider with a couple of 100k resistors, might try a third 100k on the signal wire to reduce current draw too.

Also threw in a couple more IR distance sensors for the front corners of the base, so that'll make 3 IR (2 in base, 1 in chest next to Ping) and use up 3 ADC ports, but that still leaves me with 5 for battery monitoring, 2 already used, 2 for possible future extra battery and one spare.

Should all be with me tomorrow afternoon, then I can avoid doing the arms for a while longer as I'm still not happy with how they turned out (although is the only part of the build I'm unhappy with).

Hey Rich, take a look at how Josh Starnes did his arms on his omni 2000. After I saw that I wish I had done it with my omnibot (my servo stcks out at the elbow and is not very flattering - but it works great!).

I'll take a look.

The problem with mine is the buzz from the servos when they are fixed in position so it is to do with the weight and servos. As they are the highest torque mini servos I could find and running with 6v I can't go better.

Plus they are a little floppy at the moment as I didn't want to fix the servos in too permanently until after painting but I may have to so I can strengthen the arms...

It's something I need to do a lot of thinking about and always knew I would.

The other issue is rotating the claw hand. Other than the fact I fixed the wrong hand on the wrong side (not a problem, it works), the open/close mech for the hand goes in to part of the hand assembly that I cut off of the other side to make way for the servo and horn. It's either lose the open/close or lose the rotation. Tempted to rob a hand from the omnibot I picked up as a donor for parts but he is good condition and works so it is a shame to ruin him for a hand...

My main challenge was to make the elbow move without a servo showing and I managed that, I just need motivation to get going on it again.

On my omni I replaced the original opening claw with a metal one and was able to keep rotation claw

But I have buzzing issues too. I think it is a problem if there is a load on the servo and it requires some voltage to maintain position. A little shock (like from an RC car) might work to help support the arm so it requires no voltage at a resting position. THat is the idea I am toying with.

Yeah I've thought of some kind of spring to help against the effects of gravity, I may also remove as much plastic from the inside of the arms as possible to reduce weight as much as possible (plus all the hot glue in the hand that was used as a temporary measure but ended up being permanent), I know it wont be much but it will all help.

I saw your claw, that was an idea also but goes against my whole idea of keeping it as close to original as possible. I did also have a few ideas of some hands, like the toy robot grabber hands with the mechanical handle that pulls a wire to close the hand, I saw one wired up with servos and impressive is an understatement (I may have seen it here?..)

when i have time will look at one my hearoid like your and since i design arms May be able to help build a better one,i mostly spend over a few months just designing arms

The idea is to keep it as close to the original as possible, using anything but the original arms will throw that totally out.

It seems the buzzing is louder as the vibration is being amplified by the plastic where it's all a little loose still, I plan to firm it all up at some point which will quieten it all down a fair bit. I know the servo positions to move it to before release so it sits nicely too so the noise would only be when the arms are doing something which I can live with.

Now it's just the claw hand to deal with but if it comes to it I'll just make it so the claw doesn't lock open and it should be fine.

thats very same on my projects,like the omnibot 2000,i keep the hand just way it is

building your own gear boxes like i do ,you get higher torgue and using less or the same space as a servo,and has RC input

One thing you could do on the servos (aka josh) is use dynamat sound deadening mat on the servos. I used it for the arms in my mini-B9 project and it really helps.

I never thought of that. I need some for the drive gearbox so I'll give it a try.

Yea if you get name brand dynamat its even better that "peel and seal" roofing product. You can double the layers as well. I would send you some Rich but im out myself. I sent a few squares to gunner and Bret who both had good luck.

Don't worry about it, I have some already Josh, I've used Dynamat in my cars on many occasions and have a load of small offcuts which I expect will be perfect.

I am not familiar with Dynamat, is that an automotive product? servo buzz is really annoying. I was thinking of using a building product called "blueskin" which is a waterproof material to go around window sills. Or what about using a heavy internal greese,....any thoughts Robotmaker " thats very same on my projects,like the omnibot 2000,i keep the hand just way it is

building your own gear boxes like i do ,you get higher torgue and using less or the same space as a servo,and has RC input" I would REALLy appreciate knowing how you make your gears, or a how to link! Thanks so much!

Dynamat is sound proofing. As far as I've been told it is the best of the best out there, or so the local car audio centre told me. It's used in cars to reduce road noise and to improve audio quality, I assume by absorbing the nasty vibrations through the chassis although I really have no idea. It's sticky backed, so a nice easy to use peel and stick solution too (at least mine is sticky backed, you may be able to get non sticky).

The little fella was sleeping so I thought I would grab a quick pic of how he is at the moment...

I still haven't got around to the other arm yet. I have tightened up the one that's fitted though and the vibration is a lot better. The new servos are here too although I have my doubts that they are genuine Tower Pro servos but we will see how they are when on.

IR sensor isn't staying like that, I was just testing the scripts etc.

And his neck will be covered I was just too lazy to fit it when I replaced the pan/tilt bracket.

Looking good though. Sleepy guy.

@Irobot only put grease where grease goes. Inside a gearbox. Only use lithium grease as petroleum based grease cases plastic old gears to become brittle. Yes inside a gearbox it can reduce noise. One thing ive tried is usinf analog servos over digital in cases where I want them very quiet. They don't make nearly as much racket and practically silent with dynamat. -Josh S

HOW to check the IR without a display tester like arduino ,on the output pin place a meter and to gnd ,place a dark item in front of it ,if it show voltage ,check little far away if it does more voltage then its good and EZB script not working or connection.

Aw don't they look cute when they're sleeping?

I have 2 of them ,so 1 i am going to restore to perfect condition ,on the other one will look at rich and others who made one and how i will make mine and changes

I see most of us like to make our own creation design and use some ideas of others. All of us likes to be different in our designs

I just love the way his head naturally drops slightly to the right when power is killed, I guess the weight isn't balanced but I like it like this so who cares

@robotmaker, I wasn't testing to find distance, I was testing my IR collision detection script to see if it works. The value returned to the ADC port on this IR sensor fluctuates by 1 or 2 when sitting still so accurate distance measurements wouldn't be advisable, but all I want it to do is know if something is within a few inches or so of him.It should be an 80cm range, so an adc of 255 is 80cm or above, or a factor of 0.3137254902 if I wanted to go down that route.

@Josh, I may look in to analogue servos if that's the case. There isn't much space for Dynamat in the elbows without restricting movement so none is on (yet) but having fixed the servo in properly now and closed the arm up nice and tight it's a lot quieter, I'll just have to avoid the servos as best I can when it comes time to paint it.

Just testing the charging circuit now as the battery finally dropped below 3.5v on one of the cells, but with the 800mA balance charger on a 5000mAh battery it's going to take some time (and to think, I am tempted to fit 4 more of them...) I may need to look in to a bigger charger as it can be safely charged at 6 times what I'm charging it at. But that's extra cost that I want to avoid. He was only supposed to be a cheap build and so far it's gone 3 times over budget (my fault for buying cheap rather than good).

it seems you may have on thats not great.ones i have dont fluctuates,maybe because i added a filter you said it not working,way the sensor work is distance and thats only to check it

One you have might have a long distance measurent,if you need to pick collision close you need to get another type

I use different IR sensors for different detecting this is witch one you need for 10 cm to 80 cm GP2Y0A21YK

The ones I have work fine for what they need to do. They are close range proxomity sensors.

I also got the details completely wrong, they are 4cm to 30cm so the factor is (30-4)/255 so 0.10196078431. I only need to detect a few inches at most, so around 10cm which is well within the range. I may have got the ADC value backwards too, I'm not sure without checking.

They only fluctuate over long distances and I assume that's down to the distance, light and probably a whole bunch of other factors.

The data sheet is here

The 10cm to 80cm aren't suitable, I need less than 10cm as a minimum.

@Rich Its interesting how humans "personify" inanimate objects( ships are a big example and sometimes computers/programming)... and yes I too liked Hearoid's sleepy posture, For those of us who have had children, it reminds me of my little ones falling asleep in their high chair! I really like it Rich, when you present the " ooops's/errors" of your builds because two reasons , it shows how human you are with ego not in the way .....and ...we all learn from errors as well! @robotmaker and yes I agree ,everyone like to have their own design (wait till you see mine) very much like a Zoo or peoples plates at a smorgasboard Breakfast time ..I am getting hungry now!

ok you said you needed at near 80 cm you needed to detect thats the reason i bought it up and you said they wasent working and not stable

also i dont know if you know this but one thats analog is mostly for distance,not collision but still can be set as one ,digital ones mostly for collision

@irobot I will be honest, this forum is pretty much the only place my ego doesn't follow me, or at least I try to keep it in check but this is also probably the only place where I see people who are better than me.

@robotmaker, that wasn't what I was saying but the advise will be useful in the future. I've added the one you suggested to my shopping list for another project.

soon then i get my optical bench finish and to help others besides you will put info up on all types of IR'S plus how to make the very stable,they mostly need a simple rc filter.

Need to get parts in and mill them for holders for different sensors and materials.

Awesome contribution robotmaker! A data sheet for various sensors!

i have a keen eye on your topic about the sensors as my big build will need a lot of them for sensing different things, different distances etc. It looks like it's going to be using at least 3EZBs for the servos so that's 30ADC ports (although some will be used for monitoring other things).

But that's not Melvin the Hearoid, Melvin is simple and only needs to detect objects just before he crashes in to them or detect people getting too close. The sensor I used does that very well (I have 2 more to fit to him yet though plus the sonar ping/echo sensor)

i desgn circuits to use only one EZB and many analog ,or digital or servo's

I read through your posts and I had a couple of questions:

Did you need to use high torque servos for your robot's neck?

Can you post of picture of the +6v board you referenced on this other thread: High Power Servo

I'm rebuilding a Robie Sr, so I'm very interested in your build progress.

The neck was a bit of a nightmare but only due to the height it added (and still does). I tried mini servos and the mini servo pan/tilt bracket but it wasn't anywhere near good enough. So I upgraded to full size and had high torque servos in there (although running from the EZB power not 6v). It's something I want to go back to and try to get the height down a little although it's not that bad.

Servos in the neck are Tower Pro MG995.

The board I referenced to in the other topic is very basic... A small piece of strip board (not sure if that's what it's called over that side of the pond), 0.1" spacing with a copper strip from one end to the other on each row. I took a few servo extensions and cut the red and black from the male end, soldered these to the strip board and left the other end so the servo can plug in. Male end plugs in to EZB to give only the signal wire. Power board supplies both +6V and Ground to the servo (or female end) I soldered the +6V and Ground from the 6V regulator to the strip board which gave one row +6V and the other Ground. And viola, basic break out board for 6V.

I would post photos but I never took any and it's currently buried deep inside Melvin.

Arms on the other hand need High Torque servos and +6V to lift and hold without too much struggle. I haven't finished the arms yet as I am not happy with them but the servos now do as good a job as can be expected.

Elbow joints were high torque and +6V mini servos. These still struggle at certain angles.

Hands are just cheap and nasty (suspected) knock offs, they are fine.

Perfect description, thanks!

If you're looking for a new neck bracket, I purchased this bracket from lynxmotion. The tilt range isn't very steep, but the gap between the base of the Robie neck and the shoulder is only slightly more than the height of a servo. I'd estimate about 5 to 7 centimeters altogether, but that's just from memory.

That's the bracket I have on it at the moment. I did have it with the long c bracket to begin with but replaced it for a short c at Christmas... it's still quite long.

I am playing with the idea of fixing the tilt servo flat to the base of the neck, fitting the pan servo to the tilt (with a hole cut in the base of the neck assembly to accommodate the tilt motion) and fixing the head to the servo horn of the pan bracket (if that makes sense). I think I posted a quick sketch some time ago (so back a few pages) of that idea... But it means cutting the base of the neck which I have trimmed down and filled in with fibreglass so am trying to get to like the slightly longer than it should be neck.

The other idea was to cut out all of the part inside the base of the neck (hopefully you know the part I mean) and have the pan servo lower down inside the body of the robot.

I know exactly what you mean.

Your idea in the sketch may limit your head movement...or at least make it a little awkward. If the tilt function is mounted to the body, then you'll be limited to only having a tilt along the forward facing, center line of your robot. Your robot will have a hard time seeing below/above his left and right shoulders and maintain a horizontal based line of sight. The tracking grid lines on the camera views may not function for you if you pan the head rather than turn the whole body.

I may not be describing this very well.

I understood it, I'd thought of that which probably helped understand it a bit. It's one reason I've not changed the neck for 2 months. I've learnt to deal with it the length it is, and it's a lot better with the short C bracket rather than the Long C bracket, although the bracket hits the servo but it makes for a nice stop to save the head leaning too far forwards or backwards and doesn't damage anything.

Was just wondering, did you get a chance to test out your original plan to have him move around by him self or did you move off that idea an into another?

He moves around by himself.

I've changed sensors more times than I care to remember but currently (and may change again depending on the auto charging solution) he runs with the Ultra Sonic Ping sensor in his chest area, a short range IR sensor in his chest, another IR short range low down (just above the blue bumper around the base) and another IR lower down angled at around 60 degrees to detect anything lead on the floor that may be run over (cables, rubbish, etc). These allow for him to free roam, move through doorways and avoid any object in my house.

I may also add in another IR sensor on the back just above the blue bumper for when reversing.

Then a simple script which loops around a forward command with checking the ADC ports and stopping, reversing, rotating and moving forwards again.

Tied in with the Personality Generator he also randomly does other things like dances (not perfected, he has no arms), looks for faces, nods, shakes his head and a few other things.

Do you plan on using the 3D map at all with him? I was thinking that I would use a helicopter with a wirless camera on it to fly around the house, map the house, have this stored onto a hard drive as a GPS for the house. He would fly around every so often or if a robot gets stuck or lost, he could fly to its last known location in memory, and dispatch the updated map to the lost/ stuck robot. What is your brain space on this, is this doable in your opinion? I have yet to learn even the most basic scripting/commanding, so if to you that's just not possible then I'll just scrap the idea.

No. A 3d map would take a lot of work to integrate with ARC and on this project I cannot (or will not) spend that much time and effort on. Possibly future projects may use something like a 3d map but not this one.

Something like that would probably require using the SDK, which I am not familiar with. Learning C# is on the to-do list and I have brushed on it briefly but other training takes priority (I am always learning something new which explains why I seem to know something about everything...).

Honestly, I would work your way up to that project. Start small, learn, go bigger, learn more, go huge That's my plan, this is my first project, others are planned that do more and more things all eventually leading to my big build (life size humanoid robot - years away from that!)

Everything is doable, just some is not feasible.

Thanks, I'll do my best to contain my imagination, lol. Looking forward to see what more you will do Heariod here. Anyway, I do believe it is 2:00 AM, good night.

It's 9:30am here, I can only wish it was 2am again and I wouldn't be sat at work

Adding a map it a great idea ERRUEL ,great for full autonomous robot independance ,witch i like using in all my designs

Only problem is making the map and the time and scripting,i havent look in to it yet,but will very soon

I thought EZB has a map built in,going to look at it EZB floor map

doesnt say if its a 3D MAP ,MOST LIKELY NOT,and no info on how to make a map like other software has.but may be a video on it

The Map in ARC is just a record of where the robot has travelled and will not do what @Erreul has mentioned. It is not accurate enough. The value also differ depending on what surface is being driven on etc. It's based on time not on distance.

I am thinking not about ERREUL idea,but using it with other sensors to map a room for navigation.

So far you adding those sensors in the #1 of this page looks kinda good.

but would be good to have the same on the sides both IR and sonar and 2 at a angle in the front and small cone detection,main reason is to use for detecting wall corners

IF you dont like the idea on your project its great,.just a idea for others

I have no need to detect wall corners. There is no reason to add any sensors to the sides and doing so would go against the design spec of keeping it as stealth as possible.

Not to mention, I am using almost all digital ports already, plan to use all analogue ports and this project will only be rocking a single EZB so adding in any extra sensors is a no go. Also checking that many sensors at a usable speed would bog down and flood the communications between the EZB and the PC, increase the CPU usage and has a possibility of causing major slow down. Checking every 50ms for 8 IR sensors on the ADC ports would be 160 checks per second, when I ran the benchmark I believe (from memory, I will check again when I feel like it) it managed between 70 and 80 per second so would have to be dropped to 100ms between checks for starters and that doesn't take in to account the added stress on the CPU.

Might as well skip pages 15, 16 & 17... skip to page 18

Ok great only a idea,others looking at your build may need it,

Doesnt hurt or matter to me you dont use it,ONLY A IDEA to help you out.

i found ways to add more analog and digital ports without another EZB ,looking to post my design up in a few days and its under $1.00

Unless it also improves the communication between the EZB and the PC then it will be of little use. sleep

It will be the same communcation .only uses one chip with up to 16 analog and another chip up to 16 digitals using mostly one analog port and 3 digital ports for control lines ,so speed stays the same so you are looking at total of 23 analog ports ,with 3 address lines

I hear some using up to 3 ezb in one robot,thats a big waste of money each board is about $70 ,compare to a part thats $.50 PLUS A SMALL PCB FROM RADIO SHACK thats all s needed no other parts.

WOW for once we are not arguing ,thats a first

Yes, I was about to comment that you guys are acting civil. Congratulations, maybe you two can work together to make a super robot.

Thats what i am trying,not looking to arguee and hope the same from RICH,if he doesnt think my idea is good or disagree with me no problem. still think i am correct

I will only say i disagree

I remembered a company who make custom front panels so plan to use them for the front panels of this bot for a very nice, clean, neat professional finish.

http://www.frontpanelexpress.com/

Now I just need to finalise what will be on the front panel and get it designed

P.S. I would never work with robotmaker, it is hard enough ignoring him, as stated elsewhere I will no longer be engaging in any conversation with him for reasons which are obvious. I have not said this to be nasty or to offend, I have mentioned it to clear it up. This paragraph needs no replies, but I'm sure someone will take offence despite that not being the intention.

I feel the same way too,anytime RICH says something about my ideas that DONT NOT WORK best is to just ignore him and say "you say so" and let OTHERS try my tests

BUT last comment i will make on this post ,he cant tell ME not to post about any software like robnorealm,if he doesnt like WELL TO BAD

HE cant keep me from making comments on his posts,and same i cant keep him from making comments on my posts

RICH doesnt want to work together ERREUL thats just the type of person he his

I know RICH he will comment on this

I USED frontpaneldesigner before,my company i work for had many panel i came up with,prices are high but job is good

ROBOT's most of us make dont want to put much money in to it,look at the post on the home base finder,everyone looking for a cheap sensor.

ME i may use it,since i can get a discount and money not a problem.

But depends on the robot design,any time you send to have a company to make custom parts,platEs or PCB'S its going to cost a lot,but if you have a good machine shop you can do it cheaper

Useless post removed

PLEASE IGNORE RICH,look at the prices first then compare results,many other companies make panels and i think most everyone knows anything CUSTOM cost good money,they some on this forum does not have .

do COMPARE PRICES first and see if it fits in your budget

he SAID i dont want to help,read my other comments on helping him with adding other analog ports and the idea i had for adding a sensor

RICH is the type of person who is a angry guy,if you read is other comments on me from others posts ,"saying cant wait to rip my head off" other bad word H**l and other comments

NOT once did i say any of thoses or anything badmouthed

MY my claims are true,did some i test and best thing advice to i give to others is compare or test Onthing i am not good at is programming and windows 7 software,so wont make any comments on them

REPLY AGAIN RICH like always

@EMMEUL as you can see i try to be nice to RICH in a few comments above about some idea for his project,i just could not be a nice guy and not give him any idea's ,but i was a nice guy and tried to leave what happen in past be gone

But as you can see he spit back in my face saying he didnt wont to work with me, so i guess who you can see who is a nice guy and who is not.

SO just the fighting will go on and on

Useless post removed

LIKE i said i cant make any comment on your post if i want,you cant stop me

like you said you will make comments on my posts.

I can do the same also RICH have you removed from making comments on my posts too.been windows 7 post or any other posts software or hardware.

BEST part the EZ-ROBOT team wont like are you bad comments against me,its in the rules also ,no bad words or comment the hurt others like "like you said cant wait to rip my head off"

the EZ-ROBOT team wont see any posts of mine like that

YOU can use the company if you wish also.

And in the post on your design they will see also i tried to be mr nice guy and put the past in behind and tried to help you. I WILL IF I NEED TO REPORT TO THE ez-robot team too

BEST idea i have for you dont not reply to me and i wont have to respond back,very simple

Useless post removed

Any way guys on the the custom panelboards,some good advice check other companies who make them and compare costs first.

My company uses the one rich mention,since i make in house test equipment to test the products before they get shipped,pcb testers,production testers, and qc testers and i have a lot of panels made well over 300 panels only reason we use them,

My company i work for they make 10000 panels for us a month,so when i order one or 2 panels for one tester ,cost very low and sometimes they do it free.

But when we have a panel made for one robot ,cost will be very high ,plus shipping

I see a few guys here make there own panels or decals there self ,so thats another way to save money too.

SO compare your costs first and see if it fits your budgets

As it was asked in this topic and I've just sketched it for another topic I might as well add it here too.

The external power board for servos needing 6V which I described, very simple and works great.

Snip off the red & black from the servo extension which plugs in to the EZB. Solder them to 2 tracks on the strip board. Solder the 6V supply to the strip board (not shown) Viola... 6V power distribution.

I have a PCB design which includes a 6V regulator but it's at work, I will post it on Monday if I remember and have chance.

On my board I left 2 rows between the VCC and Ground in case of sloppy soldering.

I also used a 5A 6V regulator to supply the board with a steady 6V.

Thanks, Rich. Easy to understand.

On the regulator here is the specs unless the company who makes is wrong from the one you posted

?Input Voltage - 6v to 25v ?Output Regulated Voltage - 4.8v ?Output Amps - 5A ?Weight - 19g ?Size - 20x14x49mm

it says 4.8 volts regulated so doesnt supply a steady 6 volts CHECK THE LINK IF YOU THINK I AM WRONG

Once again, please check before posting incorrect information rather than pouncing on any reason to put me down.

Quote:

This cunning little unit allows the use of any battery from 7 to 25 volts to run your RX pack.

Input Voltage - 6v to 25v Output Regulated Voltage - 6.0v Output Amps - 5A Weight - 19g Size - 20x15x49mm

This has also been confirmed by my voltmeter and the following websites;

http://www.jemodels.com/nitro-models?page=shop.product_details&flypage=flypage.tpl&product_id=11599&category_id=248 http://www.cmldistribution.co.uk/cml_product.php?productId=0000005987 http://robotbirds.com/catalog/product_info.php?products_id=1324 http://www.ebay.co.uk/itm/Etronix-LiPo-Regulator-6-0V-5A-/310562839591

RICH is so smart guys he changed his link,so at my info posted you can see it camne off his link he had before,on the new link he changed is NOW CORRECT

WOW guys he wnt very low on doing this

here is the real link he gave first http://etronix-rc.com/etronix-li-po-regulator-6-0v-5a-w-casing-20x14x49mm and specs on it too

This cunning little unit allows the use of any battery from 6 to 25 volts to run your RX pack.

?Input Voltage - 6v to 25v ?Output Regulated Voltage - 4.8v ?Output Amps - 5A ?Weight - 19g ?Size - 20x14x49mm

NOT right to call people liers RICH only open your link and posted the info i saw thats all,,didnt check with a meter only copied and pasted it on my comment all

Maybe he posted the wrong link by accident. Its not a big deal.

i think i might of done that ,that not the problem he said i was wrong

Useless post removed

I think the company is wrong on the info too.since i look at ebay on etronixs, Only i did was to look at the regulator for my design ,like anyone would do and when i saw it had a output of 4.8 volt

I just posted the info for others, to check the same If i didnt post anything i think someone would find the same link and said something too

I always check my specs first on any part first,i know companies posted some info wrong

GOOD that RICH correct the link also

I DIDNT post incorrect info (but to me it seem like i was lier) I posted your post link and what the website says ,so its no incorrect like you say

very simple is i didnt post it,someone would go to your first link and saw the link too and saw the same info as i did

OK since the regulator is good one and info is fixed,i ordered 5 of them for my designs.

i guess i could easy found one like it ,

But good for RICH to find it

IT same me some time

LIKE i said in a few posts not looking to argueeor fight with RICH any more

Only click on the link and posted what i saw ,is all i did

Useless post removed

ok sleep

Since pages 15, 16 & 17 seem to have gotten filled with some trolling etc. I will repost the useful post in the hopes that it doesn't end up being bogged down with rubbish again.

As it was asked in this topic and I've just sketched it for another topic I might as well add it here too.

The external power board for servos needing 6V which I described, very simple and works great.

Snip off the red & black from the servo extension which plugs in to the EZB. Solder them to 2 tracks on the strip board. Solder the 6V supply to the strip board (not shown) Viola... 6V power distribution.

I have a PCB design which includes a 6V regulator but it's at work, I will post it on Monday if I remember and have chance.

On my board I left 2 rows between the VCC and Ground in case of sloppy soldering.

I also used a 5A 6V regulator to supply the board with a steady 6V.

THERE is a easy way to do that too,a barrier strip or power lug strip and it has holes to mounted to a case or inside a robot

For Christ's sake... Could you two refrain from this crap? robotmaker, refraine from posting to Rich's posts and comments and Rich do not post to robotmaker's posts and comments. You two are both grown God damn adults and I wouldn't mind if you started acting as such. I don't care who is starting it, or if you guys need to make a certain post to rip on each other without it spilling everywhere else, but get it squared away. I'm not just saying this because I'm tired of this bickering, but because the community is aswell. So, just don't associate with each other, period, don't correct each other, alert another member. Both of you are smart people, so lets agree to disagree and cut some losses here. No childly "Yeah Rich!" Or "Yeah robotmaker!", lets just move on and do what we love. Robots. Lots and lots of robots.

I do apologize if this has in anyway offended either of you, however I still stand by what I say.

On a happier note: What servo is that attaching to, just the specific ones that need that specific voltage to run?

This board supplies the high torque servos in the arms, giving 6v increases the torque and reduces the humming when holding at specific angles.

JUST posting ideas,just like others do to others posts nothing wrong with that

I dont say anything wrong about his design ,only found a better idea thats all.

IN my last post only said i found a easy way to make one with out have to solder the wires.

Same in my PROJECTS designS if anyone suggest a idea that makes it easy also.

That info i posted was not for arguing or anything else.

One part of this build I was quite interested in was the sticker on the top. Sounds small and insignificant but it will make the world of difference.

The original sticker was a little worse for wear;

So I needed to make a new one. Quite simple really. Start off scanning in the original;

Clean it up and remove everything but the general size and shape of it

Sort out the outline/edge

Add in the robot's name

Then make it look almost as if it was stock (matching the blue of the Hearoid sticker)

Next stage is to print it and stick it. I have ideas on how to achieve this, the first is to use a specialist tattoo paper which is clear but takes ink very well. Until I try it and am satisfied with the results I can't really post the details, but they will come as soon as the robot is painted.

If anyone wants the photoshop file for this, well, here it is... hearoidsticker.zip

This should also fit a Robbie Sr.

I did something similar on my Gunslinger Bob, only I hand cut the letters out of self adhesive vinyl decal material. Not nearly as neat as yours though.

My standards are stupidly high and my skills with a knife are stupidly low (you should see me chopping veg for dinner - I buy it pre cut now...). I did think about vinyl, and it's a back up plan. Luckily I have contacts in the sign writing business I could call on and get the sticker made for cost price but I like to do everything myself where possible. Plus it was fun to design

label you made does look good RICH,me not great at art work

BUT as i see idea that i think is good or better i will post like on barrier strip and why its might be better,not many have circuits to cut ,second hard to mount it

on my idea you unscrew the terminal a little place wire and done.mostly high current and very low cost $1 and they have a strip of terminals at that price and you cut with a wire cutter to size second to mount it they have holes on the bottom and takes little space

Slick idea Rich

I think being a Canadian I try to be a peacekeeper in any situation, haha. I think I remember you posting on Roborad? I understand that the creator did a name as well, did you consider using his process?

See the previous page. It's plan B but as my knife working skills are next to none I thought I'd try this way first. If it comes to it I'll use a sign writer to make the sticker out of vinyl, one owes me a favour or two anyway but I prefer to do everything myself where possible. I know I wont be able to produce the sticker anywhere near as good by hand, for a start the slight gradient on the background blue would not be achievable.

True, that would take weeks by hand, and many frustrating crumpling of expensive parchments. That is from experience.

Not to mention band aids.

BRET did a really good job on the labels in ROBORAD PROJECT

Hi Rich,

Nice job on the robot! Maybe sometime I can pick your brain on some questions I have with my build.

That head is really cool...

Den

From what i see RICH design looks good,only saw a few changes i would make i have 3 of them at home and start on the someday soon,his design might give me a few ideas just i tried ti give him a few.but turn in to problems ,but thats in the past and hope over

@captor03 pick my brains any time.

I've still not finished this build; The arms need sorting and I'm in two minds to change the complete arms to something else or re-fit the connecting links that make the lower arm move. But whatever I do I need to do more dremmelling and create mess. The neck needs sorting with some grey plastic cover, I have white 100mm ducting which I can use if it comes to it but would prefer something more like Omnibot 2000's neck piece. I have some at work on the portable air con unit but it's being used... The front panel needs designing, making and fitting, all should be simple enough to get done though. LCD Display needs sourcing, fitting and programming. Sensors need fixing permanently. Paint needs to be applied. Decals need fitting.

So not much left to do. I haven't had time for any robot building time for about 3 weeks but it'll get there eventually.

Can we see a movie some time soon?

,waiting on the codes from my friends on LIDAR ,second since i dont now SDK thats the problem,but it is so very simple ,and DJ is will be very easy with a C# CODE since SDK uses C# and most the work it done already.

Plus there others who bought one from me to use on EZB and they know it needs a SDK,and most likely they are a programmer

Plus if anyone buys one from me and makes a code to EZB they get 50% back just to speed it up most of it should be very easy since it has all codes and was done easy in ROS

ANY good programmer should be super simple,i think only time holds them back

I can also let anyone borrow my board for LIDAR ,since i bought 2 of them.

@Erreul There are a few videos floating around but nothing great yet. When I get chance (and motivation) to bring him out of the spare room I'll do some videos.

In the mean time;

And

Oops, that last one isn't a robot...

I don't know which one has the smoother movement, the robots elbows, or your moonwalk!

Den

Oh! A wall! Haha. Both look pretty good, the robot and the smooth moves.

I see the camera is very fuzzy why is that and will you post a better one

I see on the head only going up and back so side to side ,dont meen head rotate ,on mine i design it so go up and back rotate 180 and side to side

ON dancing its not a robot doing it,but my nephew is a very good dancer,won a few contests ,but yours still good

The robot eye view video is just out of focus, nothing a quick tweak of the lens wont cure. it's also dimly lit which is sorted with the settings in ARC but video capture doesn't capture processed video only live video.

The head pans and tilts, tilts between something like 45 & 70 (can tilt up to about 95 but that just looks wrong) and pans the full range but I've set it for 20 each way (30 & 70).

NOT about pan or tilt ,just wounding are you going add side to side head tilt

looking to set mine like a person head does ,from shoulder to shoulder tilt down i guess its 45 deg and tilt back about 15 deg and side to side i guess about 30 deg on each side

Does need a post with a ball and socket and 2 extra servos and ball and socket design i have there are other ways too ?.

maybe this one post will explain it better

No, I thought about it but it would be too much effort making it work and having a neck that's not stupidly long (I've shortened it from the video posted) or too flimsy. The head weighs quite a bit, more than I expect it would.

That's what I've tried to replicate, a human head range. It's not 100% lifelike without the tilt but it's close enough that it's not noticeable unless you look for it.

ok just curious havent tried yet on hearoid yet,soon to be another project of mine

mine head doesnt stick up much ,omnibot 2000

IF head is heavy ,see if you can remove any plastic,i do that on mine,and still not flimsy

Few other ideas i keep all servos in the body,and the shaft with the ball is the par thar rotates and the ball moves 360 every way,can hard to expain iy

but when ready if you want to look at will have it up on my omnibot 2000 project

Looking really good Rich. You have made some good progress!

yea I said it too he did do a real good job on it,wish i had more time like he does,but in a month (counting the days) i will

The 6v on the elbows make such a difference, as can be seen in the videos... in fact I needed to slow down the arm servos. Removed the arm again recently though to get it a bit more solid and fix it to the shoulder servo a bit better.

I've neglected it for about 3 weeks now though but hopefully I'll find time to get back on it this weekend and get the sensor mounted properly, fit the servos in to the right arm and sort out the shoulder servos.

I'm really tempted to add a slight riser to the base too, just to give it a little more height so the arms can hang down straight and make the neck look more proportionately correct. What I would love to do is give him working legs and turn him in to a biped but that's a little beyond me at this stage.

BIPED desigh is hard,i ordered a kit with out servos,needs a gyro and more

You should teach it to moon walk.

I plan to!.. I have some videos planned and the frames set up, he just needs his arms before it'll work right

I was bored so decided to move my schematics over to Circuit Lab, and then got carried away by drawings the Schematic for melvin.

As you can see, I have used the EZB pretty much to the maximum possible. I think I have 1 ADC port free (planned to be used for current monitor) and 1 Digital port free. Some things may need to be combined to make way for the possible future MP3 module and 4th servo on the arm for opening/closing the claw.

cant see the schematic very clear ,both clicking on it and the link

You should try pcbexpress its easy to use and imgage can easy made any size ,plus up to 4 layers and then have a circuit board made for it

ExpressPCB costs far to much for one off PCBs, the cheapest is over £175 delivered to the UK, multiply that by roughly 1.5 and you get dollars...

I do use ExpressPCB for schematics but the problem is I can't use it everywhere, this I can hence moving everything over to it. Works on tablets and any PC regardless of OS and all stored in the cloud so it's better than ExpressPCB on that account. Plus you can simulate circuits to make sure they will work, check voltages at specific points etc.

Following the link and choosing Open In Editor, as mentioned above will allow you to view it, zoom in, pan around, copy and paste etc.

But if that's too difficult for you here's a bigger version (click it)

OK.expresspcb in usa is super cheap PCB I HAD MADE IS 19sq inches total cost me $89.12 for 2 boards including shipping, 2 layers plus holes miniboard service is 3.25 inch by 2.5 inch 3 boards with shipping and holes $51

plu on the first one for about $89 if you buy more price really goes down alot per board

wow $262 prices are high in UK

WILL TRY the EDITOR

also looking at my pcb i guess i forgot i see its 3.75 by 2.5 inch so if i move the parts together with i can easy do it will cost me $51,i need to do that tonight

Finally got the robot back out of the spare room to do some work on him tonight but that didn't quite go to plan when I checked the voltage of the LiPo and it was 3.6v for both cells... It looks like one of the cells is now under voltage if not both which sucks!.. (I may try a dangerous repair by charging it as an NiMh if I can find something fire proof and am feeling like risking burning down my house)

2 more are on order though. Now I need to decide if I want to rock two in parallel or keep one in reserve just in case I make the same mistake again. Do I really need 10,000mAh of juice?

A small set back and further expense but it'll be worth it. It just means I need to wait a little longer to buy that second EZB for project 2 (details will come shortly but want to finish this one first).

Also decided to redo the small circuits for the switching of the lamps and the battery monitor using some stripboard and adding in the Zener diode to the monitoring circuits (since I had to take the whole thing apart to get the battery out anyway I might as well).

So no new photos yet or videos. But it's the perfect opportunity to make some tutorials on those circuits. Every cloud has a silver lining and all that

I did the same thing to one of my lipos. Try recharging it outside.

Den

Looking forward to it. Wouldn't mind learning everything I can.

You need a special charger for LI-PO ,it needs CC and CV out and a balance board

On the current need its little hard,but a good rule of thumb is find out total current is needed and hours you want it on and add 25 to 30 %

Useless post removed

LI-PO dont have a memory effect,and can keep them on the charger after they are charged ON yours may be hard to tell what happen

If a LiPo goes under voltage charging them is dangerous and can cause fire and explosion, which is very well documented on the internet. Most good LiPo chargers have safety features in to prevent any potential problems.

My chargers, both the expensive all singing all dancing one and my cheapo one refuse to charge the LiPo. I had it on charge for 6 hours yesterday, when it started it was at ~4v, 6 hours later it was at 3.6v...

Charging as if it's an NiMh will bypass the safety features but can cause the battery to violently burst in to flames (check youtube, plenty of videos of that). If I do that I will not be doing it near my house I know that much.

YES mostly every li-po chargers have it built in,thats why is called CV puts outs a Constant Voltage and if goes over or under is shuts off and goes in trickle charge same with CC (constant current) is has a monitor circuit and they have a built in temperature monitor for shutdown

BUT there a few cheap LI-PO chargers that dont have this feature,NIMH chargers really dont have any monitors

WHAT model charger do you have

ANOTHER edit,i mostly make my own charging designs ,so i cant have any problems of over heating or fires

look at the NEATO XV-11 BATTERIES very well made NIMH types but with a temperature monitor plus CC monitor

I thing on you case you got some bad batteries or out of date

Useless post removed

another try on batteries is to zap it fast with a higher voltage,it may help

I think that's what they mean by charge it as if it's an NiMh. While it may help and may get it going again it's also the most dangerous thing you can do to a LiPo... before I even think about doing that I need some kind of fire proof container and do it outside away from my house, I'd hate to burn my house down all for the sake of a £25 battery...

But the 2 new ones are on their way and should be here tomorrow or Monday (depending on if the delivery service work Saturdays) and I may be able to get back to building the robot.

Rich I did the same thing and charged it as NIMH. The battery worked for a couple of charges then died. As long as the battery is outside in the dirt, away from combustable materials you should be fine. When they go bad they light up like a flare. I seen it first hand! Just charge it at a slow rate, check it every 5 minutes, then after the voltage gets back up there switch it back to lipo on the charger. If you see it start to puff standby!

Den

HOPE the new ones dont give you a problem so far i still lucky no problems with any type of my batteries,i have some robots use LI-PO and some using GEL CELLS and others using high current NIMH packs i made

The odd thing is, part of me wants to see it go up in flames Is that bad?..

WHY you did such good work on your robot

The battery not the robot!

I know you meant the battery,that poor battery hate to be in his shoes

You'll be disappointed. It's only a good flare, and smoke show. Don't try putting it out, and it will melt almost anything you put it in.

Just don't ruin the charger...

If you get a chance Rich send me an email at qkstrik@cox.net.

Den

Sorry, I did see the other message earlier but completely forgot about it. Just sent you email...

Well since my battery is toast (not literally, I didn't bother turning it in to an expensive flair) and the new ones haven't turned up yet, nor have my resistors, diodes etc. for the small circuits I want to remake I've not really done much on the robot today, but I have re-written my init script (the script which automatically runs when I start the project from the shortcut).

It still needs a little bit of work but it's getting there. A few bits I need to check but can't until I get the new battery.

Feel free to grab bits of it or ideas or whatever, here it is...

# Init script for start up
# Give verbal commentary
Say("Initiating Startup Sequence")
ControlCommand("Speech Recognition", PauseOn)
Sleep(2500)

# AutoConnect

# Configuration
$noofEZBs = 1
$maxattempts = 3
$waittime = 8000

IF ($noofEZBs = 1)
  Say("Connecting Easy Bee")
ELSEIF ($noofEZBs > 1)
  Say("Connecting Easy Bees")
ENDIF 
$attempts = 1
:begin0
IF (IsConnected(0) = FALSE)
  Print ("Board 0 not connected")
  Print ("Attempting connection")
  Print ("Attempt "$attempts" of "$maxattempts)
  ControlCommand("Connection", Connect0)
  Sleep($waittime)
  $attempts = $attempts++
  IF ($attempts > $maxattempts)
    Goto(failed)
    IF ($noofEZBs >= 2)
      Goto(begin1)
    ELSE 
      Goto(end)
    ENDIF 
  ENDIF 
  Goto(begin0)
ELSE 
  Print("Connected to board 0 OK")
  Say("Board Zero Connected")
ENDIF 
  
:begin1
IF (IsConnected(1) = FALSE)
  Print ("Board 1 not connected")
  Print ("Attempting connection")
  Print ("Attempt "$attempts" of "$maxattempts)
  ControlCommand("Connection", Connect1)
  Sleep($waittime)
  $attempts = $attempts++
  IF ($attempts > $maxattempts)
    Goto(failed)
    IF ($noofEZBs >= 3)
      Goto(begin2)
    ELSE 
      Goto(end)
    ENDIF 
  ENDIF 
  Goto(begin1)
ELSE 
  Print("Connected to board 1 OK")
  Say("Board One Connected")
ENDIF 
  
:begin2
IF (IsConnected(1) = FALSE)
  Print ("Board 2 not connected")
  Print ("Attempting connection")
  Print ("Attempt "$attempts" of "$maxattempts)
  ControlCommand("Connection", Connect2)
  Sleep($waittime)
  $attempts = $attempts++
  IF ($attempts > $maxattempts)
    Goto(failed)
    IF ($noofEZBs >= 4)
      Goto(begin3)
    ELSE 
      Goto(end)
    ENDIF 
  ENDIF 
  Goto(begin2)
ELSE 
  Print("Connected to board 2 OK")
  Say("Board Two Connected")
ENDIF 
  
  
:begin3
IF (IsConnected(3) = FALSE)
  Print ("Board 3 not connected")
  Print ("Attempting connection")
  Print ("Attempt "$attempts" of "$maxattempts)
  ControlCommand("Connection", Connect3)
  Sleep($waittime)
  $attempts = $attempts++
  IF ($attempts > $maxattempts)
    Goto(failed)
    IF ($noofEZBs >= 5)
      Goto(begin4)
    ELSE 
      Goto(end)
    ENDIF 
  ENDIF 
  Goto(begin3)
ELSE 
  Print("Connected to board 3 OK")
  Say("Board Three Connected")
ENDIF 
  
:begin4
IF (IsConnected(4) = FALSE)
  Print ("Board 4 not connected")
  Print ("Attempting connection")
  Print ("Attempt "$attempts" of "$maxattempts)
  ControlCommand("Connection", Connect4)
  Sleep($waittime)
  $attempts = $attempts++
  IF ($attempts > $maxattempts)
    Goto(failed)
    Goto(end)
  ENDIF 
  Goto(begin4)
ELSE 
  Print("Connected to board 4 OK")
  Say("Board Four Connected")
ENDIF 
  
:failed
Print("Connection Failed")
Say("Connection Failed")
$attempts = 1
Return()

:end

# Avoid HBridge random movement at start up
Forward(10,1)
Stop()

# Reset all servos
ServoSpeed(D1,0)
ServoSpeed(D2,0)
ServoSpeed(D7,0)
ServoSpeed(D10,0)
ServoSpeed(D11,0)
ServoSpeed(D12,0)
ServoSpeed(D13,0)
ServoSpeed(D14,0)
ServoSpeed(D19,0)
Servo(D1,50)
Servo(D2,60)
Servo(D7,50)
Servo(D10,50)
Servo(D11,50)
Servo(D12,50)
Servo(D13,50)
Servo(D14,50)
Servo(D19,50)
ReleaseAll()

# Start script for battery monitor
Say("Monitoring Batteries")
ControlCommand("battery", ScriptStart)
Sleep(4000)

# Start script for collission detection
Say("Starting Object Detection")
ControlCommand("irdetect", ScriptStart)
# Move sonar radar randomly then center
ServoRandom(D7,35,45)
ServoRandom(D7,55,65)
ServoRandom(D7,35,45)
ServoRandom(D7,55,65)
ServoRandom(D7,35,45)
ServoRandom(D7,55,65)
Servo(D7,50)
ControlCommand("Radar Scan", PauseOff)
Sleep(1000)

# Test the lamps
Say("Checking Circuits")
$blinkcount = 0
:blinkeyes
IF ($blinkcount <5)
  Set(D8,on)
  Set(D9,on)
  Sleep(100)
  Set(D8,off)
  Set(D9,off)
  $blinkcount++
  Sleep(100)
  GOTO(blinkeyes)
ENDIF 
Set(D9,on)

# Set servo speeds for head and arms
ServoSpeed(D1, 2)
ServoSpeed(D2, 2)
ServoSpeed(D10, 3)
ServoSpeed(D11, 3)
ServoSpeed(D12, 3)
ServoSpeed(D13, 3)
ServoSpeed(D14, 3)

# Test the head movement
Say("Checking Movement")
# Head movement code here...
# Slowly move head tilt to 70 then 50 then 60
:headtilt1
IF (GetServo(D2) < 70)
  ServoUp(D2,1)
  Goto(headtilt1)
EndIf
:headtilt2
IF (GetServo(D2) > 30)
  ServoDown(D2,1)
  Goto(headtilt2)
EndIf
:headtilt3
IF (GetServo(D2) < 50)
  ServoUp(D2,1)
  Goto(headtilt3)
EndIf

# Slowly move head pan to 35 then to 75 then to 50
:headpan1
IF (GetServo(D1) > 35)
  ServoDown(D1,1)
  Goto(headpan1)
EndIf
:headpan2
IF (GetServo(D1) < 75)
  ServoUp(D1,1)
  Goto(headpan2)
EndIf
:headpan3
IF (GetServo(D1) > 50)
  ServoUp(D1,1)
  Goto(headpan3)
EndIf

# Dance
ControlCommand("Script Manager", ScriptStart, "Dance1")

# Turn the rest on
ControlCommand("Speech Recognition", PauseOff)
ControlCommand("Camera", PauseOff)
ControlCommand("Personality Generator", PauseOff)
ControlCommand("Variable Watch", PauseOff)

# All done
Say("System Ready")

Wow, that's a heck of a start up script. Nice.

Do you have problems with your H-bridge jumping around at start up?

Sometimes it decides it wants to move in circles. I think it's the HBridge not the EZB as it'll do it before the EZB is connected (in fact it was doing it when I last tried to use it with a battery that was screwed and only had 4v in it). But a little work around solves the problem.

It's the TB6612FNG Motor Controller which I don't read about often so no idea if it's a common issue or not...

Might try another motor controller ,guess to see if you still have the problem

Everything indicates that it is the motor controller but it's fine 4 out of 5 times, maybe even 9 out of 10 times so it's not worth the extra cost or the hassle of rewiring. If it cause that much of a problem I'd throw a switching circut in there to disable VM until connected then a quick and easy Set(D#) before the Forward command.

I just decided to redo one of my TIP122 Transistor Switching Circuits (mainly because I needed a break from working - there is nothing wrong with the 2 that are already in this bot). So now's as good a time as any to do a small tutorial

Parts needed: 1 x TIP120 or TIP122 Darlington Transistor 1 x 1k ohm Resistor 1 x Small Piece of Strip Board (7x5 holes) 1 x Pin Header (1x2) 1 x servo Extension Solder Soldering Iron Cutters

The Schematic:

Underside of Board:

Method:

1. Solder the TIP transistor to the strip board so each pin is on a separate strip of copper

   

2. Solder the resistor from the Base of the transistor to a spare copper strip

   

3. Cut off the end of a servo Extension and strip back the wires

4. Solder the Black wire of the servo Extension to the strip connected to the Emitter of the transistor

5. Solder the White wire of the servo Extension to the strip connected to the end of the resistor (not the transistor end)

6. Solder the Red wire to a spare copper strip

   

7. Use a small off cut from the servo Extension and solder one end to the strip of the Collector of the transistor

8. Solder the other end of the off cut to a spare copper strip next to the Red wire.

   

9. Solder the Pin Header to the copper strips with the red and black wires soldered

   

10. Job done. Plug the servo Extension in to a Digital port on the EZB and connect the circuit that needs switching to the Pin Header, I do this with another servo Extension (as I have hundreds of them)

    

Thank you Rich for your detailed closeups of construction! ALSO, your startup script is a solder(gold) mine of ideas and, it is like your closup pics, EZ to understand...Thanks so much...I too hope to contribute something one day to this forum to assist robot builders

There is a better topic for the switching circuit in the hardware section, which I've updated a few times to now include the diode that may sometimes be needed.

I did start a voltage divider tonight but got a little confused while putting it together and it went a little bit wrong... but it's one I need to make as this project uses one (and while it's apart and I'm still without battery I'm going to remake it).

Rich,

I've been looking over your project and I am very impressed. Thanks so much for posting your code. It gives me something to work from. I am close to being ready to dive in to writing script. This is really a big help to me.

If you ever want to email me, my email is rex.gordon61(at)gmail(dot)com

Rex

I'm all about showing others how to do things and offering as much clear information as possible. It helps that for 15 years part of my job has been writing the technical documentation for post building construction and CDM regulations are a bitch I haven't posted anything close to the amount of documentation I have on this build (a lot would be pretty pointless).

Recently, since draining the LiPo to under voltage I've not been able to do much with the robot so been taking the time to re-write scripts, remake circuits, document it all, label wires and sort out the cable management.

I'm glad people find it useful and hope it helps them build their robot and program their robot.

And, as always if people need to ask things or discuss things off of the forums for any reason my email is rich@richpyke.net (I get enough spam so I don't bother with the (at) and (dot) malarkey)

When I actually scrounge enough money, with has been quite slow to be honest, would you be down if I emailed you any questions I have when I get it? I do understand you basically just said yes in your last post, I just feel obligated to ask for personal allowance, just how I am.

Yep, an open invite to all. Although if its something that may benefit the community its always best in the forum (I tend to check here a few hundred times a day).

Melvin gets a new mouth

Not much of a write up on this one, I'm too tired to explain it all but a picture paints a thousand words so...

The Evil Mad Scientist Larson scanner.

That's too big for the mouth slot so some alterations are needed. A piece of stripboard (my last piece, exactly the right size too) for the LEDs makes it smaller.

Then some servo extension cables cut in half so it can be removed, replaced and makes it just that much easier to fit.

The other half on to the circuit board

Then pop it in the mouth slot (after removing the old lamp) feeding the wires through the existing hole in to the head.

Put it all back together and voila

Unfortunately the Evil Mad Scientist Larson scanner runs on 3v, the EZB digital ports give out 5v. While there may be a 3.3v tap off on the EZB I don't plan to use it. I will be building a small 3.3v regulator circuit and placing it inline to take the 5V from the EZB down to 3.3v to power the scanner. A TIP122 circuit is used to turn it on and off. The mode select (hopefully) shouldn't need to be used.

Since I wasn't expecting delivery of this for some time as I ordered from America I was unprepared for the voltage regulator so this will come later on. For testing purposes (and for the video) it was running from 2xAA batteries.

I'm not 100% sold on the red LEDs and may change them for blue or even white. Red still looks cool though and like it belongs so I may just leave it.

Something else on today's list (which I actually did before the Larson scanner above) is the LiPo battery monitor.

Full instructions on how it's made will follow when I have tested it, tidied up the schematics and have the time to upload the photos. But for now, here is the finished product waiting to be fitted (and tested of course).

From this:

To this:

Simple design and removable. The 2S balance plug connects on to the 3 pin header on the left, two servo extension cables (plug to plug type) or servo plug to pretty much any 3 pin JST connector with 0.1" spacing fit on the other two 3 pin headers. They connect to the ADC ports on the EZB and a script monitors the voltage levels, warns of problems (over voltage, under voltage, low voltage etc.)

Basically, it's a voltage divider with over voltage protection which measures the voltage of both cells (full charge of 8.4v hence the divider), combined with a pass through for cell 2 (full charge of 4.1v), the script then calculates cell 1's voltage by subtracting cell 2's voltage from the overall voltage thus leaving cell 1's voltage.

As mentioned, full tutorial including schematics and script to follow when I have more time and have tested it.

Thanks Rich for sharing your project, this scanner led "The Evil Mad Scientist" is great, by the way I discovered the store did not know her, and lipo voltage divider for two elements is a good work, I think I will add to my list pending a scanner parts and components led to the voltage divider, good job, sure many others benefit from your project.

If anyone else plans to use the Larson scanner with their EZB you needn't use a voltage regulator (like I planned above).

Rather than adding in a voltage regulator for the 5v to 3v all that needs to be done is for the resistors to be replaced. Since the ATTiny2313 can accept up to 5.5v it shouldn't be a problem at all.

LED specs are; Forward Voltage 2.1v Forward Current 30mA

Which basically means replace the 33ohm resistors for 100ohm according to my calculations.

looks like you did agood job in the circuit RICH on mine using bar graph display circuit to work with sound not happy about beening in china,but i guess work comes first before FUN

BLue led wont be as good as RED ,since the head is blue Unless you going to repaint him

also see you added the idea i had for the ZENER ,good practice to have protection and i forgot also if the battery is wired wrong it saves the circuit,from what i remembered Havent tested the theory yet

Whats the ATTiny2313 for,does the larson project use 4017 decade counter,(one of my favorate chips )

ALSO there is a way for the larson project to move with sound,i made the design awhile back for my work,might have to redesign it a little so not the using the work circuit

I know i cant use any of there designs for profit,but willask if i can use it for my home project

It's being repainted.

Yes, added the diode but in a different position to what you suggested, I have it directly across signal and ground rather than across R2. According to my calculations and various circuit simulators it works there and other information online suggested to put it after R3.

The Evil Mad Scientist Larson scanner is totally open source. The pre-programmed micro-controller (the ATTiny2313a) doesn't support sound input, it only offers 3 speeds, 2 brightnesses, narrow and wide eye and "tie" mode but could be reprogrammed if you want. I do not intend for it to be sound activated, it just needs to be a Larson scanner so the kit (or KITT if I was going for a pun) was perfect for my requirements. There are other Larson scanner designs about but this one suited my every requirement, was within my revised budget and offered suitable means of modification to work with my project without getting too involved.

The schematics, pcb layouts, code... everything is available from the Evil Mad Scientist Wiki, as mentioned it is completely open source. That details exactly what the ATTiny2313a does.

from the EVILMAD SCIENTIST site it says the old circuit uses 4017 and 55 timer i see the new one uses a micro.they must of made a change fairly soon

ON diode it can work on ACROSS R3 or R2 but protection is not as great,mostly because of a cold solder joint to ground,it can happen ,just not much often buthaving a double ground (diode and resistor to ground ) is better then one ground (only resistor to ground )

Main reason my designs are so perfect,use a lot of protection,plus low current to save on battery life

To me is most inportant item,less weight of adding big batteries and longger run time using LDO DC to DC CONVERTER is a lot of wasted power,but iguess only like it because because easy to wire in,NO SOLDER IS NEEDED

Useless post removed

Replaced the resistors in the Evil Mad Scientist Larson scanner to 100 ohm and soldered the jumper on Opt1 in while I was at it. Now it should run off of a 5V supply (although I don't have one to test it until my bullet connectors come for the new LiPos (they come with 4mm, my old one had 5.5mm).

Not much time today with it being Mother's day in the UK so probably wont get chance to test out the LiPo balance port monitor, but it'll come soon enough

Rich, you have some great ideas! This robot is going to be nice when it's done.

Den

I love the Larson scanner, it looks really good in that mouth. This build is going very well and your work is very impressive. Also, you have a lot of good ideas. Thank you for posting them.

ONLY mention LDO because you are using dc-dc conveter in your project,just to give you another idea thats better,dont have to use it only a idea

Thanks Den & Bret.

Being a perfectionist I don't accept anything that's not going to be great. It's a curse sometimes though, which is why my robot has no arms at the moment...

An example of how picky I am...

Look at the resistors, each one around the same way even though not required. All bar one perfectly centred (all are centred now I've changed them for 100 ohm). It just bugs me so much when people don't take the extra second or two to do things properly (you should also see every screw in my house, door handles, socket outlets, everything... all heads are the same way around).