You're correct indeed, max amperage of a device defines what it will take from the power supply. However, is not a robot inductive ? and this device has no protection against any kind of power variation. I'm trying to figure out what happened. Perhaps the fact that I had not connected it to a common ground with the EZB.

@elfege, lack of common ground is very likely the cause. Mine was over-voltage I think, although like I said I was so tired I couldn't even focus my eyes properly, so I had no business playing with delicate electronics.

It was an expensive mistake but at least was not a total loss. I am able to use the USB-TTL adapter for direct connection to EZ-B and it works like a charm. (note, the tx/rx on the USB TTL are not labeled in a standard way. you DON'T want to swap TX and RX when connecting the TTL adapter to the EZ-B. TX goes to TX and RX goes to RX). One of my two EZ-B's will be in a bot with an on-board computer, so this is better than Bluetooth for this machine.

I am trying to decide whether to throw good money after bad and try the Roving Networks adapter I found, which looks to be very high quality (at least if the documentation and web site can be used to judge) or just give up and decide to stick with Bluetooth for the smaller bot without the built in computer.

Maybe I'll save my money for now until I see what EZ-B Revolution is going to be.

Alan

I think the onboard computer is a great idea since it offers tons of software based possibilities that we have to hard-wire and hard-program otherwise, such as remote control over the Internet, etc. However, from a personal perspective, I'm reluctant to do so, otherwise I'll go straight back to what I already know (I used to work as IT until I could realize my true professional dream) and it would be less fun and, moreover, less learning for me since at least now I wrote my very first C++ based programming on Arduino thanks to EZB which is the perfect introduction to the fun before the real work ; all this also allows me to discover more about the hardware itself, even when I fried the module since I have at least one credible hypothesis... I'm however so frustrated not to really understand why the common ground was so necessary. I imagine that the RX and/or TX data had to build up some kind of mass because I figure that it might work like network data, needing outstreaming of empty data, free electrons that do not correspond to actual data, like a kind of natural waste of the process that could not leave the circuit, running in circles or something. If you know about the physics behind this, I'd love to hear about it! I'm veeery curius in nature so forgive me if I'm a little bit off topic!

Can you send us a picture of your module ? Because I had to swap RX and TX in order to have it functioning so I'm wondering if it's not coming from your specific piece. Otherwise I'm really curious to know how you got it working that way.

Elfege.

Unfortunately someone else will need to explain the need for common ground. It is one of those 'things I know' but the knowledge is so old I don't know the answer origination or reasoning. It is either from being in IT long enough that I used to use a lot of serial devices, or from the little bit of electronics I learned in high school physics 30+ years ago.

I know I have seen other equipment (including boards that cost upwards of $5k destroyed by bad grounding.

II'l post a picture of the USB TTL sometime this weekend.

Alan

I learned early on that the devices connected to the EZ-B ALL need to share a common ground. You will notice that even the H-Bridge devices require the same ground treatment.

I tried the WiFi board connected directly to the 780 Roomba com port @ 115,200 k and would receive nothing but garbled characters until the WiFi board shared the same ground plane as the Roomba.

I also experienced an expensive lesson on powering the 3.3vdc required input to the WiFi board by using a 3.3 MOSFET regulator that was rated at 1amp output with a voltage input from 9 to 30 volts. The regulator worked for 30 seconds then passed the entire input battery voltage (16vdc) into the WiFi input leads.......Poof. Good thing I had ordered 3.

Common ground is a must in any integrated electronic circuit as you have to have common baseline for both circuits . In our case the Wifi module and EZB else you will not have a reference for your TTL signal levels between both circuits.

Thank you all of you! By the way, since I installed an H-bridge (before that I was using modified servos's circuits to control my Rovio's motors from EZB but they fried one after the other after a while... I know, I'm the Wizard's Apprentice kind...). Since the Hbridge is installed I get one motor running faster than the other, whether I drive from Rovio's hand or from EZB's controls. Of course I tried to swap the connections on Hbridge in order to check if it was coming from it but it seems it's not the bridge.

I think I found out why I should not use lithium batteries. The magic word is trickle charging : lithium batteries don't last long with such a charging method. So who ever has, as I do, a robot with docking station should not use such batteries.