Welcome to Synthiam!

Program robots using technologies created from industry experts. ARC is our free-to-use robot programming software that makes features like vision recognition, navigation and artificial intelligence easy.

Get Started
Asked — Edited

Voltage Regulator Vs Dc To Dc Converters

My Antonn project is having a power problem which I hope someone can help me with. I am supplying the controls with 12vdc and 5vdc from an old computer power supply. (The 5vdc runs the Io Tiny and servos which works fine.)( 12vdc runs a remote controller board.) I have two more power requirements. I require 7.2 vdc and 3.8 - 4.0 vdc power. The 7.2 vdc supplies a Scary Terry Audio Board and the current shouldn't exceed about 500 ma. The other device uses 3.8 to 4.0 vdc at about 150ma. There was no data on the smaller device, so I need to guess. (origionally ran off of a LiPo battery.)

I purchased a DC to DC converter able to supply the correct voltages and currents from the 12vdc and 5vdc supply I have available. I wired up the converters, set the proper voltages, but the converter fails to work with either device. I assume the converter supplies a type of PWM DC power. The units seem unable to tolerate the pulsing.

I have a limited electronic knowledge. I need to figure out how to build two regulated supplies, one able to supply 7.2 vdc from 12 vdc and the other 4 vdc from a 5 volt supply. Can anyone help?

AI Support Bot
Related Content
Sorry for the delay. I cheated the 4 vdc. I put a diode in series from the 5 vdc supply and got it to 4.10 vdc. Next I need 7.2 / 7.4 vdc. From 12 vdc. I will post the converter picture when I get home. Thanks for your efforts..
This is the Dc-Dc Converter

Input 3.2 to 40 v
Output 1.25 to 35 v
Current (MAX) 3A
Switching Hz 150kHz
Ripple 30mV(MAX)

User-inserted image

I won't work with my application.

I installed it in Antonn, now I will have to cut it out and try something else. There must be a 12V to 7.2V Regulator. I got to look around.

I eliminated the 4 volt problem so that problem is gone.

I am concerned with switching regulators.
Hi @ptp,

Any advice on using a LM7908?

I never used that one, but the L7805 (5v) is well known, i still use them...
is very simple to interface you need only In, GND, out.

You should add some capacitors to remove any oscillation/noise.

*** Note:

You want the LM7808 (positive version) not the LM7908 (negative version)

*** Specs:


Typical Application (page 18)
2 ceramic capacitors 0.33 uf and 0.1 uf
WOW,,, Now I am in trouble.
I was told to try the talk servo which if it works may eliminate using this board altogether. I still have the need for 12vdc to 7.2 to 8 vdc on another project so your help will not be wasted. I will be ordering a couple of LM 7808 s soon no mater if the talk servo works or not.

Thanks for the help and info. (The cap data is really helpful.)
In line switching regulators are available for servos. So you can drop from your feed power to 7.2 volts, 6 volts or 5 volts. Just whatever your servo needs. You can make them yourself too. Switching are way better than linear because you do not waste all the voltage difference as heat.
I don't know why the buck converters I have don't work? I want to try the LM 7808 s next.
@ptp and @jstarne1,

Just for future knowledge, how clean is the output of the buck converters?

Can I add filtering, or should there be enough on the converter?
Since they are for power you shouldn't need any filtering. Switching regulators are pretty efficient, usually over 90 percent. However results may vary depending on your supplier.
My application is to power an audio board. That is why I was curious about more filtering.
as long as your are only powering the board you are fine, switching regulators use between 25000 and 40000 hertz, which is likely outside the audible range a amplifier would pick up
I hope to get the scope next week. I will look at the converter input and output . It is weird the converter won't operate the board. I will confirm the power requirement of the Scary Terry board. It should work fine !


Oh Well ! I guess Antonn will have an extra cable after all !

I built the LM7808 power supply. NO GOOD AGAIN !

I still have almost the same problem as I did using the DC converter. I guess the computer power supply is introducing some kind of pulse which is screwing up the Scary Terry Board. I ended up pulling out all the wiring and supply and ran a standard 9 VDC 500 ma wall power supply. The jaw runs fine!

Well, it ain't broke so I don't need to fix it. Someday I will get a scope out and maybe see what is going on.

He is working well and soon will get a software upgrade.

Thanks to all who tried to help.
I feel your pain. I'm not working on a voltage converter, but an audio project I'm doing is driving me nuts! It's *so* frustrating when it's "supposed to work" and doesn't!

Glad you found a work-around (I did too). I just wanted to let you know I'm enjoying your Antonn build - "First Robot With A Mustache"!

I need to be honest, I still might keep trying to figure this out. One more test! I plan on trying a different power supply tap with a DC converter or the "7808" to see if the other devices connected to the original 12 VDC tap is introducing the noise. I hope the new tap eliminates the feedback. In the future I want to use a large 12 volt gel cell battery to run everything so I need to work this out. For now everything is working so I'm happy. Thanks for the sympathy on the "should work" thing.

Antonn thanks you for the compliment. He is pleased with his appearance and was surprised how fast it grew in. He hope to use Dj's API plugin in the future.


In the future I want to use a large 12 volt gel cell battery to run everything so I need to work this out.

If you have a 12 volt battery, you might want to try it and see if it resolves the issue on its own. Power supplies can inject noise into the line that batteries can smooth out.

I used to work for a company that made phone switches for contact centers. We used to run the customer's AC power into a rectifier and then through the backup batteries for 2 reasons. The primacy was for instant cutover to battery backup if the customer lost power, but the second reason was that it reduced electrical noise in the system by always running the switch from the batteries. Customers who had their own UPS system and didn't want batteries in their switch rooms always had problems with noise on the phone lines.

Hi Alan
Thanks for the tip. It is funny you mention using the battery. I use a battery system like you described on my Madame Ninndo,s EZB system and have no problems with noise.

I will try it. Thanks.
Ok time to revisit this screwy problem.

I am doing a final assembly of Antonn. I am using a computer power supply to give me 5 vdc @ 18 amps to run the IoTiny and servos and 12 vdc @12 amps to run a Scary Terry sound to servo board. When both voltages are hooked up my audio is making a rhythmic clicking sound. I tried another power supply and got the same result.

When I use the 5 volts only to the IoTiny all is fine. If I hook up the Scary Terry board again, I get the noise. If I use a plug in the wall power supply for the sound board, all works fine. I also tried to install a dc to dc converter with the computer supply to see if it would help filter it, but I still get the clicking. Everything works correctly. I will not be using the battery on this robot. He will be stationary and plug in the wall.

I don't know why the computer supply would inject this noise. Can anyone give suggestions or advise or tell me why this happens ? I want to use one supply but it seems maybe the computer supply isn't going to work.