Let's Make Robots!

Quick question - voltage regulators in parallel (sorta)

I have a 8.4V 2/3A 1100 mAh NiMH battery from an airsoft gun (charger was included!) and I read that the raspberry pi and most USB devices tolerate only up to 5.25V (okay I guessed 8,4 was way too much). I need to regulate it to 5V. I thought of using a low-dropout voltage regulator, but the shop nearby only seems to support voltage regulators of a max of 1.5A. I have calculated that my robot can have 3A surges if all systems are on and the motors are stalling (extreme scenario, but it might happen). So with that in mind, is it possible to just add two of the voltage regulators, one lets say for the Pi and USB stuff, and another for the motors, the Arduino and LED + buzzer? To my understanding it should work, but I'm no expert and I don't want anything to blow up. No magic smokey. 

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
birdmun's picture

Don't forget to tie the grounds together. :)

OddBot's picture

As bdk6 pointed out, a seperate regulator for the Pi is a good idea.

Poor efficiency and heat will be the biggest problem. You really want a switchmode regulator.

Markamas's picture

Yeah, switch mode regulators are cool. http://www.pololu.com/catalog/product/2177

Thanks for pointing that out! 

antonio.caciuc's picture

I really like the voltage regulators from http://www.dimensionengineering.com

I bought several of these: http://www.dimensionengineering.com/products/de-swadj3 for regulating power to motors or other "big" consumers (it could also easily power a PI and an USB hub)

And some of these http://www.dimensionengineering.com/products/anyvolt-micro, for powering arduinos from either 1 LiPo or 3-4 NiMH cells.

bdk6's picture

You can connect as many regulators as you want to the same battery (or other source of power) as long as you don't connect the outputs together.  so, yes, you could power the RPi from one, something else from another, etc.  That was often done in old multi-board computer systems so they could use smaller regulators (like you were doing) or have the regulator closer to the circuit to prevent voltage drops.  In fact, if you are going to run something noisy (like motors) from the regulator, that is a better idea anyway: it helps isolate the systems and helps prevent the noise from getting from one circuit to the other.  Plus, you can use smaller heat sinks.  Keep in mind that if you are putting in 8.4 volts and drawing 3 amps, you are going to have to get rid of more than 10 watts of heat.  Ten watt soldering irons get hot: I have scars to prove it.

Interesting,

I can remember the Calcomp engineer coming out to fix our 960 vertical flatbed plotter in the late 70's.  The voltage reg had blown (5A job). Took him to the local electronics place as I was into electronics even then and we had to settle for 1A regulators - so we simply connected 5 in parallel and then connected into the circuit board.  Worked for a couple of days until he replaced with the correct part.   We used to work that plotter 24 hours a day so it certainly was stress tested :)

antonio.caciuc's picture

It usually works fairly well if the regulators are closely matched in the output voltage with low ripple.

But if the output voltage is slightly mismatched current may start flowing the wrong way through one of the regulators, depending on design...

The DE-SWADJ-3 swichmode regulators from dimensionengineering worked just fine in parallel. I needed 2 of them in parallel to supply the 5 volt line in a mini itx computer. I had another two in series (powered from two separate batteries) supplying the 12 volt line and another one powering the 3.3v line.

Note that I messed up the motherboard when I powered the 5 volt line without powering the 12 volt line I guess some parts of the circuit were connecting the 12 volt line to the 5 volt line and they didn't like the reverse voltage :)

bdk6's picture

Connecting the output of regulators together is a bad idea.  Sometimes you can get away with it, but probably not for long.  The output of the regs won't be EXACTLY the same.  Say one has an output voltage of 5.1V and one has an output of 5.0V.  The 5.0 V one will be trying to pull the output of the 5.1V one to 5.0 and the 5.1V one will be trying to pull the output of the 5.0V one to 5.1.  In effect, they are trying to fight each other.  The higher voltage one will end up carrying most of the current and you will likely end up with 2 (or more ) blown regs.

Thanks bdk6, that explanation makes sense

Markamas's picture

http://www.abra-electronics.com/products/LM350K-Positive-Voltage-Regulator.html

Oh well, seems like I have found it. 3V <= Vin - Vout <= 35V and 3A max. Silly me. For the record though, would my original solution have worked?