Connect with us

How do I ground an independent dual/triple power supply?

Discussion in 'Power Electronics' started by ironspider, May 25, 2012.

Scroll to continue with content
  1. ironspider

    ironspider

    2
    0
    May 25, 2012
    Hello all, my first post here!

    Okay, so I just got started in electronics and I've just been doing very basic stuff with LEDs and learning about capacitors and transistors and such. I am a bit of a woodworker so I built a little workbench in my basement and, while outfitting it, got a "triple linear power supplies" (this one: http://www.amazon.com/gp/product/B004ISD7T6/ref=oh_details_o01_s00_i00 ) which allows for two separate variable voltage outputs as well as the fixed 5v/3a.

    Everything has been great with it so far but I just encountered my first conundrum. My brother gave me a CPU board from a malfunctioning pinball machine of his and to test it I need to put 12v, 5v and a ground on it. I thought "no problem, I can do that PLUS an additional one if needed :) ". Then I realized I have no idea how to do it. I know how to get positive voltage and ground from each SEPARATE output but not how to make it work together.

    What I mean is that the +12v and +5v on the CPU board share the same ground on the board--but I can't figure out how to make the power supply do this--heck, I'm so new to this I don't even know if it CAN.

    What I ended up doing to just get some testing in was to use an old computer PSU I took apart. That got me a 12v signal, a 5v signal and a ground. I was able to get the board to boot this way. The problems is that the voltages on those things don't always seem to be steady. Tonight, for example, the +5v rail was sending out 5.27v which is too high for the board to successfully boot with :(

    So is there any way I can get two different voltages out of this bench power supply and use a common ground betwixt the two of them? Or is this not possible and I'll have to possibly build a DIY power supply?

    If the latter, anyone got any links to a good DIY one I can build? Preferably something that could do two (or three) variable outputs?

    Thanks in advance!!
     
  2. jackorocko

    jackorocko

    1,284
    1
    Apr 4, 2010
    I would check with my multimeter to see if the grounds on the PSU you bought have a common connection. Even then, unless the manual tells you specifically not to do it, I would see no problem in connecting the two grounds together on the pinball board.
     
    Last edited: May 25, 2012
  3. CocaCola

    CocaCola

    3,635
    5
    Apr 7, 2012
    I would first try just making the two outputs use a common ground aka hook the ground from both together and then to the circuit being tested... Before you do this see if there is a warning in the manual that says it's a no-no or might offer guidance so the unit doesn't complain or shut down...

    *** jackorocko was faster on the keyboard

    Short of that it's pretty easy to build a little modular voltage regulator board you can run off the same single supply next to the test board, personally I would use a series of fixed regulators on one board with a dip switch to turn them on/off, since 99% of the stuff I do is 3.3v or 5v but you could use an adjustable one as well, and switch the value on that...

    I wouldn't even get fanny on the regulators, I would just use your basic L7805 type regular for a quick and dirty, a few caps to smooth it out and you should have a nice alternate voltage rail...

    You might want to consider the max output of the regulators, the normal 500mA or 1A ones might not suit all applications...
     
    Last edited: May 25, 2012
  4. poor mystic

    poor mystic

    1,068
    31
    Apr 8, 2011
    Hi
    I think you are right to suspect problems. I agree that there is too much variablity in the supply. You need to know why, and correct the problem.
    You are always welcome to post photographs and circuit diagrams.
     
  5. ironspider

    ironspider

    2
    0
    May 25, 2012
    Wow, thanks for the quick replies!

    Okay I set my DMM to DCV, turned on the benchtop power supply (the Mastech one not the ATX PSU), connected black the Mastech black leads (one from the 12v variable supply I set and one from the 5v variable supply I set) to the black probe. I then touched the red probe to one red lead from the Mastech and it read 11.99 volts and then touched it to the other red lead and it read 5.01v. So I think that worked! I then connected it to the pinball CPU board in that matter and the board did boot!

    Since I'm kind of new to electronics let me ask a dumb question. After I read these replies I went to my Mastech and set my DMM to continuity mode and touched both black leads--there was no sound. Is that okay? I mean to say does that mean they're not a common ground? Or is that not how you determine and the fact that I got the correct voltages off the pinball board when I tested it is all I need to know?

    Thanks again!
     
  6. CocaCola

    CocaCola

    3,635
    5
    Apr 7, 2012
    They very well might be isolated when 'off' and only common when each supply is powered up...
     
  7. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,480
    2,826
    Jan 21, 2010
    The important thing is that the outputs of your power supplies are not connected to anything else (each other, or ground).

    If this is the case, you can make any single connection between them the common rail from which both supplies are referenced.

    A common way of doing this is to tie the negative leads together causing both supplies to be referenced to a common level (called ground even if it isn't grounded!).

    The problem is if the supplies are already grounded and you connect outputs together in such a way that a non-grounded rail of one power supply is connected to the other. This can result in you shorting your power supply(or supplies) to ground via their earth leads.

    With them turned off (but plugged in) you would test this by measuring the resistance between outputs of the two power supplies (everything reading infinite ohms is good). Alternatively, you can measure the voltage difference between outputs of the power supplies when they are both turned on. Any reading indicates that the supplies are not isolated.

    There are still some gotchas, and I prefer to (after I have tested as above) to initially link the supplies via a resistor (say 1k) and measure the voltage drop across it. It should be either zero or a very small fraction of a volt.

    Once you know that the power supplies are safe to connect in this manner you don't need to subsequently test them every time...

    You may need to be careful if your load is earthed though. But that's another (although related) issue.
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-