Author Topic: Is there a simple hack to evaluate power supply  (Read 442 times)

0 Members and 1 Guest are viewing this topic.

Offline sairfan1

  • Regular Contributor
  • *
  • Posts: 203
  • Country: ca
Is there a simple hack to evaluate power supply
« on: May 12, 2021, 09:38:37 pm »
I bought some cheap Chinese power supplies 12V 10A and 24V6A I know they are not rated correctly i want to know their actual performance.

I do not have a load tester or high wattage resistor is there some way to get an idea whats their actual capacity?

if I have to buy something to evaluate these module what should i buy?
 

Offline Doctorandus_P

  • Super Contributor
  • ***
  • Posts: 1545
  • Country: nl
Re: Is there a simple hack to evaluate power supply
« Reply #1 on: May 12, 2021, 09:55:50 pm »
Are those in an aluminium box and with a shine perforated metal lid?

First thing I do with those (before first power up) is open them and check for bad solder joints and other obvious things.
Some solder balls are common and these may get loose and cause a short.

Also make the power transistors and diodes loose from the heatsink, make sure there are no burrs behind them and then mount them again with a bit of decent paste.
I once had such a power supply where some white stuff was smeared over the power transistors, but there was nothing between the transistor and the heatsink.

The next best thing would be a DC load and a thermal camera.
Then load it with some decent current and check which parts are getting hot, and how hot they get.

Also, do not assume those bricks are all the same. China is a big country, with over a billion people and there are many different factories that make very similarly looking stuff. The sizes of those power supplies is also standardized, which makes them look more alike. The build quality of these things also varies, and there are also significant differences in the way those things are built internally.
« Last Edit: May 12, 2021, 09:58:09 pm by Doctorandus_P »
 

Offline Zipdox

  • Contributor
  • Posts: 35
  • Country: nl
Re: Is there a simple hack to evaluate power supply
« Reply #2 on: May 12, 2021, 10:12:54 pm »
Are those in an aluminium box and with a shine perforated metal lid?

First thing I do with those (before first power up) is open them and check for bad solder joints and other obvious things.
Also check the thermal pads for any folds and ensure good contact. If there isn't any thermal paste or thermal pads, add some.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 14966
  • Country: us
Re: Is there a simple hack to evaluate power supply
« Reply #3 on: May 12, 2021, 10:31:51 pm »
The easiest and cheapest test load I've found for that sort of thing are automotive light bulbs. Small marker lamps of a few watts, common backup and brake light bulbs are 21W, and headlamp or fog lamp bulbs are typically 55/60W. Various combinations of these readily available lamps in series or parallel can give you a wide range of loads easily up to hundreds of watts.
 
The following users thanked this post: sairfan1, mindcrime

Offline Jwillis

  • Super Contributor
  • ***
  • Posts: 1046
  • Country: ca
Re: Is there a simple hack to evaluate power supply
« Reply #4 on: May 12, 2021, 11:20:42 pm »
Even with the cheap Power supplies its likely they won't explode into flames  under the rated load . I've noticed that the MOSFET's on some I have are just barely capable of the the rated power . And certainly not for extended periods of time . Others look just fine and perform fine .
Also power supply  may drive the correct voltage and current but be extremely noisy. Some thing you would have to check with a scope.
 

Offline JustMeHere

  • Frequent Contributor
  • **
  • Posts: 412
  • Country: us
Re: Is there a simple hack to evaluate power supply
« Reply #5 on: May 12, 2021, 11:35:19 pm »
Make sure the traces aren't too close to the mounting screws.  I had one that did give up the magic smoke.  Opened it up and found the hot trace ran right next to a mounting screw.  A little bit of vibration rubbed through the solder mask, and poof.  Thanks for my circuit breaker.
 

Offline Scottie

  • Newbie
  • Posts: 1
  • Country: ca
Re: Is there a simple hack to evaluate power supply
« Reply #6 on: May 13, 2021, 01:03:04 am »
Operating a linear power supply in constant current regulation into a short load typically puts the series pass transistor(s) into their worst case power dissipation. Hopefully your power supply has automatic transformer tap switching.
 

Offline Stray Electron

  • Super Contributor
  • ***
  • Posts: 1163
Re: Is there a simple hack to evaluate power supply
« Reply #7 on: May 13, 2021, 01:34:43 am »
  I deal with a lot used power supplies and unless it's something special, I don't usually bother with a PS that doesn't have a constant current mode or an adjustable current limit. That said, my initial quick and dirty test for used power supplies to connect a cheap but reliable DVM to the output and first check and see that the PS delivers the more or less correct voltage and that the output voltage can be adjusted. Then I turn the current limit all of the way down and while I still have the DVM attached, I turn the output voltage down to about 1 VDC. Then I connect the DVM up so that it reads as an amp meter using it's highest amp range, and I connect that to the PS output. Then I watch the amp reading as I turn the current limit up and down (keeping it within range of the DVM!).  That allows me to check that the PS's current regulator is also working.

  Accuracy is limited by the accuracy of the DVM but it still allows me to quickly check to see if the PS is basically functioning correctly.  I always do this Q&D test before connecting an expensive meter to a new-to-me PS.  Setting the output voltage to about 1 VDC before checking the current output also limits the power and heat dissipation and the possibility of damage in the meter.

  Once I have some confidence that the PS actually is controlling the output voltage and then current then I can test it at higher voltages and/or higher currents. But when measuring current always keep in mind the amount of power (volts times amps) that your meter might might be having to dissipate.  Some of the larger HP PSs that I have can easily supply 80+ volts at about 40 Amps so that's potentially over 3200 Watts and no small meter is going to be able to absorb that much power.
 

Offline Doctorandus_P

  • Super Contributor
  • ***
  • Posts: 1545
  • Country: nl
Re: Is there a simple hack to evaluate power supply
« Reply #8 on: May 13, 2021, 01:43:12 am »
A decent lab power supply should be able to handle a 1A current into an (almost short circuit) cable easily. Especially if it's for a few measurements to measure cable resistance. I do not recommend to short your power supply and then go have dinner or something.


Lightbulbs can be used as a simple load, but their cold resistance is much lower then their hot resistance and the can draw high current peaks when first connected. Especially with Halogen lamps.

This is a bit of an extreme example, but it makes it perfectly clear:
 
 

Online ledtester

  • Super Contributor
  • ***
  • Posts: 1558
  • Country: us
Re: Is there a simple hack to evaluate power supply
« Reply #9 on: May 13, 2021, 02:00:53 am »
Some videos...

- Lamps used in action to test a 12V 50A power supply:

https://youtu.be/tzoVcU9Bg_Y

- Building a load tester using lamps:

https://youtu.be/sFYHVK4ngXE

Btw - the clamp meter used in the first video is the Uni-T 210E which has gotten a lot of favorable reviews on this forum.
« Last Edit: May 13, 2021, 02:05:11 am by ledtester »
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 14966
  • Country: us
Re: Is there a simple hack to evaluate power supply
« Reply #10 on: May 13, 2021, 07:17:39 am »
A decent lab power supply should be able to handle a 1A current into an (almost short circuit) cable easily. Especially if it's for a few measurements to measure cable resistance. I do not recommend to short your power supply and then go have dinner or something.

Who said anything about lab power supplies? He mentioned "module", which leads me to believe these are power supplies intended to be integrated into some kind of equipment, not a bench power supply.
 

Offline bobbydazzler

  • Regular Contributor
  • *
  • Posts: 58
  • Country: au
Re: Is there a simple hack to evaluate power supply
« Reply #11 on: May 13, 2021, 11:42:08 am »
10w power resistors are pretty cheap, you should have some of those anyway.  I'd recommend 5-10R x 10pieces, this can easily be configured to any resistance value you'd need for psu testing purposes(for higher resistances cheap 0.25w resistors are fine).
https://www.ebay.com.au/itm/164194153463?hash=item263abbf3f7:g:QpUAAOSw161euN7K&var=464106066759
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf