jonnyGURU Forums
Home Site Search Reviews Articles Contest Links PSU FAQs  


Go Back   jonnyGURU Forums > Site and Forum Topics > Testing Methodology Discussion

Testing Methodology Discussion Questions and comments regarding the testing methodologies used on jonnyguru.com

Reply
 
Thread Tools Display Modes
  #1  
Old 07-20-2009
KeriJane's Avatar
KeriJane KeriJane is offline
Trabant Comrade
 
Join Date: Dec 2008
Location: Chicago! Run 4 your Life!
Posts: 42
Thanks: 3
Thanked 0 Times in 0 Posts
Default KeriProductions ATX338 PSU Tester

Hello.

For awhile now I've been reading the PSU reviews here at jonnyGURU.com and wondered how I could test Computer Power Supplies accurately myself.

Obtaining a SunMoon isn't really practical.
The cheap $20 testers are virtually worthless as there's no load.
An Oscilloscope is great but there still needs to be a load.
Some people have been known to use light bulbs and loose resistors but I feel that's a bad idea.

After learning how to use the scope to some degree, I noticed the spikes and noise that have to be filtered from the scope input. I learned how to do that here in this Forum.

So armed with the knowledge gained here and elsewhere I set out to make my own PSU tester.
The 300w range seemed adequate to load most PSUs as this is about the most current a normal single-videocard system is likely to draw.
I do not intend to test PSUs to the limit and make them go bang, just to see how they react to a relatively normal load.

So I obtained some wirewound resistors, a couple of fans, some voltmeters and the all important bypass caps, put them all in a box full of switches and Presto! The KeriProductions ATX338. A nice, straightforward mid-range load tester.

It seems to work as intended with 2 scope outputs being selectable to the desired rails. The fans generally keep the resistors around 70℃ or cooler.
The scope outputs are the key. It allows viewing and measuring via a 2-channel scope.

Here's some pics:



Have Fun!
Keri

Last edited by KeriJane; 07-20-2009 at 11:31 PM.
Reply With Quote
  #2  
Old 07-20-2009
Oklahoma Wolf's Avatar
Oklahoma Wolf Oklahoma Wolf is offline
Moderator of Silly Walks
 
Join Date: Oct 2006
Location: Moderator
Posts: 5,619
Thanks: 35
Thanked 108 Times in 71 Posts
Default

Still looks good. I like the carpet too
__________________
My storm chasing images: theweatherwolf.zenfolio.com
Reply With Quote
  #3  
Old 07-20-2009
KeriJane's Avatar
KeriJane KeriJane is offline
Trabant Comrade
 
Join Date: Dec 2008
Location: Chicago! Run 4 your Life!
Posts: 42
Thanks: 3
Thanked 0 Times in 0 Posts
Default

Thank You.

I HATE this Carpet with a Purple Passion!
It's GOTTA go.

You've GOTTA see the bullet points I made for the box....
It's got "Helical Ridge Fasteners for solid construction"!

Have Fun,
Keri

PS. An Analog Meter version is underway. soon. Only need the switches and some time.
Reply With Quote
  #4  
Old 01-13-2013
KeriJane's Avatar
KeriJane KeriJane is offline
Trabant Comrade
 
Join Date: Dec 2008
Location: Chicago! Run 4 your Life!
Posts: 42
Thanks: 3
Thanked 0 Times in 0 Posts
Default

OK, after a small delay caused by Trabants, language studies, getting rid of that awful carpet and other random things, I got out the PSU tester again.

This time I have a more reputable oscilloscope, a Tektronix 2235A. I happen to like analog scopes better than digital ones, especially cheap digital ones.

I also have a "known quantity" PSU to test! None other than the actual FSP Raider 750 tested here awhile ago.

My extremely modest tester is very much under-sized for this PSU, especially on the 12v but I should be able to get some idea of my testing methodology.

The scope is set for 10mv/div and .1ms/div.

I ran tests in both single channel and "add" mode using both channels with Ch B set to ground. The results were pretty close in single channel, "add" mode or add mode with a probe on the ground instead of through the tester.

I lack a differential probe as specified by the ATX specification thing so that's why I tried add mode.
A lack of significant difference could mean nothing or everything.


My PSU tester load is set for 12v at 16A, 5v at 5A and 3.3v at 6A.
This is somewhere between test 1 and 2 done by Oklahoma Wolf and probably a lot low on the 12v and a little high on 3.3 and 5v for a modern system.

So for your amusement, I present my results:

3.3v ripple- about 10mv.
5v ripple - about 5mv
12v ripple - about 20mv

Crossloading this thing by decreasing the 12v load really does make the 5v very unhappy! It looks like it got up to 70mv with 12v unloaded.

That's OK, it's not getting anywhere near a Pentium 3. It's bound for a nice, 12v-hungry modern system.

Questions?
Comments?
Suggestions?

I'll even take Ridicule!
Attached Thumbnails
Click image for larger version

Name:	12v1 load1a.jpg
Views:	85
Size:	39.8 KB
ID:	1805   Click image for larger version

Name:	12v1 load1b.jpg
Views:	87
Size:	32.5 KB
ID:	1806   Click image for larger version

Name:	12v1 load1c.jpg
Views:	87
Size:	29.9 KB
ID:	1807   Click image for larger version

Name:	12v settings.jpg
Views:	81
Size:	35.9 KB
ID:	1808  
__________________
The More You Learn, The Less You Know!
Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -4. The time now is 03:25 AM.


Powered by vBulletin® Version 3.8.1
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.