jonnyGURU Forums
Home Site Search Reviews Articles Contest Links PSU FAQs  


Go Back   jonnyGURU Forums > Computer Hardware > PC Power Supply Discussion

PC Power Supply Discussion Troubleshooting and discussion of computer power supplies

Reply
 
Thread Tools Display Modes
  #41  
Old 08-01-2013
crmaris's Avatar
crmaris crmaris is offline
Cybenetics CTE
 
Join Date: Jan 2010
Location: Greece
Posts: 1,461
Thanks: 140
Thanked 438 Times in 205 Posts
Default

I have this on the test bench and noticed something very strange.

At low loads (up to 100W) the AC Watt readings have a significant deviation although the DC load is stable. I know that this has to do with the LLC resonant converter but in this case the deviations are large (even 10-15W). And I measured them with two of the best power meters (which are calibrated) so it is clearly something going on with the design. I was able to calculate efficiency because my program takes the AVG readings through out all testing period, else I wouldn't be able to give an accurate efficiency percentage at low loads.

I noticed that the same happened to the EVGA 1000 W unit which is based on the same exactly platform.

Anyone has any thoughts on this matter (@Travis I would appreciate your input in this).

Wolf did you noticed it too in your sample?
Reply With Quote
  #42  
Old 08-01-2013
Oklahoma Wolf's Avatar
Oklahoma Wolf Oklahoma Wolf is online now
Me fail English? That's unpossible!
 
Join Date: Oct 2006
Location: Moderator
Posts: 6,421
Thanks: 116
Thanked 1,185 Times in 338 Posts
Default

Yeah, this platform is a big time power meter fooler. I can get stable readings at low loads with the old Brand 4-1850, but it's really too inaccurate at those loads to pay much attention to.
Reply With Quote
  #43  
Old 08-01-2013
crmaris's Avatar
crmaris crmaris is offline
Cybenetics CTE
 
Join Date: Jan 2010
Location: Greece
Posts: 1,461
Thanks: 140
Thanked 438 Times in 205 Posts
Default

I think it doesn't have to do with the power meter. I measured both with the Chroma AC source (which is 100% accurate since it actually provides the power to the unit) and the Yokogawa and both had the same readings. It is something strange about the design and I am curious to find out what. However I think only an SF tech could give us some reliable answers.

Ah also my sample has a white fan and all caps are Japanese! Especially the latter was a nice surprise.
Reply With Quote
  #44  
Old 08-01-2013
Oklahoma Wolf's Avatar
Oklahoma Wolf Oklahoma Wolf is online now
Me fail English? That's unpossible!
 
Join Date: Oct 2006
Location: Moderator
Posts: 6,421
Thanks: 116
Thanked 1,185 Times in 338 Posts
Default

I think you're going to need one of those bank breaking 80 Plus grade power analyzers to really find out for sure.
Reply With Quote
  #45  
Old 08-01-2013
-The_Mask-'s Avatar
-The_Mask- -The_Mask- is offline
ssssssmokin'
 
Join Date: Jun 2010
Location: The Netherlands
Posts: 4,568
Thanks: 328
Thanked 1,278 Times in 839 Posts
Default

Its already 3 weeks ago that Travis has logged in on Jonnyguru, maybe its better to send him a mail.
Reply With Quote
  #46  
Old 08-01-2013
McSteel's Avatar
McSteel McSteel is offline
Naughty
 
Join Date: Nov 2010
Location: Serbia
Posts: 2,165
Thanks: 107
Thanked 398 Times in 275 Posts
Default

Do you have any means of plotting a graph of voltage/amperage? Don't know if Chroma can do this, with or without your software prowess...

I have a strange feeling that the PSU is phase-shifting the input, possibly storing power near the current amplitude, then releasing it near the voltage amplitude, using the LLC reservoir and the main cap(s) as storage. That would be a good way to boost both PFC and efficiency. You'd need some very precise timing on the switchers, and possibly a very specific switching frequency to make the most of it, but SF probably had plenty of time for tinkering.
__________________
Careful what you wish for... You just might get it.
Reply With Quote
  #47  
Old 08-02-2013
crmaris's Avatar
crmaris crmaris is offline
Cybenetics CTE
 
Join Date: Jan 2010
Location: Greece
Posts: 1,461
Thanks: 140
Thanked 438 Times in 205 Posts
Default

I can do this but I have to modify my software accordingly and since this isn't a feature that I will use daily I don't know if it worth the trouble.
But I can log everything and transfer all data to excel for graphs. This log keeps every parameter with specified intervals (I have set 1 sec). The thing is that I don't check the log anymore and forgot to enable this option while testing (had it auto enabled but disabled this option because it consumed some system resources and thought it isn't worth it since I have all the rest data available. And who wants to keep an eye on thousand of lines logs)!! But I have the min,max and avg values registered at low load testing and the difference is large (I am referring to AC watts)!

I suspect that it may have an interleaved APFC since I spotted two boost diodes, but on the other hand there is only one PFC choke and three fets so this isn't possible.

What ever they did efficiency at low loads on the Platinum doesn't look good at least with my equipment since the older platform performed better here. Even the EVGA SuperNOVA that is based on the Gold leadex platform had better efficiency at lower loads since most likely the Platinum Leadex is tuned to offer better efficiency at normal load range and meet the 80 Plus requirements.
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 04:52 PM.


Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2018, vBulletin Solutions, Inc.
Copyright © JonnyGURU.com 2000 -