AOA Forums AOA Forums AOA Forums Folding For Team 45 AOA Files Home Front Page Become an AOA Subscriber! UserCP Calendar Memberlist FAQ Search Forum Home


Go Back   AOA Forums > Hardware > AMD Motherboards & CPUs

AMD Motherboards & CPUs Questions or comments on AMD products?


Reply
 
LinkBack Thread Tools Rate Thread
  #1 (permalink)  
Old 21st May, 2006, 09:24 AM
toaster's Avatar
Member
 
Join Date: February 2003
Location: sweden
Posts: 78

NF7-S voltage issue. Tech head question.

I am running an NF7-S v2.0 with a Hyper Type R 480w ATX12v v2.0 PSU.
This PSU does not provide the -5v however I have records of -5v being reported in MBM5, Speedfan and also BIOS as exactly that -5v.

I experienced a wierd crash about 2 weeks ago. Board level power down, leaving standby power on so that a 2 tone alarm could sound.
Rebooted o.k. but then a Prime95 torture test overheated it really quickly until alarm sounded at 60C. (BIOS shutdown is set for 65C anyway)
Removed Heatsink, cleaned everything, carefully re-applied heat transfer compound. Re-Tested. Prime95 not stable unless I increased v-core from 1.65 (stock) to 1.67 Prime95 then stable.
However, while I was in the BIOS looking around and setting the v-core I noticed that the -5v is now reported as being -61.99v
Also, MBM5 and Speedfan both see the -5v as +3.3v
Taking a multimeter to the pin out for -5v at motherboard power connector there does indeed seem to be +3v there.

All the same the board seemed to run stable for a week of pretty solid Prime95 testing.

My question essentially is this. Is it possible that the initial failure was caused by a regulator fault which is getting worse. If so, does anybody know of regulator replacement mods.
Obviously its not going to be possible to fix if its a voltage control chip as the surface mount soldering is too complex for me and the parts probably hard to find, but a few regulators may be possible.

Any other input or similar experience is welcome.

cheers chaps.
toaster. (doh, looks like I've done it again.)
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #2 (permalink)  
Old 21st May, 2006, 02:59 PM
Gizmo's Avatar
Chief BBS Administrator
BassTeroids Champion, Global Player Champion, Aim & Fire Champion, Puzzle Maniax Champion, Othello Champion, Canyon Glider Champion, Unicycle Challenge Champion, YetiSports 9: Final Spit Champion, Zed Champion
 
Join Date: May 2003
Location: Webb City, Mo
Posts: 16,178
Send a message via ICQ to Gizmo Send a message via AIM to Gizmo Send a message via MSN to Gizmo Send a message via Yahoo to Gizmo Send a message via Skype™ to Gizmo

Quote:
Originally Posted by toaster
I am running an NF7-S v2.0 with a Hyper Type R 480w ATX12v v2.0 PSU.
This PSU does not provide the -5v however I have records of -5v being reported in MBM5, Speedfan and also BIOS as exactly that -5v.
The ATX specification requires that a PSU provide, among other things, a -5v rail. It isn't used for much in modern PCs, but trust me, your PSU is supposed to be providing it.

Quote:
Originally Posted by toaster
Taking a multimeter to the pin out for -5v at motherboard power connector there does indeed seem to be +3v there.
Oh my. This is VERY not good. It suggests to me that you have had a PSU failure. It is also quite likely that you have suffered some board damage as a result, although the board damage is likely confined to a very few specific circuits (such as NIC or sound). However, that is not a certainty; as the -5v is historically used mainly for chip substrate biasing there may not have been anything damaged.
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #3 (permalink)  
Old 22nd May, 2006, 04:07 PM
toaster's Avatar
Member
 
Join Date: February 2003
Location: sweden
Posts: 78

I was looking at the old records I made of initial voltages, temps etc when I first plugged up this PSU. Luckily I had mbm5 installed and I can see from the "High / Low" log that the -5v has always been recorded as +3v with this power supply/motherboard combo. It looks likely that the -5v I noted in a text document was actually the +5.46v "Standby" voltage and I simply mistook it for the -5 reading which is directly above. (In Speedfan) and wrote it down as a negative figure.

The Hyper PSU's (I have two.) have no -5v at all. There is physically no wire connecting the pin at the connector to the inside of the PSU and the manual indicates a "Not Connected" for this value.

As the Hyper PSU itself is not providing the -5v or any other voltage to that particular pin I can only assume that the +3v is coming from somewhere else on the motherboard itself.
It would be handy to know where else on the motherboard I can measure the different voltages.

I have tried the motherboard with a Q-Tec 500w PSU which does provide a -5v line and the BIOS indicated that this was present and correct. However I didnt boot to an OS with this PSU as I'd rather not use a Q-Tec for any real PC build and in any case it doesnt have a SATA power connector.
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #4 (permalink)  
Old 22nd May, 2006, 05:59 PM
Gizmo's Avatar
Chief BBS Administrator
BassTeroids Champion, Global Player Champion, Aim & Fire Champion, Puzzle Maniax Champion, Othello Champion, Canyon Glider Champion, Unicycle Challenge Champion, YetiSports 9: Final Spit Champion, Zed Champion
 
Join Date: May 2003
Location: Webb City, Mo
Posts: 16,178
Send a message via ICQ to Gizmo Send a message via AIM to Gizmo Send a message via MSN to Gizmo Send a message via Yahoo to Gizmo Send a message via Skype™ to Gizmo

Well, I am now surprised. You are correct; it doesn't have a -5v output. Which means that either my understanding of the ATX specification is incorrect (which is possible) or that PSU is not truly ATX compliant.

In any case, I would guess that what you have going on (since there is no -5v output from the PSU) is that the -5v supply rail is being pulled up to the +3.3v rail by a resistor (pull-ups like this are common), effectively creating a voltage divider somewhere on the board (although it would be more logical to use a pullup to ground in this case, I would think). It may also be that there is a circuit somewhere in the audio or NIC stuff that uses the -5v for some kind of bias, as I mentioned previously, and since there is no -5v from the PSU, it is simply floating to +3.3v. For that matter, it might even be some strange behavior of the WinBond chip that does the voltage measurement.

Given what you've told me thus far, I suspect that we've been chasing the wrong end of the problem. The PSU does not provide -5v, and never has. MBM has always reported +3.3v on this supply rail. That means that this is most likely completely normal for this system, which brings us back to why your system seems to be getting steadilly more unstable.

My first guess would be that the capacitors on the mobo are probably drying out and getting to be in need of replacement. How old is your board?
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #5 (permalink)  
Old 22nd May, 2006, 08:51 PM
Chief Systems Administrator
 
Join Date: September 2001
Location: Europe
Posts: 13,075

From memory, ATX2 power supplies do not provide -5V. ATX power supplies do provide -5V. Thus a powersupply that does not provide -5V is not ATX compliant, but could be ATX2 compliant!
__________________
Any views, thoughts and opinions are entirely my own. They don't necessarily represent those of my employer (BlackBerry).
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #6 (permalink)  
Old 22nd May, 2006, 09:36 PM
toaster's Avatar
Member
 
Join Date: February 2003
Location: sweden
Posts: 78

A capacitor problem is a possibility. I've seen old (or simply bad) caps causing flakiness on a couple of my much older boards. - I've got an old Slot A MSI-6195 that is going to be my first full capacitor change project as it really DOES have flaky caps.
The motherboard with this +3v on the -5v rail issue is an Abit NF7-S revision 2 board and its a couple of years old. Previously I was running PC2700 DDR and didnt overclock it at all, however I've just put some Patriot XBLK PC3500 into it which allows me to bring the FSB up to 200Mhz and mildly overclock the XP2800+
It seemed to be happy running all three of the Prime95 torture tests for 12+ hours each with a very small v-core voltage tweak. (1.675v rather than the 1.65v stock)
I thought I would try dropping the voltage back down to 1.65v to see if it would be stable with stock voltages and better cooling but I kept getting error reports in Prime95. Since bringing the v-core back up to 1.675v it actually does a full power down after about 15 minutes of torture testing.

The crashes that occur generally have a pattern of a sudden power down, the fans spin down, (CPU and everything obviously stop working suddenly.) the PSU powers down also, except for the standby power which I can see is on because there is a Red LED on the Motherboard that indicates the presence of this power and also because a 2 tone alarm sounds. This 2 tone alarm continues to sound until the power supply is physically turned off at the switch.
Obviously because this is a sudden and complete power down the MBM5 log is never kept so I cant see what the very last values were.
I'm thinking of networking it with another PC and getting MBM5 to send its "Syslog" reports there, that way I can track its performance from another location and see what state it was in when the crash occured.
According to the only documentation I could find about the motherboards audio BIOS signals and alarm sounds (at the Abit forum) this 2 tone alarm is generally heat related but could also be an "Voltage out of spec" issue.
I'm thinking of looking at its perfomance if I direct a fan at the MOSFET array

I have set the overtemp protection in BIOS to pretty conservative levels. (60C case temp and 65C CPU temp which I think are the lowest settings available.) However I dont think its a heat issue because even under a full load I've not seen MBM5 or speedfan reporting a CPU temp above 54C
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #7 (permalink)  
Old 22nd May, 2006, 09:46 PM
Gizmo's Avatar
Chief BBS Administrator
BassTeroids Champion, Global Player Champion, Aim & Fire Champion, Puzzle Maniax Champion, Othello Champion, Canyon Glider Champion, Unicycle Challenge Champion, YetiSports 9: Final Spit Champion, Zed Champion
 
Join Date: May 2003
Location: Webb City, Mo
Posts: 16,178
Send a message via ICQ to Gizmo Send a message via AIM to Gizmo Send a message via MSN to Gizmo Send a message via Yahoo to Gizmo Send a message via Skype™ to Gizmo

The two-tone alarm can be generated by a number of different symptoms, but they are all related to something with the CPU circuitry, in my experience.

I have a NF7-S rev 2.0 as well, and I can tell you for a fact that if you overclock it, you need cooling on the board area under the FETs; those things get REALLY hot. Typically, this cooling can be provided by the 'blow-by' from the HSF. With water cooling like what I run though, a fan for that area is required.

You might try raising the BIOS shutdown temps. The thermal sensor under the CPU can be skewed by several degrees by heat conducted from the FETs.
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #8 (permalink)  
Old 23rd May, 2006, 04:31 AM
toaster's Avatar
Member
 
Join Date: February 2003
Location: sweden
Posts: 78

Would you believe me if I said that clearing the CMOS (using the jumper) seems to have made a difference?

Actually its not the only thing I did. I also took off the Heatsink, blasted the CPU core and heatsink base with a lubricant free contact cleaner spray. (Its good stuff, safe to use on PCB's.) then a blast of canned air. Re-applied the heatsink etc.

Anyway, cleaned up and re-seated heatsink, re-connected power to motherboard, cleared and went through the BIOS settings.

Now back to the mild overclock (200Mhz FSB, 11.0x Multi) that I initially had when first installed the PC3200 DDR, and I have been able to run Prime95 o.k. so far with vcore of just 1.65v (stock) which it has not been capable of doing without error's in Prime95 or these strange crashes for a week or more.
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
  #9 (permalink)  
Old 23rd May, 2006, 11:25 AM
toaster's Avatar
Member
 
Join Date: February 2003
Location: sweden
Posts: 78

About this ATX12v standard and -5v wierdness.
I actually stumbled across this information while looking for something about the Abit EQ software this afternoon.

"I notice the PSU doesn't have a -5v supply. Is it the case that -5v is no longer used by modern boards? My motherboard shows 0.00V for the -5V line in the BIOS and +3.55V with the ABIT EQ software. My system was showing correct values for this line with the previous PSU. Is it safe?

We have confirmed that -5V is only used for ISA slot and usually ISA can not be found on new mainboard. All new PSU was removed -5V from main connector. So it is correct that you can not detect it. Please don’t worry about it since most new PSU has no -5V."

Its from the faq pages of the Enermax coolergiant website...
http://www.coolergiant.co.uk/support/faq.asp

The faq doesnt mention the motherboard model, but obviously it must be an Abit (If using their monitoring software) and that strange +3v is there reported in the BIOS!!
__________________
Digg this Post!Add Post to del.icio.usBookmark Post in TechnoratiFurl this Post!
Reply With Quote
Reply



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Rate This Thread
Rate This Thread:

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 Off
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
8rda+ v2.1, CPU voltage issue MIP EPoX MotherBoards 38 22nd February, 2004 03:21 AM
Epox Tech-4PDA2+ rev2 issue esoteradactyl EPoX MotherBoards 1 16th October, 2003 10:32 PM
Epox 8RDA3+ ver1.1 and Abit NF-7S head to head .. Neoprene AMD Motherboards & CPUs 7 10th September, 2003 03:29 PM
8RDA+ voltage issue Forgiven EPoX MotherBoards 9 4th February, 2003 04:33 PM
Hlp Pls Epox Tech - 8k7a bios update issue compunut EPoX MotherBoards 9 7th February, 2002 10:39 PM


All times are GMT +1. The time now is 10:11 PM.


Copyright ©2001 - 2010, AOA Forums
Don't Click Here Don't Click Here Either

Search Engine Friendly URLs by vBSEO 3.3.0