P55-UD6 socket burn

**saaya** · 10-07-2009, 12:21 AM

since the contact spot between the pins and the pads definately shifts depending on the pressure, i thought this might affect the contact... but the top of the pins looks round enough to compensate for the drift, and yes your right, the seating plane prevents too much pressure...

**G H Z** · 10-08-2009, 02:23 PM

Originally Posted by raju

fundamental issue though is the lack of VCC pads - 33% fewer VCC pins than 1366 with almost the same current draw.

Where are you getting your numbers for power characteristics/comparison?

**saaya** · 10-08-2009, 05:34 PM

intel.com

its all there in the white papers...
its just quite messy cause its not clear what info is in what whitepaper

iirc its in the cpu whitepapers, not chipset, platform or electric/mechanic spec whitepaper.
theres a pinout map and details about pins

chew* · 10-08-2009, 08:14 PM

Raja,

I'm on the UD6 now with an 860, thought i would just document some stuff for ya.

Vcore about 1.40 real....did not use LLC.

VTT about 1.5 ish uncore about 4000 Cpu around 4460 for 1m and 32m.

4gig for wprime 32m and cinebench.

Will update after i go cold but here is the CPU as it sits now.

**Raja@ASUS** · 10-08-2009, 08:44 PM

Originally Posted by G H Z

Where are you getting your numbers for power characteristics/comparison?

As Sascha said and secondly with my own measurements using power clamp meters.

**Raja@ASUS** · 10-08-2009, 08:46 PM

Originally Posted by chew*

Raja,

I'm on the UD6 now with an 860, thought i would just document some stuff for ya.

Vcore about 1.40 real....did not use LLC.

VTT about 1.5 ish uncore about 4000 Cpu around 4460 for 1m and 32m.

4gig for wprime 32m and cinebench.

Will update after i go cold but here is the CPU as it sits now.

You might get lucky Chew. There are others that have pushed as hard as me and had no issues, while some have only ran 32m around 5G's and had a nasty burnout. End result I bet is that it will be erratic from user to user. I do know Intel is looking into it though.

**saaya** · 10-08-2009, 10:42 PM

got clamp?

check current draw, you might have a low leakage chip

**Raja@ASUS** · 10-08-2009, 10:51 PM

Don't think either of the benches he ran pulls too much at 4G's in terms of current (around 120-130w tops I suspect). Don't wanna tell him to torture it needlessly either. It's a case of, if you've got one that works, treat it nice.

**Raja@ASUS** · 10-09-2009, 12:38 AM

BTW forgot to add here (this is kinda important). I've had no burn outs on the LOTES sockets. The board I thought was using a LOTES (the EVGA E657) had a Foxconn socket, the second one they sent me has a LOTES socket. The single pin burn pic I posted earlier (page 1) was from the Foxconn socket board. So, just to summise, no burnouts with LOTES at all so far.

**miahallen** · 10-09-2009, 04:17 AM

From my P55-UD3R review in the "Xtreme Reviews" section...

Originally Posted by miahallen

OK...wanted to provide a brief update with more stability testing on the i5 750, as well as my first thoughts on the i7 870 that arrived today.

First I'd like to point out something interesting I found when I swapped CPUs today. Apparently, "socket burn" may have effected me as well. Time will tell, but I believe I do show some indication of an issue here...hopefully it does not get any worse

I'd really like to know if this is a mobo/CPU/design problem:

The board is still fully operational at this time, and does not seem to be suffering any negative performance effects.

**[XC] Oj101** · 10-09-2009, 05:01 AM

I just love it when a heavy hitter like George comments

**Raja@ASUS** · 10-09-2009, 05:33 AM

Originally Posted by miahallen

From my P55-UD3R review in the "Xtreme Reviews" section...

Only thing i can say for sure is no issues on LOTES thus far for me using the same CPU's at higher clocks in some cases.

This is the CPU that casued the ROG board to burn out. Stuck it in the E659 and 5.22GHz = no burnout on the LOTES socket..

**dinos22** · 10-09-2009, 05:59 AM

Originally Posted by raju

You might get lucky Chew. There are others that have pushed as hard as me and had no issues, while some have only ran 32m around 5G's and had a nasty burnout. End result I bet is that it will be erratic from user to user. I do know Intel is looking into it though.

before you said this it actually sounded you might be onto something but now it sounds like your guy's VTT claims back in P45 days

heheheh

juuuust messing with ya

**Raja@ASUS** · 10-09-2009, 06:18 AM

Originally Posted by dinos22

before you said this it actually sounded you might be onto something but now it sounds like your guy's VTT claims back in P45 days

heheheh

juuuust messing with ya

Expect nothing less from you (I always know when I'm onto something if it tweaks your input)

. Hard to nail it for sure when people have it at different levels of current draw. Seems the Foxconn socket is the common denominator (stay away from Foxconn socket boards..lol - I'm sure you'll like that)

.

chew* · 10-09-2009, 08:17 AM

Originally Posted by raju

Don't think either of the benches he ran pulls too much at 4G's in terms of current (around 120-130w tops I suspect). Don't wanna tell him to torture it needlessly either. It's a case of, if you've got one that works, treat it nice.

Hah I plan to torture it regardless.

I fully expect to have issues.

I have a killawatt hooked up and minimal graphics draw.

I will document at what point I have issues.

**Raja@ASUS** · 10-09-2009, 08:30 AM

Originally Posted by chew*

Hah I plan to torture it regardless.

I fully expect to have issues.

I have a killawatt hooked up and minimal graphics draw.

I will document at what point I have issues.

Be my guest

It'll be interetsing to see what happens if anything.

**Monstru** · 10-09-2009, 08:58 AM

Hmm...I have seen this thread only now. Of course, I do not have the high electronics knowledge that some of the speakers in this thread have, but I can tell you guys what I got from a mechanical point of view, related to the pin fragility issue.

So when I first tested the UD6 under cold, I just mounted my pot in the "regular" manner, with quite some tension on the screws. It didn't boot up, and I had to seriously loosen the screws in order for the board to boot. I have played with the same thing on Asus and MSI boards as well, and none of them will boot if the pot is to tighly fit on the CPU. So I do not know about the shape of the pins and the current (as hipro explained it in this thread), but I am really sure that this new socket is also very flimsy from a mechanical point of view.

**miahallen** · 10-09-2009, 09:46 AM

I had the same issue the first time I hooked up my water block for some 32M eff testing. I had to modify my Swiftech 1366 hold-down to work for 1156...at my first attempt to boot, I got a fast boot loop sequesnce. Nothing fixed it until I loosened up the block....everything has been fine since then

**Buckeye** · 10-09-2009, 09:51 AM

Originally Posted by Monstru

Hmm...I have seen this thread only now. Of course, I do not have the high electronics knowledge that some of the speakers in this thread have, but I can tell you guys what I got from a mechanical point of view, related to the pin fragility issue.

So when I first tested the UD6 under cold, I just mounted my pot in the "regular" manner, with quite some tension on the screws. It didn't boot up, and I had to seriously loosen the screws in order for the board to boot. I have played with the same thing on Asus and MSI boards as well, and none of them will boot if the pot is to tighly fit on the CPU. So I do not know about the shape of the pins and the current (as hipro explained it in this thread), but I am really sure that this new socket is also very flimsy from a mechanical point of view.

You might be on to something there with how tight the pot is screwed on. I was under Cascade and the thing was acting like a real dog, its a ES chip so I didnt expect much. Another problem is the eraser reallys gets into this socket badly and could be pushing the chip in strange ways. I dont mean inside the socket, just the outside areas, down in the IHS etc.

No socket burn with 1.75 vcore and 1.5vtt combos and my chip shows no signs of damage on the contact side.

**Raja@ASUS** · 10-09-2009, 10:19 AM

The whole Foxconn socket mech seems sensitive to mounting pressure (and flimsy pins), that would explain the non boots if pins are not connecting.

**INFRNL** · 10-09-2009, 07:21 PM

Sorry guys, a little off topic, but similar.
Does anyone know what would cause this? http://www.xtremesystems.org/forums/...&postcount=174
I do not feel like going through the process to repost pics on here.

I was running my X3440@1.36v@4ghz(200x19+turbo for x20) left running for only 2.5 hrs during sleep. Air cooled with Megahalem & ICD7. Temps full load mid to high 60's mobo is a UD2, need more info, let me know. Thanks

Edit: for the record, mobo is toast. End Edit

chew* · 10-09-2009, 07:38 PM

Ehh looks like you just had a bad fet. It happens. Might want to add active cooling to the fets or at least some heatsinks.

chew* · 10-09-2009, 07:40 PM

Originally Posted by saaya

got clamp?

check current draw, you might have a low leakage chip

I would say if anything it would be a high leaker. Its an ES and it runs rather warm.

I wonder if any of these cases are LLC enabled related?

**Raja@ASUS** · 10-09-2009, 09:02 PM

Originally Posted by INFRNL

Sorry guys, a little off topic, but similar.
Does anyone know what would cause this? http://www.xtremesystems.org/forums/...&postcount=174
I do not feel like going through the process to repost pics on here.

I was running my X3440@1.36v@4ghz(200x19+turbo for x20) left running for only 2.5 hrs during sleep. Air cooled with Megahalem & ICD7. Temps full load mid to high 60's mobo is a UD2, need more info, let me know. Thanks

Edit: for the record, mobo is toast. End Edit

CPU PWM FET burnout.

Insufficient cooling on that board for OC, especially if you enable Loadline calibration too. You really need a good fan pointing at those FET's to keep them cool if overclocking on this board.

**INFRNL** · 10-10-2009, 07:40 AM

Originally Posted by chew*

Ehh looks like you just had a bad fet. It happens. Might want to add active cooling to the fets or at least some heatsinks.

Originally Posted by raju

CPU PWM FET burnout.

Insufficient cooling on that board for OC, especially if you enable Loadline calibration too. You really need a good fan pointing at those FET's to keep them cool if overclocking on this board.

Thanks,
I will see if I can get it replaced and either try a better board or add some fans to blow on the board itself.

Thread: P55-UD6 socket burn

Thread Tools

Search Thread

Rate This Thread

Display

Bookmarks

Bookmarks

Posting Permissions