HP xw6200 WorkStation "randomly" shuts down

dar124

[H]ard|Gawd
Joined
Jan 21, 2012
Messages
2,004
Hopefully someone here can help me out or get me headed in the right direction with my PC. Let me know if this should be posted in another section. I just figured that it's a Multi Processor system, so here seemed to make sense??



I have a HP xw6200 WorkStation with dual Intel Xeon 3.60 GHz processors, a Nvidia Quatro FX 1800, a Samsung 128GB SSD for my OS drive and a 500GB WD spinner for the programs drive. It has 8GB of DDR2 PC2-3200 memory running Windows 7 Ultimate 64-bit in a Cooler Master case. My desktop PC in my signature.

I put the PC together about a year ago and it's run like a charm, up until about a month ago. The PC normally runs 24/7 but about a month ago I started randomly finding that the PC was off. At first I didn't think too much of it, just figured that it did some updates and didn't restart correctly or we briefly lost power at home, etc. But it happened more and more and eventually I had it power off while I was using it.

It seems to run at idle without issues. But when transferring data to my server, scanning the PC, running multiple programs, etc it will shut down. My first thought was that it was dusty inside and during these tasks it was overheating. So I opened it up and cleaned it out real good. It wasn't really dusty, but figured I'd still clean anything that I found. I used a shop vac with the hose switched around so it can be used as a blower. I also checked in BIOS and nothing seems strange there (I can post screen shots if necessary).

I also just installed Speed Fan and HWMonitor and attached screen shots below. The screen shots were taken when using the PC to scan for new media to transfer to my server, an IE session with multiple tabs open, an open PowerPoint slide and watching a video in VLC.


I also listed a handful of things that I've done below.

Removed additional SATA port card
Disconnected DVD drive
Checked for blown capacitors
Ran malware bytes scan (PC shut down during scan)
The WD HD passed HD DPS test & error checking
Checked the OS SSD with SSD Life (results are attached below)
I did a system restore back to mid June
There aren't any errors in device manager
All 4 fans (2 on the heat sinks and 2 case fans) are running


The PC will boot up, run for a while and then shut down seemingly when under a decent load. I did recently adjust the virtual memory to a custom size 8191MB as the initial and 14336MB as the max. Not sure if that would effect anything??

I'm still going to try to remove each of the 4 memory sticks one by one and see how it runs. And I can switch out the video card (the motherboard doesn't have any onboard video, so I'll have to temporarily put another card in there).

Hopefully that's a decent explanation of what's going on and what I've done so far. Thanks in advance.




28062470910_c3dce7b965_z.jpg




27728615533_4c1c40e7c9_z.jpg




And the SSD Life info.



28351182735_14ae60bf1c_z.jpg




27734970224_546ac7885e_z.jpg
 
Last edited:
Run memtest to test ram, but im thinking it may be power supply related. do you have a spare you can test with?
 
Run memtest to test ram, but im thinking it may be power supply related. do you have a spare you can test with?


Ok, I downloaded memtest and will get it on to a CD today to test the RAM.

I'm pretty sure that I have an extra power supply around somewhere. I did put in a new Corsair CX500 PSU when I build the PC a year or so ago, but I'll look for another one to swap out to test.
 
I booted to the memtest CD twice and each time the test started and probably got 80% done, but PC shut down before it completed.

So I figured I'd swap the PSU out and if I could get the PC to stay running I'd do the memtest again. I found the old Cooler Master PSU that came with the case that I'm using. And man it was a chore to get the Corsair unit out!!! I had tucked cables out of the way everywhere!!! But the PC wouldn't boot with the Cooler Master PSU installed. It would power on and then immediately turn off. It was a 400W unit (instead of the Corsair CX500), so I'm not sure if it wasn't big enough or if it was a bad PSU to begin with?? The only other one I had around was a small 250W unit. So that definitely won't help out.


Would it be worth it to try to RMA the Corsair unit?? Just figured that I've got it out and it does seem like a power issue with the PC.
 
I'd try a larger power supply if I were you...something 600W or bigger. Those Nocona Xeons weren't the most energy efficient processors. The fact that your system runs for a while and then shuts off makes me think that you issues are the result of an overworked power supply.
 
I'd try a larger power supply if I were you...something 600W or bigger. Those Nocona Xeons weren't the most energy efficient processors. The fact that your system runs for a while and then shuts off makes me think that you issues are the result of an overworked power supply.


Ok, the PC has worked for a year and a half with out any issues. And before I originally got it I worked up some kind of online calculator to tell me I needed a 500W PSU. Maybe the PSU was just overworked and as a result has failed early?? But I assume bigger is better (at least with a PSU)!!!

I wonder if Corsair would give me a refund instead of sending me a new PSU?? That way I could get a 600W + unit.
 
Ok, the PC has worked for a year and a half with out any issues. And before I originally got it I worked up some kind of online calculator to tell me I needed a 500W PSU. Maybe the PSU was just overworked and as a result has failed early?? But I assume bigger is better (at least with a PSU)!!!

I wonder if Corsair would give me a refund instead of sending me a new PSU?? That way I could get a 600W + unit.

A PSU will run much warmer the closer it gets to its peak capacity. They also "age" over time (meaning that they'll put out less power as time goes on). A year seems quite quick for a power supply to deteriorate (usually this happens over a longer period of time) but temps play a big role. The lower end power supplies aren't known for being able to handle high temps for extended periods of time and typically use lesser quality components.

I just looked up the requirements for the 8800GTS...NVidia recommends a minimum of a 400W power supply. That's with a regular single CPU system though...you're running two power-sucking Noconas (or Irwindales, which are worse), easily exceeding the 500W limit. Consider the memory, motherboard consumption and drives and it's likely you are right at the limit and tripping the PSU on overload (a number which drops slightly when the PSU is warm).
 
Last edited:
A PSU will run much warmer the closer it gets to its peak capacity. They also "age" over time (meaning that they'll put out less power as time goes on). A year seems quite quick for a power supply to deteriorate (usually this happens over a longer period of time) but temps play a big role. The lower end power supplies aren't known for being able to handle high temps for extended periods of time and typically use lesser quality components.

I just looked up the requirements for the 8800GTS...NVidia recommends a minimum of a 400W power supply. That's with a regular single CPU system though...you're running two power-sucking Noconas (or Irwindales, which are worse), easily exceeding the 500W limit. Consider the memory, motherboard consumption and drives and it's likely you are right at the limit and tripping the PSU on overload (a number which drops slightly when the PSU is warm).



So I actually have a "Nvidia Quadro FX 1800" graphics card. Was having some issues with the 8800GTS and hadn't updated my signature since I swapped it out. Will take care of that right after this post (y)

And I called Corsair to follow up on my RMA from yesterday. They processed / approved the RMA and I asked about upgrading the PSU. They said they can only replace what I have. I could, however, talk to Customer Service and they might be able to authorize a cash refund, but they wouldn't give me the full amount that I paid because the PSU was over a year old. So my plan is to get the new RMA'd CX500 from Corsair, sell it, and then use that money towards upgrading to either the CX600 or CX750 PSU.
 
I had a xw6200 with a pair of 3.8ghz Xeon's. That thing would heat the room that it was running in.
 
I had a xw6200 with a pair of 3.8ghz Xeon's. That thing would heat the room that it was running in.


Yea, this thing does get nice and warm. When I built the PC I replaced the case fans in the Cooler Master case with Noctua fans, hoping they'd be better than the stock case fans. So it's got 4 fans (2 case fans and 2 CPU fans) and it still gets pretty warm.

Not sure if there is much else that I can do to keep it cool?? I try to periodically clean the dust out of it and if I know it wont be used for a while (if we're going out of town, etc) I'll power it down. I also replaced the CPU thermal paste when initially building the PC.
 
I ended up ordering a new Corsair CX750 PSU while waiting for the RMA on the CX500. I received the new CX750 PSU a couple of days ago and the PC has been running good ever since.

I updated some programs and did some downloading yesterday and stayed powered on. Glad it was just the PSU and hopefully with this one being a bit larger it'll have enough power to keep it running!!!
 
Back
Top