PC Won't Start

I had a problem with my PC a week ago. It wouldn't stay on. I figured it was the power supply gone bad, it would run for a while, then just randomly shut off for what seemed like no reason at all. I bought a new power supply, and plugged everything in today. When I turned it on, the computer has one solid beep and it does not post. Now it was working fine a week ago, minus of course the power supply problem, I mean I've never had it beep like this before, and it always used to post. I'm assuming it's not the ram, because that is usually 3 beeps if I'm not mistaken. Do I have something plugged in wrong, a DOA power supply, a software problem of some type, or a blown processor? OR is it something else, I don't know, you tell me. Thanks for reading!

 

I reseated my two graphics cards and it works fine now, thanks!

 

The beeping has stopped and the PC starts fine, but I still get random shutdowns. Any reason you think this may be happening?

 

What are your system specs?

I had one of the HS lugs pop out recently, CPU got really hot and PC shutdown. Long story short the CPU was hitting the upper critical threshold of the temp range and shutdown to preserve the hardware.

Use EVEREST Free Edition 2.20 , Computer>>>>Sensors and see what sort of temps you are generating.

 

System Specs, it's an Antec Alienware case with an ASUS Striker Extreme MB. Intel QX6700 processor, 2 Sata II hds, one 400 one 500 GB. 2 EVGA 8800 Superclocked grapics cards, 750 W Corsair Power supply, was running a 600 W Ultra power supply and had same problem, figured the power supply was bad thus the replacement. 2 GB of Corsair DDR2 RAM, Pioneer CD/DVD burner, Lite On DVD Reader. floppy drive, and wireless card that's everything. I know that this is overkill as far as the info provided I mean floppy drive? but hey, I'm the one with a computer problem here I figure I'll tell you everything.

Today I sprayed out all the dust, it wasn't very dusty, but I sprayed out the heatsink really well and fired the thing up. I went into the bios and am monitoring the temps. As of a few min ago when I last looked the CPU was at 61 Degrees Celsius and the MB temp was at 37 Degrees Celsius. This seems really hot to me, but it is a quad core processor. I'll check to see if the temps increase any more before the thing shuts down, if it does this time. This isn't with a load at all on the machine. This is just the bios. I wonder how hot the thing will get if I actually try to do something on the machine. What's the temp this thing needs to reach before it shuts off?

 

Max temperature before the Thermal Monitor kicks in is 64.5C according to the Intel design pdf.

I think that at idle, yours should be running in the low 30's with a good cooling setup.

 

REALLY? 64.5 degrees? It was running at around 65 when I left and a MB temp of 41. I'm using the stock fan, any suggestions? I'm obviously gonna have to get some after market device.

 

Stock fan should be fine in a case with decent through-flow of air.

Check that the heatsink is correctly seated and that there is just a 'film' of Arctic Silver or similar, enough to fill any micro holes in the CPU heat spreader and heatsink base - any more will likely cause temps to rise.

EDIT - by stock fan I mean the Intel-supplied heatsink/fan combo if you have the retail CPU (if you do have a retail CPU, using another cooling solution may lead to invalidation of warranty).

 

Yup I have the stock fan that came with the processor. It's always run really hot. I'll try to reset the heatsink when I get home later and I'll repost temps at that time. Should this still be an issue are there any other fans you would suggest? Maybe this one is just faulty, or on its last leg. Also, what are your thoughts about liquid cooling? I'm not looking to spend a fortune here, but I am looking to solve my problem. I'll do whatever it takes to make this thing run.

 

Just redid the thermal grease and the heatsink on the CPU I'm still running at 60 degrees C. This is far too hot for this computer to be running. I have two intake fans on the front two exhaust on the back a top fan and a side fan, I assume there is good air flow. there doesn't seem to be an excuse for this heat. Any liquid cooling systems you suggest?

 

Liquid cooling is really really risky, and expensive. id go for an aftermarket cooler. Id reccommend anything by zalman.

 

I've heard of the brand before. But to be honest, I'm not certain which fan is going to provide more cooling, I've checked newegg and it's not easy to tell. The LGA775 chipset isn't really all that common anymore and there are few fans for it. I need something that is going to solve my problem and hopefully keep my cpu temp under 50 degrees. Please reccommend a specific model and not just a brand as a whole, I have brand loyalties too for certain things in my computer, but I've never had to replace a heat sink before so I need a model not a brand. And I thank you so very much for taking the time to help me.

 

Bluntly, I'm surprised you haven't killed that CPU. Running dual "superclocked" Video Cards will generate a lot of system heat and then you only have the stock HS. Is the CPU OC'd?

The CPU cut of temperature is normally under the Hardware Monitoring section in your BIOS. While in there I would be changing all fan profiles to "Performance" as you can't expect to keep quite fan profiles while running that hardware.

The best air cooling will outperform the low end and average water cooling systems. I have air and don't know much about water cooling but I do know thr is a lot of "gimmicky" crap out thr so you would have to do your homework for water.

Air Cooling links and references can be found here The Big Air Cooling Sticky! , you will see some of my "old" efforts in this thread Air Cooling Project: 101. Lowered CPU load by 9-10C.

From the 2008 shootout links you will find some "trends" on the better performers, here are 2 that were doing really well:

Budget OCZ Vendetta 92mm Ball CPU Cooler - Retail

Quiet Noctua NH-U12P SE1366 120mm SSO CPU Cooler - Retail

Both of those should get CPU temps to low 50's during load, but if you have the chipset cooling fan on the mobo remove it or you will get a "backwash".

 

Let me be clear about the first comment, it is not you but those cheap and nasty lugs that come with the stock Intel HS. They can look like there in place but they have a habit of popping out, bending and or wearing out and therefore not making a good contact with the CPU which is what I expect you are experiencing. For you to be idling at 60C means a poor contact with the CPU and I would estimate that loading would add another 10-15C on a Quad, i.e. upto 75C which would mean it's death over a short time period. The only way I have made sure of good contact with the stock HS is to install them when the mobo is out of the system and you can see that the white lugs are properly spread and locked in place by looking at the back of the mobo.

 

Finally these are the Zalman's but the other 2 I linked are better now, the Zalman's were king in 2006-7:
ZALMAN CNPS9700 LED 110mm 2 Ball CPU Cooler - Retail

ZALMAN CNPS9900LED 120mm 2 Ball CPU Cooler - Retail

 

I took the MB out and the heat sink is on correctly. I know what you mean about it being a less than crappy design but it's on correctly. I will check out the fans you reccomended tomorrow morning and likely purchase one. I want this thing to be as cool as possible could you reccomend the one that will make it run the coolest? at this point I don't give a shit about noise the stock fan is noisy as hell I just want this to function. thank you for all your help once again.

 

Well the king for 2008 for a long time in many reviews was, google it to satisfy yourself;

Thermalright Ultra 120 Extreme (aka TRUE)
http://www.thermalright.com/new_a_page/product_page/cpu/u120ex/product_cpu_cooler_u120ex.htm

You will have to search for one of those in your local area and I feel confident that you will be pressed to find anything better from it's era (2008). You may want to link it so we can assess whether you want an after market Fan with some serious CFM.

Sadly newegg doesn't have it in stock anymore?

Thermal Interface Materials (TIM) shootout's will more often than not mention and or crown:

Arctic Silver 5 or Ceramique

If you get both of those (HS & TIM) you should be fine.

 

I clicked your link in the last thread, and once again I thank you sincerely. But am I correct that there is no fan on this heatsink? It appears that it is just a huge hunk of metal which is awesome. is there no fan?

 

That's just the manufacturers site, if u goto a traders site and can find it in stock it should link or show a fan as well. If not we will get a decent fan for it as well. That is the "bee's knee's" for it's time and I bet most would be using a Noctua Fan, will have to choose the right one of course

 

Found one for sale, but it does not come with a fan, would like to purchase everything at once, is there a fan that you suggest? thanks again for your help.

 

Okay first things first:

Motherboards Compatibility List; Striker Extreme.....check

Installation Guide Ultra-120 eXtreme;....check

Fan reviews from very trusted source (imho):

XS Fan Review - Part 2 and the king of the hill is:

Sanyo Denki San Ace 1011

US suppliers:

Sanyo Denki US.....check

Distributors/Reps....Stock Check.....check.

Hopefully you can find a rep/dist point near you, should be smooth sailing from there. Keen to hear results afterwards.

 

Okay that info was a bit lacking and I was curious myself. Further reading of XS thread reveals at least 2 suppliers but more importantly the Product No. Sanyo Denki 109R1212H1011:
120mm
http://store.thermalfx.com/merchant...&Product_Code=109R1212H1011&Category_Code=120
80mm
http://www.frozencpu.com/products/7..._x_25mm_Hi-Speed_Fan_w_RPM_109R0812H4011.html
Manufacturers
http://www.sanyo-denki.com/sanace_sdaspec.aspx
Goggling part number
http://www.google.com.au/search?hl=...q=Sanyo+Denki+109R1212H1011&btnG=Search&meta=

Frankly I feel very confident if you get the HS, AS5 and this fan you will address your CPU heat issues. The main thing I really like is the HS is attached with 4xspring screws and not "cheap and nasty" plastic lugs. I would be very impressed if anyone can come up with a better Air Cooling solution. Yes it is a little bit more than you might pay but this is arguably some of the best parts you will get.

 

Just want to be sure, all I need is the heatsink, the fan, and some grease correct? Will purchase upon confirmation.

 

Well looking through the thread again we find that your mobo temp is 41C whereas your CPU was at 65C, that is a 24C difference between the two. Normally I take mobo temp as a "coarse" indicator of "system environment" temp (the average heat inside your case) but I only expect my CPU to be about 10C higher under load, your getting 24C higher. Even in BIOS you noted 60C at system idle after refreshing the TIM. To the best of my knowledge this is indicating poor contact and thus thermal transfer of heat with the CPU and will lead it to fail in the near future. It is quite clear that your PC is shutting down beacuse it is trying to save your CPU from frying.

As a final check you should have a look at your Video Card temps and report back using GPU-Z 0.34.

With all that said you need to address your CPU overheating issues and the HS, TIM and Fan will resolve this very well. It may seem expensive compared to other air cooling alternatives but they are the best parts (thus best results), cheaper than water (and will outperform many of the "gimmicky" water cooling kits) and can be readily applied to any future system builds if on the mobo is on the compatibility list (including I7 systems).

 

Heatsink came in the mail today. I thoroughly cleaned the processor and heatsink with rubbing alcohol before installation. My temps were in the low 50s. Temps were much cooler with the side of the case off naturally so I intend to drill a large enough hole to fit an extra 80mm fan on the case to provide more airflow. Once I buy the drillbit required to do so I'll report back with results. This should lower my temp by a degree or so I hope. In case you're not familiar with the case I use I have 6 case fans. All of them are 80mm. 2 intake on the front 2 exhaust on the back a side fan which seems to be blowing out and an top fan which also seems to be blowing out. They all came stock with the case, perhaps replacing them with different fans would provide better case temps and thus lower cpu temps? To me, a fan is a fan, and I know little about what to shop for.

I am typing this message on my desktop for once rather than my laptop so this is exciting. it's not shutting off! However... I'd love to know if there's a program that could monitor temps of the CPU, MB and GPU and anything else whilst in windows rather than in the bios. This way I could try to run some programs and such to see what it does under a load of some type. Also I'm really still not up to snuff on all the "lingo" for computers. You guys are friggin geniuses compared to me. I can tell what the parts are and where they go, but I just kinda figured out now that OC'd meant overclocked and no. I'm not overclocking at all I don't think... and do not know how to do so, so it's unlikely that this is causing the error. Would love to learn for the sake of learning but would also like to understand how to keep my temps down too. I'm not a newbie but treat me as such as I only really understand how to match up the pieces and parts. I know what I need to build a functioning PC and know where parts plug in and that's about it. Yes I built it myself, but I'm by no means an expert.

The case I have is an Alienware case it's the Super Alien. It has fan speed controllers on the front which don't seem to do much of anything, I don't know if I haven't set it up right or not, but they don't seem to control squat. There's probably a cable somewhere that I haven't plugged in. It's also got 2 temp devices, one that reads the HD temp and one that reads the CPU temp, neither of which I know where to mount the device that would actually tell me such things. I see the end of the wire and all, but where to put it is beyond me. I stuck the CPU one in the heatsink and it says 39 degrees Celsius but we all know that is a lie, and the HD temp I stuck well... on one of my HDs and it says 33 degrees but I don't know how accurate that is either. I assume that at the moment these provide me nothing more than a round about temperature for where they are sitting.

Like I said, it's staying on now so problem solved kinda, but we all know a cool PC is a happy PC and if there is any more that you could tell me I'd be absolutely ecstatic. Let's get this thing cooled down as much as possible and see what this baby can do! I bought expensive parts at the time of it's construction a few years back and would love to finally get this right since I've apparently had things so wrong for so long. Thank you so much for all your help thus far. I look forward to receiving a reply from you.

 

Use Everest for your temp monitoring as previous linked.

I would run OCCT for a while and it will plot ur temps and volts for 30-60mins to push the system a bit, do a 5min at the start and look at the graphs and make sure u feel comfortable b4 a decent 1 hour stability run!

OCCT (OverClock Checking Tool) 3.0.0.

Your not OC but this is an incredibly powerful and informative tool for "stability" testing.

 

OK I am running Everest now. My temps are as follows.

MB 33C (91F)
Aux 22C (72F)
GPU 60C (140F)
Seagate ST3400620AS 35C (95F)
Seagate ST3500630AS 35C (95F)

It does not tell my my cpu temp in there. I went to Computer and Sensor. First time I've ever used the program. What's the plan now?

 

Downloaded OCCT. CPU temps seem to be listed on Startup
CPU 1 60
CPU 2 59
CPU 3 58
CPU 4 57

GPUS are also listed they read out as
GPU 1 60
GPU 2 62

I'm not sure why but in this program some of the links on the left are grayed out and they are.
GPU: OCCT
GPU: Memtest
Power Supply

It might be useful for me to be able to monitor these items as well. It mentioned I needed to update something for them to work, I didn't know what it was or where to get it. But you guys probably do.

 

Other decent CPU temp monitoring apps are:

Core Temp 0.99.5

Real Temp

USe either or both of those and post back the temps.

 

I used both the first one I used was Core Temp and the temps are as follows.

Tj. Max: 85C
Core #0 44C
Core #1 44C
Core #2 43C
Core #3 43C

All of which are fluctuating by a degree in both directions except the Tj. Max. that is a constant.

The second program was Real Temp, and the temps are as follows

GPU 62C
Temperature C 49 48 48 48
Distance to TJ Max 41 42 42 42
Minimum 49 48 47 47
Maximum 52 49 49 49
Thermal Status ok ok ok ok

Hope this helps.

 

Okay now you want to run a 10-15 min run with OCCT and keep either of those apps open. Then look at the temp plots from OCCT and keep an eye on the CPU temps from the other app. Let's see how it goes after some stress testing. 44-49C is a lot better than 60C and you should be feeling a bit more comfortable. The restarts should cease entirely.

 

Testing as I type. The test was a 20 min test the test mode was Large Data Set and the priority was set to High. I assumed these options would push it to the limit. OCCT says that the cpu is at 98-100% load right now so I'm assuming I was correct in using these settings. Being that the different programs I've downloaded per your instructions gave me different temps, I left them all open and reported all the temps in this post. The temps are all over the place and very inconsistent from program to program. and they are as follows. And yes, the shutdowns have stopped. THANKS!!!

OCCT
CPU 1 77
CPU 2 75
CPU 3 73
CPU 4 73
GPU 1 60
GPU 2 62

Core Temp
Tj. Max 85
Core #0 62
Core #1 59
Core #2 58
Core #3 58

Real Temp
Temperature 67 64 63 63
Distance to TJ Max 23 26 27 27
Minimum 48 47 46 46
Maximum 67 65 63 63
Thermal Status ok ok ok ok

EVEREST
MB 33
Aux (fluctuates btwn 36 and 41 degrees)
GPU 60
Seagate HD1 35
Seagate HD2 35

 

Okay firstly we can note that both of those apps use TJ Max as a way to calculate your current temps. We can further note that Core Temp (CT) defines TJ Max as 85C and we can note that Real Temp (RT) is defining TJ Max as 90C (temp + distance to TJ Max = TJ Max). If we take into account that RT is defining TJ Max 5C higher than CT we can see that subtracting 5C from all values presented by RT gives us the same values as CT.

From the 2 apps it seems more rigorous testing has been done on RT:

http://forums.techpowerup.com/forumdisplay.php?f=65

Being of interest is a comparison here using CPU-z to assess the Processor:

http://www.xtremesystems.org/forums/showpost.php?p=2883315&postcount=573

I would be doing a Register Dump with CPU-Z 1.52.2 and then comparing with either Monitoring app to see which is being most accurate. Run CPU-z>>>>>goto About tab>>>>>Save Report as Text or HTML>>>>>scroll down report to Hardware Monitor and then compare those outputs with the apps. Which ever is closest you could infer is the most Accurate.

Can you take a pic of your current PC with the side cover off we may want to discuss the Fan set up.

 

I meant to say all 3 apps were giving exactly the same readings for my Q6600 after a 10min run of OCCT.

 

I will gladly take pictures for you tomorrow of whatever you like, just a few questions first to assure that I get you what you need. And also some questions just because I'm new to this whole thing. As I've said, I can put them together no prob, but the lingo is all new to me and I really need things spelled out for me to understand. I hope this isn't a burden. But, I know that most tech people are happy to help and share their knowledge or this site wouldn't exist. It's fun for me to talk about what little I know, so I'm hoping this isn't becoming a chore for you.

1. I'm assuming the insert image button here is how I attach pictures. can I attach more than one?
2. I have 2 intake on the front, 2 exhaust on the back an exhaust on the top and another on the side that also seems to be exhaust. They are the stock fans that came on the case, I'm sure looking up the case will give you all the same pictures I could take as I've not modified the fans in any way. So what precisely would you like my lens to focus on that these stock pictures can't show you? Or is it the hardware positioning in relationship to the fans that you need clarification on? I don't know the specs on these fans, or what the specs on fans even really mean, purchasing better fans would probably help if this is an issue, what would you recommend if you feel this is a necessary purchase?
3. I understand the 5 degrees difference in the programs as it's a simple estimate of your max temperature. But I still have no idea what TJ max is. Your explanation states that temp + distance to TJ Max = TJ Max. this confuses me. What do the T and J stand for? And what does this mean?
4. Do the video card temps seem hot to you? Would a PCI card slot fan do any good or would such a fan simply work against the fans on the video cards themselves and cause more problems? Also there would be extra wiring involved that would undoubtedly inhibit air flow. To many fans, with too many wires can do more harm than good, I've tucked away as many wires as possible and kept things neat in there to provide for the best cooling and airflow but this would put some wires in places that I couldn't easily tuck them away.
5. I've just realized that my hard drive that I store my data on such as music, movies, games, and my numerous pro tools files is completely full and bought another drive. This would undoubtedly cause temps to rise as my hard drives are positioned right in front of the two intake fans on the front of the case and the new drive will be further blocking the 2 intake fans and of course cause heat of it's own. Predicting the temps is of course impossible but I figured it should be noted.
6. The shut downs have stopped. However being that you are still trying to help me out which is awesome by the way, should I still assume that I am running too hot for comfort? What do you feel my temps SHOULD be?

I'll get you pictures tomorrow, but for right now it's 4:30, Goodnight and THANKS A TON!

 

1. Yes you can insert images that way or you can also host your images on another site and "insert link" which will allow you to add as many as you want.

2. I assume your case should be here and a link will provide enough info:
Antec

3. Tj MAX= Thermal Junction Maximum and it is supposed to be a more accurate measurement of CPU Temp than the DTS method. When you hit the Max value your CPU will throttle down and or shutdown to preserve. Tj max acknowledges that the greater the Tj Distance the less accurate the CPU temp readings are (because of so much "natural" variation) but as Tj Distance decrease the more accurate the temp readings become. Here is a discussion of a white paper by Intel on it but you are most interested in page 14-15 (which will list your CPU in tables), you can assess your CPU stepping by using CPU-z linked above:
http://intel.wingateweb.com/taiwan08/published/sessions/TPWS002/FA08 IDF-Taipei_TPWS002_Nov_1006.pdf
You can then define the right TJ Max in either of those apps I linked and this will give you the most accurate temp reading of your CPU and should ensure they are matching.

4. The Video Card temps look reasonable as they can generally handle a lot more temp than CPU. To assess your Video Cards you can run GPU-Z 0.34 . You can use OpenGL Fur Rendering Benchmark and run that for 1-2 mins and see what sort of temps you are getting in GPU-z. My card is idling at about 65C at the moment and just hit 92C during a 1 minute run of Fur. This is an extreme test but Video cards handle a lot higher temp ranges.

5. If you get a 2nd HDD make sure one is at the very bottom of the mounting cage and one is at the very top to optimise to inflow window.

6. Once we look at the case and other factors I will try and give an estimate of temps for you. I am concerned that your CPU may have already suffered a bit of damage and we have installed the HS in the "knick of time". If you can run CPU-z and tell me the stepping of your CPU we can then define the right Tj Max value in the monitoring software and then assess the most accurate value.

P.S. The EVEREST *Free Edition is a very old app and does not recognise your CPU.

Finally in your next post can you give me a idea of the average daily temp where you live as any and all heat will be correlated with ambient temps, i.e. your temps will get hotter in summer compared to winter.

 

WOW! heavy stuff mate. OK thank you for bearing with me. My exact case is listed here.

http://www.newegg.com/Product/Product.aspx?Item=N82E16811144004

that's the link from the invoice from newegg. the product is no longer listed on their site, but that's what I have. By the way, as far as the HD goes, this will be my third one, not second. I'll make sure to space them out as much as possible.

Thank you for the info on my video cards, I guess my case temps are normal then as well and I won't be needing to drill any holes which is a relief.

As far as my CPU taking damage as you've said. What does this mean? did I break the thing? I sure hope not, it seems to be running. And if I'm to have to buy a new one, I'm going I7 which would require a new motherboard and I just don't have the cash right now. I hope this thing isn't broken or I wouldn't have bought the heatsink.

As far as an average daily temp. I live in Pittsburgh PA. it gets fairly cold during the winter and fairly hot during the summer, but an average daily temp, I don't know it really fluctuates. but my house is usually around 75 degrees.

I think this answers most of your questions for now. Thank you so very much once again for all of your help.

 

I am sure I have asked but is this system overclocked in any shape or form? If so your going to have to back it down to stock for now.

Well I'm concerned that your CPU die may have been heat damaged slightly and may not be giving the most accurate temp readings using the traditional methods.

Can you run CPU-z and list the specs of your CPU:
Specification
Stepping
Revision

We can then define the most appropriate TJ Max value to be used with Real Temp and or Core Temp and therefore get better estimates of your actual CPU temp.

Finally I would suggest a good 10 min run of OCCT with Real Temp or Core Temp open and see what the temp ranges indicate. A final run will be a 1 hour stability session with OCCT to make sure the system wont shut down.

The HDD could go 1 above in the "External 3.5" Bay" and then one at the top and the bottom of your lower HDD cage, or if using a floppy top, middle, bottom.

 

No I'm not overclocking I don't think. OCCT says I was overclocked by .5% or something, but I'm sure that was just because I was testing. I don't know how to overclock and don't think I need to with this processor. It's fast enough. As for heat damage, how do I know if it is damaged, and if it is, what's that mean for performance and the life of the cpu? If it's gonna die on me in a few months I might as well just start saving for a new PC.

Yes I have a floppy drive, so I'll space the 3 out in the bay as best as I can. I'll run this program tonight when I get home from work. and post back then. Thanks!

 

I don't know how to tell if heat damaged unless you had been doing a lot of testing before it was overheating. You can at least see that all fours cores are loading with OCCT. You really just want to get the temps sorted out and make sure the shutting down has stopped.