• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

e6600 + evga 680i

B

BloodyIron

2[H]4U
Joined
Jul 11, 2005
Messages
3,439
I recently installed a new HSF (Thermaltake 120 IIRC) on my rig with the hopes of having a higher oc headroom. However it has become a lower headroom.

I want to explore options on the software/bios end before opening my case, as the setup I have is a pain to get in and out, and takes a long time to do so.

I have a C2D e6600 on an EVGA 680i SLI. This is the first version of the motherboard as I picked it up a week after release. I also have 2x1gig Patriot 1200mhz DDR DIMMs. Also a nVidia 260 GTX 216 core (don't recall manufacturer).


I replaced the retail intel HSF with the Thermaltake one. In the process I cleaned up as much dust as I could. I doubt that I dislodged the Northbridge HSF, but I am uncertain. The MCP (Northbridge) temp reports 70-77oc. When applying my thermal compound (arctic silver 5) I followed the instructions on the arctic silver's site.


Previously I was able to safely OC the northbridge to 1333mhz, up from 1066. Now, I cannot even reach 1100 stably. Windows shortly crashes after loading the desktop.

As it stands, I am out of ideas. Can someone toss me some bones pls?
 
yeah 70C is a WEEE bit high.. check make sure you didnt accidently bump up the northbridge voltage or something.. or try defaulting the bios and starting over again..
 
FYI, you might want to redo the thermal compound on the North and SouthBridge of those boards. A lot of people have reported drastic temp drops after doing so.

You do have the NB fan installed right?
 
I actually replaced the thermal compound on the NB and SB (since theyre connected via the cooler) last time I was serious about OCing it.

It does have the fan on it.


As for the avoltages, this temp occurs both on stock and upped voltages. Also, if I OC to say 1100a and leave the CPU voltage on auto, it goes from about 1.28 to 1.4. This also seems like a rather excessive increase.
 
So, I reseated the NB/SB HSF. Quite a lot of dust between the NB HS and the F. I thoroughly cleaned the dust out of the fins. Alcohol swabbed the NB and SB chipsets, and re-applied thermal compound (and swabbed the HSF's themselves).

I'm now down to about 62oc with the FSB at 1300.

HOWEVER, I still had the crashing.

So, I got creative. Went into safemode, and stopped all nvidia autoloading apps. It is now running stable.

I suspect the issue originally was that I was autoloading a profile on launch which jerked my FSB back to 1066 (scary stuff, im sure).

So far, so good. Thanks for your help so far guys!
 
Keep those NB temps down. 62C is still too high, really... I'd look at replacing the fan with something higher CFM. Those boards are known for killing themselves over time when OCed and not very well cooled. Speaking from experience.
 
the bios excessively increases voltages when overclocking. lock all chipset voltages to where they are at stock cpu clock speed. increase them as necessary to achieve a stable fsb overclock, but don't let the board adjust them.
 
Although those temps will not kill your 680i mobo, your MCP temps can be much lower. Another thing, MCP is your southbridge while SSP is your northbridge on any Nvidia board. People get this mixed up all the time.

Which thermal compound did you use to reapply on the NB/SB heatshink? I've used MX-2 and it's worked out great for me.

I've also removed the worthless and loud stock fan off the heatsink and replaced it with a much quieter 70mm x 15mm fan that moves more air.

DSC00021.jpg


DSC00022.jpg


DSC00019.jpg

I did have modify a bit to ensure that it would fit. A little dremel did the trick.

Some double sided was all it needed to ensure it stays in place. My MCP temps rarely goes over 45c and usually hovers around upper 30's and lower 40's.

Another thing to note is that your MCP temps will be directly affected by the temps of the video card since they are so close to each other. Keep the temps of the video card low and you should see a few celcius drop in MCP temps.
 
In all of this im using arctic silver 3.

Wicked responses guys! and great pics! Is there anything wrong with ziptying a replacement fan for the NB? I'm wary of using double sided tape since I haul this thing to lans.

As for GPU, it runs about 60oc idle, or just under (57?). As for in use, i'll have to get back to you.

I OC'd the FSB to 1333 and ran prime95 for like 45 mins straight while i went to go get groceries and it didnt fail, but the temps were stable with the NB running in the low 60's or so (IIRC). The CPU is 55-60oc.

Should I move my GPU down one socket? (it's currently at the top) As for cooling, my case has a expansion slot exhaust fan. 120mm, on the quiet end I think.
 
AS3 is good but a bit outdated. Not sure how much you could improve your temps by using MX-2 but try it if you can obtain some.

Your GPU idling at 60c?? That's freaking high man. My GTX 280 SLI'd doesn't idle more than 40c on my 780i mobo. Do you use Rivatuner? If not, use it to set your idle fan speeds a bit higher without noticeable increase in fan noise. You can also set the program up so that it uses that fan setting at startup.

The CPU temps seems high for e6600 with FSB at only 1333. Can you provide a link to your CPU cooler so that I can see what it is? You may need to reseat the CPU cooler to ensure that it is seated properly.

About moving the GPU... as far as I know, moving it to any other slot will cause an error upon boot on the 680i.

Also, zip ties will work well for the replacement fan... as long as you can tie it on there somehow.

Posting your BIOS voltages will help.
 
ahem, Arctic Silver 5

My CPU is idling now about 45-48 now that it's been running for a while.
 
AS5 is good, but one downside to it is that it requires inordinate amount of curing time... 200hrs in order to see its potential benefits. MX-2 virtually has no cure time and it performs just as good, if not better, than AS5.

Your CPU idle temp is still too high for relative mild OC. Nothing detrimental but high nonetheless. It's fine If you can live with that.

I have my e6600 OC'd to 3.53ghz and its idle temps rarely break 35c. I'm currently using Scythe Ninja Mini as its cooler. At load, using Orthos or Prime, I have a hard time breaking 55c. I'm using Realtemp as the temp monitor.

Your CPU runs hot and your GPU runs hot... this equates to your mobo running a bit hotter than it should.

Post your BIOS voltages so that we can let you know if you're running your voltages too high or not. Increase in voltage = increase in temps.
 
I'm running auto for all except CPU, which is set to 1.35v.

I'll have to get back to you on the exact numbers. I will force them to a setting later (whatever it is at the time).
 
As rennyf77 stated, you do not want to let your mobo run all of the voltages on auto. It will set it to more than your system needs, which in turn creates more heat.
 
Wow those are some big pics. :D

Your MCP (the southbridge - this is found just right of the first PCIE slot) is showing 42c, which is very good temps. I'm guessing that you took the pic right after you turned your system on since the board temp is only 23c.

FSB - you can try to lower this to 1.3v. It should work but if you are getting stability issues, you can raise this back up to 1.4v. Shouldn't need no more than that.

SSP - You can lower this to 1.35v unless you are running your memory at its rated 1200mhz. If that is the case, you probably want to keep it at 1.45v.

Everything else looks good except for the memory voltage. 2.3v is high and it's not in your best interest to run your DD2 memory at that voltage, especially on the 680i.

How fast are you running your memory right now?
 
Here are temps. Ran the system for a while and shut down and went into BIOS immediately.

temps1.jpg
 
Yeah, that is fairly high. I think you should try remounting the chipset heatsink with thermal paste other than AS5 that does not have such long cure time.

Here are few more suggestions that can help in lowering not just your MCP temp but temps of rest of your components as well. I've already mentioned some of this before.

- Make sure the pins that hold the chipset down are pushed all the way through. If the pins are not holding the chipset down properly, it will obviously have an effect on the temps.
- Make sure the pins are not FUBAR'd, in other words, damaged. Damaged pins will obviously not hold it down properly.
- Your CPU temps idle a bit high for relative mild OC on e6600, which will no doubt get your load temps up in the 60-70c range especially when you are OC'ing. I suggest you remount the CPU cooler and check pins/adapter that hold the cooler down for damage. You may need to invest in a different cooler all together if this doesn't help.
- Your GPU temps idle rather high at 60c. Download Rivatuner and manually set the fan speed to 50%. The GTX 260 fan at this speed should be barely audible and you should be able to get a good drop in idle temps. Rivatuner has many customizable features that are very useful for overclocking, temp control, and monitoring. You can find a comprehensive Rivatuner setup guide if you google it (unless you are already using the program... which I don't think you are).
- 2.3v is way too much for DDR2 ram on the 680i. This puts quite a bit of strain on the chipset as well the as the memory. Regardless what the memory is rated at, I highly recommend that you lower this voltage and do not go above 2.1v unless you want to test your luck. You will need to lower your memory frequency if you actually have it running at 1200mhz when you lower the voltages.

Good airflow within the system itself is also very critical in keeping your temps in check. Get some more fans (or more powerful fans) on your case if you can. Hopefully you have performed some cable management within your system since unkempt cables can hinder airflow.
 
-When I rethermal'd my NB/SB I made sure the pins were properly holding it down.
-I have only one damaged pin but it is retaining it as well as the others
-I have no interest in remounting anything at this point. It is the worst pain to get this mobo in and out of my case. I would only take it out if the system were to immediately damage itself.
-Ill have to get back to you on idle temps, but ive cranked my fan before, and it made little difference (cranked to 100 according to the nvidia utils)
-I really do not want to lower the frequency, as that was the purpose I purchased the RAM for. It is rated for 1200mhz, so why go lower?

I have already replaced my stock intake fans with higher airflow ones. My case is a LianLi PC 1000 Plus II, so it's not a common airflow design. As for cable management, I have tried my best with this case, but it has it's limitations :/
 
You must log in or register to reply here.
Back
Top