• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Chipset cooling

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
well here are some more pics for you.

1
2

Ok so what i did, I cut out 1/2 the fins like you can see. Then i used a router to make it flat to fit under the GPU. Then i had to cut holes for the capacitors to fit under. Counter sank the pins from the stock hsf. And finnally lapped it with 220,500,1000,1500,and 2000 grit sand paper.
 
You mean rotate it 90 degredds? It wouldnt fit that way. My friend and i spend 30 min planing out the best way to do this.
 
To be as OT as possible, I have a couple questions if you don't mind twitch. We have really similar set-ups so I was wondering what RAM you had origonally, it looks like redline, why did you switch to the OCZ, what rev is the OCZ and what kind of speeds are you getting out of it and the proc. I've got 2x512's of OCZ VX 3200 that I can barely get to do default timings and I have to run at 2t if go to 201 FSB. I need to replace it ASAP and I'm looking at 2GB kits. Thanks.
 
well originally i had 2x512 sticks of patriot 3200 ram, the non TCCD verison. I switched to 2gb or ocz platium ram. I am able to get 230mhz out of them but i have losen the timmings to 3-4-3-8. I have goten my cpu up to 2.8 280x10 with a 3/4 divider and 1.55vcore. Havent really tried past 1.55vcore cuz im not sure what is too much input on this would be great. Hope this helps.
 
I've taken my 3700+ San diego to 1.62v but I've found that anything above 1.55 didn't help much if at all. I have yet to be able to break 3ghz. On 1.55v I've run the multiplier range and gotten to 2988 perfectly stable but just can't great through 3ghz. I'm sitting at 270 x 11 right now, 1.45v, but 296 x 10 works fine too, at 1.55v. I'm very ram limited as the OCZ vx will not do 225. I'm running the DDR333 multiplier to keep it under 220, 212 right now I believe.
 
i cant get mine stable over 2.8 i was just up to 2.9 @1.67v just too see if i could. then i went messing around to see how high i can get my ram, long story short, bios got reset and i lost my raid0 array......
 
your array should come right back up if you set your bios back to using raid. The raid information with NF4's internal raid controller is stored at the end of the second drive and should not be gone from just a bios reset.
 
Originally posted by: DerwenArtos12
your array should come right back up if you set your bios back to using raid. The raid information with NF4's internal raid controller is stored at the end of the second drive and should not be gone from just a bios reset.


QFT.

Go into the integrated peripherals part of the BIOS, and in RAID config, enable RAID, and then enable it for whieva two sata ports you're using.

Should spot it again no problem 🙂
 
Originally posted by: letdown427
Originally posted by: DerwenArtos12
your array should come right back up if you set your bios back to using raid. The raid information with NF4's internal raid controller is stored at the end of the second drive and should not be gone from just a bios reset.


QFT.

Go into the integrated peripherals part of the BIOS, and in RAID config, enable RAID, and then enable it for whieva two sata ports you're using.

Should spot it again no problem 🙂

GFY if he set it up once, he should know how to do it again, i was just commenting that it should come back, and I believe I did say should, not will. So, yeah, I'm sticking to GFY.
 
yea well with the raid i just did a whole fresh reinstall. I was on the internet withing a hour of starting the format. I was really impressed with that. but all it good now.
 
Twitchee2:

I'm still trying to figure out how you secured that hunk of aluminum to the mobo. Various classes of mobos are different. My system -- a little out of date -- uses a P4P800 SE motherboard. I kept the stock ASUS aluminum heatsink, monitoring the temps the best way possible by taping a thermal sensor to the bottom of the heatsink -- as close as possble to the Northbridge chipset.

The air-cooling from my SI-120 ThermalRight would keep the ducted motherboard fairly cool, and the thermal sensor near the Northbridge never showed a temperature above 96F under load at room temperatures above 72F.

But, alas!! I had run the FSB on my system up to 1,000 Mhz with some OCZ DDR500 Gold EL's. Memory errors began to crash the system after six months. I still have some "validation" to do with the replacement memory modules, but it appears most likely that the chipset had degraded.

So now I'm wondering if I needed a better cooling solution for the Northbridge, or if it was just the luck of the draw.

In the meantime, I replaced my 3.2E (see sig) with a 3.4E, set the DDR500's aside, replaced them with some OCZ Platinum DDR400's with 2, 3, 2, 5 latencies, and tuned it up to about 3.689 Ghz at DDR434 -- no errors in MEMTEST86+ or S&M 1.7.6. Incredible bandwidth for the memory.

I'm just wondering if putting a hunk of aluminum on my Northbridge would save me the heartache I just went through. It just doesn't seem like bumping the 200 Mhz external frequency up to 217 is going to lead to that happening again. . . . .
 
Nice job. I've been thinking about doing the same thing with an old Taisol HS I have on an NForce4 mobo.
 
I'm correct in assuming that the motherboard came from the factory with the holes for those pins? I guess that would probably be the case for a Northbridge.

Either way, if it had the hook-and-loop spring steel retainers, that sort of thing wouldn't make modding the heatsink too difficult to use that approach.

 
Back
Top