• 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.

Replace Northbridge paste - worth it?

tuzz

Member
So I upgraded my e7200 stock CPU cooler to a Xigmatek s1283 which works great, but now my motherboard components get less airflow. The northbridge in particular is running hotter than before, ~48C at idle (tested with a digital multimeter) and instead of buying a new NB cooler I was wondering if replacing the stock pad/goop that is under the NB would be worth it? I'd replace it with AS5.

Oh for some background info, the system is currently running stock but I will definitely be OC'ing it when I get back into gaming and programming @ home (i.e. more than just browsing the web & listening to music). For hardcore OC'ing later on I'll definitely be looking into a new NB cooler, probably one from Thermalright, Xigmatek or Noctua.

Anyway, has anyone replaced the stock thermal pad/paste before and know if the temperature drop (if there is any) is worth it?
 
If they aren't put on with adhesive, I'll open them up and clean then apply a good grade of goop. Odds are that they weren't very carefully done at the factory. It certainly won't hurt to do that - unless you don't know how to apply your goop. The basic idea of goop is to fill all air spaces without applying too much - if the mating surfaces are nicely flat (remember, flat and shiny are very different attributes) it helps.

.bh,
 
OMG, aigo! Say it isn't so... :roll: My $50.00 ASRock was done much better than that, but I changed it out anyway.

.bh.
 
How long does it take you to put paste on a chip and mount the cooler?
If they spent that much time on every motherboard in their production line, they would go out of business in a month.

If the choice is between too much or too little paste, too much paste is the way to go.
If they set up the production line for just the right amount of paste, many of the boards will end up with too little paste due to production spread.
So, they set it up for more paste. Then, the majority of the parts will have enough paste. Some may have too much like what you see in the picture.

The objective in a production line is to assemble parts that meet spec and do it fast to minimize product returns and cost.
It is a business after all.
 
i personally think it is quite worth it. i just got a new board today to replace my old NF47-SLI from EVGA and the first thing i did was take off the NB and SB coolers and begin lapping them (not done yet). one thing i noticed is that the thermal pads they used were UTTER CRAP. i mean these things were horrible. they seemed more like some cheap rubber sticky pads than actual thermal pads, and i had to mix up a special mix of AS5, 2 different generic white TIMs i had lying around, and a tiny bit of isopropanol, all just to disolve the pads off of the northbridge and southbridge. standard alchohol wouldnt work, so i needed something which could act as an dual purpose abrasive/diluting agent without damaging the chip long story short, the chips are both clean, and once the coolers are lapped properly (one was covered in chrome on the outside 😵) im going to rebuild my system.
 
OK, so I went ahead and did it as I had some spare time the other day. The southbridge was easy to replace as it had a small thermal pad that was sticky on one side. But the northbridge was a PITA to take off, the paste that was on there had hardened and bonded the heatsink to the NB chip (which is surprisingly small). On second thought it could have been one of those wax pads similar to what AMD used to use on their stock heatsinks.

Anyway after replacing the paste, putting the mobo back in and firing up the computer, I let it sit at the desktop for a bit so it could warm up, maybe browsed a few pages. And idle temps were about the same, maybe 1 degree in difference from before. It'll take a few days and a few thermal cycles before the AS5 cures properly, but even then it should only be 2-3 degrees lower than currently.

I guess, in the end, it *might* be worth replacing the paste depending on how great/shoddily the original paste was applied, as it seems the stock stuff that was on there was OK after all. For me it wasn't really worth the effort, but hey you gotta try it to know for sure. It's still stable at 400MHz FSB (up from 266MHz stock) but for 24/7 operation at that OC I'll be looking at a new NB heatsink.

Cheers for the replies guys!
 
Originally posted by: Navid

The objective in a production line is to assemble parts that meet spec and do it fast to minimize product returns and cost.
It is a business after all.

no they have prepacks like this:

http://www.sidewindercomputers.com/sh35thpadamd.html

And they plop it over. The thing is, they use way too much. Its not the perfect amount, its called minimizing labor usage by using a generic TIM.

Taking the board apart and reduing the TIM can only be in your favor unless you dont know what your doing and you break your board.
 
Like I said in my reply: Don't bother if the sink is put on with adhesive. IOW, if the HS is not EASY to remove. All pastes are easy to remove the HS. Some pads are gummy messes and a solvent like naphtha (lighter fluid like Ronsonol) will be needed to clean it off. A minor twist should free many pads and all greases. Even if one can break the grip of an adhesive TIM, it will usually be very difficult to clean it off well enough to replace with normal grease. I'd expect to find adhesive TIM only on heatsinks that lack mechanical fastenings (push-pins, clips, etc.).

.bh.
 
Originally posted by: Zepper
I'd expect to find adhesive TIM only on heatsinks that lack mechanical fastenings (push-pins, clips, etc.).

That's what confused me when I was removing it - the heatsink had push pins so it didn't need adhesive. Which would be silly since there were push pins. And on second thought (to correct my post above) I'd say the TIM was a pad, not paste, as it didn't come off the same way paste does when cleaning with isopropyl.

Removing the NB heatsink felt just like removing my mates A64 stock heatsink that had a wax based pad, the pad had melted and bonded the HS to the CPU heatspreader in his case. Maybe for my mobo, the thermal cycles and much higher temps from high FSB OC'ing bonded it more than it usually would.
 
Back
Top