Q6600 and big tower coolers

O

OlIv0rIolI

Gawd
Joined
Dec 2, 2005
Messages
634
The reason I started a new thread is because of a problem I don't see any other people address. I am thinking of getting a thermaltake 120 extreme with the q6600, but I am afraid of the weight bending my mobo.

In my previous build. I had a copper zalman 7700 and my the top edge of my mobo around the cpu area is warped in due to the stress of the weight. My main concern is that will something similar happen with the thermaltake since its heavior and alot taller and therefore puts more bending moment on the mobo?

I am thinking of getting a p180 case as well. Anyone with experience and solutions? Thanks
 
While the tower coolers do induce a bigger static moment on the mounting holes of the motherboard than low/er profile coolers it's still ways off total destruction of the motherboard.

I've seen some installation of LGA775 stock Intel coolers which induce some high stresses on the pcb when people are trying to get the push-pins in when the motherboard is on the standoffs.

I recently installed a Thermalright Ultra 120 extreme. It does induce a bigger than usual moment on the motherboard, but I'm not worried that anything will break off.

Once installed in the motherboard tray i could see the motherboard was already flexing a bit (bigger gap between pcb and case than other edges).

I guess one solution would be to have a piece of string (or two for lateral stability) from the top of the case to the heatsink. With this you could reduce the force couple acting on the motherboard through the mounting holes. (not a very practical solution.. probably a pain to implement.)
 
I guess one solution would be to have a piece of string (or two for lateral stability) from the top of the case to the heatsink. With this you could reduce the force couple acting on the motherboard through the mounting holes. (not a very practical solution.. probably a pain to implement.)
Click to expand...

An extreamly good idea, I have seen a couple of solutions, rubber bands (big thick suckers) tye wraps etc.

OP you are right to be concerned those things in general violate the Intel spec for total HS weight (around 575 grams if memory serves) and "hanging" a big tower with some support from the crossbar that is just above is a very good idea.

Somthing obivous but I will mention it anyway, I would unmount such a big heatsink before I transported the computer or at the very least lay the machine down on its side so the HS was vertical.
 
I figured as much and I know that it is a pain in the rear end to implement this. I don't like it when my mobo is bent a little even if its safe epecially since I am getting an expensive mobo. Also I will probably have to move the system around once every two month. It gets me worried.

Do you know any other heatsink that cools as well as the ultra 120 extreme or should I start considering my first water cooled setup?

I am going for 3ghz overclock with a cool temp of around 36c not under load
 
36 degrees Idle at 3Ghz might be possible. I lapped both the Ultra 120 Extreme and the Q6600, right now I'm getting 42-44 Idle on Core 0, and 37-41 on the other cores. This is with a yate loon. I'm hoping once the AS5 sets in it drops a bit.

Now I haven't yet optimized the overclock, so cpu voltage might be too high for nothing. I'm hoping to find the sweet spot over the weekend.
 
What about just having the case on the side? Air flow is still the same. If I ever got one of those giant heatsinks I would definitely put the case on the side. Those motherboards are flimsy.
 
Can't put them on the side. Just too impractical. Takes up floor space and does not look good for a antec p180
 
Unfortunatley the towers are the kings of cooling atm.

Artic freezer pro seems to be the most popular other than the huge towers for good cooling. Should get 3.2GHz out of it but thats based on what I read, not what i have done.

Will need $150 to WC ouchies,
 
PCB is designed to flex... there is no real issue to worry about.
 
OlIv0rIolI said:
Do you know any other heatsink that cools as well as the ultra 120 extreme or should I start considering my first water cooled setup?
Click to expand...
From what I've read, the Ultra 120 extreme is THE best air cooler. So no, there is no other that you should consider.

If you are truly worried about your mobo, you might conisder water. However, like has been said, there is little cause for concern.
 
1. find piece of string
2. tie one end of string around heatpipe of tower cooler
3. tie other end of string to somewhere at the top of the case
4. ?????
5. profit
 
One more question. I read some reviews of ultra 120 extreme and it seems like alot of people complain about quality control specifically the uneven and ruff heatsink surface. So my question is how do you lap it to make it smooth and even? I never done it before, and what tools should I get?
 
I really like the idea of the string. I have a tuniq 120 and it's quite heavy. I'm afraid that it'll just break off from my motherboard one day.
 
The base of my Ultra 120 extreme was very curved. the edges were high the middle was low. Almost as if they try and get extra pressure in the middle on the cores...

As for the tools, look around for lapping guides. It was my first time at lapping anything and it went fairly well. I bought sandpaper at the local hardware store (some 220, 400, 600 and 800 grit, waterproof). I then proceeded to taking apart a framed picture on the wall to get the glass to use as a flat surface. From there, the next 2 hours were spend sanding down the base of the Ultra 120 and the Q6600 heat spreader.

I was pretty surprised by how the edges of the heat spreader were quite a bit higher than the center.


As for the bracing of the towers... I slept on it and it would probably be possible to implement some bracing which uses the ATX motherboard mounting holes as anchorage points. This might sound crazy, but would work well with my removable motherboard tray.

I'll take a better look at this option tonight.
 
These large coolers won't break the motherboards. I don't see any real reason for concern here. People used to panic concerning the size of the old Alpha heatsinks during the Athlon (Thunderbird) days. Now look at us, the heatsinks we have now dwarf those things.
 
Solder joints are not designed to flex and there is a real concern with damaging the socket and its electrical and physical integrity with the board (read "rip the sucker right off:) )

Specification for heatsink weight, flex, clamping force and socket specifications are available on Intels web site. Reading them (the thermal mechanical guidelines ) is highly recommend so informed comments can be made.

Force = mass x accleration Intel sets a maxium weight to the cooler based on what the socket can withstand based on common accleration values (moving the machine around) and then probally adds an engineering margin for safety. Going from the max of about 550 grams to a 700+ gram HS and then plopping it down on a desk could be trouble.

As mentioned in Section 2.1, the heatsink mass must take into consideration the package and socket load limits, the heatsink attach mechanical capabilities, and the mechanical shock and vibration profile targets.
Click to expand...

The recommended maximum heatsink mass for the ATX thermal solution is 550g. This mass includes the fan and the heatsink only. The attach mechanism (clip, fasteners,etc.) are not included.
Click to expand...

Package pull-out during mechanical shock and vibration is constrained by the LGA775 socket load plate (refer to the LGA775 Socket Mechanical Design Guide for further information).
Click to expand...

These large heatsinks are out of the design envelope, the backplate makes up for a lot of the "sins" but frankly, IMO, you would be "less than smart" to not take further easy steps to mitigate the risk if the machine is ever moved around. Now if its going to sit on your desk or floor and once in a blue moon you gently pick it up and set it aside to dust or vacume, no there is probally not a issue. But I find it impossible to argue against a well done brace or additional support. I will use the same reasoning we dont run our processors 2 deg C under the PROCHOT trip point, it just aint smart.
 
I have read those same technical documents and while I understand what you are saying, the fact remains that I have never seen a heatsink destroy a motherboard due to it's weight. Nor have I seen one damage any processor that has an integrated heat spreader. Now there is a concern I have (others may have this concern as well) in reference to these gigantic heat sinks that use the Intel stock push-pin style mounting mechanism. This just seems stupid to me. I know it's stronger than it looks but I don't trust that on anything significantly heavier than the stock coolers.
 
Dan_D said:
I have read those same technical documents and while I understand what you are saying, the fact remains that I have never seen a heatsink destroy a motherboard due to it's weight. Nor have I seen one damage any processor that has an integrated heat spreader. Now there is a concern I have (others may have this concern as well) in reference to these gigantic heat sinks that use the Intel stock push-pin style mounting mechanism. This just seems stupid to me. I know it's stronger than it looks but I don't trust that on anything significantly heavier than the stock coolers.
Click to expand...

I have seen a couple of posts (and only a couple (2) since I started hitting the boards again last summer) cant remember if here or XS where the socket did rip right off yeeouch! However this was during the computer being transported and in one instance it was during a household move so apparently it got extra rough treatment and the other was a new system shipped by UPS, which will do in even a "normal" machine :eek: unless packaged very well. The push pin only are the worst I agree, the backplates help a lot. Obviously this is a very rare occurrence or we would have heard all about it. However when asked I will always revert if possible to the tech docs, we engineers tend to be a conservative lot and like to worry.

If someone designed a little adjustable brace that clamped onto the the top fin and then attached to the case crossbar, they could probally sell me one if I used a big tower.

One thing I noticed even with the stock cooler is that with the pushpins (I didnt try a HS with a backplate) the board flex (board flex due to clamping force preload on the cpu) which is accounted for and actually if within specs is desired by Intel, caused the MOSfets to "move downward" away from the base of the heatpipe designed to cool them. I could have put a credit card in the gap. i had to fabricate a "board stiffener" to bring them back in contact as the same thing happened when I mounted my waterblock. LOL we cant win, if its not one thing its another.
 
BillParrish said:
I have seen a couple of posts (and only a couple (2) since I started hitting the boards again last summer) cant remember if here or XS where the socket did rip right off yeeouch! However this was during the computer being transported and in one instance it was during a household move so apparently it got extra rough treatment and the other was a new system shipped by UPS, which will do in even a "normal" machine :eek: unless packaged very well. The push pin only are the worst I agree, the backplates help a lot. Obviously this is a very rare occurrence or we would have heard all about it. However when asked I will always revert if possible to the tech docs, we engineers tend to be a conservative lot and like to worry.

If someone designed a little adjustable brace that clamped onto the the top fin and then attached to the case crossbar, they could probally sell me one if I used a big tower.

One thing I noticed even with the stock cooler is that with the pushpins (I didnt try a HS with a backplate) the board flex (board flex due to clamping force preload on the cpu) which is accounted for and actually if within specs is desired by Intel, caused the MOSfets to "move downward" away from the base of the heatpipe designed to cool them. I could have put a credit card in the gap. i had to fabricate a "board stiffener" to bring them back in contact as the same thing happened when I mounted my waterblock. LOL we cant win, if its not one thing its another.
Click to expand...

Yeah I hate using the amount of force required to mount those stock Intel coolers. They really need to switch to a better design. The only thing I like about that method is you can remove the heat sink with the board installed in the PC.
 
With my Si-128 using the original push-pins retention mechanism, the board is flexing so much I was afraid to pop something. And when putting the case standing, the stress seems a lot with the top of the cooler weighting so much. I decided to buy a Ultra-120 bolt-down retention mechanism (not designed for the SI-128 but both heatsinks share the same base) and since then, it's much better and improved the temps a lot.

With big heatsinks, a bolt-down mechanism should be the only way to hold the cpu...

 
Fortunately on mine, the weight of the 120Ex is helped by one end of it resting somewhat on the fan controller edge, so not as much flexing as it normally would if any. Hasn't seemed to hurt it all from what i've observed. Would go with something to hold it up if i had to do anything else with it though.
 
Xilikon said:
With my Si-128 using the original push-pins retention mechanism, the board is flexing so much I was afraid to pop something. And when putting the case standing, the stress seems a lot with the top of the cooler weighting so much. I decided to buy a Ultra-120 bolt-down retention mechanism (not designed for the SI-128 but both heatsinks share the same base) and since then, it's much better and improved the temps a lot.

With big heatsinks, a bolt-down mechanism should be the only way to hold the cpu...

Click to expand...

I am a little confused. Can someone explain to me or show a picture of what exactly is bolt down mechanism and push pin retention mechanism and which cooler/heatsink has which?
 
OlIv0rIolI said:
I am a little confused. Can someone explain to me or show a picture of what exactly is bolt down mechanism and push pin retention mechanism and which cooler/heatsink has which?
Click to expand...

If you go to http://www.thermalright.com/a_page/main_product_accessories.htm#acc_lga775, you will see a bolt-down mechanism with screws and a back plate to hold it together. A push pin mechanism is another one without the back place, relying on springs and locking mechanism to hold on the holes. It's a pita to install due to the sheer force required to lock the push pin.

 
You must log in or register to reply here.
Back
Top