s1rrah

I like the mounting system on the Thermalright Ultra 120 Extreme I installed this past week ... but I think the pressure could still be improved somewhat.

After removing my TRUE the other day to remount, I noticed the AS5 "footprint" on the CPU heatspreader wasn't as spread out as it used to be on my high pressure modified scythe infinity (using backplate with bolts that are not thread distance limited like the stock TRUE's).

I'm thinking of hacking the mount on my TRUE this weekend to look something like this:

...



...

This way, when using a decent backplate, one could crank the pressure up way higher than the default TRUE mechanism, which really only lets you tighten to a certain point before running out of thread on the screws.

Anyway ... I personally think it might positively impact temps in the 3 to 6 degrees (load) range ...

I'll post a mod/work log this weekend after I do it ...

...

But otherwise ... have any TRUE owners modified the mount pressure any?

???

__________________
...
Aumotocnic: "An unfortunate member of the overclock.net insomnia club"
...
OSAMT: "Member of OCN Songwriter and Musician Thread"
...
OHPC "Member of OCN Headphone Club!"
...

i_ame_killer_2

This is just another way of doing it.

Source if you want.

__________________

fergsonfire

is that a penny? thats such a cheap hack! awesome! +rep for you my friend

Flower

I shall have to try this...

__________________

Gollie

I don't even think my screws are screwed down all the way. I have MORE than enough pressure applied between the base of my TRUE and the IHS on my Q6600.

This is not the first thread I have seen regarding this issue which makes me wonder why people would rather exert (in my eyes) extreme pressure on your CPU, Motherboard and your TRUE instead of just taking an afternoon to lap the IHS/HSF flat.

__________________
Home Setup: Forte > Zu Firemine > Cambridge 840C > Soloz reference IC > Singlepower MPX3 (Brimar CV1988/Tung Sol 5687's) > Zu Mobius > HD650

"Overclock.net headphones club: Because perfect hair is overrated."

"No Highs, No Lows, must be Bose"
-Duckydude

afzsom

The thing is, after lapping both flat, you usually have less material being squeezed with the same mounting pressure, so it's likely to be looser than before you lapped.

__________________
Watercooling Loop: MCP655 Vario -> Black Ice GTX360 -> Stinger Diamond Max -> MCW-60 -> MicroRes
Loop Cost: $395.03

Best. Thread. Ever. || C1E/EIST/SpeedStep Fix || Undervolt and Overclock your CPU! || GPU-Z Benchmark Settings || GPU-Z Stock Settings || Lian Li Owner's Showcase || Lian Li Group

Gollie

If you have done enough research to want a TRUE, you should know that they only work to their optimal potential when lapped. If you buy one without the intent to lap it, you should just go with a Tuniq Tower; they will perform the same.

__________________
Home Setup: Forte > Zu Firemine > Cambridge 840C > Soloz reference IC > Singlepower MPX3 (Brimar CV1988/Tung Sol 5687's) > Zu Mobius > HD650

"Overclock.net headphones club: Because perfect hair is overrated."

"No Highs, No Lows, must be Bose"
-Duckydude

afzsom

Uh, I think you misunderstood me. I have an Ultra 120 Extreme. I have lapped it and the E6600. After doing so, the performance was much better (thermally) than before lapping. However after lapping, the same mounting mechanism was used and the heatsink was looser (when properly mounted) than before I had lapped both. Increasing the mounting pressure (with a washer) helped to lower the temps another couple of degrees. It's not a replacement for lapping, its in addition to lapping.

__________________
Watercooling Loop: MCP655 Vario -> Black Ice GTX360 -> Stinger Diamond Max -> MCW-60 -> MicroRes
Loop Cost: $395.03

Best. Thread. Ever. || C1E/EIST/SpeedStep Fix || Undervolt and Overclock your CPU! || GPU-Z Benchmark Settings || GPU-Z Stock Settings || Lian Li Owner's Showcase || Lian Li Group

s1rrah

Thanks for the comments.

My TRUE and CPU are both lapped and I did indeed find that the mount was a tad looser than before. Temps are fine for the moment but I'd like to experiment with getting even better temps if possible.

I know the mount hack/mod dropped my Scythe Infinity temps well over 6 degrees (it uses lame push-pins stock) .... so I figure I can get a few more degrees with a very high pressure mount with the TRUE.

With my old Infinity, with both it and my CPU lapped ... and when using the tiniest dot of AS5 imaginable, the compound would still be compressed to the very edges of my CPU heatspreader.

With the TRUE, the compound is only compressed across about half of the CPU surface which to me indicates that it could be improved some.

The TRUE is doing fine so far ... idling around 33C @ 3.5ghz ... but I hope to improve it even more ...

I *am* sort of upset that my Q6600 requires 1.49 volts to run at 3.6ghz, but that's how it goes I guess . I've read others running the same overclock with much lower voltage settings:

...



...

I'm going to try the bolt-through mod this weekend.

Thanks again for the comments.

__________________
...
Aumotocnic: "An unfortunate member of the overclock.net insomnia club"
...
OSAMT: "Member of OCN Songwriter and Musician Thread"
...
OHPC "Member of OCN Headphone Club!"
...

afzsom

Try the washer mod first, its ten cents at your local ACE Hardware store, and much easier to use than a complete bolt through. However if the washer mod is insufficient (it shouldn't be), then you can move upwards

__________________
Watercooling Loop: MCP655 Vario -> Black Ice GTX360 -> Stinger Diamond Max -> MCW-60 -> MicroRes
Loop Cost: $395.03

Best. Thread. Ever. || C1E/EIST/SpeedStep Fix || Undervolt and Overclock your CPU! || GPU-Z Benchmark Settings || GPU-Z Stock Settings || Lian Li Owner's Showcase || Lian Li Group