ICD 7 Test Results

[ICD7]

Dropped 5C pretty impressive

 

[Insidious]

It surprised me.

I didn't expect to get that much gain (4-5 degrees)

Best part is that I was able to go back to the winter-time overclocks (the ones in my results) even though the house is warmer by several degrees.

-Sid

 

[Cutthroat]

Sid, what do you mean by you added a washer to each mounting screw? Can you explain how you did this? Right now I can wiggle my TRUE even with it fully tightened, I'm worried that after lapping I won't get good contact anymore.

 

[Insidious]

The ultima-90 has two screws with springs attached. These screws go through the clamp and screw into the board. I just put the washer on the screw before I put the screw through the clamp. It rests on the lip of the clamp and makes the spring compress farther. I tighten the screws untilt hey bottom out.

Here's a picture of it

-Sid

edit: The hole on the washer is big enough for the screw, but smaller than the spring. The washer rests on raised lip of the clamp to make the spring compress more (tighter)

 

[ICD7]

Insidious Could you put together another before test with the original paste? Then we could compare/ link a full thermal set to the digital mapping

 

[Cutthroat]

Originally posted by: Insidious
The ultima-90 has two screws with springs attached. These screws go through the clamp and screw into the board. I just put the washer on the screw before I put the screw through the clamp. It rests on the lip of the clamp and makes the spring compress farther. I tighten the screws untilt hey bottom out.

Here's a picture of it

-Sid

edit: The hole on the washer is big enough for the screw, but smaller than the spring. The washer rests on raised lip of the clamp to make the spring compress more (tighter)

OK, I get it. I doubt that will work with my TRUE because it's really hard to get the screw started in the first place, they already seem too short.

I did my pressure tests and lapping today. I don't have time to post the results right away, but it's fair to say that my lapping was a disappointment, my before and after temps are the same, I can probably look forward to a couple of degrees difference after curing. I'll post the full results with scans of the film tomorrow maybe.

ICD7, do you only want us to send back the print copy, or the donor copy too? And btw, for future reference don't send out self-addressed envelopes with postage paid out of the country, because Canada Post isn't going to accept that stamp. No big deal, I'll get a stamp and send it back to you shortly.

 

[Insidious]

Originally posted by: ICD7
Insidious Could you put together another before test with the original paste? Then we could compare/ link a full thermal set to the digital mapping

I would if I could but I don't have anymore of the original paste (AS5). That's why I had to beg for a new tube of ICD7 to do the pressure tests.


-Sid

 

[ICD7]

Here's a lapp and edge job, little disappointed he did not get before numbers

http://www.ocforums.com/showthread.php?t=557899

 

[Cutthroat]

OK here's my disappointing lapping results, I'm not sure what to do next.

Lapping results:camera::brokenheart:

I didn't reverse the pressure tests, it actually does look better before lapping. Possible problem is mounting the TRUE, it takes quite a bit of pressure to get the screws started and the heatsink kinda rocks back and forth. That may be putting pressure on only the edges of the CPU? I'm going to try another test, the plan is to place the TRUE on the film, but not mount it, I will use a weight, or hand pressure to hold it down flat and see if I get better results.

I'm also not sure if the heatsink makes really good contact to the CPU since I can still wiggle it after fully mounting it. What i might try is removing the clasp that holds in the CPU, then use Sid's trick with the washers to make up the difference.

I'm really disappointed with the temps after lapping, they really didn't improve at all. I don't know if I used to much TIM or what, but I'll do some experiments with less, different, and no TIM to see if I can figure it out.


Any thoughts or suggestions?

Pics of the lapping job to come... (need to install camera software first)

 

[ICD7]

Originally posted by: Cutthroat
OK here's my disappointing lapping results, I'm not sure what to do next.

Lapping results:camera::brokenheart:


I'm really disappointed with the temps after lapping, they really didn't improve at all. I don't know if I used to much TIM or what, but I'll do some experiments with less, different, and no TIM to see if I can figure it out.


Any thoughts or suggestions?

Pics of the lapping job to come... (need to install camera software first)

Yep, read comments at xtremesystems for a nearly Identical result

http://www.xtremesystems.org/f...=2955822&postcount=483

 

[Cutthroat]

Originally posted by: ICD7

Originally posted by: Cutthroat
OK here's my disappointing lapping results, I'm not sure what to do next.

Lapping results:camera::brokenheart:


I'm really disappointed with the temps after lapping, they really didn't improve at all. I don't know if I used to much TIM or what, but I'll do some experiments with less, different, and no TIM to see if I can figure it out.


Any thoughts or suggestions?

Pics of the lapping job to come... (need to install camera software first)

Yep, read comments at xtremesystems for a nearly Identical result

http://www.xtremesystems.org/f...=2955822&postcount=483

Interesting, even if I don't 100% understand. Does that mean If I can manage a few more MHz OC my load temps would likely not increase?

I'm going to try the pressure test without mounting later, if I don't like those results maybe I'll also edge lap the CPU as well and I'll do a pressure test after that too.

I keep wondering how much sanding it would take to virtually remove the IHS, and what would happen if you did reach the core or holes appeared in the IHS?

Anyway, here's some pics of my lapping job, sorry I'm not good with my camera.

Before Lapping

Before removal of ICD7, good coverage IMO, pea method used.:camera:

After ICD7 removal, no staining, but some scratches in the center from twisting the HSF.:camera:

Partially Lapped

TRUE partially lapped, note the moat.:camera:

CPU partially lapped, same moat as the TRUE.:camera:

After lapping

After 1000 grit and polished with ICD7.:camera:

 

[ICD7]

Interesting, even if I don't 100% understand. Does that mean If I can manage a few more MHz OC my load temps would likely not increase?

I keep wondering how much sanding it would take to virtually remove the IHS, and what would happen if you did reach the core or holes appeared in the IHS?

As to Q1 I am not sure, your idle temp delta's temps should only increase as you increase the load Watts which would mean the the diodes are not linear if so would indicate you probably could get some numbers like Insidious did but unfortunately he did not record the idle load temps.

Q2 you probably have .050-.060 of metal there. you probably were no more than .005 taken off.

 

[Narse]

Finally got around to applying the IC Diamond compound.

q6600@3.6

Before
Idle 43C
Load 74C

After
Idle 37C
Load 61C


I am very pleased

Simply an amazing product. I was a long time AS5 user but from now on I will be sure to only use IC Diamond in the future.

 

[ICD7]

Hey Narse, thanks for posting your result. Much appreciated.

What was the the "before" compound and how long was it installed?

 

[ICD7]

I sent out the first round of film for digital mapping - and I will email your results monday or tuesday only 9 went out this week.

I did talk to the rep about the part of the film that shows voids.- The white or unexposed portions "voids" are a go-no go type of situation on the pressure side and no indicator of defects. Voids could be .0001 or 10 miles deep as far as the film reading is concerned. Just something to be aware of.

 

[Narse]

Originally posted by: ICD7
Hey Narse, thanks for posting your result. Much appreciated.

What was the the "before" compound and how long was it installed?

Arctic Silver 5, had it installed about 5 months.

 

[alfa147x]

9.5x
417 x 9.5
3.96Ghz
1.284V->1.3

 

[ICD7]

I am posting the images from the pressure film results as I have time on all the forums that we had a giveaway test on.
you can view first result OCF only 60 more to go.

http://www.ocforums.com/showthread.php?t=538919&page=13

 

[Narse]

I would be interested in trying out the film.
e-mail on the way

 

[ICD7]

Here you go, Intel spec on flatness for you lappers

http://download.intel.com/design/processor/datashts/31685201.pdf

page 32. flat .05mm -parallel .203

 

[Insidious]

Originally posted by: ICD7
Here you go, Intel spec on flatness for you lappers

Automated the linky
page 32. flat .05mm -parallel .203

All I know is (with a lapped HSF), the edges of the CPU heat spreader seemed to monopolize the contact area until it was lapped too.

:beer:

 

[Assimilator1]

Just to let you know I haven't forgotten about the film test I've just got way-layed, hopefully I'll do it by next weekend.

 

[ICD7]

Poll - what do you guys think of the Raw contact film as a diagnostic tool for contact/pressure issues?

 

[Fullmetal Chocobo]

I have the film as well, and we are in the middle of a move and I'm trying to get it done. One good thing is that the CPU that I need to test is in the computer that is too heavy to move, so it is gong to be at the house for a little while longer (until Liz's brother arrives to help move it).

 

[Insidious]

Originally posted by: ICD7
Poll - what do you guys think of the Raw contact film as a diagnostic tool for contact/pressure issues?

It's nice as a tool to see where contact is being made, but I don't think it's of any use for pressure testing. We have no point of reference. Also, the shades of red are ambiguous at best.

-Sid