So far, so good... DS5 prep

[bilbat]

Ready for TIM application testing, & reassembly...

[Lsdmeasap]

Re: So far, so good... DS5 prep

Which DS5 did you get? Good luck with the rest of the Build!

I'll see you tonight, Bout to jet

[bilbat]

Re: So far, so good... DS5 prep

X45-DS5...
Will tidy up harnessing at install...

[Dweirdo]

Re: So far, so good... DS5 prep

im noob, but isn't that 2 much paste ? :S just curious how much i should put when i build my new rig, anyways good luck!!!
and be careful :D

[bilbat]

Re: So far, so good... DS5 prep

They look 'blobby', but they're actually pretty thin - I scooped 'em out with a dental pick to tailor 'em to the area - tried twice under plexiglass for practice...

[Lsdmeasap]

Re: So far, so good... DS5 prep

Ouch!

I hope that is not conductive paste you put on your mosfet chips is it?

And I agree that is a TON of paste on the CPU, the method Is fine I suppose, but the amount is way to much. A X shape with the proper thin lines, similar to your SB but less, will make a square that will cover the whole CPU fully so you would not need the other lines thru there. Just giving you some pointers and my thoughts

WOW, I sure do hope that is a NON Conductive paste on your mosfet's! Looks just like Arctic Silver 5 to me, hopefully it's just the pics

[bilbat]

Re: So far, so good... DS5 prep

The paste on the MOSFETs is ArticSilver5, which Artic says is non-conductive: "Not Electrically Conductive: Arctic Silver 5 was formulated to conduct heat, not electricity. (While much safer than electrically conductive silver and copper greases, Arctic Silver 5 should be kept away from electrical traces, pins, and leads. While it is not electrically conductive, the compound is very slightly capacitive and could potentially cause problems if it bridges two close-proximity electrical paths.)" Figured while 'capacitive effect' might conceivably be a problem on high-speed, low amplitude signal paths causing skew, couldn't see how it could be a problem w/a MOSFET - that's specifically why I used it there (applied it with a dental probe, and as gooey is it is, the line between 'any', and 'too much' seemed impossible to bisect), and TIM Consultants TC0098 on the bridges.

Did several applications under plexiglass so I could see the spread - didn't seem to get full coverage 'till I tried the 'double-cross' pattern; here was my thinking on the northbridge:
If you notice in the picture, the thermal pad is raised about 90 thousandths from a surrounding three-eighths wide 'ledge'; my thought was to get complete coverage, and let the 'ledge' act as a catch-basin for the excess 'squeezed out'; didn't think it could possibly leak over onto the board, as I could have actually put the entire volume of TIM I had on the NB sink onto the 'ledge' without it even touching the surface of the heat spreader/sink. Checked all around w/a dental mirror after seating it - seems to have worked as I figured.
The TC0098 is the stuff I quipped about you 'costing me $7' - the TIM application article you pointed me to had a link to "33-Way Thermal Interface Material Comparison" : 33-Way Thermal Interface Material Comparison | best thermal compound,thermal interface material,thermal grease,thermal compound,TIM,heatsink compound,33-Way Thermal Interface Material (TIM) Compound Benchmark Performance Test Comparison Product Revie which included the product - it's thixotropic: the viscosity drops when subjected to shear, so a good 'twist & wiggle' gets excellent spread...
Took about eighty pictures of the process of: building up the silicone pads to anchor the NB/SB fans w/o blocking much fin cooling area, or transmitting fan noise to the fins themselves (which I figure could as easily be called 'resonators'!), removing and reinstalling the heat pipe assembly, modifying the RAMFAN to replace its sixty cent fan and relocate the replacement to the top, mounted on rubber grommets, etc. Will sit down & try to work it up into a 'how-to'...
If you really think that the AS5 on the MOSFETs might be a problem, I suppose I can always unscrew the assemble, use some Q-tips to clean it up, & try again...
Thanks in advance for any feedback...

Bill

[Lsdmeasap]

Re: So far, so good... DS5 prep

Arctic silver 5 has silver in it for sure, and will short out your Mosfets. If you are lucky you can clean that off, but I would be scared for sure

As you quoted >>>

Not Electrically Conductive:
Arctic Silver 5 was formulated to conduct heat, not electricity.
(While much safer than electrically conductive silver and copper greases, Arctic Silver 5 should be kept away from electrical traces, pins, and leads. While it is not electrically conductive, the compound is very slightly capacitive and could potentially cause problems if it bridges two close-proximity electrical paths.)

That is just their disclaimer to keep them out of lawsuits. As you can see it is very vague and contradicts itself. There IS metal in it, and I have seen pictures of the havoc caused on Mosfets (I tried to find you the pages, but could not as it has been a bit and I dont remember where they were posted)

Here is another warning for you if you do not like mine >>>
How To: Applying Ceramique to Mosfets

I think this says it more precise Capacitive not so much conductive >>>

(3) Apply thermal compound to the SPP die. I recommend either Arctic Silver Ceramique

or Tuniq TX-2 compounds. Do not use AS5 or any metal bearing thermal compound

as the are capacitive (not conductive) so any compound squeeze out that get on the

bypass, noise suppression or de-coupling SMC's on the chip substrates will change

their capacitance values and characteristics.

NCIX.com - Buy Viperjohn 790I Spp Chipset Water Block & Vcore Mosfet Heat Sink Kits G1/4 Support *No Barbs* - VJ790i-SPP&Vcore-MOSFET In Canada.

As for your NB that pad is there so when the push-pins push down on the heatsink the chip does not crack. It should be there and allow the NB to get contact, if you would fell better you can replace the Push-pins with Screws, but leave the pad and be careful you do not tighten to tight and crack the chip

[bilbat]

Re: So far, so good... DS5 prep

Not talking about the grey thermal pad(s) in the original setup - yanked 'em out & tossed them. The metal thermal contact area of the NB itself is the part that's raised 90 thou, and the blank phenolic surface below caught the excess TIM... Used the original spring loaded pins for the MOSFET sinks; tied the NB & SB sinks down with screws, but rather than tighten them 'one half turn before the snapping sound', put the original springs between washers on the 4-40 x 3/4 nylons I used, and torqued 'em down a little bit more compressed than the original pins had them, AFTER having done the 'twist & wiggle' (with the board supported by anti-stat foam on an antistat worksheet) to shear the TIM into flowing out...

Ahh, the picture you linked actually seems to have a bit more TIM on the MOSFETS than I wound up with - like I said, the choice seemed to be none or too much - it's really gunky to try to work with in small quantities in inaccessible spots... Can see all the MOSFET traces at the rear of the board (facing the I/O plate) & nothing leaked down onto them; also made a 'blob' of ArticSilver5 & stuck a pair of probes in from high-impedance meter on high resistance range - reads 'dead open'...

[Lsdmeasap]

Re: So far, so good... DS5 prep

Ya, I was talking about the Northbridge pad that is there around the outer edge of the heatsink, that is there for cushioning so you do not crack the bare silicon chip.

And do be sure you also do not use conductive paste on the NB as there is MANY bare contacts under that heatsink as well right beside the chip itself

Ya, Too much is ok I suppose as opposed to too little, if you are using the right material. Do keep in mind also if you warp your board any at all you may make the middle of the heatsinks not make contact with the top of the chips. So also keep that in mind when tightening down your HSF

I wont keep telling you, just remember you were informed in this thread when you have a burnt Mosfet chip