No one has mentioned this, so I will chime in, especially after doing some reading after asking GG707 about his thermal pads and confusing myself that much more...
At DigiKey, when you order your Cobs, there is a few links at the bottom, and with the Vero's were these Loctite thermal pads for $.86
http://www.digikey.com/product-detail/en/1886887/1000-130-ND/4898949
ds:
http://media.digikey.com/pdf/Data Sheets/Loctite PDFs/TCF_ 4000_PXF.pdf
1st, I can't seem to find a thermal conductance rating, BUT the thermal impedance is the 1st thing above the text....
Notice @30psi, the thermal impedance is about the same as PK3 thermal paste [
0.013 ℃-in²/W]
The whole range of PK1,2,3 respectively is...0.017 * 0.015 * 0.013 ℃-in²/W and all have a thermal conductance of
10.2 W/m-℃
This isn't a knock on PK3 by any means, in fact, with all the positive data provided by Supra and everyone else emulating, I think that these pads have some serious potential...........
Why Care?
Well.....lets start with impedance which might turn out to be the one of the more important #'s for calculating thermal "stuff".....
http://www.arlon-thermal.com/thermal_prop/
Thermal Impedance (q, m2*K/W):
Thermal impedance is defined as the
total opposition that an assembly presents to the flow of heat. In practice the assembly represents the material itself and the interface between the material and the surfaces it is in contact with.
Thermal impedance can provide the designer with more information than either conductivity or resistance because impedance includes the material interaction with a surface. However, the designer must remember that different surface topographies and pressures will affect impedance. Therefore the impedance measured by a standardized test method like ASTM D5470 will not necessarily translate into the impedance of the material in application. In fact, the impedance measured by one ASTM D5470 test fixture may not equal the impedance measured by another due to differences in the interaction of the material with the test fixtures surfaces.
The thermal impedance is the sum of the material resistance and contact resistance.
So Two question arise out of this.....
1st, the phase change for the thermal pads is 45C. I have seen two builds plus my own, where the thermal paste, looks like it doesn't get hot enough i.e go thru phase change. Now I don't know enough to be making a valid opinion here, but it is interesting to see.....
I have a punctuated theory, that says the impedance is too low initially when firing up the COB's and can cause may enough distortion to prevent phase change or the thermal paste possibly has a higher phase change temperature than the silicon pads, Which seems somewhat unlikely, but impedance is based on thickness and the thermal pads seem to have less variability because of tolerances and a low phase change..........
II keep coming back to the advice I first got when installing CPU's, "thinnest line possible"......
#2
Can we measure the psi we apply and how....granted I still don't fully understand the relationship between conductance and impedance and I don't have a thermal cond. # to compare to the PK conductance rating.....
but is there a way to measure the pressure across the Cob surface? Without some research on my part I am clueless, but interested.
For instance, Ideal says to mount the ChipLoks down @ 3.5 "/Lbs. Could you then calculate the surface pressure with the diameter of the hardware and the mounting angles?