Heatsink Mounting Paste Question. / Chemistry / Forums

Forums

4hv.org :: Forums :: Chemistry

« Previous topic | Next topic »

Heatsink Mounting Paste Question.

previous 1 2

Move Thread

LAN_403

Sulaiman

Mon Jul 18 2011, 11:47PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

OOPS !
posted in the wrong thread....goto 2N3055 heatsink

Pinky's Brain

Tue Jul 19 2011, 01:55AM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

Ash Small wrote ...
I'm building some similar, larger ones for a bridge rectifier, and that will probably use up several syringes full.

Why not just use bits of soft solder, clamp it and put it in the oven? Much better thermal conductivity.

magnet18

Tue Jul 19 2011, 04:46AM

Registered Member #3766 Joined: Sun Mar 20 2011, 05:39AM
Location:

Posts: 624

Zinc oxide in oil... sounds like sunscreen...
I might give that a shot in the future, but I think I'll check to make sure it's not too conductive first.
I wonder how it would compare to titanium oxide?

Bjørn

Tue Jul 19 2011, 06:23AM

Registered Member #27 Joined: Fri Feb 03 2006, 02:20AM
Location: Hyperborea
Posts: 2058

Some did get squeezed out though, but I thought it was best to use slightly too much than too little.
I'm building some similar, larger ones for a bridge rectifier, and that will probably use up several syringes full.

The paste should only fill the microscopic voids on the surface. so a syringe should last almost forever if you use parts that are properly flat. Using more than required will give you worse performance since the thermal conductivity of the paste is lower than that of the heatsink itself.

Sometimes when I am just testing something I use a drop of water, it works well until the water evaporates.

Ash Small

Tue Jul 19 2011, 08:54AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Pinky's Brain wrote ...

.Why not just use bits of soft solder, clamp it and put it in the oven? Much better thermal conductivity.

I had considered doing just that if I was to use copper plates instead of aluminium alloy.

I was under the impression that you can't solder aluminium, (hence the Al MOT wire being 'useless')

I'm not sure if sunscreen will be ideal, due to viscosity, etc. (I'm just guessing here, I imagine it is more viscous than heatsink paste, but I've never had to use it, I live in England)

I'm tempted to try DOT 5 silicon brake fluid and zinc oxide nano particles (French system).

DOT 5 is hydrophobic, but does absorb a small amount of air. it does have a high boiling point though, and a relatively low viscosity.

I imagine it also has pretty good heat dissipation qualities, due to it's intended use.

Any suggestions/comments?

Forty

Tue Jul 19 2011, 11:47PM

Registered Member #3888 Joined: Sun May 15 2011, 09:50PM
Location: Erie, PA
Posts: 649

I wonder if any polishing compounds might work, as they have small particles in an oily suspension. I've got a bucket of cerium oxide powder at school. I think a soft metal might be the best for heat transfer though. even if the solder won't stick to the aluminum, it will fill up the voids. you could use straight lead and hammer it into little premade heatsink squares.

Neet Studio

Sat Jul 30 2011, 01:36AM

Registered Member #4037 Joined: Fri Jul 29 2011, 03:13PM
Location:
Posts: 86

Here is one of the classics: Thermal transfer compound comparison.

They tried the usual thermal compounds from different makes and just for fun: toothpaste, Vegemite and tap water.

>Firing up my Minty Fresh Thermal Transfer Experiment produced, rather hilariously, an excellent score - 0.47Â°C/W
>Vegemite's runny compared with thermal goop, as well.. [stuff deleted] But the stuff still delivered a solid 0.48Â°C/W result.
>To see whether these substances work so well simply because they're wet, I tried some plain tapwater between the heater and the heat sink. And lo, it scored a magnificent 0.41Â°C/W - possibly at least partly because, without any gelling material, it was free to move heat by convection as well as conduction.

So ultimately the right goop is the one that give you the long term reliable and performance...

Chip Fixes

Sat Jul 30 2011, 02:13AM

Registered Member #3781 Joined: Sat Mar 26 2011, 02:25AM
Location:
Posts: 701

I heard toothpaste works very well... until it dries up

Ash Small

Sat Jul 30 2011, 06:30AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Neet Studio wrote ...

Here is one of the classics: Thermal transfer compound comparison.

That's an interesting article, Neet. The difference between the best and the worst is 0.02 degrees C per watt, and they are all at least 0.16 degrees C per watt better than no paste. I think that means that the difference between best and worst is around 12.5%, but as using heatsink compound only gives a 25% increase in performance than no compound, 12.5% of 25% is an overall increase in performance of only around 3.125%. This will obviously vary from application to application depending on other factors (the most important being size/design of heatsink, rate of airflow (cubic feet/minute), flatness, etc., but indicates that in a well designed system, assuming the heatsink and component concerned are cool enough to touch (the general rule used by most here), the type of compound used makes difference of a couple of degrees C at most in the 'worst case scenario'.

Apparently, the silicone based DOT 5 brake fluid I mentioned above has a sufficiently low vapour pressure that it can be used as diffusion pump oil. I think (but I've not checked) that it's boiling point is probably higher, so it shouldn't 'dry out'. It is hydrophobic, so doesn't absorb water. The only disadvantage I can find is that it will absorb a small amount of air, but given that the surface area exposed to the atmosphere is virtually non-existant, this shouldn't be a huge problem (I think all silicone based oils do, even the ones they use in the 'proper' compounds.)

I'll just mix it with zinc oxide until I get the same consistency as the 'real thing'.

previous 1 2

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob