Mica and Goop

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Yes, the surfaces of these alumina insulators are very smooth, like glass (they also look very fragile, though I haven't broken one yet. At $0.70 each shipped, it wouldn't be a crisis if one did though).

With the alumina it isn't as easy to tell when you have enogh goop on, compared to mica, since the insulator is the exact same color as the goop. Is Arctic Silver silver colored? That might help.
 
Are the alumina insulators large enough for TO-247 or TO-3P devices?

Jims Audio sells TO-220 alumina insulators that are 14x20mm. I just measured some Fairchild MOSFETS with the metal tab and looks like they just make it around 15mm wide. Is the hole in the right place length wise?
 
Last edited:
expected semiconductor life is said to double with every 20C decrease in junction temperature.

Arrhenius equation : AF = Exp [Ea/k x (1/Tn - 1/Ta) ]
Tn is 298K (273+25), Ta is accelerated temperature
(Ea=0.7eV, k is Boltzmann constant)

Half-life expectancy temperature raise is :
<10C for Ta < Tn +100
>10C for Ta > Tn + 100

Examples :
10C above 25C : ~7.84C
20C above 25C : ~8.08C
100C above 25C : ~10.1C
150C above 25C : ~11.37C

Wet thumb average : 10C
 
Last edited:
I'm looking into going to Kapton Tape instead of pads. I have both mica and silicone pads out of recycled electronics, which come in handy, but for building new projects, it would be easier, especially for parallel, to just tape your surface, use grease on the transistor, and you are done. It would also come in handy if you wanted to use a nice large flat piece of aluminum or copper bar on a row of parallel devices with no insulators, and just insulate that bar with tape before connecting the heatsink. More instant thermal transfer from transistor to metal that way.
 
Administrator
Joined 2004
Paid Member
Hi EWorkshop1708,
You can sometimes find Mica in a sheet. Grease both sides and you are off to the races. But since you are only a DIY'er and not a manufacturer, individual mica insulators aren't very much of a big deal.

Bwaslo is dead on the money too. Kapton was used in some early car amplifiers, then they switched sides to use as heat protection while Mica returned to the heat sink side. You really can't afford the hit in the thermal resistance department.

-Chris
 
In my project, I used AAVID Thermalloy Aluminum oxide pad + Arctic Silver. Worked great. AL pad is quite expensive, but this small Arctic Silver tube can be used for many project, so I don't think it is expensive at all. Easy to apply with old credit card.

I think Thermalloy Thermalcote is better for amplifiers. I have Arctic Silver too which I use for CPUs because it's stiffer.
 
It may have something to do with the way they clamp down.

A very thin goop will spread out easily as the clamp load is increased. One ends up with a very thin layer and with sufficient clamping force there will be metal to metal contact at the peaks of the irregular surface.

A thick goop will resist being squashed and may leave a thicker layer where the clamping pressure is lower.
A single bolt/screw fixing through a slightly flexible plate will have high pressure in the area around the screw fixing, but the pressure will rapidly fall off as distance from the fixing increases. In the extreme the plate may flex enough to leave a gap to the sink.

Going back to the thin vs thick, the thin may have a lower proportion of solids compared to a very high solids content in the thick goop.

The solids are what gives the permanent and good thermal transfer. The liquids may evaporate to be replaced with air.

If the clamping method can achieve a more even and adequately high pressure, then a high solids goop may perform better in the longer term.
Whereas a low solids goop may test better at the time of assembly.
 
I think Thermalcote has better specs than Arctic Silver and I've been using it for ages.

I find that single use thermal pads work well for CPU heatsinks. Less mess. Sort of like Kerafol pads. When using Thermalcote, the heatsink tends to slide around after it heats up. I guess because of the clamp design. Arctic Silver is a little better.
 
It may have something to do with the way they clamp down.

A very thin goop will spread out easily as the clamp load is increased. One ends up with a very thin layer and with sufficient clamping force there will be metal to metal contact at the peaks of the irregular surface.

A thick goop will resist being squashed and may leave a thicker layer where the clamping pressure is lower.
A single bolt/screw fixing through a slightly flexible plate will have high pressure in the area around the screw fixing, but the pressure will rapidly fall off as distance from the fixing increases. In the extreme the plate may flex enough to leave a gap to the sink.

Going back to the thin vs thick, the thin may have a lower proportion of solids compared to a very high solids content in the thick goop.

The solids are what gives the permanent and good thermal transfer. The liquids may evaporate to be replaced with air.

If the clamping method can achieve a more even and adequately high pressure, then a high solids goop may perform better in the longer term.
Whereas a low solids goop may test better at the time of assembly.

Sounds very reasonable to me.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.