Thermal Epoxy

You didn't know his Super Hero personality it Captain Obvious?

Even the metal-filled stuff is mediocre thermally. Diamond-filled epoxy is mediocre! The epoxy between the grains dominates theta. Getting the interface very flat and very thin is the best way to reduce theta. If the fill grain size is such as to increase the gap, the filler usually makes things worse.

John

PeterD wrote in news: snipped-for-privacy@4ax.com:

Plus you get the protection of circuit breakers and UL approval... OTOH,you just made a nice path for lightning to enter your home. ;-(

--
Jim Yanik
jyanik
at
localnet
dot com

Nope. That would require using 110ac and I want to run relatively small 2-conductor cable with a hefty wall wart 24vac that doesn't have to meet any code.

Jim

Then step it up at the work site. Watts are watts. The step up would not lose much to inefficiency.

Rain gutter or mobile home water pipe tape is good for this application. You only need to increase the temp to a point slightly above the freezing point for water. That's where the rain-x would come in.

Also, there are DC versions of such heat tapes available for the hot tub boys.

Set your PC clock, dipshit.

Not fair, John, With that idiot, it's like shooting fish in a barrel.

Bob

--
== All google group posts are automatically deleted due to spam ==

There is *nothing* in your post indicating why this is

*not* a viable solution.

A "nearly flat" fastener would have un-measurable impact on the dish's geometry. Likewise, it would provide little opportunity for ice to "stick" (any moreso than any other place on the surface).

Even if the dish is relatively small (high curvature per unit area), that just means:

- the fasteners must be located closer together (e.g., straddle the device instead of axially)

- use smaller heating elements (for above)

- use *longer* fasteners and accept small gaps between the dish and the heating element at its extremes (the other gaps proximate to the tanget point can be filled with a thermally conductive compound *if* you think the heat will dissipate to the atmosphere more than assisting in the melting of the surrounding ice/snow)

I would have explored a CalRod operated at lowered voltage to both distribute the heat over a larger area, prolong its life *and* more readily conform to the (back) surface of the dish.

But, I'm sure you already thought of that

Smaller wattage tends to mean smaller devices. Smaller devices have mounting holes (axial) closer together. Closer together more closely "fits" a curved surface.

But I'm sure you already thought of that

As did:

"Dabs, that's a technical term, you'll get used to it."

But I'm sure you already thought of that

Hmmm, if we wrapped all that with shiny aluminum foil tape even fewer photons would escape our evil clutches.

John

If I had ever worked for a dumbfuck like you, you would no longer be with us, as I like ridding the world of vermin.

More retarded even than the first reference was/is.

Pretty brave words for a guy who is too chicken to even use his real name.

John

IFYPFY

If you knew what CTR meant, you'd be impressed!

John

What else do you expect from NoNads?

--
Lead free solder is Belgium's version of 'Hold my beer and watch this!'

I learned that when bonding some LM-35's to various pipes. I used a epoxy from Tri-con, ISTM. But even the best stuph was still FAR worse than direct contact.

--
A host is a host from coast to coast.................wb8foz@nrk.com
& no one will talk to a host that's close........[v].(301) 56-LINUX
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433

I helped to engineer opto-isolated HV SSRs about eight years ago. I think there were two patents. It was for 6kV, and was for a high speed color laser printer/copier. The supply was as big as a full sized motherboard, and had six discreet HV supplies on it with about 3 different voltages for the corona wire, and the drum. It had encapsulated little HV daughterboards all over it and HV coupling links using 30kV HV SPC wire, and was computer controlled (microcontroller). That was an interesting supply to "come up with" The printer was a monster I guess,like the ones I now see at workplaces. Really fast tabloid sized high speed high res full color laser printer/fax(not color)/copier, etc.

For the 6kV we tried 15kV wire, but the Teflon allowed the corona to worm its ways to the surface, boring a hole along the way. That is why we had to use the 30kV stuff even though we were only pushing 6kV.

Capacitive loads are fun to have to design a supply to feed into.

I know that the Dow white thermal silicone grease squeezes down below

100 micro-inches, my resolution limit. So the fill particles must be pretty small. If the surfaces are really flat, an unfilled grease, something that truly flows out of the gap when compressed, is probably best.

You can buy an LM35 in a TO-220 package, handy for bolting to things.

John

No shit. It is a simple series circuit. You have the thermal source, and taking say a CPU as an example, that gets adhesively bonded to a medium that has a smooth. flat outwardly exposed surface that the user of the device has to attach a heat soaking device to so that the source can be continually cooled. Each of those interfaces between good thermally conducting materials is a thermal resistor, as is the chip die-to-substrate interface, and then the heat sink substrate we attach our heat sink to and the heat sink. I count two interfaces that are essentially thermal resistors. Since your sacred intimate contact is all but impossible in the real world of cheap, mass produced heat sinks and other variables, we put a material between said interfaces.

One hopes the gaps is as near nil as can be achieved, so that the 'gaps' are mere 'slivers' of void and most of the two surfaces ARE very closely mated.

That is the heat sink case.

There are cases where one cannot rely on the profiles of all the heat generating components being the same, and needing to conduct heat from multiple sources on the same card. In such cases, the design calculates circuit operation based on a mean, relatively hot operating temperature for all the components, and uses conductive cooling methodologies, so large gaps can be tolerated for some components, and the entire assembly ends up rising up to about the same temperature, where the sources are, of course, definitely identifiable 'hot spots' when examined in the IR. The whole thing ends up running pretty hot, but it is best when one does not know what the air density will be and cannot rely on convection to bring assemblies down to known, repeatable temperatures.

There are thermal "putty" sheets that are applied to the circuit, and then vacuum attach that to the filly machined can it goes into until it sets. You have 100% shielded, 100% reliable, environmentally sealed, known operating temperature assemblies.