Heat sink question

I've searched and searched and all I find are tests about exotic thermal pastes that PC overclockers use.

What I'm wondering is: In general, is zinc oxide thermal paste or these silicone pads better at heat transfer? This is for a TO-220 voltage regulator. One problem is the pads I have are from a surplus distributor so I don't have any technical info on it. For anyone interested, here they are here

formatting link

There is no need for electrical isolation between heat sink and the device. My instinct is the thermal paste is better.

I really don't like when silicon gets hot to the touch, but maybe I'm being overly cautious. It's a 7805 regulator and it's dissipating max of 1 Watt. My heatsink is about 20C/W no forced convection.

Reply to
hondgm
Loading thread data ...

formatting link

Silpads are awful. Grease is best, mainly because the gap can be very small *if* the mating surfaces are flat.

A 7805 dissipating 1 watt doesn't need great heat sinking. If you can tough it' it's fine.

I can touch 50C forever, 60C for about 1 second, and can interpoate pretty well between.

John

Reply to
John Larkin

"John Larkin" wrote in message news: snipped-for-privacy@4ax.com...

Hmm, but a 7805 running 1W might be unbearable initially, but with repeated touchings, can be touched for longer and longer, until your fingers are heatsinking it quite nicely.

I typically use a small clip-on heatsink in that size. Grease or pad is not required.

I also recall being able to hold a 30W soldering iron in my hand -- if I start it from cold and keep rolling it around, I can dissipate 30W pretty much forever. Let go for a few seconds and there's no going back, it's too hot to touch.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms
Reply to
Tim Williams

info.asp?number=3DG15622

yeh if theres no need for isolation grease can't be beat, it'll basically be metal to metal with the grease just there to fill imperfections in the surfaces

in a TO220 Rthja is only 50C/W so it could do with nothing but the package

-Lasse

Reply to
langwadt

formatting link

--
IME, silicon pads are crap.
Reply to
John Fields

Repeated touchings? I'm not some sort of silicon molester.

But the theta of a fingertip should be measured. That would be good to know. Sloman could do it for us; he's a peer-reviewed expert at thermal stuff.

Someone mentioned that a lot of heat sinking wouldn't be needed at 1 watt.

I've held irons as they warmed up and always had to let go. The 30 watts is interesting. A person at rest dissipates about 100, so the iron is increasing your overall dissipation by 30%.

John

Reply to
John Larkin

here

formatting link

I've tried to measure how thin Dow-Corning filled thermal grease will squish down. It's below my measurement resolution of about 100 micro-inches, so the fill particles must be very small.

Somebody with a decent microscope could measure some fill particle sizes for us.

Sil-pads are many mils thick. Not to mention that most all their specs are optimistic by 2:1 or so.

John

Reply to
John Larkin

Use the LDO version.... it will run much cooler :) (LM2940)

Reply to
TTman

TTman:

Why?

--
Saluti
Reply to
F. Bertolazzi

formatting link

At 1 watt, he really doesn't need any thermal interface material.

John

Reply to
John Larkin

I know this isn't true. All an LDO is good for is to get by with putting less voltage in.

It would run cooler if I did that, but my power transformer is what it is, and the regulator input voltage isn't changing.

Reply to
hondgm

Magic, of course.

--

Tim Wescott Wescott Design Services

formatting link

Do you need to implement control loops in software? "Applied Control Theory for Embedded Systems" was written for you. See details at

formatting link

Reply to
Tim Wescott

formatting link

Engineering likes silicone grease because it works. Manufacturing likes Sil-pads because they're much less messy.

Silicone can handle high heats quite well -- you'll be frying your skin long before you'll damage the silicone grease.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" was written for you.
See details at http://www.wescottdesign.com/actfes/actfes.html
Reply to
Tim Wescott

formatting link

The compromise is to use the phase-change stuff. Nobody likes it.

John

Reply to
John Larkin

Silpads insulate the parts from the heatsink. This avoids a lot of problems.

VLV

Reply to
Vladimir Vassilevsky

"John Larkin" wrote in message news: snipped-for-privacy@4ax.com...

Be sure to collect lots of datapoints, especially high temperature points to check for nonlinearity.

Yeah, seems like a lot. Admittedly I've never done it for very long, principally because I've got soldering to do.

I've done it with 6L6GC's too, which run around 30W. They only get to ~200C in operation.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms
Reply to
Tim Williams

Sil-pads are interesting stuff. You wouldn't think a rubbery material would be all that conductive, but you can wrap a sil-pad around a finger, and lightly brush a hot soldering iron across it, and feel it almost immediately. You might drop the pad.

It's conductive stuff alright, it's just that metal is a way better conductor.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms
Reply to
Tim Williams

If you use LDO version and keep the input voltage just over the needed level, the unit will operate cooler because everything inside is at a much lower R state.. Less Heat because of Less R...

Reply to
Jamie

formatting link

Sil-pads are crap. If you don't need electrical isolation, then just use some of the heat sink goop. When I was in the USAF, we used DC-340; lately, every time I've bought a new CPU and heatsin, I've cleaned off the heat goo that comes with the unit, and slathered DC-340 on to it.

Presumably, that's the same as the "zinc oxide" goo, but isn't zinc oxide the stuff that lifeguards smear on their nose? ;-)

Cheers! Rich

Have Fun! Rich

Reply to
Rich Grise

Depends on how quickly your skin can form a callouse. ;-)

Cheers! Rich

Reply to
Rich Grise

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.