1. Home
  2. Electronics Design
  3. glob-top thermals

glob-top thermals

I was wondering what might be the thermal effects of lumping a heap of glob-top epoxy on a surface-mount part.

I have an interesting circuit that unfortunately dumps 0.8 watts into a 1206 resistor. That's pushing things a bit even by my standards.

I might be able to push stuff around to fit a 1210 or even something bigger, but not much. The entire board is 0.77 square inches and there are some big parts wasting space already.

The board will be glob-topped, so maybe the epoxy will spread the heat around some.

Experiments may be required. I could glob-top a 1206 platinum RTD and let it self-heat and measure itself maybe.

This is nice epoxy:

formatting link

PANACOL STRUCTALIT 5891 C

but the thermal conductivity is only 0.4 w/mK, still a lot better than air.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics
Reply to
John Larkin
Loading thread data ...

It even has its own Youtube show.

formatting link 

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics
Reply to
John Larkin

Yeah it's hard to know if a layer of 'insulation' would help. I was once thinking that several layers of heat shrink tubing around a through hole R's might allow for higher powers. And though the 'insulator/ goop' is better than still air, is it better than convection in air? Who knows, and may be orientation dependent.

I never tried the resistor + heat shrink tubing.

George H

Reply to
George Herold

mandag den 11. februar 2019 kl. 18.07.59 UTC+1 skrev John Larkin:

formatting link

Reply to
Lasse Langwadt Christensen

I've used them, but they are not stocked in the value I need (lots of

50 ohms available) and a 1206 would get awfully hot. They need lots of copper to cool the end caps, and I don't have space for that, nor can I afford the capacitance. 
--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com
Reply to
John Larkin

FWLIW I once used electrolytic caps to cool some small germanium transistors. Just tied them together. It worked :)

Dunno if you can get some copper powder in that epoxy - not too much of course.

NT

Reply to
tabbypurr

There are 1-W 0612 resistors. I'm using these ones:

R15-100H-00, RES,, FIXED, 100.0m, 1.0%, 1.0W, 100.0p, THICK, 200.0,, Wide 1206 hi pwr , 0612 , ERJ-B2CFR10V , PANASONIC , $0.1174 , 1000.0 , @EACH, DIGIKEY

Cheers

Phil

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
https://hobbs-eo.com
Reply to
Phil Hobbs

That only comes in 0.1R. The 0612 orientation does allow better conduction cooling, hot spot to end caps.

Susumu has a 1 watt 1206

formatting link

but it looks like they just have more pad area. I can do that with any

1206. 
--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com
Reply to
John Larkin

There are some 0612s in my resistance range, rated 3/4 watts.

formatting link 

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com
Reply to
John Larkin

Now I'm thinking that two 0805s in parallel will be a lot better than one 1206.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com
Reply to
John Larkin

Or series. Maybe you can separate them. I'm using CRCW1210-HP e3, rated 0.75W.

--
 Thanks, 
    - Win
Reply to
Winfield Hill

Good idea, series: less capacitance.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com
Reply to
John Larkin

A watt for a square inch is troublesome. Surely the board MOUNTS to something, can you position the hot thing near a (heat-conductive) structural element?

Reply to
whit3rd

It will be a baby board that surface-mounts on a bigger board.

1 watt isn't bad for the whole baby board. It is a potential hot spot on the little board. 
--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com
Reply to
John Larkin

Good tunes, too. Panacol The Disco!

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Design 
Website: https://www.seventransistorlabs.com/
Reply to
Tim Williams

Needs a bunch of copper in the middle though.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
http://hobbs-eo.com
Reply to
Phil Hobbs

How about this:

formatting link

That roughly multiplies the pad areas by 5x. I'm thinking that the glob-top epoxy will pull heat out of the resistors and the topside traces and spread it out some to the surface. Heat will also go down through the FR4 to the ground plane, layer 2.

Capacitances shouldn't be bad. L2 helps isolate things.

Using two 0805s spreads out the heat a bit too.

I still wonder what glob-top does to the hot spot on a resistor. Maybe I'll do the thinfilm RTD experiment, or get a scut bunny to do it.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics
Reply to
John Larkin

There's a lot of copper missing on layer 1 that could remain there. And on other layers IIRC.

There needs to be a tiny heatsink one can pick & place onto those sm parts.

NT

Reply to
tabbypurr

Why not use a thermally conductive layer of glob or a thermal pad spreading the heat to a larger amount of glob. You've used thermal pads before. Bu t then I guess they need to be pressed into good contact which would requir e some sort of clamp. How about a small aluminum sheet on top mounted to y our screw holes?

================= Al Sheet ============= Thermal Pad [===] [===] Resistors ================= PWB

Yeah, instead of a very, very easy glob top that doesn't conduct well, use something a little harder that conducts very well.

Or you could use a little math to calculate the thermal conductivity of the glob top. I expect this is not a case where more is more since the conduc tivity is so poor, but the math will tell all.

Rick C.

Reply to
gnuarm.deletethisbit

I don't think I'd want to rely on it, because it seems likely to delaminate at some stage on account of the thermal shear stress.

The alumina substrate of the resistor is about 35 W/m/K, so going to

0.85 for lateral heat conduction on the top probably won't change the temperature much unless there's some aluminum or something on the other side.

Maybe you could extend the pads towards the middle of the resistor and use the epoxy as underfill.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
http://hobbs-eo.com
Reply to
Phil Hobbs

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.