Solder mask and heat?

M

mkr5000

Guest
This is going to sound like a silly question but it's the perfectionist in me.
I did a pcb with a rather large copper zone area (like a ground plane but not grounded), as a base for a couple of those adhesive backed led strips that I mount on it. Had the boards done and there is solder mask covering those areas, which can be easily fixed but it may not matter at all. Just did the areas for any (tidbit) of added heat dissipation but probably best to let it go to straight plated copper area than a layer of solder mask in between?
Having solder mask looks nicer though.
Does the mask transmit heat pretty well or do you think I should get rid of it ?
 
On Thursday, June 13, 2019 at 11:37:50 AM UTC-4, mkr5000 wrote:
This is going to sound like a silly question but it's the perfectionist in me.
I did a pcb with a rather large copper zone area (like a ground plane but not grounded), as a base for a couple of those adhesive backed led strips that I mount on it. Had the boards done and there is solder mask covering those areas, which can be easily fixed but it may not matter at all. Just did the areas for any (tidbit) of added heat dissipation but probably best to let it go to straight plated copper area than a layer of solder mask in between?
Having solder mask looks nicer though.
Does the mask transmit heat pretty well or do you think I should get rid of it ?

There are at least two effects to consider. One is that the solder mask layer acts as an insulator. Since it is pretty thin this should not be a large effect. The other is the emissivity of the solder mask vs. the copper/tinning/gold surface layer you would otherwise have. I would expect the solder mask to have better emissivity. Which is greater? Beats me, but I'm betting on the emissivity being a larger factor.

--

Rick C.

- Get 1,000 miles of free Supercharging
- Tesla referral code - https://ts.la/richard11209
 
"Tim Williams" <tiwill@seventransistorlabs.com> wrote in
news:qdtvfr$evm$1@dont-email.me:

Easy calculation. Look up the product, find its thermal
conductivity, and find the typical application thickness (LPI
soldermask, typically around a mil). Plug the numbers into Google
Calculator to do whatever unit conversion for you if need be. And
there you go.

I think you will find the tape is dominant, unless it's
specifically a thermally conductive type. But that the air is
even more dominant still (typically ~150 in^2 °C/W), so it really
doesn't matter.

Tim

I would make a small, copper strip stand off and solder that onto
the PCB pad, thereby placing an air gap between the heat sources
and the PCB and that air (and thicker Copper) would definitely
dissipate better than putting them onto the PCB. You could even
make smaller ones that only carry one or two of the LEDs each and
they would then be separate little stands and be better
mechanically.

Any thick Copper foil or hobby shop sheet material. It would only
add a few mm to the profile height. The air gap is the primary
cooling gain. Otherwise the PCB should be fine.

IOW, if heat is a problem, he wants a method such as that which I
describe. If he is simply trying to max out thermal transfer when
the devices never will exhibit much to speak of anyway, the PCB is
fine mask or not.

Find some old thrown out motherboards with Copper heat sinks still
on them. Those thin fins are great for repurposing for protos.

Machined Copper is expensive because it gums up the tooling even
more than Aluminum does.
 
On Thursday, June 13, 2019 at 1:06:21 PM UTC-4, Tim Williams wrote:
"Rick C" <gnuarm.deletethisbit@gmail.com> wrote in message
news:46cbf5c1-e880-4a85-9278-6d3796bcb57e@googlegroups.com...
There are at least two effects to consider. One is that the solder mask
layer acts as an insulator. Since it is pretty thin this should not be a
large effect. The other is the emissivity of the solder mask vs. the
copper/tinning/gold surface layer you would otherwise have. I would expect
the solder mask to have better emissivity. Which is greater? Beats me,
but I'm betting on the emissivity being a larger factor.


For moderately high temperatures in still air, it's a measurable difference
(if not a significant one). 2-10% range.

I would also be more concerned about the bond strength, as cheap proto
glossy mask tends to be not as strong as full spec usually-matte mask is.

Conversely, how does that compare with the strength of the adhesive? Sticky
tape on shiny metal tends to not be very good. Sticky tape on plastic tends
to be better.

I don't think I've ever had even VHB tape pull mask off a proto board, so it
should be good enough either way.

I didn't read the OP well enough and didn't realize this was a contact mount through the solder mask. So emissivity is pretty irrelevant and given the other layers involved, I think the solder resist thermal resistance will be in the noise given the thickness and the area involved.

--

Rick C.

+ Get 1,000 miles of free Supercharging
+ Tesla referral code - https://ts.la/richard11209
 
"Rick C" <gnuarm.deletethisbit@gmail.com> wrote in message
news:46cbf5c1-e880-4a85-9278-6d3796bcb57e@googlegroups.com...
There are at least two effects to consider. One is that the solder mask
layer acts as an insulator. Since it is pretty thin this should not be a
large effect. The other is the emissivity of the solder mask vs. the
copper/tinning/gold surface layer you would otherwise have. I would expect
the solder mask to have better emissivity. Which is greater? Beats me,
but I'm betting on the emissivity being a larger factor.

For moderately high temperatures in still air, it's a measurable difference
(if not a significant one). 2-10% range.

I would also be more concerned about the bond strength, as cheap proto
glossy mask tends to be not as strong as full spec usually-matte mask is.

Conversely, how does that compare with the strength of the adhesive? Sticky
tape on shiny metal tends to not be very good. Sticky tape on plastic tends
to be better.

I don't think I've ever had even VHB tape pull mask off a proto board, so it
should be good enough either way.

Tim

--
Seven Transistor Labs, LLC
Electrical Engineering Consultation and Design
Website: https://www.seventransistorlabs.com/
 
Easy calculation. Look up the product, find its thermal conductivity, and
find the typical application thickness (LPI soldermask, typically around a
mil). Plug the numbers into Google Calculator to do whatever unit
conversion for you if need be. And there you go.

I think you will find the tape is dominant, unless it's specifically a
thermally conductive type. But that the air is even more dominant still
(typically ~150 in^2 °C/W), so it really doesn't matter.

Tim

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

"mkr5000" <mikerbgr@gmail.com> wrote in message
news:e5cb3648-ba71-488a-bf95-94f1c6886240@googlegroups.com...
This is going to sound like a silly question but it's the perfectionist in
me.
I did a pcb with a rather large copper zone area (like a ground plane but
not grounded), as a base for a couple of those adhesive backed led strips
that I mount on it. Had the boards done and there is solder mask covering
those areas, which can be easily fixed but it may not matter at all. Just
did the areas for any (tidbit) of added heat dissipation but probably best
to let it go to straight plated copper area than a layer of solder mask in
between?
Having solder mask looks nicer though.
Does the mask transmit heat pretty well or do you think I should get rid of
it ?
 
"Tim Williams" <tiwill@seventransistorlabs.com> wrote in
news:qdtvq8$gva$1@dont-email.me:

"Rick C" <gnuarm.deletethisbit@gmail.com> wrote in message
news:46cbf5c1-e880-4a85-9278-6d3796bcb57e@googlegroups.com...
There are at least two effects to consider. One is that the
solder mask layer acts as an insulator. Since it is pretty thin
this should not be a large effect. The other is the emissivity of
the solder mask vs. the copper/tinning/gold surface layer you
would otherwise have. I would expect the solder mask to have
better emissivity. Which is greater? Beats me, but I'm betting
on the emissivity being a larger factor.


For moderately high temperatures in still air, it's a measurable
difference (if not a significant one). 2-10% range.

I would also be more concerned about the bond strength, as cheap
proto glossy mask tends to be not as strong as full spec
usually-matte mask is.

Conversely, how does that compare with the strength of the
adhesive? Sticky tape on shiny metal tends to not be very good.
Sticky tape on plastic tends to be better.

I don't think I've ever had even VHB tape pull mask off a proto
board, so it should be good enough either way.

Tim

He said "adhesive backed LED strips" So, if the LEDs exhibit heat
and the designers of the strip knew that, the adhesive should be
thin and one of the 'thermal' types. VHB sticks GREAT, but is
usually pretty thick stuff, and I don't think it moves heat to well.

If he truly expects a heat issue, I would detach each LED from the
strip and use thermal epoxy to attach them to a sinking mass.
 
as I expected, 6 to 1 half dozen to the other. the mask looks better so I'll just leave it. and may be a better bonding surface.

fact is, these strips do fine 24/7 without adhering to anything.
 
DecadentLinuxUserNumeroUno@decadence.org wrote in news:qdu1j8$1lig$1
@gioia.aioe.org:

Any thick Copper foil or hobby shop sheet material. It would only
add a few mm to the profile height. The air gap is the primary
cooling gain. Otherwise the PCB should be fine.

BTW, the 'air gap' I speak of is not in any way the interface between
the LED and the sink. It is the elevated sink that allows air to pass
under it I was referring to.

Heck, he could attach them to a Copper heat tube (readily available),
and let that pull heat off the set.

Heat tubes are cool <sic>. I designed a charging device using a
peltier array core and the heat tubes made the difference (because
there needs to be a difference). It was one of my first 3D CAD
layouts. Now I are a ME too. :) (me and my CAD package)(wow).
 
On Thu, 13 Jun 2019 08:37:43 -0700 (PDT), mkr5000 <mikerbgr@gmail.com>
wrote:

This is going to sound like a silly question but it's the perfectionist in me.
I did a pcb with a rather large copper zone area (like a ground plane but not grounded), as a base for a couple of those adhesive backed led strips that I mount on it. Had the boards done and there is solder mask covering those areas, which can be easily fixed but it may not matter at all. Just did the areas for any (tidbit) of added heat dissipation but probably best to let it go to straight plated copper area than a layer of solder mask in between?
Having solder mask looks nicer though.
Does the mask transmit heat pretty well or do you think I should get rid of it ?

Unless the copper area is heroically cooled (a zillion vias to a heat
sink on the back side, or to an internal plane) the solder mask won't
increase thermal resistance much.

Solder mask over solder coated copper tends to bunch up and crack.
Solder mask over ENIG is fine.

But a solid rectangle of ENIG, gold, looks pretty good too.




--

John Larkin Highland Technology, Inc

lunatic fringe electronics
 
On 6/14/19 12:25 AM, John Larkin wrote:
On Thu, 13 Jun 2019 08:37:43 -0700 (PDT), mkr5000 <mikerbgr@gmail.com
wrote:

This is going to sound like a silly question but it's the perfectionist in me.
I did a pcb with a rather large copper zone area (like a ground plane but not grounded), as a base for a couple of those adhesive backed led strips that I mount on it. Had the boards done and there is solder mask covering those areas, which can be easily fixed but it may not matter at all. Just did the areas for any (tidbit) of added heat dissipation but probably best to let it go to straight plated copper area than a layer of solder mask in between?
Having solder mask looks nicer though.
Does the mask transmit heat pretty well or do you think I should get rid of it ?

Unless the copper area is heroically cooled (a zillion vias to a heat
sink on the back side, or to an internal plane) the solder mask won't
increase thermal resistance much.

Solder mask over solder coated copper tends to bunch up and crack.
Solder mask over ENIG is fine.

But a solid rectangle of ENIG, gold, looks pretty good too.

We've taken to using bare ENIG for our logo and board title. Bling bling!

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
 
fredag den 14. juni 2019 kl. 06.25.56 UTC+2 skrev John Larkin:
On Thu, 13 Jun 2019 08:37:43 -0700 (PDT), mkr5000 <mikerbgr@gmail.com
wrote:

This is going to sound like a silly question but it's the perfectionist in me.
I did a pcb with a rather large copper zone area (like a ground plane but not grounded), as a base for a couple of those adhesive backed led strips that I mount on it. Had the boards done and there is solder mask covering those areas, which can be easily fixed but it may not matter at all. Just did the areas for any (tidbit) of added heat dissipation but probably best to let it go to straight plated copper area than a layer of solder mask in between?
Having solder mask looks nicer though.
Does the mask transmit heat pretty well or do you think I should get rid of it ?

Unless the copper area is heroically cooled (a zillion vias to a heat
sink on the back side, or to an internal plane) the solder mask won't
increase thermal resistance much.

Solder mask over solder coated copper tends to bunch up and crack.
Solder mask over ENIG is fine.

ENIG is added after soldermask, so there is no gold under the soldermask
 
fredag den 14. juni 2019 kl. 19.19.07 UTC+2 skrev Phil Hobbs:
On 6/14/19 12:25 AM, John Larkin wrote:
On Thu, 13 Jun 2019 08:37:43 -0700 (PDT), mkr5000 <mikerbgr@gmail.com
wrote:

This is going to sound like a silly question but it's the perfectionist in me.
I did a pcb with a rather large copper zone area (like a ground plane but not grounded), as a base for a couple of those adhesive backed led strips that I mount on it. Had the boards done and there is solder mask covering those areas, which can be easily fixed but it may not matter at all. Just did the areas for any (tidbit) of added heat dissipation but probably best to let it go to straight plated copper area than a layer of solder mask in between?
Having solder mask looks nicer though.
Does the mask transmit heat pretty well or do you think I should get rid of it ?

Unless the copper area is heroically cooled (a zillion vias to a heat
sink on the back side, or to an internal plane) the solder mask won't
increase thermal resistance much.

Solder mask over solder coated copper tends to bunch up and crack.
Solder mask over ENIG is fine.

But a solid rectangle of ENIG, gold, looks pretty good too.





We've taken to using bare ENIG for our logo and board title. Bling bling!

https://hackaday.com/wp-content/uploads/2017/03/benchoffnickel.jpg

5 colors; gold, bare fr4, soldermask over no copper, soldermask over copper,
and white silk screen
 
On Fri, 14 Jun 2019 11:11:53 -0700 (PDT), Lasse Langwadt Christensen
<langwadt@fonz.dk> wrote:

fredag den 14. juni 2019 kl. 06.25.56 UTC+2 skrev John Larkin:
On Thu, 13 Jun 2019 08:37:43 -0700 (PDT), mkr5000 <mikerbgr@gmail.com
wrote:

This is going to sound like a silly question but it's the perfectionist in me.
I did a pcb with a rather large copper zone area (like a ground plane but not grounded), as a base for a couple of those adhesive backed led strips that I mount on it. Had the boards done and there is solder mask covering those areas, which can be easily fixed but it may not matter at all. Just did the areas for any (tidbit) of added heat dissipation but probably best to let it go to straight plated copper area than a layer of solder mask in between?
Having solder mask looks nicer though.
Does the mask transmit heat pretty well or do you think I should get rid of it ?

Unless the copper area is heroically cooled (a zillion vias to a heat
sink on the back side, or to an internal plane) the solder mask won't
increase thermal resistance much.

Solder mask over solder coated copper tends to bunch up and crack.
Solder mask over ENIG is fine.

ENIG is added after soldermask, so there is no gold under the soldermask

I guess not. When I scrape a trace to make contact, I do see copper.
Well, the copper won't bunch up either.


--

John Larkin Highland Technology, Inc

lunatic fringe electronics
 
On 6/14/19 2:10 PM, Lasse Langwadt Christensen wrote:
fredag den 14. juni 2019 kl. 19.19.07 UTC+2 skrev Phil Hobbs:
On 6/14/19 12:25 AM, John Larkin wrote:
On Thu, 13 Jun 2019 08:37:43 -0700 (PDT), mkr5000 <mikerbgr@gmail.com
wrote:

This is going to sound like a silly question but it's the perfectionist in me.
I did a pcb with a rather large copper zone area (like a ground plane but not grounded), as a base for a couple of those adhesive backed led strips that I mount on it. Had the boards done and there is solder mask covering those areas, which can be easily fixed but it may not matter at all. Just did the areas for any (tidbit) of added heat dissipation but probably best to let it go to straight plated copper area than a layer of solder mask in between?
Having solder mask looks nicer though.
Does the mask transmit heat pretty well or do you think I should get rid of it ?

Unless the copper area is heroically cooled (a zillion vias to a heat
sink on the back side, or to an internal plane) the solder mask won't
increase thermal resistance much.

Solder mask over solder coated copper tends to bunch up and crack.
Solder mask over ENIG is fine.

But a solid rectangle of ENIG, gold, looks pretty good too.





We've taken to using bare ENIG for our logo and board title. Bling bling!


https://hackaday.com/wp-content/uploads/2017/03/benchoffnickel.jpg

5 colors; gold, bare fr4, soldermask over no copper, soldermask over copper,
and white silk screen

And the result is waaaayyy ugly. ;)

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
 
lørdag den 15. juni 2019 kl. 08.28.04 UTC+2 skrev Phil Hobbs:
On 6/14/19 2:10 PM, Lasse Langwadt Christensen wrote:
fredag den 14. juni 2019 kl. 19.19.07 UTC+2 skrev Phil Hobbs:
On 6/14/19 12:25 AM, John Larkin wrote:
On Thu, 13 Jun 2019 08:37:43 -0700 (PDT), mkr5000 <mikerbgr@gmail.com
wrote:

This is going to sound like a silly question but it's the perfectionist in me.
I did a pcb with a rather large copper zone area (like a ground plane but not grounded), as a base for a couple of those adhesive backed led strips that I mount on it. Had the boards done and there is solder mask covering those areas, which can be easily fixed but it may not matter at all. Just did the areas for any (tidbit) of added heat dissipation but probably best to let it go to straight plated copper area than a layer of solder mask in between?
Having solder mask looks nicer though.
Does the mask transmit heat pretty well or do you think I should get rid of it ?

Unless the copper area is heroically cooled (a zillion vias to a heat
sink on the back side, or to an internal plane) the solder mask won't
increase thermal resistance much.

Solder mask over solder coated copper tends to bunch up and crack.
Solder mask over ENIG is fine.

But a solid rectangle of ENIG, gold, looks pretty good too.





We've taken to using bare ENIG for our logo and board title. Bling bling!


https://hackaday.com/wp-content/uploads/2017/03/benchoffnickel.jpg

5 colors; gold, bare fr4, soldermask over no copper, soldermask over copper,
and white silk screen

And the result is waaaayyy ugly. ;)

the result or the subject? ;)
 

Welcome to EDABoard.com

Sponsor

Back
Top