How to protect circuit boards that must be exposed to weather?

[John Robertson]

I have another problem - display panels on an outdoor hammer game - the
style that you swing a hammer at a pedestal and it lights up a column of
LED panels with 8x8 graphics.

The one I am repairing was made in 2001 and it is breaking down. The
surface mounted serial drivers are loosing connection to the PCBs from
being thermally stressed over a couple of decades of outdoor use and I
want to make replacement panels (the original company has abandoned the
product) that will be more durable.

Currently on the rear (somewhat protected) side of the board is the
electronics, covered (or course) with a conformal coating. The outer
side is an 8 X 8 matrix of LEDs in groups of 8 for each pixel of the 64
pixels on the panel.

There are 40 - 8 x 8.75 inch panels going up around 30 feet...

Is there a better way to solder on SMD devices to help avoid thermal
solder failure? After connection corrosion, the biggest problem is
solder failure of the SMD pads to the PCBs.

Thanks!

John :-#)#

[Don Y]

I have another problem - display panels on an outdoor hammer game - the style
that you swing a hammer at a pedestal and it lights up a column of LED panels
with 8x8 graphics.
The one I am repairing was made in 2001 and it is breaking down. The surface
mounted serial drivers are loosing connection to the PCBs from being thermally
stressed over a couple of decades of outdoor use and I want to make replacement
panels (the original company has abandoned the product) that will be more durable.
Currently on the rear (somewhat protected) side of the board is the
electronics, covered (or course) with a conformal coating. The outer side is an
8 X 8 matrix of LEDs in groups of 8 for each pixel of the 64 pixels on the panel.
There are 40 - 8 x 8.75 inch panels going up around 30 feet...
Is there a better way to solder on SMD devices to help avoid thermal solder
failure? After connection corrosion, the biggest problem is solder failure of
the SMD pads to the PCBs.

If you are "reimplementing" the design, you might opt for thru-hole
devices and leaded solder. (Or, perhaps a piggyback board that
does this and ties into the existing board)

But, 20+ years for something exposed to the out-of-doors is a good
run. When I was designing KWHr meters, we targeted a 30 year life
but the LCD displays invariably shit the bed early (because you can't
ensure the device will be installed in a "protected" space).

Is there anything wrong with just "refreshing" the product and hoping
for ANOTHER 20 years?

[Jan Panteltje]

On a sunny day (Thu, 1 Aug 2024 17:50:29 -0700) it happened John Robertson

Post by John Robertson
I have another problem - display panels on an outdoor hammer game - the
style that you swing a hammer at a pedestal and it lights up a column of
LED panels with 8x8 graphics.
The one I am repairing was made in 2001 and it is breaking down. The
surface mounted serial drivers are loosing connection to the PCBs from
being thermally stressed over a couple of decades of outdoor use and I
want to make replacement panels (the original company has abandoned the
product) that will be more durable.
Currently on the rear (somewhat protected) side of the board is the
electronics, covered (or course) with a conformal coating. The outer
side is an 8 X 8 matrix of LEDs in groups of 8 for each pixel of the 64
pixels on the panel.
There are 40 - 8 x 8.75 inch panels going up around 30 feet...
Is there a better way to solder on SMD devices to help avoid thermal
solder failure? After connection corrosion, the biggest problem is
solder failure of the SMD pads to the PCBs.

I have this Motec satellite dish positioner, bought around year 2000,
sits outside in the very cold (freezing) and very hot (glowball worming) weather,
has GHz electronics and connectors, and motor drive electronics.
Never ever failed.
I once opened it up, it was covered (the electronics) in what looks like candle wax...
I wanted to see something about the motor drive (curiosity) and pressed my scope probe point through the wax.
Did I melt it gain? Dunno, but it still works 100% every day.

[John R Walliker]

Post by Jan Panteltje
On a sunny day (Thu, 1 Aug 2024 17:50:29 -0700) it happened John Robertson

Post by John Robertson
I have another problem - display panels on an outdoor hammer game - the
style that you swing a hammer at a pedestal and it lights up a column of
LED panels with 8x8 graphics.
The one I am repairing was made in 2001 and it is breaking down. The
surface mounted serial drivers are loosing connection to the PCBs from
being thermally stressed over a couple of decades of outdoor use and I
want to make replacement panels (the original company has abandoned the
product) that will be more durable.
Currently on the rear (somewhat protected) side of the board is the
electronics, covered (or course) with a conformal coating. The outer
side is an 8 X 8 matrix of LEDs in groups of 8 for each pixel of the 64
pixels on the panel.
There are 40 - 8 x 8.75 inch panels going up around 30 feet...
Is there a better way to solder on SMD devices to help avoid thermal
solder failure? After connection corrosion, the biggest problem is
solder failure of the SMD pads to the PCBs.

I have this Motec satellite dish positioner, bought around year 2000,
sits outside in the very cold (freezing) and very hot (glowball worming) weather,
has GHz electronics and connectors, and motor drive electronics.
Never ever failed.
I once opened it up, it was covered (the electronics) in what looks like candle wax...
I wanted to see something about the motor drive (curiosity) and pressed my scope probe point through the wax.
Did I melt it gain? Dunno, but it still works 100% every day.

It might be worth using 3528 automotive LEDs. These have a leadframe
which gives some mechanical compliance. Also, make the SM pads larger
than usual to get more solder into the joint.

John

[Buzz McCool]

Post by John Robertson
Is there a better way to solder on SMD devices to help avoid thermal
solder failure?

Underfill

[Glen Walpert]

Post by John Robertson
I have another problem - display panels on an outdoor hammer game - the
style that you swing a hammer at a pedestal and it lights up a column of
LED panels with 8x8 graphics.
The one I am repairing was made in 2001 and it is breaking down. The
surface mounted serial drivers are loosing connection to the PCBs from
being thermally stressed over a couple of decades of outdoor use and I
want to make replacement panels (the original company has abandoned the
product) that will be more durable.
Currently on the rear (somewhat protected) side of the board is the
electronics, covered (or course) with a conformal coating. The outer
side is an 8 X 8 matrix of LEDs in groups of 8 for each pixel of the 64
pixels on the panel.
There are 40 - 8 x 8.75 inch panels going up around 30 feet...
Is there a better way to solder on SMD devices to help avoid thermal
solder failure? After connection corrosion, the biggest problem is
solder failure of the SMD pads to the PCBs.
Thanks!
John :-#)#

If the column of lights is attached to the same base as the hammer target,
there could be significant solder fatigue from the vibrations traveling up
the light column.

Switching to lead-free will improve fatigue resistance. While there are
lead-free alloys specifically designed for good fatigue resistance, and
any solder mfgr app engineer could make a recommendation, the standard
SAC305 is a significant improvement over tin-lead according to numerous
test results published in the industry trade magazines like SMT
Technology.

Stiffening the boards with a thicker material or with stiffeners will
reduce thermal bowing from sunshine. Leaving all unused area covered with
copper ground plane and white or light colored soldermask might help.

Flexible mounting to isolate vibration might help, perhaps clamp between
silicone foam strips with rubber bushings around the screws or similar,
could also provide sealing of the back side.

Flexible conformal coating will prevent the coating from stressing solder,
as can happen with rigid materials due to differential thermal expansion.

As Buzz mentioned, underfill can help if your assembler has the
capability.

Regards,
Glen

[John Robertson]

Post by Glen Walpert

Post by John Robertson
I have another problem - display panels on an outdoor hammer game - the
style that you swing a hammer at a pedestal and it lights up a column of
LED panels with 8x8 graphics.
The one I am repairing was made in 2001 and it is breaking down. The
surface mounted serial drivers are loosing connection to the PCBs from
being thermally stressed over a couple of decades of outdoor use and I
want to make replacement panels (the original company has abandoned the
product) that will be more durable.
Currently on the rear (somewhat protected) side of the board is the
electronics, covered (or course) with a conformal coating. The outer
side is an 8 X 8 matrix of LEDs in groups of 8 for each pixel of the 64
pixels on the panel.
There are 40 - 8 x 8.75 inch panels going up around 30 feet...
Is there a better way to solder on SMD devices to help avoid thermal
solder failure? After connection corrosion, the biggest problem is
solder failure of the SMD pads to the PCBs.
Thanks!
John :-#)#

If the column of lights is attached to the same base as the hammer target,
there could be significant solder fatigue from the vibrations traveling up
the light column.

No, the tower is on its own, no risk of shock at least!

Post by Glen Walpert
Switching to lead-free will improve fatigue resistance. While there are
lead-free alloys specifically designed for good fatigue resistance, and
any solder mfgr app engineer could make a recommendation, the standard
SAC305 is a significant improvement over tin-lead according to numerous
test results published in the industry trade magazines like SMT
Technology.

Thanks for that information!

Post by Glen Walpert
Stiffening the boards with a thicker material or with stiffeners will
reduce thermal bowing from sunshine. Leaving all unused area covered with
copper ground plane and white or light colored soldermask might help.

Copper ground plane is planned.

Post by Glen Walpert
Flexible mounting to isolate vibration might help, perhaps clamp between
silicone foam strips with rubber bushings around the screws or similar,
could also provide sealing of the back side.

It is not so much vibration as much as thermal changes over a day in the
sun...times many days!

Post by Glen Walpert
Flexible conformal coating will prevent the coating from stressing solder,
as can happen with rigid materials due to differential thermal expansion.

Conformal coating is part of the problem for repairs. While it is
flexible it doesn't peel off easily at all - which I guess is the intent
- but this does make it harder to repair. Are there conformal coating
that respond well to a solvent that isn't toxic?

Post by Glen Walpert
As Buzz mentioned, underfill can help if your assembler has the
capability.

I will look into that too.

Post by Glen Walpert
Regards,
Glen

Thanks for taking the time!

John :-#)#

[Phil Hobbs]

Post by John Robertson

Post by Glen Walpert

Post by John Robertson
I have another problem - display panels on an outdoor hammer game - the
style that you swing a hammer at a pedestal and it lights up a column of
LED panels with 8x8 graphics.
The one I am repairing was made in 2001 and it is breaking down. The
surface mounted serial drivers are loosing connection to the PCBs from
being thermally stressed over a couple of decades of outdoor use and I
want to make replacement panels (the original company has abandoned the
product) that will be more durable.
Currently on the rear (somewhat protected) side of the board is the
electronics, covered (or course) with a conformal coating. The outer
side is an 8 X 8 matrix of LEDs in groups of 8 for each pixel of the 64
pixels on the panel.
There are 40 - 8 x 8.75 inch panels going up around 30 feet...
Is there a better way to solder on SMD devices to help avoid thermal
solder failure? After connection corrosion, the biggest problem is
solder failure of the SMD pads to the PCBs.
Thanks!
John :-#)#

If the column of lights is attached to the same base as the hammer target,
there could be significant solder fatigue from the vibrations traveling up
the light column.

No, the tower is on its own, no risk of shock at least!

Post by Glen Walpert
Switching to lead-free will improve fatigue resistance. While there are
lead-free alloys specifically designed for good fatigue resistance, and
any solder mfgr app engineer could make a recommendation, the standard
SAC305 is a significant improvement over tin-lead according to numerous
test results published in the industry trade magazines like SMT
Technology.

Thanks for that information!

Post by Glen Walpert
Stiffening the boards with a thicker material or with stiffeners will
reduce thermal bowing from sunshine. Leaving all unused area covered with
copper ground plane and white or light colored soldermask might help.

Copper ground plane is planned.

Post by Glen Walpert
Flexible mounting to isolate vibration might help, perhaps clamp between
silicone foam strips with rubber bushings around the screws or similar,
could also provide sealing of the back side.

It is not so much vibration as much as thermal changes over a day in the
sun...times many days!

Post by Glen Walpert
Flexible conformal coating will prevent the coating from stressing solder,
as can happen with rigid materials due to differential thermal expansion.

Conformal coating is part of the problem for repairs. While it is
flexible it doesn't peel off easily at all - which I guess is the intent
- but this does make it harder to repair. Are there conformal coating
that respond well to a solvent that isn't toxic?

You can solder right through acrylic conformal coating.

Cheers

Phil Hobbs

[Glen Walpert]

Post by John Robertson

Post by Glen Walpert

Post by John Robertson
I have another problem - display panels on an outdoor hammer game -
the style that you swing a hammer at a pedestal and it lights up a
column of LED panels with 8x8 graphics.
The one I am repairing was made in 2001 and it is breaking down. The
surface mounted serial drivers are loosing connection to the PCBs from
being thermally stressed over a couple of decades of outdoor use and I
want to make replacement panels (the original company has abandoned
the product) that will be more durable.
Currently on the rear (somewhat protected) side of the board is the
electronics, covered (or course) with a conformal coating. The outer
side is an 8 X 8 matrix of LEDs in groups of 8 for each pixel of the
64 pixels on the panel.
There are 40 - 8 x 8.75 inch panels going up around 30 feet...
Is there a better way to solder on SMD devices to help avoid thermal
solder failure? After connection corrosion, the biggest problem is
solder failure of the SMD pads to the PCBs.
Thanks!
John :-#)#

If the column of lights is attached to the same base as the hammer target,
there could be significant solder fatigue from the vibrations traveling
up the light column.

No, the tower is on its own, no risk of shock at least!

Post by Glen Walpert
Switching to lead-free will improve fatigue resistance. While there
are lead-free alloys specifically designed for good fatigue resistance,
and any solder mfgr app engineer could make a recommendation, the
standard SAC305 is a significant improvement over tin-lead according to
numerous test results published in the industry trade magazines like
SMT Technology.

Thanks for that information!

Post by Glen Walpert
Stiffening the boards with a thicker material or with stiffeners will
reduce thermal bowing from sunshine. Leaving all unused area covered
with copper ground plane and white or light colored soldermask might
help.

Copper ground plane is planned.

Post by Glen Walpert
Flexible mounting to isolate vibration might help, perhaps clamp
between silicone foam strips with rubber bushings around the screws or
similar, could also provide sealing of the back side.

It is not so much vibration as much as thermal changes over a day in the
sun...times many days!

Board stiffeners to reduce warping, compliant mounting to allow thermal
expansion without restraint, and cooling with a fan blowing up the column
should all help some with that.

Post by John Robertson

Post by Glen Walpert
Flexible conformal coating will prevent the coating from stressing solder,
as can happen with rigid materials due to differential thermal expansion.

Conformal coating is part of the problem for repairs. While it is
flexible it doesn't peel off easily at all - which I guess is the intent
- but this does make it harder to repair. Are there conformal coating
that respond well to a solvent that isn't toxic?

The non-toxic part is the issue there. Miller-Stephenson sells a
conformal coat stripper which will remove all coatings except epoxy which
they claim is environmentally friendly but I haven't used it or read the
MSDS:

https://miller-stephenson.com/chemicals/conformal-coating/

I have used their MS-460H - soft silicone, solder thru (messy), easily
scraped off. easy poke-through with a sharp probe, as well as the similar
Chemtronics Konform SR which strips with xylene or acetone:

https://www.chemtronics.com/conformal-coating-selection-guide

Post by John Robertson

Post by Glen Walpert
As Buzz mentioned, underfill can help if your assembler has the
capability.

I will look into that too.

Underfill is generally epoxy, glues the components to the board, great for
shock and vibration but terrible for rework.

Post by John Robertson

Post by Glen Walpert
Regards,
Glen

Thanks for taking the time!
John :-#)#

[John Robertson]

Post by Glen Walpert

Post by John Robertson

Post by Glen Walpert

Post by John Robertson
I have another problem - display panels on an outdoor hammer game -
the style that you swing a hammer at a pedestal and it lights up a
column of LED panels with 8x8 graphics.
The one I am repairing was made in 2001 and it is breaking down. The
surface mounted serial drivers are loosing connection to the PCBs from
being thermally stressed over a couple of decades of outdoor use and I
want to make replacement panels (the original company has abandoned
the product) that will be more durable.
Currently on the rear (somewhat protected) side of the board is the
electronics, covered (or course) with a conformal coating. The outer
side is an 8 X 8 matrix of LEDs in groups of 8 for each pixel of the
64 pixels on the panel.
There are 40 - 8 x 8.75 inch panels going up around 30 feet...
Is there a better way to solder on SMD devices to help avoid thermal
solder failure? After connection corrosion, the biggest problem is
solder failure of the SMD pads to the PCBs.
Thanks!
John :-#)#

If the column of lights is attached to the same base as the hammer target,
there could be significant solder fatigue from the vibrations traveling
up the light column.

No, the tower is on its own, no risk of shock at least!

Post by Glen Walpert
Switching to lead-free will improve fatigue resistance. While there
are lead-free alloys specifically designed for good fatigue resistance,
and any solder mfgr app engineer could make a recommendation, the
standard SAC305 is a significant improvement over tin-lead according to
numerous test results published in the industry trade magazines like
SMT Technology.

Thanks for that information!

Post by Glen Walpert
Stiffening the boards with a thicker material or with stiffeners will
reduce thermal bowing from sunshine. Leaving all unused area covered
with copper ground plane and white or light colored soldermask might
help.

Copper ground plane is planned.

Post by Glen Walpert
Flexible mounting to isolate vibration might help, perhaps clamp
between silicone foam strips with rubber bushings around the screws or
similar, could also provide sealing of the back side.

It is not so much vibration as much as thermal changes over a day in the
sun...times many days!

Board stiffeners to reduce warping, compliant mounting to allow thermal
expansion without restraint, and cooling with a fan blowing up the column
should all help some with that.

Drat, I didn't think of that as a problem. These panels are currently
screwed (4 screws on each of two sides) to an aluminum column...it makes
sense that the column expands with temperature changes and if the boards
are secured tightly (8 screws/panel) then this could be part of the
issue. Flexible mounting needs looking into!

Post by Glen Walpert

Post by John Robertson

Post by Glen Walpert
Flexible conformal coating will prevent the coating from stressing solder,
as can happen with rigid materials due to differential thermal expansion.

Conformal coating is part of the problem for repairs. While it is
flexible it doesn't peel off easily at all - which I guess is the intent
- but this does make it harder to repair. Are there conformal coating
that respond well to a solvent that isn't toxic?

The non-toxic part is the issue there. Miller-Stephenson sells a
conformal coat stripper which will remove all coatings except epoxy which
they claim is environmentally friendly but I haven't used it or read the
https://miller-stephenson.com/chemicals/conformal-coating/
I have used their MS-460H - soft silicone, solder thru (messy), easily
scraped off. easy poke-through with a sharp probe, as well as the similar
https://www.chemtronics.com/conformal-coating-selection-guide

I'll take a look into that.

Post by Glen Walpert

Post by John Robertson

Post by Glen Walpert
As Buzz mentioned, underfill can help if your assembler has the
capability.

I will look into that too.

Underfill is generally epoxy, glues the components to the board, great for
shock and vibration but terrible for rework.

I think with new panels that securing the chips to the board with
underfill might well be a good idea.

Post by Glen Walpert

Post by John Robertson

Post by Glen Walpert
Regards,
Glen

Thanks for taking the time!
John :-#)#

Thanks!

John :-#)#