Potting a large area PCB

[qwaarjet]

I'm looking to pot are large PCB ~1 square ft. I've potted small boards before but never anything this big. The PCB has a mix of smd and through hole parts. I'm concerned about mechanical stress of the curing potting over such a large area. Anyone ever potted something that big, tips/ tricks, things to watch out for.

[JJalling]

I think that Mike (mikeselectricstuff) has some experience in this area - maybe he has some tips.

BR Jonas

[T3sl4co1l]

Use a rubbery or gooey mix.

If you must use rigid fill, coat the board with a thick layer of gooey/rubbery conformal.

Redesigning the board to use smaller PCBs, joined together with flexible connections, may become an expensive necessity... this should've been foreseen in the design phase... 

Tim

[RobK_NL]

Potting with what and for what purpose? There's a lot of different stuff around.

Epoxy is out of the question, as it set hard. Polyurethane could be an option as it is much softer. Then there's silicone (the non-corrosive stuff!!) and several different types of gel.

[uncle_bob]

Hi

What sort of environment is this going to "live" in after it is potted? (temperature, humidity, shock, vibration, salt spray, solvents ...)

What is the objective of the potting? (Vibration, shock, high voltage, salt spray ....)

How long does it have to survive (what is the expected life of the product)?

What connects to the board and how?

1 square foot is a *lot* of parts. Do you have anything in the mix that is not "potting friendly"? If so, is there provision for covering that part of the board before potting the beast?

How is the assembly going to be supported? You could be up around 1/4 cubic foot of potting ... that's heavy.

Why not put something this big in an enclosure? That's how it's done in  99% of the cases I've seen on stuff this size.

Lots of questions.

Bob

[qwaarjet]

I want to pot a the innards of large lcd tv for use in extreme vibration environment I used an unmodified consumer unit and it lasted a few months. (even with isolation mounts)  This is mostly an experiment since even the best industrial units I could find most likely would fail. I had some other ideas but they all start getting fairly complex the potting idea was the least crazy.
 I was thinking using a Shore A 60 urethane since it relatively cheap.

[uncle_bob]

I want to pot a the innards of large lcd tv for use in extreme vibration environment I used an unmodified consumer unit and it lasted a few months. (even with isolation mounts)  This is mostly an experiment since even the best industrial units I could find most likely would fail. I had some other ideas but they all start getting fairly complex the potting idea was the least crazy.
 I was thinking using a Shore A 60 urethane since it relatively cheap.

Hi

Well, you *do not* want to use anything that is not specifically designed for potting. There are all sorts of exciting corrosion / electro migration / creeping crud issues. My guess (based on the insides of most TV's) is that you will not gain much in reliability. The connectors they use aren't up to the task. That's true before or after you pot them ...

Bob

[T3sl4co1l]

I'd be very concerned about anything that has to dissipate power, which is most TVs!

It could be reworked to handle the power (say by replacing heat sinks with heat spreaders and goop or thermal pads, and ultimately bringing that heat to an outer surface), but you're probably looking at a lot of work...

Tim

[acolomitchi]

I want to pot a the innards of large lcd tv for use in extreme vibration environment I used an unmodified consumer unit and it lasted a few months. (even with isolation mounts)  This is mostly an experiment since even the best industrial units I could find most likely would fail. I had some other ideas but they all start getting fairly complex the potting idea was the least crazy.
 I was thinking using a Shore A 60 urethane since it relatively cheap.

So, the reason is mechanical primarily.

Shoot this idea if it's stupid, but... why do you need full potting or even conformal coating when gluing the components should be enough? Full potting will increase the mass considerably, I imagine.
A drop of very adhesive resin over each component or only the connections... wouldn't this be enough?

I've done it once using acrylic stuff - ebay, acrylic nails monomer and powder kit (in my case, for chemical protection mainly but it should withstand moderate abrasions - connection to probes to be sunk in soil)

[batteksystem]

I want to pot a the innards of large lcd tv for use in extreme vibration environment I used an unmodified consumer unit and it lasted a few months. (even with isolation mounts)  This is mostly an experiment since even the best industrial units I could find most likely would fail. I had some other ideas but they all start getting fairly complex the potting idea was the least crazy.
 I was thinking using a Shore A 60 urethane since it relatively cheap.

So, the reason is mechanical primarily.

Shoot this idea if it's stupid, but... why do you need full potting or even conformal coating when gluing the components should be enough? Full potting will increase the mass considerably, I imagine.
A drop of very adhesive resin over each component or only the connections... wouldn't this be enough?

I've done it once using acrylic stuff - ebay, acrylic nails monomer and powder kit (in my case, for chemical protection mainly but it should withstand moderate abrasions - connection to probes to be sunk in soil)

A more plausible solution would be conformal coating. Use hard potting for LCD TV which is thin and wide is basically asking for trouble.

[uncle_bob]

Hi

Why does a TV fail in vibration?

The cute little parts on the pc board(s) don't care much about vibration. You can put a driver chip into 100 G's RMS random for 12 hours and (by it's self) it could care less abut what's going on. The SMD stuff isn't going to die in vibration. It could *crack* from board flexing, but that's a different sort of issue. Leaded stuff, unless it's big and floppy ... still not likely to be an issue.

What does fail:

1) The wires between this and that.
2) The wires to the connectors
3) The connections in the LCD panel its self. (same thing, just burried)
4) Mounting brackets for sub-assemblies
5) The mounts for the LCD panel.
6) Big heavy floppy parts (aluminum electrolytics)  on the boards.
7) Giant boards that go into resonance and crack off of their mounts.
Ceramic capacitors that crack from board flexing (resistors sometimes, but a lot less often)

Some of that you can fix with RTV gobbed here and there. You still are unlikely to have a reliable device.

Bob

[MosherIV]

I want to pot a the innards of large lcd tv for use in extreme vibration environment I used an unmodified consumer unit and it lasted a few months. (even with isolation mounts)  This is mostly an experiment since even the best industrial units I could find most likely would fail. I had some other ideas but they all start getting fairly complex the potting idea was the least crazy.

For 'extreme vibration environments' you should be looking at aeronautical or military grade equipment.
Yes they are very expensive. 
That is because of all the engineering (both design and lots of testing), sadly the audit trail also adds significant cost that is not necessary for you

Potting may help with some vibration but will add problems with heat dissipation 

[qwaarjet]

I want to pot a the innards of large lcd tv for use in extreme vibration environment I used an unmodified consumer unit and it lasted a few months. (even with isolation mounts)  This is mostly an experiment since even the best industrial units I could find most likely would fail. I had some other ideas but they all start getting fairly complex the potting idea was the least crazy.

For 'extreme vibration environments' you should be looking at aeronautical or military grade equipment.
Yes they are very expensive. 
That is because of all the engineering (both design and lots of testing), sadly the audit trail also adds significant cost that is not necessary for you

Potting may help with some vibration but will add problems with heat dissipation 

yea the closest thing I could find were displayed made for inside army tanks but they were like 16K a pop

[acolomitchi]

yea the closest thing I could find were displayed made for inside army tanks but they were like 16K a pop

At this price, I'd get very angry would someone drive a RPG through it.

[Wirehead]

Any different way of mounting the lcd, so the mount absorbs the shocks?

[elecman14]

A more plausible solution would be conformal coating. Use hard potting for LCD TV which is thin and wide is basically asking for trouble.

I would second the advice for conformal coating. Getting potting to work on a board that large seems like a nightmare. You may still want to glue the large components as well if you conformal coat.

[CatalinaWOW]

Any different way of mounting the lcd, so the mount absorbs the shocks?

I second this suggestion.  If you can identify dominant frequencies in your vibration environment a tuned suspension system will probably help far more than the potting.  If not, a simple mechanical low pass with damping is worth a try.

If you don't have room for this (don't give up too easily, the simplest systems are not much more than a few cubic inches to support a TV size object) I would look into the past failures and see if there is any weak point that you can locally armor with a small amount of potting or other support.

[MosherIV]

Conformal coating and potting is just trying to work around the problem.

Fundamentally, the display has not been design to handle vibration.
There are many elements that may fail:
Wires may break due to metal fateage
Large mass parts might brake free.
Screws may shake loose.

You probably want to try a combination of things:
Multiple damping stages
Thick rubbery conformal coating
Screw lock all screws.
Look carefully at large masses (large components, heat sinks etc) how can they held in place and damped

[ConKbot]

Yeah, wiring flopping around, flappy pcbs that resonate, and flappy components on the flappy pcb are your main concern. Use neutral cure silicone (ge silicone II, Not silicone I) or other electronics silicone (sensor safe RTV) to secure the suspect things.

Keep wires from flexing where they go into connectors with a small bead down the back of the connector. Also put a blob under the pcb to keep it from resonating. If the gap is big, mix silicone and corn starch to make a putty. The moisture in the starch makes the silicone cure rapidly, and it won't be as runny.   Stake wiring down to secure it to the frame, and even stake and dampen any metalwork.  Finally make sure to stake any electrolytic capacitors, and 2 or 3 lead through hole components that can bend at all.

Between damping any resonance in the pcb, securing vulnerable components,  and improving the shock mounting,  the high frequency vibrations that would break solder joints should be reduced greatly.