Capacitor glue?

[Axtman]

What glue or sealant do you guys use to adhere capacitors to boards? Thanks.

[Mick Bailey]

I have some decent hot met in a small gun that I use for radial caps, but I've always wanted some of the off-white/yellow stuff that some of the Japanese manufacturers once used that really holds tight.

For axial caps ii use a non-acetic, neutral-cure silicone. Sometimes when I have it, an acrylic sealant/adhesive, though it has to be left in the warm overnight to cure.

[xtian]

None. I prefer zip ties. Far easier to service.

[J M Fahey]

I use contact cement since forever.

Not as contact cement proper but simply as a standard glue.

It holds tight to everything, including glass and polished metal, and it keeps getting stronger after time, because besides plain solvent evaporation, most include a slow curing/vulcanizing agent.

The beauty is that it keep some elasticity (silicone does the same) which makes it impact resistant, while hot glue snaps off cleanly under impact.

At least the regular hobby type.

[Mick Bailey]

I'm down to my last few sticks of rubbery hot melt that stays flexible. Agreed, regular hobby stuff breaks away easily. I think the Japanese manufacturers used a rubber-based contact adhesive.

[tedmich]

Commercially polyimide hot melts are commonly used to attach caps, like Loctite/Henkel Hysol 7802. It can handle 120C.
Kapton is a type of polyimide, but not the melty kind, as its MP>400C

I like Permatex Ultra Black*, as I usually have some left over after attaching Toyota oil pans...

* current PUB formulation is neutral cure! (thanks Pix!)

[Pixel]

There are at least two kinds of neutral cure silicone, oxime and alkoxy. You want the alkoxy kind. If you look at the package you can usually see a very long chemical formula and somewhere in there you will see something like methyloxy or ethyloxy. The methyl and ethyl refer to the type of alcohol that is released during curing. I have been using GE Silicone 2 lately, but check the package because you never know when they might change their formula.

[g1]

I think the Japanese manufacturers used a rubber-based contact adhesive.

Some of this stuff ended up getting conductive over the years. With a new product it would be difficult to real-world test for long term degradation. From the manufacturers standpoint, I guess outlasting the warranty period by some factor would be acceptable. But it can be a nightmare for long term serviceability.

[Pixel]

Some of this stuff ended up getting conductive over the years. With a new product it would be difficult to real-world test for long term degradation. From the manufacturers standpoint, I guess outlasting the warranty period by some factor would be acceptable. But it can be a nightmare for long term serviceability.

It is interesting how an insulator could become conductive. You would need some kind of charge carriers that were not present before or were not free. Perhaps something to do with capacitor electrolyte?

[Helmholtz]

It is interesting how an insulator could become conductive. You would need some kind of charge carriers that were not present before or were not free. Perhaps something to do with capacitor electrolyte?

I guess what it takes is chemical decomposition (maybe caused by oxidation, heat or electric fields) of the glue setting free ions. Add some humidity and there's conductivity.
E.g. PVC likes to release HCl (hydrochloric acid).

Rod Elliot hates the" yellow glue": https://sound-au.com/articles/yellow-glue.htm

[The Dude]

Yes, the "yellow glue" has become a problem in recent years. I've had several amps that I've had to remove that glue because it become conductive. Not too much of a problem in some amps, but others require removing components to fully get rid of it. There are some threads about it here on the forum.

[Tom Phillips]

It is interesting how an insulator could become conductive. You would need some kind of charge carriers that were not present before or were not free. Perhaps something to do with capacitor electrolyte?

I wonder if it was slightly conductive from the start?
I recently needed some non-conductive compound to "pot" the end of a tube. I did a test using JB Weld two part product. Measurement with my DVM did not detect any conductivity of the cured product. I am not suggesting using JB Weld to hold caps in place. Just a memory triggered by the thread as I am always interested in adding knowledge about adhesives.

I have some of the electronics grade silicone. I don't use a lot of it because that stuff is very expensive compared to the common silicone product.

[tedmich]

You are spot on, its usually ions which allow charges to pass, especially if water is present. In the case of the infamous Marshall conductive PCBs its just repeated thermal cycling and moisture ingress that allows their bias to drift, with catastrophic results.
HV stuff really is different and tube amps are the dinosaur backwoods of electronics design sadly!

Chloroprene/neoprene glue make sense for Rod's hated "yellow glue" as SEA has lakes of the stuff to make the petro sneakers so loved by the West...I see no reason it should breakdown under about 150C but these formulations are likely HIGHLY impure.

[Pixel]

I wonder if it was slightly conductive from the start?
I recently needed some non-conductive compound to "pot" the end of a tube. I did a test using JB Weld two part product. Measurement with my DVM did not detect any conductivity of the cured product. I am not suggesting using JB Weld to hold caps in place. Just a memory triggered by the thread as I am always interested in adding knowledge about adhesives.

I have used JB Weld to stick LED emitters to heatsinks for aquarium lighting and it has worked quite well in that application. I also have used JB Weld to repot thermistors that lost their coating and then glue the repaired thermistors to heatsinks.