Hot Glue Woes

[riz]

I've been growing fond of using hot melt glue, for sticking in extra caps--maybe I worked around a cap-can--and such. I usually clean the area of the chassis with solvent, scuff it a bit with sandpaper, and heat the area with my nifty little butane pen torch (also great for heat-shrink tubing) before sticking my part with the glue, all in an effort to get a "permanent" bond. I've had a problem with getting a good, strong bond when using it to stick these little 1:1 signal transformers--the things pop loose too easily for my comfort. Am I asking too much of the glue? Should I just stick (PUN!) to super-lightweight parts? Any tips on using the stuff? I know some of you here are fond of using hot glue while modding/repairing and I can see why--it's one of my favorite new tools! Just hoping I can get it to do what I want it to do, which is stick these parts solid enough for me to feel confident in the work. Otherwise, I'm thinking it might be better to just rivet in some little boards.

[J M Fahey]

Hot glue is a poor adhesive and has no impact resistance ; works fine as a "space filler" though.
I've used contact cement for ages with excellent results, with the only caveat that it really needs to sit overnight.
But it sticks very well to non-porous materials as aluminum, Electrolytic sleeves and the like.
Excellent impact resistance because of its built-in elasticity.
Silicone sealant is even better, but needs at least 24H drying and is *very* tough to remove for servicing.

[riz]

Thanks for the reply--I had honestly put the alternatives out of mind! I like the immediacy of the hot glue, not to mention every time I open a tube of silicone, the whole thing seems to go bad before I use much of it. But I agree silicone would be better suited to the job on my bench. It's hard, if not impossible, to stick things to a thin metal chassis without a very flexible and aggressive adhesive like silicone. Metal just expands/contracts at a different rate than most other stuff. O.k., off to dig up a tube of the nasty stuff...

[Gtr_tech]

Silicone that does not emit acetic acid when curing works well. If it smells like vinegar, its not a great idea to use it as it will continue to emit small amts of acetic acid even when fully cured. That is corrosive....
AnaLog Services RTV Silicones

Electronic grade silicone is what to look for.

[riz]

Good to know, thanks... for next time! I knew about possible conductivity issues, but never thought about issues with acetic acid. Don't tell anyone, but I just stuck a part in with black vinegar RTV--all I had on hand! I think it will survive...

[hrbay]

You are performing what is called staking...if you don't do a conformal coating, expensive, you stake the circuit board components before putting them thru shaker table abuses. The key is to degrease the surfaces as best as you can and use as little adhesive as possible. We use commercial epoxies and electronic grade non acetic RTV's. The idea is to minimize the motion of the components that are higher above the board, as they are the first to move when loaded.

[riz]

hrbay, you don't have problems adhering to metal chassis with epoxy? My experience (not in amps, with other stuff) is that the epoxy-metal bond eventually fails, even if prepped properly. Don't get me wrong, I'm a big fan of West-system epoxy... The silicone is working very well, btw, and has solved my problem--why I started this thread--for now. I just wanted the hot glue to be the solution!

[jrfrond]

Hot glue doesn't do well in extremes. Too hot (over 90C) and it will flow. Too cold, and it will break, which is why most techs use a blast of freeze-spray to remove it. However, for most applications, it seems to work just fine, and is used by many manufacturers to lock down components, connectors, etc.

If you are not going to use it on leads, acetic-based RTV works fine, but DOES have a 24hr. cure. Otherwise, you really needs to use electronic-grade RTV. As stated, it is a pain to remove cleanly.

Another thing that works well is rubberized CA adhesives. Straight CA is too brittle and has low shear-strength, so rubber is added to the mix. This is the type of CA used a lot in speaker reconing, and they come in black and clear. The black has carbon black added for cosmetic purposes, and the cured result is non-conductive.

This is a bit of a hot topic, as there is nothing that is perfect in this application. Those of us who've worked on vintage Asian electronics, both pro and consumer, will remember staking components that carbonized with age and wreaked major havoc with electronics. Of course, the ubiquitous Yamaha SPX90 power supply comes to mind, but there were many others.

[riz]

Hmmm, rubberized CA... I know the stuff, and I haven't been all that impressed with it. It's another one of those products that seem to kick off in the bottle! What I have found it useful for is repairing cracks in Brazilian rosewood, where it's color match is almost perfect for the spider-webbing of the grain, hence the repair looks good. I use lots of thin CA, but have never found much use for the rubberized as an adhesive, although I still seem to try about once a year... I had never considered it for staking. I guess the hot glue goes back to the Halloween costume department!

[jrfrond]

Hmmm, rubberized CA... I know the stuff, and I haven't been all that impressed with it. It's another one of those products that seem to kick off in the bottle! What I have found it useful for is repairing cracks in Brazilian rosewood, where it's color match is almost perfect for the spider-webbing of the grain, hence the repair looks good. I use lots of thin CA, but have never found much use for the rubberized as an adhesive, although I still seem to try about once a year... I had never considered it for staking. I guess the hot glue goes back to the Halloween costume department!

You need to keep CA in the fridge if you want it to last, rubberized or not. Yeah, the black CA is even great for inlays on ebony or dark rosewood, but you can also dye clear CA with Fresco powders for inlay filler, etc.

[J M Fahey]

The problem with Epoxy is that it's weak if you *flex* the metal part, that's why most normal chassis are by definition poor partners.
I have epoxied *very well* thermal sensing diodes or transistors to backpanels used as heatsinks (2mm aluminum) or straight to finned heatsinks, typically from 3 to 9 mm, not much flexing possible there.
I have used them for ages (as well as anybody else) to glue ceramic magnets (*impossible* to flex) to iron disk speaker plates, typically from 6mm to 12 mm; almost as rigid.
Or in a few words: epoxy works well if you avoid parts flexing; traction/shear/torsion efforts are well tolerated.
I have glued to chassis big filter capacitors with excellent results, and even *power transformers*; in this last case as anti-vibration backup to bolts, with good, thick contact cement.
It's the one that shoe makers use to glue soles, it's the best by far, very hard to remove.
I have pulled all bolts from a transformer and found it impossible to pull it by hand, had to use a large screwdriver as a prying lever.
I started using this after repairing a few amps with vibration-loosened bolts, and hanging_from_the_wires transformers smashing whatever they could reach.

[imaradiostar]

We often use a 3m "low melt" adhesive at work. It melts at lower (safer) temperatures and seem to stick better than regular hot melt glue. I guess that probably has less to do with being low melt and more to do with being a quality 3m product rather than something from Wal-Mart.

jamie