Any way to get the sand out of a transformer?

Thread Starter

strantor

Joined Oct 3, 2010
6,782
I've got a big 150lb (68kg) 6kVA encapsulated transformer and I want to get at the windings. Looks to be potted with epoxied sand. Google indicates these type transformers are encapsulated in "silica sand and resin mixture":



Has anyone successfully removed this crap from a transformer? if so, how?

I took a hammer and chisel to it briefly and made decent progress. I think with a pneumatic chisel tool I might be able to free the innards in just a couple of hours, but not without ruining everything. I wonder if heat from a torch might help. Actually I'm hoping there's some easily accessible chemical that will soften whatever "resin" might be in there (without eating the enamel off the copper). Some sort of paint/lacquer thinner maybe? acetone?
 

SLK001

Joined Nov 29, 2011
1,549
Heat will most likely soften the resin (but use a heat gun, not a torch). Probably any chemical that you use that will soften/dissolve the resin will also degrade the coating on the copper.
 

jpanhalt

Joined Jan 18, 2008
11,087
There are so-called epoxy solvents. Here's one: http://www.dynaloy.com/products/epoxy

In reality, removing epoxy means denaturing it as one does with an enamel, rather than dissolving it as one might do with a lacquer coating. The key ingredient in many of the mild denaturants for epoxy is NMP (1-methyl-2-pyrrolidinone), which is also in many paint removers that are not based on methylene chloride (CH2Cl2). Depending on the insulation used for the winding, NMP may be safe, but it will be a very laborious process. It is best to remove as much of the potting material mechanically before attacking it with such solvents.

John
 

Thread Starter

strantor

Joined Oct 3, 2010
6,782
Ok I have tried nearly every chemical in my garage from solvent stripper to pvc primer. I tried heat gun and blow torch. I even tried fire and chemicals at the the same time:

20160115_181807.jpg

Here are my findings so far...

Acetone works beautifully, for the first 2 inches. It turns the stuff into sand sludge which can be scooped out by hand. But below the 2" layer, it does absolutely nothing. Here's a pic where you can clearly see the layer :

20160115_183650.jpg

I don't know what the deal is. Maybe the liquid epoxy settled at the bottom and the stuff above was just loosely bonded. Maybe instead of dipping the xfmr in lacquer before potting in sand, they poured lacquer into the sand. I suspect the latter.

The stuff on the bottom is a giant PITA. No chemical I have tried does anything. I'm going to get some lacquer thinner tomorrow and try that. Heat works well, but is kinda haphazard. it's become more of an issue now that I'm down to the actual xfmr. I've found that the blow torch is superior to the heat gun because its wrath is more easily focused on the target area. It also heats the sand faster, meaning that I can quickly fry a layer and scrape it off without appreciably heating whatever is underneath.

In 2hrs I got from this:

20160115_172147.jpg

...to this:

20160115_192138.jpg

I figure I'm about 25% done. Labor intensive, yes. But if I'm able to get what I want out of it, I think it still makes more sense than paying $1000+ for a similar xfmr that isn't entombed. (Well, I simply wouldn't pay for it, period. I got this one out of a dumpster.)

It isn't all that bad work though, kinda gratifying to watch it come along. I think I might enjoy being an archeologist.
 

jpanhalt

Joined Jan 18, 2008
11,087
If acetone actually dissolves it, the matrix may well not be epoxy. With heat, epoxies soften. You can check the glass transition temperature (Tg) for common formulations. Again, what you are most likely to see is softening, not liquification. The softened material is easier to peel away. Epoxies are common, but are relatively expensive compared to some older polymers (e.g., urea-formaldehyde or phenolic).

Your description and the solubility characteristics make me wonder whether the matrix is simply a varnish. If acetone works, that is great. Methylethyl ketone (MEK) is similar to acetone and less volatile, so it sticks around longer. Lacquer thinners will likely have one or both of those solvents along with some esters,e.g., butyl acetate, ethyl acetate.

My first experience with NMP (mentioned above) was with an "environmentally safe" paint stripper called Peel-Away 7 (http://www.sherwin-williams.com/hom...lies/solvents-removers/removers/&N=1816535833) ). The S-W price is quite inflated, but you can find that chemical in many other products. In my experience it makes a soft goo rather than nice liquid when applied to old polymers, but the soft goo is easily removed with acetone or MEK. If you go in that direction, be sure to read the label. Peel-Away 1 is quite different chemically and may not be what you want. It is basically a poultice of lye in chalk. You smear it on, let it sit damp for several hours, then scrape it off. Lye is cheap, so you may want to test some of the chips you have removed with strong lye to see whether it has an effect. As with any chemical, be careful, but be particularly careful to avoid splashes that might get into your eyes when working with lye.

John
 

Thread Starter

strantor

Joined Oct 3, 2010
6,782
Well I've unearthed a transformer-shaped object.



That's another 2hrs worth of work. I've found a trick that really speeds things up...
Cut all the sides off and then heat one area of the sandpoxy super hot. Pick a corner or blank side, somewhere that isn't too close to anything vital and heat the crap out of it with the blow torch. Heat one spot (like a 2"dia spot) until it turns black, smoky, and on fire. Keep heating past black, and until it turns red and glows (the sand, not the steel, don't heat the steel). After the sand in that spot glows red/orange, keep heating it another 2min or so. Altogether probably 5min in one spot with the blow torch on full blast. This lets the heat soak deep in through inches of epoxy and get everything nice & soft. Once the heat reaches the transformer steel, the steel acts like a big heat sink and the steel stays cool while the epoxy is hot. I think this causes a "rift" in the sandpoxy wherever it touches steel. In most cases, the sandpoxy will start to lift away from the steel all on its own,but if not, all it takes is a little persuasion from a putty knife and you can peel chunks or sheets away from the steel , and find the steel cool to the touch.

I did not get to the store to buy any of the chemical products whch have been discussed. Not sure I need them now.
 
Top