Leak options

[Andyroo]

That's the usual approach for repairing leaks in silicone, as all you're doing is repairing the skirt & not the structure. 5200 will be stronger, I expect. In recent fixes I've used a little triangle of glass to reinforce & just make damn sure as we get an occasional quake too.
My worry here is adhesion between silicone and, well, anything including 5200, and then I'm loving the idea that I don't have to drain & dry for 5200 - just drain snorkel-depth. 5200 apparently grips very well to cured epoxy, so should be good so long as it goes from clean epoxy to clean epoxy, across the silicone.

I've written to 3M on all of this, including water-pressure sending material into the leak-gap itself, and will keep you posted.

 

[fishdance]

Have used bentonite before but not in a glass aquarium. The slow drip works against you here as not much would get in between and you would be battling cloudy water for many many months. It works very well for earth ponds and dams. And good to de-sticky adhesive fish eggs if you want to tumble in an incubator (zoug jar )

 

[jjohnwm]

It all comes down to the inevitable love/hate relationship that most DIY tank-builders have with silicone. When it's wet, it sticks to everything...and when it's dry, nothing sticks to it.

If that tank has a wooden frame, which it appears to, then I hate to say this but I think your only completely safe option is to drain, dry, scrape and clean all old silicone from the inside (not the structural bond between the glass and wood, just the "sealing bead"), re-seal, wait for cure and then re-fill. I have built a fair number of plywood tanks and had to deal with a couple of leaks. Two of them I treated as I outlined above, and the leaks were corrected.

One other time I tried a Rube-Goldberg-style repair without draining. The leak appeared to be be fixed...but in actuality it was merely slowed down so much that there was no visible water that reached the floor, or even ran down the outside of the tank and frame. The tiny, extremely slow leak that remained simply soaked into the wood, and naturally did so in a location that was impossible to see and examine. Almost a year later, I found a 6" X 12" area of plywood in a bottom corner that was soft, rotted and useless. I shudder to think what would have eventually happened if not for that completely accidental discovery. There was no coming back from that one, repair was impossible.

 

[Andyroo]

jjohnwm - is that glass to wood directly, or glass to PondShield? I've found the connection between epoxy & silicone to be annoyingly variable - some places it sticks nicely, but never as well as it does to glass.

I expect your suggestion is where I'm headed: drain, clean, dry, cut, clean, seal, wait, test. The question right now is whether silicone is in that equation vs 5200 or another material, plus possibly an epoxy&fibreglass semi-structural overlay bridging the silicone. Still no word from 3M.

There's a plywood facade around this tank; the glass is set into an angle-iron frame and the back&bottom&sides are rebar'd cement. Steel & cement are sealed in WestSystem epoxy and there's silicone at the glass-steel and again at the steel-cement (min 1" thick/wide sealing bead, nearly 2" at cement-steel). There are steel structural tabs with 1/4" anchor screws & epoxy into the cement also - I'm paranoid that way. I remove the facade later this week/weekend to see exactly where the water's coming through, before draining; I expect it'll be at the cement-steel.

I'm somewhat terrified that the carpenters managed to crack the cement in screwing around with that stupid facade, but let's see. Such a thing would "just" be further fibreglass & resin in the repair... theoretically. I hate fibreglass... nasty, hellish, itchy stuff.

 

[jjohnwm]

Silicone to Palguard epoxy (probably a long-obsolete product by now...) was outstanding. Heavy sheets of glass up to 94 x 23 x 1/2-inch, adhered for years to it with silicone; no braces or supports to hold up the glass, just that heavy piece glued to the inside of the plywood cut-out front panel. Lasted years or decades without issues.

Past 2 years I have built a couple with Pond Shield. Silicone adhesion seems just as good, but it's not really a long-term test yet.

Just finished an experimental tank where I sealed the front glass (46 x 16 x 1/4-inch) to bare, sanded, dry plywood, using Silicone II. I then painted the interior surfaces with another epoxy paint product (Interseal) and ran a bead of Silicone I all around the glass to isolate the Silicone II and bare wood from water contact. Don't know if it'll last long-term...or even short-term!...but results seemed promising.

I can't help but wonder if your inconsistent results with silicone might be related to stale-dated product. Are you checking the expiry date on each tube as you use it?

Edited to add: sorry, just noticed in your other thread that you are aware of this last item; disregard.

 

[Andyroo]

UPDATE:
Found a black 5200 supplier in Kingston & so removing the steel-concrete gap's silicone - started last month and have been continuing piece-piece.
The gap is anywhere from <0.5 to maybe 4mm = headache. However, as the silicone mostly didn't stick, it's mostly coming out in long, clean chunks. The few "stuck" bits, lumps and residues are a worry, though...

AND(!!!) the WestSystem seems to be delaminating from the steel and even from the concrete - it almost seems like there was a reaction with the silicone, but that's really not supposed to happen. At the glass-steel (via Epoxy) it's holding fine and not delaminating, suggesting that it didn't like the shear forces. What a palaver...

However, now that I don't have to baby the epoxy I'm in there with steel scraper & hacksaw - 5200 is good on steel so long as it I'm on top of any rust. There's then some question of sticking to exposed cement, but there appears to be an epoxy layer/residue remaining regardless.

Considering that the 5200 is to be somewhat structural, any further suggestions on getting the last of the silicone out will be greatly appreciated. I am using ethanol (100%) in a little spray-bottle - yes, it's the hand-sanitizer from my car. Also suggestions on getting any loose epoxy out... beyond raw rage & scraper, that is.

Option #3 is to cut the frame (& glass) out of the wall & start again, but that would destroy the welded steel tabs which couldn't be (safely) replaced without removing the glass from the frame. As it is all I have to do is seal it and these tabs will do the heavy-work. Alt would be to rely on the sealing material (5200) to hold against shear, which it should do, but... paranoia.