Profile

unixronin: Galen the technomage, from Babylon 5: Crusade (Default)
Unixronin

December 2012

S M T W T F S
      1
2345678
9101112131415
16171819202122
23242526272829
3031     

Most Popular Tags

Expand Cut Tags

No cut tags
Tuesday, March 30th, 2010 04:44 pm

So.  The toilet tank in one of our bathrooms has been leaking for a while, and I haven't gotten to fixing it because for so long we didn't have money for replacement parts.  On Sunday, one of the toilet tank bolts fractured because it had corroded away due to the leakage (and due to not being made of anything remotely corrosion resistant, like, say, stainless steel or phosphor bronze).

Yesterday, I buy a replacement bolt/washer kit, dismount the tank, replace both bolts (a significant endeavor in itself, requiring the application of a methylacetylene torch because the broken bolt and its upper nut are solidly corroded into a single unit that cannot be freed with the torque possible on the badly corroded bolt head), and refit the tank.  I discover after doing so that the tank is still leaking from the bolt holes.  The new bolt/washer kit is not sealing the bolt holes because of a tolerance problem (the inner rubber washers are not thick enough to seal between the tank and bowl).  I also discover that, having been disturbed during this process, the stopcock is now leaking as well.

Today, I get a new, different bolt/washer kit, a new tank-to-bowl gasket, a new flex hose to replace the non-reusable 1/4" semi-rigid copper feed pipe, and eventually find the only possible new stopcock that has the correct fittings on both inlet and outlet.  I remove tank, turn off the water to the house, and remove the old stopcock - by the only feasible means, cutting it off the end of the pipe, since it seems to be a single-piece fitting integral to the pipe and chromed after assembly.  I clean up the cut end of the pipe, go to install the new stopcock, and discover at this point that the pipe coming out of the wall is not ANY standard copper pipe size.  It's just oversize enough that it will not in any way go into a 5/8" OD compression fitting.

I curse, scratch my ehad a lot as I try to figure out what the hell to do now, and eventually discover that, by chance, the proprietary-weird-size pipe fortuitously happens to be just large enough ID that by reaming it out a fraction, I can just barely insert a piece of standard half-inch (which is to say, neither the OD nor the ID is half an inch) copper water pipe into the oddball pipe.

I cut and prep a suitable 2" piece of half-inch copper pipe.  I clean and scour both old and new pipe thoroughly, dry the inside of the old pipe as far in as I can, insert my pipe stub and solder it in place, using my rosin-core electronics solder because the lead-free plumbing solder that's all you can buy in hardware stores is utterly worthless crap.  I fit the new compression-fit stopcock to the new pipe stub, turn the house water back on, and leak-test the new stopcock.  The joints are good, so I fit the new bolts kit and gasket to the tank, combining washers from both bolt kits to get the tightest possible seal, and adding some RTV silicone for good measure.

I reinstall the tank, tighten all the bolts, connect the new flex water line to the tank, and leak test again.  It's still leaking, from the other side now.  At first I think it's the pipe joint, but on closer examination I discover that the leak is from a hairline crack in the bottom of the tank.  A crack which originates, to no surprise whatsoever to any remotely competent mechanical engineer, from one of the mounting bolt holes, and which is invisible from inside the tank.

This is the point at which I inquire, yet again, of the universe in general what festering inbred excuse for the illegitimate offspring of an illiterate plumber's drunken apprentice and the village idiot's syphilitic pet monkey came up with the brilliant idea of mounting a toilet tank to a toilet by putting bolt holes THROUGH THE BOTTOM OF THE TANK.  What could possibly go wrong with that stroke of genius?

Tags:
Tuesday, March 30th, 2010 09:15 pm (UTC)
epoxy or something with fiberglass tape inside the tank? outside as well on the botyom out if sight? sounds like an application for marine grade boat fix...

don't get me started on those nasty wax seals on the bottom... afaik, toilet science has not evolved in 1000 years or at since the inventor

#
Tuesday, March 30th, 2010 09:37 pm (UTC)
Actually, the idea occurred to me to try patching the tank from the inside with leftover hydraulic cement from the house repairs last fall. It's waterproof, will cure wet, and will bond to wet surfaces. The only drawback to this plan is the discovery that my memory was in error and we don't actually have any leftover hydraulic cement ... the landscaper must have used the last of it when he sealed the new drain pipe to the septic tank.
Tuesday, March 30th, 2010 09:48 pm (UTC)
Various kinds of caulk will cure wet. However, that crack is likely to propagate in the way of all cracks in brittle material.
Tuesday, March 30th, 2010 09:55 pm (UTC)
Yup. But if we still had any hydraulic cement, it might have bought us a little time. As it is, I guess I'm shopping for a new tank tomorrow. I'm guessing I won't be able to find one that actually matches the existing toilet in color.
Tuesday, March 30th, 2010 09:57 pm (UTC)
And ew, yeah ... there are few jobs nastier than cleaning up the remains of a years-old wax ring. Hello, people? Neoprene rubber?
Wednesday, March 31st, 2010 12:55 am (UTC)
i was thinking of a threaded screw in number or locking seal...

#
Tuesday, March 30th, 2010 09:17 pm (UTC)
So, design a fully workable better idea, patent it, and end your worries where finances are concerned.

Edit: The thought occurs to me that (in a similar way to what many commercial/public bathrooms now do) you could eliminate the tank completely by putting a flow meter in-line between the water supply and the bowl. *poof* no water storage problems.
Edited 2010-03-30 09:46 pm (UTC)
Tuesday, March 30th, 2010 09:46 pm (UTC)
There's better ways already. They were quite widely used in Europe and the UK. The simplest way is that you have a lip along the bottom edge of the tank, and a clamp lug that holds the tank down by that lip by a bolt that doesn't have to pass through the tank. I understand these days, many European toilets are designed such that the tank is a single piece with the bowl, eliminating the entire problem in the first place.

The other problem is that the way the system works right now, patents on your own inventions are all but worthless. If the invention is actually worth anything, and you're an individual inventor without backing, the odds are some large company will simply steal the idea and stall you into bankruptcy in the courts if you try to sue them for violating your patent.

I have several patentable ideas kicking around, ranging from the kitchen to disaster relief. But I can neither afford to develop them myself, nor to patent them on speculation of ever making any money off them.
Tuesday, March 30th, 2010 09:54 pm (UTC)
The thought occurs to me that (in a similar way to what many commercial/public bathrooms now do) you could eliminate the tank completely by putting a flow meter in-line between the water supply and the bowl. *poof* no water storage problems.
The problem with that idea is that most American household plumbing systems simply can't flow enough water fast enough to flush a toilet completely without relying on a tank. However, the Swedes came up a while back with a toilet that uses vacuum-assist to flush a toilet from a small tank that fits entirely behind the toilet and is cast with it in one piece.

You might be able to manage an entirely tankless toilet in the UK. But in the UK, typical household water plumbing is — or was, when I last lived there — about twice the diameter of typical US household water pipes, and capable of flowing more than four times as much water for the same pressure head. (Not only does the pipe have four times the sectional area, it also has lower impedance; if memory serves, a one-inch pipe will flow about six times as much water as a half-inch pipe at the same pressure differential.)
Wednesday, March 31st, 2010 01:19 am (UTC)
It isn't the flow rate, but the pressure. You have to have a certain minimum pressure for a flush valve toilet to work. US residential water pressure varies a lot.
Wednesday, March 31st, 2010 01:57 am (UTC)
True, I've been on well water systems that wouldn't have had enough pressure to flush a tankless toilet.
Wednesday, March 31st, 2010 05:47 am (UTC)
Yet another solid proof that Murphy is Lord of this Universe -- and that His wife, Eris, is probably laughing Her sides off over this somewhere.
Wednesday, March 31st, 2010 06:48 am (UTC)
Boy this is horribly familiar. I've got a tank with very similar problems and I'm doing my best not to look at it strangely because my current fix works...
Wednesday, March 31st, 2010 06:03 pm (UTC)
Yes. Yes, it is. When we bought the house in Tracy, CA, his parents bought us a new toilet tank as a Christmas present.