Double Rad weld failed and sticking/leaking valve

[Fitz]

Two problems.

1. Double rad off the wall for decorating behind it. Now it's back on I have discovered that one of the welds that joins the two panels together has failed. This was probably during one of the many dragging, lifting, moving experiences it has suffered whilst off the wall.

Can anyone recomend an appropriate substance for fixing it? Current thoughts are araldite or some metal putty/ instant metal type stuff.

1. Another radiator has a leaking valve. This is an old radiator and valve on the original system which has been converted into pressurised combi heated system about 18 months ago. It worked fine before it was shut off which was about 6 months ago. I cracked it open the other day and it started slowly bleeding water from the valve. Closing it again stopped the leak but means the radiator is unusable. Is there any alternative to draining the system and replacing it? IS this a simple task which could be learnt? My assumption is that to drain it there will be a valve of some kind at the lowest point in the system. I'd turn the boiler off, drain the water out via a hosepipe into an outside drain. Do the work. Refill by putting some inhibitor into the system and then turning on the filling loop. Is it that simple? Do I, by asking if it's that simple, betray some fundamental lack of knowledge which means I should just get someone in?

Any help appreciated

-- Steve F

[sPoNiX]

JB Weld should do it. Make sure the surfaces are clean and DRY before applying.

Failing that I believe you can buy special "Waterweld" JB Weld that sticks to wet surfaces...haven't seen any though...maybe worth trying a chandlery?

[Ian Stirling]

I would be a bit reluctant on relying on this, solely because it's going to be bloody difficult to prepare the surface adequately. I would be tempted to find a bit of metal of the right size to sit at the end of the two panels, and sand off the paint in that area, before gluing the strip along the whole length.

[Dave Plowman (News)]

Have you priced up a new rad? Properly done welding just doesn't fail like this - sounds to me like corrosion has started. And other welds will follow.

[Malc]

You have tried tightening the nut on top of the valve where the spindle pokes through? I've seen a couple of valves where the nut (gland nut?) loosens when the valve is opened and they start weeping.

-- Malc

[Steve Firth]

I agree, it's not the end of the world to replace a radiator, especially since you have removed it from the wall already.

[ritchieaber]

snip request for advice re repair....Steve I would never be happy with this repair. the thing is expanding and contracting all the time, you'll never sleep for fear of it failing. Get a new rad.

snip leaky valve saga.... Where is it leaking Steve? up past the spindle? this can be cured just by nipping up the gland nut that the spindle passes through. No draining required.

hope it does.

[chris French]

Indeed, I replaced rad that I took off recently as it was rusting on the base. It was probably going to be ok, but it looked a little dubious, and the cost and hassle of fitting a new rad was nothing compared the cost and hassle of stained new hallway carpet....

[Steve Firth]

The last time I bought a two component epoxy sealer (to repair a terracotta chimney pot) it was IIRC about eight quid from the local DIY shop. That's a fair chunk of the price of a radiator for a fix that may not fix.

[Fitz]

There have been a couple of similar replies and I think I may have described the problem badly. The radiator is a new(ish) double panel radiator. The panels are joined at either end near the top by a piece of metal about 2cm x 5cm x 2mm. This seems to be purely for bracing the two panels and the weld that has failed does not form part of any seam. There is no sign of corrosion where it has failed. I think the failure is purely down to the radiator being dragged and lifted whilst full of water upside down. I imagine that's not what the designers had in mind for the joint.

A couple of people suggested this so I'll have another look. From the brief period it was open I think it _was_ coming past the spindle. It was only on for a few seconds though. If this is the case I'll be delighted.

cheers all.

-- Steve F

[Dave Plowman (News)]

If a weld has failed through pure mechanical reasons, I'd not give any other repair method much chance of succeeding. If the rad is off, why not just take it to a welder? Shouldn't cost much.

Look for a MOT repairer in yellow pages.

[Rob Morley]

So weld it :-)

[Ian Stirling]

If an adequate weld....

It's quite easy (with some processes) to make welds that look perfectly sound, and don't do much more than cover up the paint.

[Dave Plowman (News)]

Could you run that by me again?

[Ian Stirling]

It is quite possible with some processes, to make a weld that looks perfectly adequate on the surface, but only actually welds a tiny fraction of the underlying material. You get a beautiful looking joint, but with only maybe 1% of the area under it actually attached to the metal.

[Huge]

When I was learning to weld, we had to cut open the practice welds, polish them flat and etch them with acid to check the penetration.

[Rob Morley]

It's a good job MOT testers don't do that :-)