Odd tank leak

[hzbloke]

1986 R65. I've had a small amount of petrol running along the seam on the bottom left side of the tank for a month or so. I assumed it was either the seam or a nearby rust hole. I finally got around to taking the tank off today and found that it was neither. It appears to be something to do with the welds that hold the bracket on that supports the rear of the tank.

This is looking up the rear of the tank on the left side:


A closer look at the welds holding the bracket to the tank:


A little talc and some petrol in the tank:


The leaks, unlikely as it seems, appear to be coming from the welds themselves. If the hole were behind the bracket I would have thought that the fuel would leak from the centre of the bracket between the two welds. Is it likely that the tank is weeping from both welds?

More importantly: how do I fix it without cutting the bracket off?

ps While we're down there, which way do the grommets go? Thick side up or down? I've always put the thick side down but they could go either way.

[Bob Bennett]

G'day Mate

I've had the same problem in the past,,I had fatigue cracks beside the welds, due to poorly tuned
motor vibration in my case. radiator/fuel tank putty fixed it for a while but some embarrassing
moments when it let go made me take the tank off and get it welded up.
Then a few years later I had to clean it out and reline it due to the original lining being ruined
by the welding.

[Duane Ausherman]

I have fixed such leaks by welding them up. I have welded up many with gas still in the tank.

[bbelk]

I have fixed such leaks by welding them up. I have welded up many with gas still in the tank.

Sounds like a Darwin moment....

[PAS]

I had the same leak on my 81 R100 tank. First I tried "Seal All" which lasted about one season. Last time I cleaned the area well then fluxed and flooded the weld, especially under the bracket with solder. That is till holding up.

[Duane Ausherman]

I have fixed such leaks by welding them up. I have welded up many with gas still in the tank.

Sounds like a Darwin moment....

Sounded like that to me too, but this old guy taught me how to do it........... and it worked......... so far.

Of course, this is coming from a person who willingly rode bikes with high speed wobbles.

I have always known that I am lucky. Lucky is better than good.

[hzbloke]

Thanks for the replies. I've had a look at the welds with a loupe and can't see any cracks so they must in the 'inside' of the bracket. I'm guessing that means that welding won't help. Putty doesn't seems to be a long-term fix so I like the idea of flooding it with solder.

Thanks again.

[Ken in Oklahoma]

Thanks for the replies. I've had a look at the welds with a loupe and can't see any cracks so they must in the 'inside' of the bracket. I'm guessing that means that welding won't help. Putty doesn't seems to be a long-term fix so I like the idea of flooding it with solder.

That sounds worth a try, but I'm a bit pessimistic. Solder is substantially softer than steel, so the stresses/vibrations that caused the crack to propagate in the first place are likely to flex the soft solder 'covering' the crack. That is to say that there will still be a 'stress riser" underneath the solder. (Imagine a putty covered crack).

Soldering the crack might work, especially if the solder is thick over the crack and then tapers to thinness. Personally I would be inclined to stay away from solder and use something stronger and more flex resistant. Silver solder I think would be a good choice. Using an acetylene torch, when the silver solder comes up to temperature (with appropriate flux) the silver solder aggressively wicks to wherever it is hot enough to liquify the silver solder. Conventional brazing rod (cu/zn) seems to be a bit more touchy to use. You have to get the rod to a higher temperature and it doesn't wick very well unless the surrounding steel comes up to red-hot temperature. Recently I've used a 15% silver, 80% copper, and 5% phosphorous rod which I got from a plumbing supply and which I believe is intended for sweating the joints of copper tubing. The flux for that rod looks very much like regular borax brazing flux. That rod is my new rod of choice for closing up the pin-hole leaks that show up in air compressor tanks (because the owner-me-wasn't prudent about letting the water out of the tank.)

Oh, whatever you do, you will need to have the surrounding area squeaky clean. That means a wire brush removing paint, scale, and rust. Luckily silver solder and the brazing rod I'm talking about is very good at wicking through rust when brought up to temperature.

Caveat: This stuff originates mostly from my fertile mind. There's a slight chance I don't know what I'm talking about.

And perhaps ME109 or some other experienced welder will come along and give you better advice.

Ken, modest in Oklahoma

[Rob]

The one time I used silver solder, I was surprised at how much more heat it took to work with than rosin-core solder used to electricals.
Soldering iron didn't come close. I had to pull out the propane torch. I didn't have a pencil torch at the time. I have one now, but I have yet to find a need to use it.

I was building my own throttle cables, as I couldn't find any OEM that would work. Something was messed up, but I couldn't find it. Nipples on cable need silver solder.

[Ken in Oklahoma]

. . .I was building my own throttle cables, as I couldn't find any OEM that would work. Something was messed up, but I couldn't find it. Nipples on cable need silver solder.

The trouble with using silver solder for cables that the temperatures needed to flow the silver solder will at the same time anneal the cable's inner strands. Those strands, even if not subjected a specific hardening process will have been work hardened by the manufacturing process of drawing the wire to the specified diameter through a series of successively smaller dies. Silver soldering or any other such process will by degrees soften (anneal) the wire strands. You might be able to use the cable for for a while, but not for long. That's why you so often see cable ends with swaged steel tips. No heat involved.

You can lead/tin solder the cable ends, but the accepted process (at least in the olden days) is to crimp and otherwise 'fray' the very end of the cable which fits into a correspondingly enlarged recess at the back of the barrel. The soldered tip will then 'freeze' the crimped up wires such that they can't pull through the barrel. Without the crimped ends the nicely soldered cable will simply pull through the barrel. (I've done that a few times.)

At least that was the conventional wisdom back in the olden days when I was too poor to afford new clutch and brake cables for my Brit bikes.

The other down side is that if the solder wicks too far up the cable that makes that area stiff and forces the adjacent non soldered area to bend and work fatigue at another location.

Or so I have come to believe.

Ken