Soldering a tractor radiator

I had a problem radiator that came with my Cummins L423D engine. The radiator was damaged due to physical impact. I took it to a radiator shop and they fixed it for $40 or so. I highly recommend taking it to a radiator shop.

Here's a video of it running and fixed:

I'd rather fix it myself if I can, not just to save money, but because hopefully I can learn something from it too.

Best wishes,

Chris

If you want to learn, ask the radiator shop to let you watch them fix it. Problem is if you try to solder it and don't have everything exactly right and have experience at it, when you heat it enough to solder the leak, every single core piece will desolder on you and it's pretty well toast.

Pete C.

That's why I was considering leaving it almost full of water.

Chris

On Sun, 10 Jun 2007 17:22:03 -0700, with neither quill nor qualm, Christopher Tidy quickly quoth:

Radiator repair is an art, so good luck. One of the artists I watched back in the 80s had a dripping brush full of flux in his torch hand and he rotated it into use every few seconds as he worked.

Clean the hell out of it, flux the crap out of it, and gingerly solder with the smallest tip which will do the job so as not to unsolder the rest of the tubes.

- Metaphors Be With You -

Good idea, that's usually my approach to problems. I'd take it out but lay it horizontal and put water in it but prop it up a little so the area to be soldered is dry. Clean the area well with sandpaper or steel wool, apply a good acid solder containing zinc chloride and use lead bearing solder. It flows better than nonlead solder. heat the area until it's hot enough to melt the solder rather than melting the solder with the torch. Your choice of torch is optional, I once saw a person successfully solder a radiator with an oxyacetylene cutting torch. Engineman

It probably is soft solder, but maybe mostly lead. I'm no expert, but maybe 95 -5. I have heard that solder changes its melting point as it amalgamates with whatever it is in contact with, so it's melting temp may be affected by the copper. So, I suppose you could use some 60-40 and maybe not melt too much of the existing joint. Having sid that, google: radiator solder There is a lot of good info there, on the first page.

Don't tell anybody I said this, but have you tried egg-emulsion stop leak yet? I had a small leak in my Ford 8N radiator several years ago and the stuff has stopped the leak ever since. Worked on my Case 800, too OTOH, when my Ford 4000 industrial backhoe sprung a pretty good leak in a spot similar to yours, I chose the "buy a new one" route because I only wanted to go to all the effort of getting all the sheet metal off once.

Have you considered epoxy? The kind used for gas tank repair. Not 5 minute stuff, the longer the cure, the better. Work hard to get the suspect joint as clean as possible. Then wrap a piece of heavy string around into outside of the joint. Work epoxy into the joint and the string. The string will whet with epoxy and help retain alot more epoxy in the joint area.

Pete Stanaitis

----------------------------

Christ> Hi folks,

If you have a good torch and have everything clean, a few seconds of heat will fix it without affecting the rest of the rad. Use a cut acid flux and soft solder, then clean well to remove any left over flux.

Fix it once - fix it right. I've never seen epoxy last on a rad, and the stop-leaks are less than effective on a "stressed joint". Either solder it yourself or get a good rad shop to do it for you.Zinc Chloride (cut acid) is hydrochloric acid with as much zinc as will disolve added to it.

Have you tried Barr's Stop Leak or what ever equivalent exists in your part of the globe?

Wes

Acid flux and eutectic solder along with a old-fashioned soldering copper would be my approach. Actually, that would be my approach if I just didn't haul it to a radiator shop where they've got skilled guys who do the work each and every workday. If you use a torch, you're guaranteed to loosen other parts. The soldering copper provides localized heat sufficient to make the joint and not loosen everything else in the vicinity. Might want to use some wet rags around the area. Best to get it out of the tractor before working on it, too. Eutectic solder is the lowest melting point lead-tin alloy, presumably you'd have a better chance of NOT loosening everything else up using it. Did I mention that the area has to be CLEAN? No way you're going to make any sort of pressure-tight joint with old crud in the seam. Probably the radiator shop guy would take the joint apart, clean it and then reassemble with fresh solder.

Stan

Wire brush it off real good and put some epoxy on it.

John

Clean to shiny metal by scraping or sanding, apply flux and heat very carefully. It is very easy to overheat thin metal. Using a torch requires more care and skill than a soldering iron. I have a 300 watt iron which works great for this kind of work. When the flux starts bubbling it is about hot enough. Do not apply flame to the solder. Wave the torch over the metal a few times and then rub the solder on the clean metal. If it does not melt readily, heat a little more. When you get it just right the solder will flow smoothly all over your cleaned area. If unsure of your ability, practice a little with some thin brass to get the hang of it.

Don Young

I don't have a copper soldering iron, and I suspect I'd need a really big one to get enough heat to this joint. The pipe is about two inches in diameter.

So it looks like it's the propane torch. I have a couple of burners for the torch: 5.2 kW and 12.9 kW maximum power, but they can be turned down a lot. I'm tempted to try to solder the joint in-place with the water level just above the top tube joints. At first I was inclined to use the smaller torch to keep the heat input down, but it's a very narrow flame so I doubt it will have the spread to solder the whole joint at once. The alternative is to use the larger burner but turned down a lot. It produces a nice, controllable flame at lower powers. What do people think?

Best wishes,

Chris

I really can't see epoxy lasting on a stressed joint which is subject to a lot of vibration.

Best wishes,

Chris

I had a 72 Oldsmobile that I fixed the radiator. What I did was to scrounge up some braided ground strap and I was able to flex this so it fit over the filler pipe and used this as a sort of a gusset. I used a regular propane torch and lots of flux and sort of made a nice strong connection between the tank and the fill tube. I don't really now if this was the right way, or the way that a radiator shop would have done it but it held pressure for about a year before the car died from some other cause.

If you can't find a braided cable a piece of stout copper wire should accomplish the same thing.

The reason I adopted this approach was that I had failed on the several attempts previous to trying this where I kind of buggered the shape a bit, hence my resort to gap filling.

You'd be surprised at how well copper conducts heat to nearby water. Unless there is a lot of tank above the top of the tubes, you will have a real problem getting the tank hot with water in the tubes. If the filler pipe is just a fraction of an inch above where the tubes join the tank, it will be impossible to get it hot enough to solder with water in the tubes. You may be able to do what you want with wet rags wrapped around or pressed against the tubes, but water-filled copper tubes just below the area to be soldered will make it absolutely impossible to heat to soldering temperatures. I have had to use a torch to boil out a foot of water below where I needed to make a solder joint before I could get a simple pipe hot enough to melt solder. With a whole row of tubes joining the tank, it has to be several times worse. It may be possible to use an Oxy-Acetylene or Oxy-MAPP torch to spot heat the area, but that will put a lot of thermal stress on old material, so not a great idea, either. I just don't think you have a chance to melt solder with water nearby.

In fact, if you are used to soldering copper water pipe, you may find your torch is too small to melt solder on an EMPTY radiator. All those copper parts suck the heat away with great efficiency.

Jon

You might be able to fix it yourself, but you can also mess up the main solder joint where the top tank meets the top tube header, and then the problems multiply...

When was the last time you had the radiator boiled out and/or rodded out? While they have the radiator all apart, it's easy to fix that spot. The problem is, if it's soft solder at the neck join AND the tank join, you can mess one up heating the other.

The best trick would be to braze the neck to the top tank, and then the soldering at the tank joint won't get the neck joint anywhere near hot enough to come apart or sag and leak. If there are any smart people running the radiator factory, that's how they did it.

(Which of course means that's NOT what they did - the Bean Counters in the Accounting department can ruin perfectly good ideas by pointing out they cost another 25c to produce, no matter that the end product is measurably much better than the competition...)

-->--

I've fixed a few radiator tube pinholes through the years with JB Weld epoxy.

All appeared to be gravel punctures, and were in accessable non stressed areas.

I drained the coolant, cleaned the puncture areas well with steel wool, then followed with laquer thinner on a clean rag. I worked the epoxy into the hole as best I could and left about a 1/32" even layer spread as wide a possible. Then let them cure at least overnight.

I even fixed a GM aluminum timing chain housing with a coolant passageway casting flaw (luckly leaking externally) that must have been .125" in diameter. Pretty much the same thing, only I cleaned up the area with 240 silicon carbide wet or dry paper, then the laquer thinner treatment. It ran more than 15 years that I know of, and might still be.

I think the key is use common sense, don't try to epoxy anything stressed, get the site 100% oil/grease free and dry, and give the patch loads of cure time.

Erik