Soldering Aluminum

Last updated - 07/27/08.

Ok, everyone knows that you cannot solder aluminum. Solder is a lead/tin based alloy used to attach copper to copper, as well as gold, tin, and probably a few other metals. Solder does not stick to aluminum, so what do I mean by aluminum soldering? Well, I have found two readily available products that will attach aluminum to aluminum using aluminum alloy rods and a propane torch, much like sweat soldering copper pipes. Note, that the HTS-2000 website uses the term "Aluminum Brazing", so this is probably a more appropriate term than "soldering".

The results that I have gotten experimenting with these products are very promising, so here's what I have found. The first product is HTS-2000 from a company called New Technology Products. Someone on the rotary list brought this to our attention a year or so ago, so I ordered the one pound starter package for about $65. When I got it I followed the instructions, and I was able to attach a couple of pieces of scrap 2024T3 angles to each other without too much trouble. I was pretty impressed, because all I needed was a propane torch.

My work was not perfect, but I was pretty sure that with a little practice, I would be able to use this stuff to fab an intake manifold somewhere down the road. In fact, last year, when I screwed-up my forward top skins with a dreaded figure 8, I was able to use the HTS-2000 to fab a patch that I'm pretty sure will do the job.

Sometime after my first experience with the HTS-2000, I found Alumiweld Rods at Harbor Freight. I didn't expect a lot, but the cost was only $13, so I bought a tube. There's only 8 rods in a tube, but that's plenty to experiment with. The instructions inside their tube, which I didn't read until now, lists their website as AlumiWeld.com. Their website lists the price as $30 for a one pound starter kit, so this seems to be half the price of the HTS-2000.

Anyway, fast forward to July 12, 2008. My engine mount should be done soon, so I thought I would practice a little with the HTS-2000 brazing rods on different types of aluminum. My previous experience with 2024T3 was positive, so I thought I would try some 6061 angle. I found some 3/4" aluminum angle at the aviation department of Home Depot that looked like 6061, so I thought I would give it a try.

I cut off two 3" pieces, clamped them in my vice, and tried to weld them together with the HTS-2000 using my propane torch. No matter how much heat I applied the stuff just would not attach itself to the HD 6061. I figured this HD 6061 just could not be brazed, which was a disappointment, but who knows what it really was, anyhow. To be sure that I knew what I was doing, I decided to play with some 2024T3, again. I clamped two small strips of .025 2024T3 in my vice. Again, I could not get the HTS-2000 to stick to the aluminum. So, I gave the HF AlumiWeld a try, with the same results! What gives?

Frustrated, I shut down the fans, and closed up the shop. Nothing like 100° weather in central Texas to make you want to quit when things do not go right. What am I going to do, now. My experience with trying to learn to arc weld did not go too well at all. No way can I rely on learning to Tig weld in order to fab my aluminum parts for the engine build. Things do not look good.

Later that night, I thought, surely, I must be doing something wrong, so I went to the New Technology Products website, and reviewed their instructions again. This stuff only requires 700° to melt. Why couldn't I get it to flow? Let's see, aluminum pieces clamped in a steel vice. Maybe I could not get the parts hot enough to melt the rods because the vice was sinking all of the heat. So, I made a couple of wooden vice jaws to replace the metal jaws--problem solved.

Below are some pictures of a few pieces that I have brazed. I do not have any pictures while I was actually doing the brazing, because it was too hard to take pictures with a rod in one hand, and a propane torch in the other. Just go to the website to have a look at how the professionals do the job. Below is just my experience with the brazing rods on a few test pieces.

Here, I have two pieces of HD 6061 angle, back-to-back. The brazing rod is HTS-2000. This looks a lot better than last week. The metal has filled the grove between the top of the two pieces and attached itself to both pieces. I didn't get any pictures from last week, but the aluminum rod never did flow nicely. I just got a few globs that only seemed to stick to one part, and the globs could be broken off with pliers fairly easily. Note the paper clip. Without the clip, the two pieces just flop open when you close the vice.
The other side flowed all over the parts. This is because I used the brush a little too much and the parts had a little residual metal from last week's disaster. I'll work on getting this right, although, this is probably not really a problem.
Here's the finished part after being filed-down a little, and polished on the 3M wheel.
Bottom side.
Here, we have two more pieces of HD 6061, using the AlumiWeld rod. The AlumiWeld rod does not flow quite as nicely as the HTS-2000, but it still seems to work.
I tried to get a better bead here than I did with the first pieces.
Upon close examination, it looks like I have a slight gap here where the metal did not fully attach to the part, but the rest of the bead looks pretty good.
Both parts, side-by-side, with the HTS-2000 brazed part on the left, and the HF AlumiWeld brazed part on the right. Both seem to work ok.
Bottom sides.
Ok, we can braze 6061, how about doing some sort of aluminum box? Here, we have two 2"x4" pieces of .025 2024T3, bent to form the outside of a 2"x2"x1" aluminum box. Looks like I have already brazed the other end and the two sides. Note, I am using the AlumiWeld for this test.
After brazing both ends and sides, I have quite a bit of metal accumulating on the sides. I think it would have been better to braze each end separately. However, it is hard to do one end without affecting the adjacent sides.
Closer look at the build-up. Here, you can see a few holes that need to be filled (asuming that we want an air tight box).
This is after rotating the box and reflowing the AlumiWeld. After this photo, I got rid of the excess, which was accumulated when I pushed the glob off the end with the wire brush. The HTS-2000 tutorial showed them using a wire brush to help spread the metal once it was melted. I think that helps a bit.

Note, after reading the AlumiWeld instructions, it says you need to use a scraper of some sort (i.e., a stainless steel brush) to break the ozidation layer in order to get a good bond. The oxidatized aluminum should rise to the top of the pool. So, this agrees with the HTS-2000 tutorial. I need to practice this, since it seems to be key to getting this right.

Here's the box, after reflowing the big chunks of AlumiWeld. I think I have done a bit of filing here, also.
The other side .
Here's the finished box after filing and polishing. Note, the "X" marks are spots where there is actually a hole. The box passed the blow test, but failed the suck test, so if this was for real, I would have to do a little more work to get this right.
The other side.

So, what have we learned? First, it is possible to "weld" aluminum (or braze it), without having to buy a TIG welder and learning how to use it. With a little practice, even a software guy can do this. Second, both the HTS-2000 and the AlumiWeld Rods seem to do the job.