The Casting Experiment


I had a Lyon & Healy 5-string with 30 brackets that I wanted to rehabilitate. Unfortunately, 9 of the brackets were missing. A search of my spare parts stock (a.k.a. "junk box") revealed no reasonable matches, so I needed to make them. A query to Banjo Hangout forum was unfruitful. I have made some of a very similar bracket before (like the “Nice No Name”, see second photo). I made them from brass round stock with saws, files, and a Dremel tool. This is laborious at best and not very easy when you include the shaping, drilling, tapping, polishing, and plating. These particular brackets have a pleasing, delicate shape that is hard to duplicate.


If I just needed one or 2, I would just do it my old way, but for nine I thought a different way was in order. The originals are cast brass, of course. So, I have been reading about small-scale casting. Sounds like fun (and with good potential for burns and setting stuff on fire) but there is a fair start-up curve which includes mold making, a melting furnace, and selection of the appropriate metal. Obviously, the smart person would simply sand cast brass. The concept is simple but the details and equipment needed are much more complex. I very well might do it that way but I wanted to try some other things first.

OK, I’ve been reading about silicon rubber molds. These are easy to make and can be used over and over again, easily enough times for my bracket needs. The main problem here is that brass casts at around 1700° F. and most of the silicon molding materials have a maximum temperature ceiling of 350° - 600° F. After a lengthy search, I located a casting compound used for casting pewter jewelry called High-Temperature RTV Mold Compound sold by Rio Grande Jewelry & Findings in Albuquerque, NM. This is their own brand and they claim it is “(d)esigned to withstand metal temperatures up to 1100°F”.

Next was the issue of selecting a metal. There are a variety of alloys now available with relatively low melting points, known in industry as “low melting alloys” or “fusible alloys”. Most of them melt well below 1100°F. So far, I have ordered several different types and will experiment with them to see which ones, if any, work for my particular application. My main doubt is whether any of them will be strong enough to use for brackets. By the way, there are many different alloys of brass (copper and zinc) but generally the tensile strength is on the order of 30,000 – 50,000 psi.

So far, I have ordered:

1. A zinc alloy known as Z-27 or Zemak from Budget Casting Supply, the same stuff that we sometimes call “pot metal” which has been used for banjo parts such as flanges and (rarely) tension hoops, tuner buttons, and sometimes certain forms of brackets (such as the exotic Regal “elephant foot” brackets. Budget Casting Supply says “Zinc Alloy ZA-27 is an under-appreciated metal for foundry use. This metal is 1.5 to 3 times stronger than cast aluminum, and can have the tensile strength of grey or malleable cast iron! It's casting temperature is in the range of 950-1,100 Deg F as compared to aluminum at 1,450 Deg F”.

2) From Micro Mark, a lump of “CT CASTING METAL” [Type CT Almost pure lead. Melts at approx. 500 degrees F. Can be melted with propane torch (plumbing type torch sold at hardware stores). Keep away from children.] and

3) “CASTING METAL TYPE 280” [Type 280 tin/bismuth alloy; lead and cadmium free. Melts at approx. 280 degrees F. Can be melted with small butane torch.]

It is unfortunate how limited the Micro Mark product descriptions are but I guess that encourages experimentation. There are some other alloys to look at such as CerroBend and like products. You may know of this stuff as being used to facilitate bending metal tubing. It is not very easy to find in non-industrial quantities. But I did! A cool company called Rotometals (specializing in non-ferrous and custom alloys) of San Leandro, CA, sells all sorts of interesting metals for casting in small quantities. I got a couple of ingots of a low melting point alloy called 217-440 (see RotoMetals) described as:

217-440 Degrees Fusible Alloy - Originated by GE for anchoring punches in dies, anchor non-moving parts in machinery; hold-down bolts in concrete floors, locator parts in tooling docks, split-jaw chucks, jigs, fixtures, metal forming dies, form blocks, joggle jaws, repairing broken dies, Filling Low Melting Point holes in castings.

The composition is 9% Antimony, 48% Bismuth, 28.5% Lead, and 14.5% Tin. Other sources indicate the tensile strength is about 13,000 psi, the strongest of readily available fusibles.

Onward! Mixing the silicon rubber is easy enough, not unlike any other 2-part epoxy or latex. There is one problem that I encountered early on. Everyone who does casting is familiar with this but I didn't think it would be much of a problem (as opposed to the clear warnings in the instructions). It was simply BUBBLES. After you mix the silicon, there are a lot of bubbles released. Some come to the surface but many hang throughout the body of the mold and float into contact with the work piece.

Here you can see my first clumsy try at this. The lump of metal is obviously the CT (see above). The 2 whitish chunks are the cured silicon molds. The one in the center still has the bracket (with screw).
They don't look too bad but after the molten metal is poured, you see the awful results.

Yeech, clearly there are a LOT of bubbles in the silicon. Just for grins, I drilled and tapped these things and put them on the rim. This is quite unacceptable.

There are two ways to remove the bubbles, either by holding the latex in a vacuum chamber for a while - the bubbles expand and rise to the surface where they are not a problem - or place the latex under high pressure so that the bubbles collapse or get very small. I thought vacuum is safer than high pressure so I pursued that route. Ideally, you would have a bell jar and good vacuum pump. Of course, I didn't want to invest in that stuff, so I went to Harbor Freight and got a little hand-operated vacuum pump used to bleed automotive brake lines.

Easy enough, any moron could operate it, hard to harm oneself, and cost about $20. I mixed the latex as before and poured it right into the reservoir included with the pump.  After the specified time, the bubbles behaved as they should and I set the bracket to be cast into the mold material.  I used silicon spray mold release and it worked very well.


The blurry photo at left shows the cured mold and two of the newly cast brackets. The black stuff is powdered graphite (found in gun cleaning kits) that helps the casting separate from the mold and extends the life of the mold a bit. Just below is a much better photo of said mold and the two castings.

All of the metals that I've been using can be melted with an ordinary propane torch.  Since I'm only doing one or two at a time, I simply use a stainless steel table spoon as a hand-held crucible.






So, we're getting somewhere.  The castings need a little cleaning up and still need to be drilled for the hook and the rim bolt, but that is ordinary stuff.  I did learn that the CT casting metal is way too soft and weak.  It would be OK for fishing sinkers but I can't think of any banjo-related uses.  The "Casting Metal Type 280" is easy to work with but the strength is marginal.  I did make some brackets out of it, put them on the banjo, and tensioned those hooks the same as the brass originals.  They held up OK but it remains to be seen how they will look in 20 or 50 years from now.

The Z-27 or Zemak melts at around 1000° F and the silicon molds handle up to 1100° F.  I thought that thi might be pushing the envelope but the molds held up fine for several castings.  The Z-27 is very tough stuff - very hard to drill and tap but easily as strong as brass, perhaps more brittle but so far it is the best alloy for this purpose that I have used (note: I still have not tried the 217-440).



I made still a third mold to fix some other imperfections and this one worked well.  Here it is with one of the raw castings - using the Type 280 straight out of the mold.  The bar-shaped bit of metal coming off the side of the bracket is the sprue channel.  This is where the molten metal actually enters the mold and allows over-filling to offset any shrinkage.





Here is one of the Type 280 casts that has been drilled and tapped next to the original (left):

Below are some shots of the finished replacement set, ready to be mounted.  Keep in mind that these were nickel plated (using Caswell electroplating equipment) and polished.  I think they look pretty good.



Congratulations on making it to the end of this far-too-long article!  I've been working on it occasionally for the better part of a year.  Casting is whole field in itself and I have just learned the bare minimum to make these relatively small, simple items.  Casting larger things like flanges and tone rings is way beyond me at this point.  Your comments are appreciated.