Reshape a Tang Band 1931 speaker module?


“Where there’s a mill, there’s a way!” Sometimes.

After reading James (bigtrainjames) post “Seeking advice/specs on a first DCC install” in which he is fitting a Tang Band 1931 into a kit bash SW1500, I was inspired to see if there was any way to fit that module in the fuel tank area of a Red Caboose GP9. The first step was listening to a stock module and a QSI 2.07" speaker I previously installed in a fuel tank. The QSI sounded really good, but the 1931 was even better, revealing deeper bass notes.

It is certainly too wide and deep to fit inside the car body. Because of the module’s length, one could fit if the centers of the fuel and air tanks were removed. Extra plumbing around the fuel tanks might hide bits of the speaker. Frankly, this would be the easiest path, and one I might take. Let’s call that Plan A.

So here is what Plan A might look like. First, a photo looking straight up at the speaker and fuel tank. The perspective makes it seem that the speaker is longer than the fuel tank and air tanks, but that’s not the case. The speaker is under 3.5" and the fuel + air tanks just under 3.6". About 0.125" could be milled off the passive end of the enclosure (which was learned while testing Plan B, but I get ahead of myself.) At least the ends of the air tanks can remain in place. The cabinet is just a bit wider than the simulated frame members. They could easily be moved outboard a tick to clear the speaker.

What about the height of the module? It is 0.76" with the speaker cone at rest. The tank is 0.96" so there is plenty of clearance for the driver. The air tanks extend 0.81" below the chassis and the passive radiator about 0.6" at rest. So there should be enough room for at least part of the fuel tanks all the way across the bottom without having to saw them completely in half.

Then another thought came to mind: What if I cut the module into pieces, harvested the organs, then transplanted them into the stock fuel tank, mounting the passive radiator sideways? I already had an old 1931 module sitting on the shelf that was DOA. Too many months had transpired to send it back to Parts Express. :frowning: Let’s call this Plan B. The autopsy follows.

Here’s the second saw cut. I wouldn’t attempt this freehand since it would be all-too-easy to slice something important inside. Or a finger.

After slowly working around the enclosure, it was time to “part the waters.”

The passive radiator came off cleanly. It was a wise move to offset the cut closer to the passive. The mill bit almost caught one of the leads to the voice coil. If I attempt this again, it would work better to release the passive by milling from the top, cutting closer to the long sides and ends. But I didn’t know what was inside so edge cuts were potentially less damaging.

Here are all the parts spread out showing the bottom of the passive radiator. That’s a tiny wad of acoustic stuffing in the lower right. The cabinet walls and bottom are about 0.078" thick and the top about 0.1".

Tang Band sure was stingy with the stuffing! How can this little fuzz ball kill any resonances?

This suggests how a speaker might be reassembled. The passive needs to be trimmed a lot narrower. Everything might just barely fit in a modified GP9 fuel tank.

Quick measurements of external dimensions suggest that the original cabinet had a volume of about 2.7 cubic inches. The fuel tank would be about 5 cubic inches. Often bigger is better for smallish speaker enclosures. Would it hurt to be a bit more generous with acoustic stuffing?

Plan B would be nice as it minimizes unprototypical bits of a speaker cabinet visible even if only at just a few low viewing angles. However, this would be a lot of work and it would be all-too-easy to slice one of the wires to the speaker cone. Maybe if I only had one locomotive to build, it would all be worth it. But I have about 6!

So after the autopsy, I’m thinking Plan A. Here’s a prototype photo of an Omnitrax GP7 in Fort Collins. Most of the side skirt has been removed so we can see more of the frame. But there are a lot of straps and pipes blocking a clear view through the fuel tank area. It would not be too difficult to detail this area so that nothing can be seen unless looking nearly straight up.

One final thought. With reasonable effort a stock 1931 module could fit inside Atlas SW8/9 car bodies, if the PowerPack migrates to the interior cavity of the fuel tank. But there would not be room for a scratch-built radiator core. The 567-8 and 567-12 sound projects sound sweet through a 1931, and they don’t have to be played at an insane volume.


This drawing approximates Plan A. Some dimensions are off but one should be able to get the basic idea. With the speaker cones recessed deep inside the fuel tank, the tank itself can act as a protective barrier against damage. A bit of black silk glued inside could prevent stray track spikes and other detritus from being picked up by the speaker magnet.

Styrene baffles attached to the sides of the speaker might be prudent to prevent the hollow sides from producing resonances. Alternatively, one could pour a low temperature alloy there as a weight.