I have a set of Williams Crown Line FA Alcos (Dual Motored A and Dummy B), #AC-29 and AC-33 (PRR) in silver boxes. These are the only Williams locos that I own and have serviced. Judging from my research, the model of reverse unit board must have been pretty early, as there are no relays in the design. It's purely based on a H-Bridge motor driver circuit using Bipolar Junction Transistors, D44H11 and D45H11. This looks like Williams' version of Lionel's & K-Line's similar board, though the motors operated from those are much smaller. Similar to early Lionel and K-Line boards, there are no external heatsinks on the transistors.
When I purchased these, used, two power transistors for the motor control circuit had either been removed from, or had fallen out of, the reverse unit board. Consequently, only one direction functioned. I reinstalled the D44H11 transistors and the locomotive ran smoothly in both directions. After attempting to operate with a short train of light cars, I noticed a distinct reduction in speed after a few laps. Raising the applied voltage seem to compensate for it, but I got curious. The loco shell is warm to the touch over the reverse board and the distinct smell of hot electronics was easy to sniff, up close. I suspect that, in the past, the lower transistors became so hot that they loosened the solder joints of the transistors above them. The need to raise track voltage might also explain the heat dimple that was created in the shell over the headlight bulb. I plan to massage-out the dimple with some heat and I've installed a LED bulb to prevent further dimpling.
Has anyone else experienced this heat condition with older Williams units? I suppose this was a driving factor behind later revisions to their reverse boards.
This isn't sustainable, and I'd like to solve it, if possible. Has anyone added heatsinks and/or changed the internal orientation of the board? I have a spare JVDE board in my stash that would work, as well. https://jvde.us/eeu/
For those interested:
The D44 and D45 transistors are rated to carry 10A continuously, and operate up to 150 degrees C (302 F !!) of internal junction temperature. Of course, the internal resistance is going to increase once temperature rises, so this would explain the reduction in speed over time for a given track voltage.
I can add photos, if anyone wishes to see my Williams board.
Thanks for your input.
