LOLbooster Update: Power Supply Design Issues

I got my bigger breadboards, and I began prototyping the bit of the design of which I am least confident: The voltage supply to the track. Recall that the design was meant to monitor the track voltage, and adjust the supply voltage to keep the track voltage steady, to compensate for voltage drops due to large loads on the tracks. The design I created reflected a fundamental misunderstanding of how adjustable voltage regulators work, and so failed. Unsurprising, really. But, what’s more, at some point I had two pins on the LM338 swapped, and, although it is hard to be sure, I may have damaged it. It still mostly works, enough that I cannot reliably diagnose the problems with the subsequent modifications of the design—I can’t tell if the problem is the new design or the regulator itself.

And of course, I didn’t order any spare LM338s.

So tonight (if I can) I begin afresh with some new ideas and a couple of spare LT1084s (a fancy drop-in upgrade for the LM338, of which I just happen to have a stash of samples lying fallow).