jsnospam@cimmeri.com wrote:
What's buzzing / ticking is the big transformer in the switching converter section.That's what my original hypothesis was. But you say that the computer is working fine with this ticking going on, and that seems very unusual to me. The "ticking" is the usual sound of the switch-mode power supply starting up and shutting down immediately, so it usually indicates: 1) shorted output (also sometimes no load at all), 2) overloaded output, 3) internal short fault in one of the output filters, or 4) special case of 2), in which the "overload" is actually a normal load, but the off-line supply (120VAC-->170VDC) is delivering insufficient power to the switch-mode converter. (This is often the result of a low-capacitance filter cap in the off-line supply's filter, but could also be caused by an open diode in the bridge rectifier.) There are other possibilities, like trouble in the feedback circuits, but I expect those are the most probable. If your machine continues to operate with the ticking going on, then the ticking is something quite different from repeated power supply shutdown--for example, the switching transistor could be periodically failing and recovering or otherwise "dropping pulses", which seems very improbable in a self-excited oscillator of this type. If you can observe the outputs with a scope, do you see anything on any output that is synchronized with the ticking? (Avoid probing around the transformer, since there may be very high peak voltages.)A few more clues: - if I lower the input VAC via variac to about 100vac, the sound diminishes into a softer hissing (ticking so fast and and soft, it's a hiss/sputter). - if I adjust the single trimpot that provides for fine adjustment of the dc output voltages (+5,+12,-12), I can change the frequency of the ticking or make it go away almost completely... but only if the input VAC is at 100v. This trimpot is in the feedback control circuit. I can also get the hissing to go away but then the output voltages are a tad too high.
Hmmm. It may be that the low-pass filtering in the feedback circuit has diminished (another, lower-value electrolytic capacitor) to the point where it is causing "squegging"--in which the feedback circuit is acting like a relaxation oscillator. This would be visible on the output as a small amplitude sawtooth oscillation with the edge of the sawtooth coinciding with each "tick". If you set your scope to AC coupling and turn up the gain, you should be able to see this. This mechanism would explain the dependency of tick frequency on voltage setting. If this is the case, look at the few components in the feedback circuit that form a low-pass filter. I'd bet that there's a low-value electrolytic capacitor (1-10uF) that has decreased in capacitance. Again, the easiest way to confirm this is to bridge the capacitor(s) one at a time with a good capacitor of comparable value and comparable or higher voltage. Once you've confirmed the bad cap, you can get a near-exact replacement.
As an aside.. I'd just like to say how grateful I am for resources like those of you who reply to my posts.. especially Michael who has spent so much time explaining things to me. You guys make life better.
We are a small band of folks who share an enthusiasm for these great old machines. It's a real pleasure to help a colleague. -michael NadaNet file server for Apple II computers! Home page: http://members.aol.com/MJMahon/ "The wastebasket is our most important design tool--and it's seriously underused."