By the way... Using the exact same hardware, but replacing the 851s with a pair of TM-100s and an old PC power supply works fine, regardless of who is powered on first.

A non-text attachment was scrubbed... Name: not available Type: text Size: 2430 bytes Desc: not available Url : http://www.classiccmp.org/pipermail/cctalk/attachments/20000425/3ec4d9a3/at…

It seems to me that the problem involves the two power supplies, and probably not any of the S-100 hardware. If the enclosures are connected with a substantial protective ground, that's probably OK if the protective ground subsequently goes to a genuine "earth" ground. If you examine the ground lines on the FD cable, and there are plenty of them, they should all be connected to your system ground at both ends and bypassed/AC-coupled to frame ground. If your drives have signal and frame ground connected together yet not connected together at the FDC, i.e. in the Imsai box, that could be contributing to the problem. If I were in your position, I'd hook the 'scope up to the ground line with the reference ground at the S-100 bus, not at its power supply. Then you will probably see an oscillation. If that's the case, disconnect the drives and any of your S-100 hardware from frame ground, paying particular attention to the serial I/O boards, where it's common to connect pins 1 and 7 of the DB25's together. If you don't want to interrupt that connection, replacing it with a 0.1uF bypass cap, then I'd recommend you connect all your supply grounds to frame, using a 1"-wide copper braid, and connect that braid to a ground like a water pipe. If you have plastic plumbing, tear it all out and replace it with iron (snicker!), otherwise use a gas pipe or some other adequate ground attached with a really low-impedance connection. Since your ground isn't a single point, you need a really low resistance between them all. That's why it's actually better to isolate frame and power supply (signal) ground. The comparison with the TM-100's doesn't really show anything because it uses a different supply, not to mention that it uses a small fraction of the current. There's probably just a single ground line floating somewhere between the drives and their supply ground. It may come up stabile enough, but suddenly jump into oscillation as soon as a surge occurs on the +24 Vdc supply. When you hook the 'scope up, look at it while the system tries to move the heads. If the regulators for the FDD's remain solid, it's got to be the connection between the FDD box and the Imsai box. If not, then look for inadequate grounding or inadequate bypass at the regulator. This is particularly critical if you have a pass-transistor regulator, since there's a phase shift there that often causes oscillation. There should be a solid frame ground, though both supplies should be connected to the "earth" ground at their frames regardless of whether they're well connected to one another. ----- Original Message ----- From: Bill Sudbrink <bill(a)chipware.com> To: <classiccmp(a)classiccmp.org> Sent: Tuesday, April 25, 2000 1:19 PM Subject: Shugart 851 power problem