Screeching & terminal gibberish

Craig Johnson vince at tc.fluke.COM
Fri Nov 10 07:55:08 AEST 1989


What exactly was screeching?  This is terribly important to answering
your question.  Possible sources are, the hard drive, the monitor, the
power supply, and the speaker.  I'm going to guess it was the power
supply.
 
The scrambled display is indicative of a processor which has lost its
mind (due usually to misread opcodes), or a video memory which has
suffered a power hit.  In either case, a sagging Vcc (5 volt supply) on
the motherboard will cause this kind of nonsense.  I'm suspicious that
your power supply experienced a brownout, either because the AC main
had a problem or because the power supply itself is failing.  Further, 
I'm guessing that the noise you were hearing was coming from the power
supply due to its switcher oscillating at the wrong frequency.  Its
entirely possible that the switcher became ill because of a momentary
drop on the line voltage in.  Once it started oscillating at the wrong
frequency it very well may have gotten stuck there.  And, a switcher
which is switching at the wrong frequency will undoubtably have
problems properly regulating its output (ie. you no longer get 5V out).
 
If it ran in this mode very long, its possible that the power supply
overheated some of its components.  I would be suspect of its long term
reliability.  Now would be a good time to get a spare power supply from
Halted Specialties if you haven't already.  [thanks Thad, I've got mine]
 
Speaking of power supply misbehavior, some of you have undoubtably
experienced the same thing I've seen on a couple of 7300's.  That is,
upon applying power nothing comes on, especially apparent if you've 
still got the 12V fan(s), but instead you hear the power supply
clicking.  It usually clicks at slightly faster than a 1 Hz rate.
This appears to be due to the power supply's oscillator failing to
oscillate at the right frequency, and may in fact be stuck at the 1 Hz
rate.  Switching the power off and back on normally fixes this; the
second time the oscillator starts up correctly.
 
Well, I could be all wrong about your problem Tanya, but this is at
least one good possibility.
-----
	Craig V. Johnson		...!fluke!vince
	John Fluke Mfg. Co.
	Everett, WA



More information about the Comp.sys.att mailing list