GREAT SCOPE BUT:
Does my venerable Philips PM3266 oscilloscope (100MHz + memory; analog) show a 200 KHz oscillation because the power supply switching regulator (chopper) runs at that frequency? But then why is any input signal always clean on VIEW TRIGGER?
ONE HOT SUMMER:
To start at the beginning, everything has always worked perfectly. The more savvy members of this forum should make me "milk the front panel" for logical things to check, but, as far as I can tell, the scope works perfectly. It went into a single summer of storage, in a locker hot enough to disintegrate the foam surrounds of two AR38s woofers, and came out with a thick trace.
I was wrong to assume my signal traces were thickened with noise from baked-out capacitors -- random noise, high impedance, something floating that should have been bypassed. Wrong. Turning up the sweep speed resolves a 200 kHz signal. Nothing changes that frequency, and nothing changes its amplitude of 1 box on the screen. So the signal is injected after the head stage signal amps and attenuators.
Most small, pulse-width modulation (PWM) power converters and power supplies today run at around 50 KHz. Is it possible this 1983 scope runs at 200 KHz?
This scope isn't built with the kind of op amp chips that can go into parasitic oscillation, and those oscillations are in the megahertz range, where the Bode plot fall-off produces phase shifts (they tell me; I'm at the limit of my knowledge here). But there's digital logic in the scope, there must be comparators in the time base generator . . .
IT AIN"T EVERYWHERE
VIEW TRACE provides a clean sweep of any signal on either input amp. This kills me. It means the head stage and attenuators, and the VIEW TRACE processing of the signal for triggering purposes are all clean. And the scope's grids are clean -- it can display a clean signal! So where the devil is this fixed amplitude crap getting injected? Maybe only one power supply rail is bad, or maybe one path (a short) that shouldn't be there has somehow been created, and it does not go to the early-stage signal amps/attenuators, and it does not go to the trigger/VIEW TRACE circuitry, and it does not go directly (irrevocably) to the deflection plates, but . . .
Is every power rail really good? Besides 1kV and 1.5kV outputs, the main supply (UNIT 5) has bipolar DC outputs at 6, 12, and 60 volts. An ADDITIONAL PWR SUPPLY (UNIT 24), besides a 6.3V heater supply, has DC outputs at 500, 200, 160, and plus/minus 130 volts. But please ask for schematics, ask for anything, but don't ask me to look at all these voltages -- I don't have a 2nd scope and the power supply is buried in a steel box inside the middle of the instrument, so I'm discouraged at the thought of doing a supply rails check. Does a company that writes a 3-lb manual (OK, it's in 3 languages) without **any** principles of operation or useful test-and-debug information provide technician test probe points for its power rails? Should I be looking for a row of gold posts somewhere?
Meanwhile, I bought in on the baked capacitor theory for the scope and bought my first ESR meter -- it already saved one instrument (paid for itself) by finding cooked capacitors at over 10x their max acceptable resistance. Stand by for the next easy fix. I opened an oscilloscope for the first time in my life. Let's just say after unending square footage of densely packed circuit boards, dozens of electrolytics, scores of individually-socketed discrete transistors, let's just say no audio amp or PC motherboard will ever seem daunting again.
CAPACITORS & CABLES ALL OK
There are no bad electrolytics. The scope shows its age, capacitors we would judge dead at 2 or 3 ohms are 0.7. 0.8 0.9 ohms today. But every capacitor from the dual 220ufd x 350V mains capacitors on down is OK. Surrounding the sea of circuit cards is a ring of subsystem interconnect cables, sometimes plugging into card sockets standing three deep. I treated every one with Stabilant, which has saved my butt with old equipment many times in the past.
One summer of hot storage permanently softened the "Nortel suede coating" on the case. (The scope was in a carton whose walls were stiffened with styrofoam sheets. I speculate that plasticizer volatilized out of such plastics changed the case coating.) I scraped off the suede gunk with a razor blade. Nortel deserves its bankruptcy for this coating alone.
With written notes, I got everything reassembled, every screw with its own lockwasher. I had paid attention to kilovolt voltages and to the power cord to be sure it was unplugged whenever it was supposed to be, but stopped paying attention to it after that. It was now wound around my leg, and I turned around to see the scope falling three feet onto the concrete floor. It didn't show a trace on the screen after that. Taking it apart was faster the second time around. I removed two of the three capacitors I had piggy-backed onto weak spots. You can't just add capacitance anywhere -- I had disturbed the transformer-isolated drive to the main chopper circuit. The scope now works fine, flood gun memory and everything else, but with traces thickened by a 200kHz oscillation. After hurling it onto the floor, my amateur technician's license is suspended, so I'm writing for help.
A $500 digital scope would be better in some ways than this one, and the voltages it takes to run an LCD screen can't kill you. But I don't have authorization to buy **any** scope. I'll have to accept living with a relic that costs $14,000 in 2011 dollars, but why do I have to accept phantom oscillations that thicken the display?
If you're going to flip a big piece of equipment around on the bench, bundle the power cord up with a fat rubberband.
Yours truly,
--Hen's Best Effort
Home
Help
Search
Login
Register
