Flicking white screen and buzz from PSU - steps to troubleshoot?

[Lunet]

Old Vic board (2 prongs PSU), symptoms on power on: 1) white screen with rolling shaded black bar, sometimes it looks like it tries to align the screen and jumps, PSU buzzes - quite an audible sound. I do have a working board, although tad a bit newer and the very same PSU when connected there makes almost no sound at all. I've tried swapping the Vic chip - no dice. Could buzzing sound mean that PSU is overloaded and we draw too much? Any good way to test voltage regulator or caps that would be most prone to failure in such a situation? Ideas to troubleshoot from here welcomed.

Thanks!

[Mike]

You own a multimeter and know how to use it?

I've tried swapping the Vic chip - no dice.

You hopefully only tried to test the suspect VIC chip in your working unit.

The VIC is the only irreplacible chip in the VIC-20. If you check a working VIC in the faulty VIC-20, whatever might have killed the VIC chip there could also kill the working VIC chip, and then you have *two* dead chips!

Could buzzing sound mean that PSU is overloaded and we draw too much?

That is quite likely and not an operation mode the PSU endures for too long.

The symptoms indicate that either the +5 VDC supply rail or the 9 VAC supply are shorted inside the VIC-20. Is the power LED rather dim or even off? Please check the two voltages.

When the +5 VDC rail shows excess voltage, turn *off* the VIC-20 as soon as possible! The voltage regulator then is fused and that can kill all chips within short time.

When the +5 VDC rail measures too low a voltage (also power LED dim/off) - if you keep the VIC-20 powered on for more than 1 minute, does any of the chips or the rectifier get so hot that you can't touch it anymore? Then that one is likely the culprit. A fused rectifier will also show as partial short and low reading on the 9 VAC power rail.


Edit: see also here: http://www.sleepingelephant.com/ipw-web ... php?t=8303

[Lunet]

Yes, I do have a multimeter and I know how to use it. Fortunately faulty board did not fry working VIC chip, I've moved it back to the "test working board" and all is fine there.

Power LED looks fine, AC coming to the bridge rectifier is 11.15V and DC coming out of it is 8.2~8.6V - too low? I would expect the same voltage here as on the AC side of the bridge, right? After voltage regulator it is 4.6V (also a bit lower than 5V)

EDIT: Bridge rectifier heats up like crazy, I guess changing it would be the first step.

[Mike]

Fortunately faulty board did not fry working VIC chip, I've moved it back to the "test working board" and all is fine there.

O.K.

Power LED looks fine, AC coming to the bridge rectifier is 11.15V and DC coming out of it is 8.2~8.6V - too low?

Yes, that's definitely too low. The original voltage regulator needs a higher input voltage to work properly (around 10 V at least).

I would expect the same voltage here as on the AC side of the bridge, right?

Not quite. The AC reading you get from the multimeter is the RMS value, i.e. the equivalent DC voltage that would result in the same power figure in an ohmic load.

Now, for a sinusoidal AC voltage, the peak value is sqrt(2)*VRMS, i.e. if your multimeter tells you 11.15 VAC, the peak value actually is 15.8 V. That is also the voltage you would get on the stabilizing capacitor if rectifier and capacitor had no load attached.

With a load after the capacitor, there's also current, and this current adds two diode voltage drops in the bridge rectifier, i.e. around 2 x 0.7 V. Thus you could expect a maximum voltage of 14.3 V = (15.8 - 2 x 0.7) V on the capacitor. In practice, you'll see a lower DC figure as there is significant ripple on that voltage and 14.3 V then is really only the theoretical peak value.

After voltage regulator it is 4.6V (also a bit lower than 5V)

That's definitely outside spec.

EDIT: Bridge rectifier heats up like crazy, I guess changing it would be the first step.

Yes.

Good luck!

[Lunet]

Thank you! I've ordered the new bridge rectifier an as soon as I get my hands on it I will give it a try and report back. Fingers crossed it's only the bridge!

[Lunet]

I've swapped bridge rectifier and now I can see clear 5V after voltage regulator and 11.80V before filter capacitor - that seems to be good and the two problems got eliminated 1) no more buzz from PSU 2) Screen does not jump, however, screen only shows full screen white... any suggestions where to go from here?

EDIT, tried testing VIC chip and 6502 - it is not it

[Mike]

tried testing VIC chip and 6502 - it is not it

From what I had gathered, you already proved the VIC chip of your defective VIC-20 working in your working VIC-20, not much gained by exchanging two known good components ... ... it is more informative to see if a fault goes along with a suspect component.

Next you should try replacing the UB9 7402 chip (master oscillator). Note this must be a standard TTL chip, not LS.

Then, the next candidate for replacement is the UB10 74LS74 chip - if present: with PAL VICs, this chip serves as toggle-FF to halve the oscillator frequency if a quartz with twice the colour carrier frequency is used.