Philips PM3218 repairs

[samnmax]

Hi all,

Here's my repair of a Philips PM3218 35MHz oscilloscope, bought "as is" on German eBay. For a 35 year old device (TTL chips are from 1979) the condition was not bad, however it needed some repairs.

If you have to repair one, get also the service manual of the PM3217 (50MHz model), as it is much more complete with detailed explanations, and contains some components that were bodged and are missing on the PM3218 manual. As far as I can tell the component references are the same.

Vertical section:
-  Channel A not working (no trace) => Open BF450 transistor, reference V508. Replaced both V508 and V509 with Siemens-made replacements.
-  Channel B signal too small in 10mV, 5mV, 2mV positions => Faulty reed relay K601, replaced reed switch inside.

Horizontal section:
-  AUTO trigger mode not working (no flat trace when no signal present). Triggered when a signal was present, although not very stable. Also trigger slope switch not working in - position => Shorted BAW62 diodes, references V1016 and V1203 (this one is bodged, does not appear in the PM3218 manual but does in the PM3217). Replaced with 1N4148.
-  X magnification not working => Faulty reed relay K1401, replaced reed switch inside.
 
Controls and switches:

- Power on switch failed while repairing, I though I killed the power supply  => just open the power switch and bend the metal inside a bit.
- Funky controls => Sprayed contact cleaner, much better. However the input AC/DC coupling switches are still unreliable.  They are a pain in the ass to remove so I guess they will stay that way.
- Corroded delayed timebase vernier => Submerged knob parts in dissolved hydrochloric acid. Not as shiny as new but usable.
- B channel invert and DTB trigger slope switches completely stuck => Again some hydrochloric acid in the shaft to dissolve corrosion. In the attached pictures it's still missing some knobs, they are in place now.

Questions:

- Any idea how can I do a simple calibration? I don't have a signal generator, just computers, Arduinos, stuff like that. I already adjusted the MTB sweep speed to better match a frequency generated by the computer. The 1.2 Vpp calibration signal seems to be OK.

- Can the oscilloscope measure its own voltage rail ripple? For now I haven't replaced any electrolytics, voltages appear to be stable and my meter on AC does not measure significant ripple.
This is my first scope, I'm waiting for the probes to arrive. For the moment I've only been playing with it. Youscope and Oscillofun look great!

Thanks!

[Cyberdragon]

It shouldn't need to be calibrated too much, they are usually fine once they are working. But if you want to you will need a signal generator and the probes. As for power ripple, yes, you can see it because it messes up the CRT trace. Just turn it down to it's lowest input and highest zoom if you have it. (don't connect anything to the input) It should be perfectly flat, otherwise there is ripple.

The only thing you should need to calibrate is the probes, which the manual should tell you how using the CAL terminal.

[jancumps]

I used to have the Philips PM3305. That is in essence your scope with dual time base taken out, and digital section with two additional chanels added.

[samnmax]

It shouldn't need to be calibrated too much, they are usually fine once they are working. But if you want to you will need a signal generator and the probes. As for power ripple, yes, you can see it because it messes up the CRT trace. Just turn it down to it's lowest input and highest zoom if you have it. (don't connect anything to the input) It should be perfectly flat, otherwise there is ripple.

The only thing you should need to calibrate is the probes, which the manual should tell you how using the CAL terminal.

OK, thanks. Since I don't have the equipment, I think I'll just calibrate the probes once they arrive. As for the ripple, it is less than 1/5 div at 0.2mV/div, I guess that's good enough.

I used to have the Philips PM3305. That is in essence your scope with dual time base taken out, and digital section with two additional chanels added.

Nice! There is a lot of room inside of mine, so I guess other models added other functions like digital functions.

[oldway]

Well done...
You can calibrate the dc gain with a power supply, a precision resistive voltage divisor and a multimeter.
For AC calibration, you need a square wave generator (fast rise time) with a 50R load.