Salvaged HP1201A oscilloscope repair

[ceoxrad]

So I managed to get hold of this T&M history for free!

The HP1201A was the first solid state analog memory oscilloscope ever sold I think. It's an interesting beast with a very small bandwidth (only 500kHz, yes you read right!) but a somewhat unusual 100uV/div vertical scale and differential input on both channels.

I was planning to restore this thing just for the fun of it but upon inspection I found a fault on the mode selection knob: the screen blanks out on every setting except AvsB (XY-mode) and channel A.
These two entries works fine but sometimes I have to twiddle a little the knob to get a trace for channel A mode (the AvsB works 100% of the time).

After a little bit of fiddling I observed that on the other modes I can get a dot on the screen if I put the timebase in "external horiz", in this mode the dots are responsive to vertical position.
In this mode the dots are strangely out of focus, this is very strange since the focus was previously adjusted while on the working modes and was perfect (?!)

Since XY mode and channel A mode are fine I thought there's likely a problem with the mode rotary selector. I popped the hood and in fact the switch contacts were very oxidised. I cleaned them with contact cleaner but sadly this did not cure the fault and the behaviour is still the same.

I'm starting to think that this fault is much worse than a simple dirty contact. Maybe I did not clean the selector enough but this seems strange.
I'm asking there for an opinion about the next step. Should I dismantle the switch completely to further clean it or should I look elsewere?

BTW i checked the voltage rails and they are all in spec so no problems there.

I was able to find a full service manual for this scope, http://bee.mif.pg.gda.pl/ciasteczkowypotwor/HP/1201A_B.pdf. but I'm not terribly sure on where to start looking for the fault: I think that the vertical amplifiers are fine and so are the power supply and CRT drive/memory, but at this point I'm uncertain on the entire horizontal section.

I'll attach some pictures of the oscilloscope and of the rotary switch

Any help will be appreciated and thank you in advance

[Tomorokoshi]

Rotary switches of that time are usually silver plated. The silver tarnishes to silver sulfide, which has very high resistance. Removing that switch is possible, but it will be difficult with everything soldered in.

I don't know what the best method is at this time to remove the tarnish.

[bob91343]

It would be a lot of work, but you could, with the aid of another oscilloscope, verify each switch contact to see if the same signal is on both sides of a pair of contacts.

But signal tracing is still probably your best approach.  Put in a signal and see where it goes.  If the sweep doesn't trigger, then trace the trigger signal.

[Khaveer]

You may have to pull the switch out and clean each contact individually. A few years ago I've rescued a scope like this one and it's older brother - the 1200A. Both scopes had serious issues with most of the switches. It took quite a while to clean every single contact with a cotton swab and some IPA.

Unfortulately I'm pretty sure that in order to reach the mode selection switch you'd have to tear down most of the scope.

[ceoxrad]

Thank you all for the answers, these days I was busy with work but i'll check out soon and at this point I think I'm going to dismantle the switch and try to clean it further  . If this does not cure the problem I'll try to trace the signal while in fault.

For the cleaning I've done earlier I used an old bottle of "Electro CRK" contact cleaner for electronics, it's an italian brand and I think is not sold anymore, when I tried to clean the contact with a Q-tip tbe gray-black tarnish came right off and the surface was shiny, so I think I'll use that and IPA for the final cleaning.

I forgot to add this in my first post: when I operated the selector with the scope opened I saw really tiny sparks on one of the wafer contacts. Do you think it's normal or maybe another fault? (the sparks are really small and best seen in darkness, but sometimes you can hear the crackling)

[jdragoset]

Be very careful of anything that rapidly removes oxide/sulfide, as it is likely a strong electrolyte and must be completely removed with IPA or de-ionized water rinse.
A better bet would be De-oxit followed by a very vigorous workout.
Those double-tension contact fingers, unless corroded and weak, will likely exert enough pressure to clean the contact track that counts of the rotor disc.

[ceoxrad]

Some updates to this repair: I think I've found the (hopefully only!) fault.
I tore down the board on which the selector switch resides, it wasn't that bad apart the ton of unlabeled wires I had to label beforehand. The tiny setscrew holding the shaft of the knob were completely stuck but I eventually removed those with a bit of WD40 (put on with a Q-tip, being very careful to not stain the board).

After the board was out the inspection was way simpler, on it I found some fingerprints on the underside, and the screws holding the selector switch were a bit rough, someone had a go before me  .
Anyway I buzzed every contact on the switch and to my surprise there is a short on one of the wafers. I found that one of the spring contacts is badly bent and makes contact even when it should not!
For now I cleaned the switch better and I'm thinking of removing it from the board to try straighten up the leaf contact. I'll keep this thread updated

[ceoxrad]

I desoldered the switch, tore down the contact further and straightened the spring.
I also took the opportunity to clean the wafers with contact cleaner and then IPA as suggested.
Upon testing I'm glad to say that now ALT and CHOP works well (no sign of life before) and on channel B i get a dot but no sweep, at this point I'm very confident that the fault is in the switch, sadly that spring was not the only.
Next days I'll disassemble the switch again and check the contacts better, maybe I missed something

[ceoxrad]

Okay so I disassembled the swith again and as expected there is another problem that I missed.

While the spring I straightened last time is still fine there is another wafer that is very intermittent.

I hope that this description is readable (not my native language...):

There are two spring contacts on it that I think are supposed to connect to the central part of the rotary plate, that just barely touch it in all positions. From what I can see the mechanism is slightly eccentric (may be tolerances of the switch) and so the contact gets worse over the course of rotation, the extreme position is channel B and there is no contact at all. I can also see a little score mark on the outside of the rotary plate where those springs slide, maybe this poor contact was there for a long time and eventually reached the failure point.

Now I'm a little uncertain on what to do next.

I was thinking about trying to extend the springs by compressing them so they make contact every time, but those are very fragile and before making damage I want to be certain that is a good idea.

I also want to be sure that the contact to the central plate is right, I based my assumption on the fact that there are two different sets of spring contacts: the short ones connects to fingers protruding from the central disk and the longer ones connects to the disk itself, the problematic are long ones.

I attach some microscope pictures of the problem (wafer in Channel B position), I hope that are clear enough (i used a crappy USB microscope)

[Tomorokoshi]

For the switch section that is failing, which position and side is it? See the illustration on page 78.

Can you also post a picture of the other side of that switch section?

I am having trouble downloading pictures. Everything is from the last post.

[ceoxrad]

Sorry for long absence, I have a lot going on with my job and could not spend a lot of time. The wafer is 5F w/ P78 notation
About the pictures IDK what happened, I think the forum somewhat messed up and only saved the pictures of the last post. I'll put a link to google drive with everything here
https://drive.google.com/drive/folders/1E8LJasilnMkmHv6qvXZoT2rv0Gy0Kzfj?usp=sharing

[Tomorokoshi]

Ok, that seems to match 5F and 5R on page 94. Those contacts were either too short originally or they wore down over time.

I see a few options, none of which are easy. I also haven't tried any of them. Perhaps others have tried? There might be something in here:

https://www.eevblog.com/forum/repair/vintageclassic-renovation-techniques/

1. Replace the wafer with a spare from ebay or some other place. Availability is difficult.

2. Extend the contact by soldering something underneath. Difficult.

3. Replace the extended contacts with parts scavenged from other switch wafers. Requires similar parts. You would need to drill out the rivet, and somehow install the new contact. A solid wire wrapped around the rivet location and soldered in should work.

4. Adjust the position of the worn contacts. Similar to #3. Drill out the rivet. Extend the mount hole in the wafer. Reinstall the contact. Need a way to keep it in place without moving back. Use a soldered solid wire to replace the rivet.

[ceoxrad]

Thank you for the answer, I had still the wafer on my bench and I tried to gently move the leaves of those contacts back onto the central ring.

That was quite easy to do with a toothpick and some patience, they went into contact to the ring and upon reassembly of the switch were still having contact (checked with multimeter).
So I took the gamble and reassemble everything, sadly seems to have worsened the problem , now I don't have sweep on Alt, Chop and ChB.

I think I'll start to look for a replacement switch, if I have a little time these days i'll tear it down again to recheck