Uncased oscilloscope and somehow broke it

[allanw]

I opened up my old Philips PM3256 scope to try and fix the power switch. Now when I turn the scope on, the trace only takes up a third of the screen, and I think that everything is compressed now. When I measure a 60Hz signal, it shows up as about 180Hz.

This is different from the issue I was having before, where the trace stopped showing at the edges of the screen, because in this situation, I can still use the x pos knob to move the trace around to most of the screen.

I undid the change I made (soldered the contacts of the switch so it's always on), but that didn't fix anything. I don't think I disconnected any wires or broke anything when I did the soldering, so I'm not sure what went wrong here.

I'm guessing it's an X deflection problem, but I was looking through the service manual and I couldn't find any obvious thing to adjust that would solve this. I can put up a copy of this service manual if anyone wants to look.

[djsb]

Hi,

Please download and read this tektronix training material.

http://www.sphere.bc.ca/test/tek-parts/troubleshooting-scopes.pdf

I have followed and used this manual on all my repair projects.

David.

[alm]

Agreed on the recommendation about the above link (the fact that Tek wrote it doesn't preclude you to using it on other scopes, they almost all work the same as long as they're analog).

The most likely problem is a connection in my opinion, you might have accidentally have touched a wire, and loosened a connector that just barely made contact. Verify all connections, remove and re-insert them in case they're oxidized (make sure to use correct orientation). Also check transistors/IC's in sockets. Suspect would be power supply or something in the horizontal chain, including the deflection wires to the CRT.

[allanw]

Ahah, I took a second look at it and very obviously, one of the five wires going into the CRT was disconnected. That fixed it!

[allanw]

Also, I thought this was pretty curious: the front panel knobs are connected to a plastic tube which turn pots that are all the way in the back of the scope:


I wonder why that's done instead of the pots being right where the knobs are at? I guess just for space reasons?

[djsb]

Hi,
Maybe also to prevent capacitive coupling of noise signals from the users hands. Problem is when those plastic couplers break they are impossible to replace.

David.

[EEVblog]

Those rods are often easier than taking bundles of wires to the front panel (along with all the associated signal integrity issues of doing that).
It can also be used for high voltage adjustments which are best kept on their board.

Given that these scopes were (and still are) analog devices throughout, many of the knobs and switches carry the signal under test, so you can't just run the wires willy-nilly.

It was a very common technique in scopes, and is still used in high quality gear. Very common for say the mains power switch.

There's a real art in designing analog oscilloscopes that doesn't just stop at the schematic!

Dave.

[DJPhil]

Oddly enough, I just got my first scope less than a week ago (Tek 465B) and had to fix a problem with one of those rods. The A channel variable gain knob was unresponsive. When I opened it up I found that the rod was fiberglass, and the coupler that tightened it to it's pot was broken on one side. I saw that I could fix it rather quickly if I drilled a hole in the rod, but I wanted to find a solution that didn't compromise the original parts so that it could be reassembled properly if I ever got a proper replacement coupler.

I used a . . . I'm not sure what to call it. It's used in cooking, wooden, about 1/8" diameter by 12" long, kind of like a kebab spear or a giant toothpick. I trimmed it to length against the fiberglass rod, used a pin drill that I use for circuit boards to drill a hole in it, and fastened it to what's left of the coupler. Works ok if I'm gentle with it, and it was enough to verify that the circuit works.

I assumed that if I used a metal rod (which I have a pile of) it would have caused problems with stray signal pickup. If I'd wanted to do a bit better job I'd have headed down to the hobby shop nearby and chosen a stronger wooden dowel of the right thickness, or perhaps even cannibalized a kite kit for an appropriate fiberglass rod.

What really stood out to me when I had the scope open was the high quality of construction and care given to details. I saw, for example, the exact device I had in my head when reading about temperature compensated transistor pairs: a small metal heatsink/clamp like device that thermally bonded two adjacent transistors together. It's obvious that Tek started with a good design and didn't allow heavy nickel and dime cost cutting to compromise the design. I suppose it's worth considering that (according to what I've read) the B series wasn't produced for long due to it's high cost of construction.

[djsb]

Hi,
Just for general interest and as a useful resource have a look at this yahoo group

http://tech.groups.yahoo.com/group/TekScopes/?yguid=330951138

Ok, it's for Tek scopes but it could be useful if anyone finds something on Ebay that needs fixing.

David.