Scope on the bench ready for refurb.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
These old scopes are works of art. I have four 545's and my main scope is a 585A. These are not really beginners scopes to work on, still, you may get lucky and find there's not much wrong.

As mentioned like any bit of test gear this old bring it up on a LL and variac, slowly, to reform the big electrolytics. Some will need changing as will maybe a few of the Sprague black sausage shaped caps.

I'd give all the switches a squirt of cleaner first. Any soldering must be done with hot iron with the right shaped bit and 3% silver solder, there might be a little roll inside the scope.

Here's some video's that might help - YouTube not your model but similar circuits, he did another video covering the X amp. This one covers the wafers and solder - YouTube

Try and go slowly with scope and not make too many mistakes, they are complex scopes and any damage you do, you may not be able to fix.

Good luck, Andy.
 
Thanks Andy. Just a couple of quick clarifications. Is the silver solder needed for all connections or only on the ceramic strips? Is a 30W iron good enough or should I pick up a hotter one? Would it be bad form to pigtail solder (clip out old components and solder to the original leads without disturbing the original joints.?

My guess is that the main filter caps are not leaking badly yet or voltages would (I think) be running on the low side. I will start by trying to set -150V. If it won't set or requires extreme adjustment I could try changing out the tubes in the error amp (12AX7 differential amp and the 6AU6 I presume) and see if that makes any difference.

I suppose that replacing all the PS caps is probably warranted fairly early in the process.

Curious that the -150 being numerically high would cause the pos supplies to go high. I would think the more negative -150 ref would tend to pull thm lower. Even the unregulated 235 is high but only by about 4% which could just be line voltage creep since 1960s.

Oh, one more thing. Is it OK to.operate with the unit on its side with bottom plate removed to check voltages (eg. Plate voltages of 12AX7)?

Just got the chance to check out the videos. Turns out they answered the above questions by and large. Probably ought to invest in a beefier iron.

Actually the videos answered most of my questions (except the pigtail) probably ought to get a bigger iron.
 
Member
Joined 2004
Paid Member
You should not need a bigger iron for most connections. You do need the silver solder for the ceramic strips. There should be some in the scope already. A bigger iron would be for heavy connections but those are unlikely to be high on the list for now.

The supplies will track as in a higher negative voltage will lead to a higher positive voltage. The caps are old but replacing them won't be real easy. If you have an ESR meter you can find the worn out caps pretty easily. The 400V and 500V caps will need some searching and will most likely be in a different package and 1/4 the size. I just dealt with that on a Tek 5112 where the 5V regulator was down to 3V. The cap had "evaporated" and the DC was 6V where it needs to be 9V. 35 years and it failed! The others are OK.

Most likely once the supplies are correct and you undo whatever amateur adjustments were made it will work fine. The tubes are not stressed in Tek designs.
 
I've found that on old valved scopes changing the valve/tube fixes the problem most times.

Wouldn't change the PS caps unless there is a lot of ripple. Check your rails for ripple, this could be the cause of dodgy readings. Just put your DMM onto AC volts.

It's ok to put it on it's side for a short while, but the scope needs a lot of ventilation. there's an auto cutoff if it gets too hot.

Re your iron, you might get away with it, but I've found a hotter iron better. Wouldn't pigtail unless I had to, be a shame to put a bodge in amongst all that beautiful hand soldered artistry. Should be ok to get things working though.

As 1audio says, there should be a little roll of silver solder inside the scope, on one of the inside walls somewhere.You can use ordinary solder for a temporary fix to tack another component in place if your quick and don't heat the joint up too much.

Good luck, Andy.
 
Turned the -150 adjust 1/4 turn either direction with absolutely no change in output voltage.

Basically I just have to take a few minutes each day before I shower and get ready for work so we are kind of picking at it. I did not see any heater glow on the front row so changed out the 12AX7 fo a new Chinese 12AX7B. It lit up fine but reading only changed by 1 V and still no response to adjustment. My guess is that the original Telefunken just has the heater buried where you can't see it. The 12B4s in the back all light up fine.

I am not very good at regulator circuits but my take on this one is that the 12AX7 and 6AU6 comprise the comparator and error amp with the 6AU6 driving the 12B4s which provide the actual voltage drop. One possibility that occurs to me is that the adjustment pot itself may have failed. I think that I can reach the terminals of the pot with an ohmmeter so I plan to check the resistance to ground of each terminal (power off). I should be able to detect open wiper, open or shorted resistance element and value if R615.

There are a lot of carbon comps in there so it is possible that some R values are way out.
 
Member
Joined 2004
Paid Member
You can operate the scope with no plugin but the verticle will have no control. Keep the intensity down. It's a quick way to confirm that the plugin is not overloading the supplies.

The reason I suggested checking vert vs horz is to see if the excess deflection is in both axes. That would indicate an HV supply problem. Otherwise it may be an amp problem.

If the verticle 1V deflect 2V and horz 1 mS defects 2 mS then its HV supply. Do you have a high voltage probe? And be very careful around it.
 
I don't have a HV probe. My VOM only handles up to 1000VDC. The vert is completely off screen. I can bring the horiz on to the screen with pos all the way to the right. So at the moment I can't really compare deflection. The only reason I can tell that the horiz is (I think) getting somewhat on screen is that there is an out of focus tail and the off screen indicators for horizontal go out. The off screen indicator for vertical never goes out.

Since the horizontal is all contained in the scope itself I was thinking that tackling that would be an easier starting point than vert.

At early on in the debugging process I did briefly get a short trace that appeared and then processed off the screen to the top but after a short period of time even that trace stopped showing up and it just went immediately to an off screen indication.
 
I started looking at voltages. The schematic shows two volatges for some nodes. Perhaps this is a range of voltages dependant on switch positions. For example the 1st stage I looked at (V343B) lists +80 and+103 at the cathode. For their 225V B+ I get 229V and the -150V is spot on. I get 106V at the cathode.

Capture+_2018-07-27-10-37-37-1.png

The plan is to go through in like manner mapping as many important node voltages as I can reach.
 
Last edited:

PRR

Member
Joined 2003
Paid Member
...My VOM only handles up to 1000VDC....

You should own a DMM. (And IMHO a VTVM.) However a professional passive VOM will work for many stages of this 'scope.

All the deflection voltages will be under 1000V. If you have any spot brightness, the EHV is not zero, let it be for now.

Yes, one side may be 80V or 104V. But the *differential* voltage across two sides of one push-pull stage should be quite small to the final deflection amp, only ~~100V at the final stage to the CRT plates. Wild off-balances are strong clues.

I would (as I think has been said) get the main regulated voltages right *first*. Most 'scopes derive all the stage voltages from predictable regulated voltages. In some H-P this was a "-105V" reference, multiplied around to give +350V etc etc (even the 1200V).

If it has a neon gas reference, go there. These go bad in mere decades. Fail to start. This may cause a high voltage, but designs vary. Shining light ON the gas tube may start it. Brushing a little radioactive on it would help but is not really a home remedy. Vintage tubes may also be old/cranky. At some point it may make sense to drop a Zener in.
 
I should be more careful with terminology. My meter is a digital mulimeter.

The LV PS uses a 5651 reference I did notice several neon tubes in there to however. The PS is dialed in. The 545A manual unlike the B manual has some troubleshooting tips and they suggest a 100ohm shorting jumper to find the offending stage by jumpering across the output of each stage starting at the finals and working back. Does that seem like a worthwhile process?
 
Last edited:
I decided to change it up a bit and take a stab at the vertical this morning. It was fairly easy to get to the tapped inductors through which drive is applied to the CRT plates so I started there. As expected the voltages were asymmetrical one lower that the schematic value the other higher. The next convenient spot to check was the bases of the transistors which comprise the input to the stage. These voltages were only slightly off from the schematic value and they were identical. That seems like strong evidence that the problem is in this stage.

Vert: (sb=should be)
L599 tap at 280 (sb 305)
L589 tap at 335 (sb 305)
Base of Q584 at 175 (sb 171)
Base of Q594 at 175 (sb 171)

Capture+_2018-07-28-07-45-56.png

Next I swapped the two output pentodes to see if the imbalance followed the tubes. It did not but remained the same. My conclusion is that either there is a problem with one or both of the transistors or with some passive components in the stage. The next logical step is to switch the transistors around but I didn't have time to trace out the collector and emitter connections with sufficient certainty (obviously transistor sockets are not keyed like tubes) so instead of risking misconnection that will have to wait for tomorrow.
 
Member
Joined 2004
Paid Member
I would start at the input on this circuit. The neon lights are saying that the trace is off screen and they are in the second stage. The input to the followers is supposed to be at 67V (so much for running with the plugin removed). I would see what the voltages there are and if they can be adjusted. Much of that circuit is low impedance current mode stuff so small voltage changes. The delay line is obvious for the triggering but the inductors are part of a shunt peaking circuit Tek used to boost the HF response. If you can check the E-B voltages and they are around .7V then the transistors are probably OK. You may find a lousy connection at a transistor socket so wiggling them is valid.
 
Good point on the indicators. I noticed the 67V inputs. My first thought when I saw that was that a 67V source would have to ne jumpered in there if one wanted to remove the plugin. I should be able to get to the emitters on Q584 and Q594 faily easily. The current mode does make this a bit trickier.

Since the input stage is a follower the input voltage should be meaningful. I wondered about all those inductors. They were quite baffling to me.

If those input voltages are out it of course points to the CA unit but if they are right it gets tricky. Normally with a CC amplifier stage I would take the voltages at the plates to be a reliable indicator but the second stage (V514) looks more like a Common Grid amplifier with the input applied to the Cathode and it is driving less than 2.2k load p-p (R552 in parallel with the delay line) and I am not sure whether Q513 and Q523 might not be providing some drive as well. As I said a bit tricky circuitry for me.

I appreciate the expert guidance. I will follow up on your suggestions.
 
I did some checking at the input of the vertical amp today. Had a real scare. I am still not sure that I didn't blow something. I measured the voltages at pins 1 and 3 where the plugin connects which are the inputs to the cathode followers and also at the bases and emitters of the transistors of the following stage with the following results.

Vert amp input pin 1 (top triode)72.9V (sb 67) pin 3 lower triode 73.6V (sb 67) = 0.7V difference

Base of Q514 76.2V (sb 71) and Q524 77V (sb 71) Q524 emitter 76.1 Q514 emitter 75 = 0.8V difference (1.1 at emitters).

In measuring Q514 emitter I used an insulated clip thinking I would be being more cautious but when the relay kicked in I heard a snap and saw a wisp of smoke. Blast!!! I shut down right away. After a couple of minutes of kicking myself I ditched the clip, powered back up, and probed by hand getting the above measurements. Also still got the same base voltage measurements as before the smoke signals so hopefully didn't fry a transistor.

Voltages are running consistently about 5V high but it doesn't seem like the differential voltages are unreasonable but I am a rank green horn with these devices. Unfortunately I don't have any of those super special pricy 7119 triodes laying around to sub in. For that matter I am out of 6922s as well (do have a few 6N1Ps).

Any thoughts on the results thus far? anything to be gained by shorting the input pins 1 and 3 together (with or with out CA unit installed)?
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.