Oscilloscoe Failed

[nevetslab]

A while ago, I disconnected the output from the Power Supply Filter/Rectifier assembly that has all of the buss caps from the Regulator PCB assy (nice Gold-plated Tektronix ribbon cable connectors between the assemblies everywhere). The DCV on C811 (14,000uF/25V) remained at -16.6VDC. That suggests to me C811 isn't healthy. I measured the DCR of the power supply xfmr winding feeding that full wave Bridge Rectifier that feeds both C811 and C813/C814, and get 0.22 ohms on both sides. 35mm dia x 85mm tall electrolytic cap can. Has four Ground support terminals and a single (+) terminal. I might be able to fit a snap-fit 10mm spacing 15,000uF/35V Electrolytic Cap in that area. Fully unloaded, I would have expected the DCV to then be equal/opposite that of hte (+) side at C813/C814.

[nevetslab]

I've now managed to extract the power supply module from the 7623A chassis, got all but the AC Mains cable disconnected. Not sure if this ends up being disconnecting the fast-on terminals from the multi-connection pushbutton-extened power switch, or instead, unsoldering the multiple wires of that cable. But, it's at least now in a position where I can do final test on the 14,000uF/25V Sprague Filter cap that's measuring low and suspect. I'll try connected my ESR Square-wave test to it and see what I find.

I haven't seen this filter cap format outside of Tektronix, and, fat chance finding any in that format out of distribution. There appears space where this suspect cap is positioned where I could install a current format snap-fit cap in there, with some shaped spacing to get it to set into place.



Some of the shielded cables that I had to remove were plugged into the BNC PCB assy on the rear panel assembly of this power supply module. One had to be extracted from the long Storage PCB on the inside of the CRT shield wall. So, careful notes and photos to aid in reassembly have been taken.

Onward.......

[Tom Phillips]

...There appears space where this suspect cap is positioned where I could install a current format snap-fit cap in there, with some shaped spacing to get it to set into place...

You may know this but I'll add this reminder that some Tek board layouts use the four ground terminals of the can caps as jumpers connecting different ground points on the PCB. Therefore, if you fit snap-in caps, be sure to check if you need to add jumpers to connect all the appropriate ground points.

[nevetslab]

You may know this but I'll add this reminder that some Tek board layouts use the four ground terminals of the can caps as jumpers connecting different ground points on the PCB. Therefore, if you fit snap-in caps, be sure to check if you need to add jumpers to connect all the appropriate ground points.

AH.....VERY GOOD Maintenance tip! I hadn't thought about that. AND, of course, getting to the bottom of that board is tedious.

This morning, I verified my suspicions regarding that 14,000uF/25V -15V Supply Buss Cap C811 which was only measuring -16.6VDC on it, while the two caps C813/C814, each 9600uF/30V measure 23.5VDC (disconnected from the regulator board in this case for both supplies).

Some time ago, I had set up an ESR Square Wave test, using 20V P-P as a source voltage, 5kHz, and applied that across the capacitor under test. You can both measure ESR (at that frequency) as well as Capacitance. Today, I was only using the test setup as a frame of reference to compare what appeared to be a working cap against one that is suspected to be bad. I then used my GenRad 1658 Digi-Bridge to measure Capacitance/DF on the two caps used in the +/- 15V Regulator circuit for the scope on the power supply buss cap assembly. The Test Signal/Counter and the source waveform on my Tek SC504 scope:



Then, applying this signal to the presumed 'good' Filter Cap bank C813/C814, each 9600uF/30V in parallel:



The 5kHz Square Wave source was 20V P-P, so there's a fair amount of current from this 50 ohm source. Now looking at that signal across the filter bank, we see on the 50mV/Div scale around 130mV P-P. Now look at what I get with that same source applied across the 14,000uF/25V buss cap C811:



Here, the scope is still at 5V/Div, and here across this 14,000uF/25V cap C811, we see around 11V P-P. Huge difference. Like there's very little capacitance here. Checking what I get using the GenRad 1658 Digi-Bridge at 120Hz test frequency, 1st image is C811,showing 68.3nF & DF of 2.76, then the next image is that of C813/C814 (22,323uF/1.13DF):



Just to check what square wave waveform I get with a 15,000uF/50V buss cap, I've applied that 20V P-P waveform to this used cap I had:



Here we see around 150mV P-P across this 15,000uF cap. I didn't take a photo of it connected to the Digi-Bridge, but it read 14,200uF with a DF of 0.74. I didn't record the ESR readings on these caps, which I can get on the Digi-Bridge, but this confirms C811 is shot. That would explain why there's only 16.6VDC on the main filter cap for the -15V Regulator Circuit. Gotta replace this 14,000uF/25V buss cap. For all I know, that might even make this 7623A scope with it's horizontal sweep system function properly. It's all out of whack at the moment, but this exercise with this 7623A's power supply is part of the step to get to pulling the failed scope out of the equipment stack and tearing into it, to see if perhaps I can do a module/board swap to restore the 7633 back to life. Ain't life grand?

Onward to the next step. Selecting a suitable replacement buss cap for that 14,000uF/25V cap. And see if what Tom brought up is involved in this step while replacing that cap.

I'll paste that earlier ESR Post I had done using this square wave source, as it's a handy procedure and quickly reveals problems such as this.

https://music-electronics-forum.com/...sr-quest/page2

In this link to a thread posted in 2014, I had found some measurement methods posted on the internet using a square wave and being able to both measure capacitance and ESR with reasonable accuracy (for maintenance work) such as this.

[nevetslab]

Just as Tom Philips warned me about, having 'temporarily' installed the 15,000uF/50V Snap-in Cap as a substitute for the bad original part, which has the four Ground terminals, and at this C811 location, IS USED to pass the -23.5VDC Unregulated supply voltage over to the regulator PCB assy on the other side of the instrument, behind the CRT assy....I forgot to install a jumper there. I had put the power supply assembly back into place, confident I had restored all of the wiring and hardware as it came out. and, right there on the PCB pattern for C811, there's a 4.7k bleeder resistor to ground from that cap-can connected negative voltage trace. So, it didn't arrive at the regulator board. DUMB!

Powered down, discharged the cap, prepared an insulated jumper and soldered that into place, cleaned up the flux, and checked again. I now have the -23.5VDC there at the bleeder resistor, but on the output of the -15V regulator? 0V. Before, I had +0.6VDC or close to that.

Sigh.........would it have been -15VDC if I had placed the jumper there as I should have? It WAS reading -12.5VDC before I found the bad C811 filter cap.

I unplugged the Time base and the two vertical plug-is, just in case, but no difference. Stupid is as stupid does.