B.W.D. 521 oscilloscope

BWD Electronics were a Melbourne, Australia, company that made test equipment in the sixties and seventies. Many years ago this unit was saved from a fate as landfill, and served as my introduction to "serious" test gear. Never having had any training with this sort of thing, I was never sure if it was fully working, but it was certainly good enough to signal-trace audio, helping me fix synths and FX units, and teaching me along the way.
And then one day it went blank.

It was a daunting task for a novice, but I had some hand-holding from the SOS DIY forum where a couple of regulars wised me up to EHT (extra high tension) circuits and how they work with CRTs. Unfortunately not before I cooked my multimeter poking around the 3000 volts area. Oops! 

The EHT board...

.. and the reverse, with the twin coils

the coil mounting

Luckily, just a couple of resistors took the brunt of that. I was advised to build a high voltage test probe, by soldering a chain of large resistors and covering them with an insulating tube (I used an old biro shell) to prevent arc-overs. This would allow me to safely measure the EHT circuit points to see if anyone could suggest a fault.
Several people did mention they thought it an "unusual" circuit while helping me trouble-shoot.

I plodded along testing, replacing, testing, replacing... It was difficult for me to understand exactly what voltages to expect on the transistors 40 - 42, perhaps because they make up the high frequency oscillator, so I probably replaced a few passives in and around here that did not need replacing.
And then I had a breakthrough - under the mounting for the twin coils I discovered a broken connection. Fixing that brought back 2500 volts to the CRT and the traces returned. The spec said it should read 3000 volts, but it seemed to work much as before, perhaps one trace was a little less bright now.

Emboldened, I took a look at the other side of the machine, where the various low voltage supplies were derived, and the plug-in frame, at the rear in a delightfully inaccessible area near the massive transformer.
I was hoping to fix a few quirks in the operation, notably in the timebase selection.
A few voltages were off-spec, and a few associated electrolytic caps were showing high ESR readings. Given the age of the machine I bit the bullet and replaced every single electrolytic cap - where I couldn't get the approximate values I combined various values in parallel or series.
There are still a few flakey functions, but generally it's a joy to use.

The rear board for the plug-ins

The low voltage board...

... now with a hodge-podge of electro caps

... some even making it to the flip side of the board. Lucky there was plenty of room!