I thought it was about time for an update.
WARNING: POST CONTAINS MODERATE SCENES OF DODGY HACKS AND CRUSTINESS Ok, so since last time you saw it I disassembled it completely. All the metal surfaces were polished with Brasso using a drill fitted with a polishing wheel. If I were to do it again I would probably try out Silvo, although Brasso did work well. The faceplate was just washed with soap and water then waxed using a light car body wax. The black plastic body was just washed then coated with ArmourAll.
For the BNC connectors I used a q-tip/cotton bud to apply the Brasso. Then I just added a small amount of DeOxit Gold to the BNC and rotary switch contacts.
I rewired the internal mains wiring with 600V, 18AWG hookup wire as discussed previously. I then coated all the live parts that could not be heatshrunk with Liquid Electrical Tape and replaced the fuse cover with a thicker PVC one.
I then made an adapter for an IEC C6 socket to replace the old Oval/PH-163 type. But there is quite a large gap between the socket and the outer cover, so I had to do some 'precision adjustments' to the IEC power cord for it to reach into the socket.
I know it looks pretty dodgy, I'll probably redo it one day when I have a crimper etc. However, I didn't modify the chassis at all so it can be reversed at any time.
I didn't do much to the circuit boards, just replaced the suspect resistors. Then I gave them a light rinse with Methylated Spirits, dried them on cold with the hairdryer and put them in the Australian sun to dry.
But they later got little flecks of dry Brasso on them... ergh... oh well, close enough I tried.
Oh, and I got a whole box of replacement lamps for cheap, so no need to convert to LED, plus these look warm and retro.
On to the test results. At first it was clipping the tops of the sin waves for some unknown reason, but then it fixed itself. A problem that fixes itself is my favourite type of problem (as long as it stays fixed). Also, unfortunately the only BNC cables I have are cheap and dodgy, so they might have affected the results.
Operation at
0.2Hz (which is the minimum limit):
1KHz:
10KHz:
2MHz (max limit):
The rise times of all the square waves from 1kHz-2MHz was about 36nS, so was getting a bit less square at the higher frequencies.
The sweep function seems to work, but the rotary switch for 'Sweep Time' doesn't rotate past 100ms. It looks like there is actually a detent to stop it rotating past that point. I'm not sure why, maybe it was replaced with the incorrect part at some point?
Still to do:
-> Find 3 suitable replacement knobs and repaint the rest
-> Fix the Sweep Time rotary switch
Anyway, at least I now have a function generator that 'sort of' works. Let me know if you have any advice or want me to test it under certain conditions.
Edit: Here's a scan of the front panel in case it's of use to someone.
https://www.dropbox.com/s/s66s8b2d80kscm9/WAVETEK%20134%20FRONT%20PANEL%20SCAN.jpg?dl=0