H 8569B Spectrum Analyzer: some progress, and a few more items to fix

A brief summary of the somewhat tedious job of fixing the 8569B front controls, which are a great feat of engineering but the plastics are prone to aging:

First, replacing the gear, it is a 48-pitch, 48-tooth spur gear, glad that I had one spare, but they are still available:
8565a gear
8569b gear

Next step, some contact cleaning, using a soft eraser, and some isopropanol.
8569b cleaning contacts

The frequency control – for some reason, there are different versions around, one that has wires attached, and one the has pins – interestingly enough, the 8565a control fitted has wires, but the 8569b requires pins – also here, good to have a spare assembly around, with the pins…
8569b frequency control

Defective bias pot – also here, a spare fitted.
8569b bias pot

Some of the contact fingers broken off-also there were fixed, and everything put back together.
8569b back together

Finally, noticed that the 10 dB segment of the 70 dB input attenuator (5086-7365) is stuck. Nothing dramatic, just one of the little O-rings holding the contact actuators broken off, and parts of it stuck in the contact.
8569b 5086-7365

Finally, a quick test – everything seems to be working-
8569b display

-unfortunatly, still missing something – the sweep time indicator is not showing a time denominator (µs, ms, sec), and the analyzer remains in the digital mode for all sweep time settings – it should switch to mixed analog-digital mode (at the time the 8569b was designed, there was no easy way to run analog-to-digital converters any faster than a few kHz without causing exorbitant cost, so the 8569b uses the digital storage mode only at the slower sweeps).
8569b display missing

Remaining items:

(1) Identify the issue with the sweep time indicator and missing transition to mixed analog-digital mode. Maybe related to the 8565a control fitted to a 8569b? – Checked by substituting a control assy from a good 8565a – working just fine with the 8569b – accordingly, this is not the reason for the fault. There seems to be a defect on the main front panel board. Maybe a bad switch contact, or a broken trace. Will be quite a pain to test will all the cables, switches and screws.

(2) Manufacture the control knobs for the frequency-bandwidth-span-atten-ref level setting. These were missing and don’t have spares at hand/these wheels are getting brittle anyway. Will make a new set from aluminum alloy, during next stay, in Germany, the only place with access to adequate machine tools.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.