Hewlett Packard 5090A Receiver

[factory]

The HP 5090A was the first instrument designed by the HP Bedford site, it was an off air receiver that used the Droitwich 200kHz radio signal as a frequency standard, of course the frequency changed to 198kHz decades ago, making this and other off air receivers unusable without a converter.

The 5090A has a 100kHz oscillator that locks to the 200kHz signal and a multiplier to provide a 1MHz output, it can also be used for phase comparison of 100kHz or 1Mhz signals.

It's mentioned in the Nov 1963 HP measure newsletter.
https://www.hp.com/hpinfo/abouthp/hi..._11.pdf#page=5

This is actually the second of these I acquired, the first had shorted turns causing the transformer to overheat.
For this one I started with checking the Zener diodes in the power supply, that had failed in the 5090B & first 5090A, they were OK in this one. Next I checked the red Plessey capacitor on the same PCB, unsurprisingly it was open circuit, it's one of those plastic cased ones.



After this I noted in the manual, that there were more inside the oscillator box, I ended up with this mess on the bench to gain access and you've guessed it more dead plastic cased Plessey capacitors, these were all replaced with the exception of the orange one, which tested fine.



A bit of reassembly later I was able to test to repairs, all the power supply rails tested good, but I was having trouble getting a signal to lock, this turned out to be the frequency of the Marconi signal generator drifting too much for testing (even after hours warming up). Also I made a slight error when reconnecting the 1Mhz & 100kHz BNC outputs, somehow I managed to swap them.



Some time ago I created a converter PCB, it's based on the design by G7EAH in the Radcom magazine from Nov 1989. I had lots of trouble trying to get this to work last time I tested it, which turned out to be related to the wiring of the ferrite rod antenna and it's tuning capacitor.

Finally I got it assembled ready for testing & was surprized it was working fine, the RF alignment of the 5090A is spot on and the 5327B (with oven option) measuring the 1Mhz output settled to exactly 1MHz, I've never been able to check my counters before.



Hopefully I can get the 5090B working too, but I might not be able to if it needs anymore repairs, as I haven't got enough free time to finish reverse engineering it, before the end of transmissions on 198kHz and no manuals seem to exist for the 5090B.

David

[Julesomega]

What a lovely little unit, and delightfully "Home-made" with the Eddystone box at the centre. That it won't lock to a free-running signal source seems like an endorsement of the electrical design.
Droitwich is scheduled to remain in service for several years - how long do you need?

[factory]

5090A

It is a nice unit, not too big, there is another smaller Eddystone box at the back that contains the RF amp section. Guess I thoroughly tested the failsafe function too.



5090B

For the 5090B, I've been sent some diagrams from a member (who scanned the 5090A book ) on the HPAK groups.io, originally sketched by a member from the Netherlands.
Also time is no longer an issue, but for all the wrong reasons, currently out of hospital recovering, since having a heart attack at work last Tuesday. Can't play with the heavy boat anchors, but some drawing won't do any harm.

These are a few bits of the reverse engineering I did previously on the 5090B, the RF amp is much the same as the A version but with Si transistors instead of GE ones and a bit of comparison of the PSU layout. Sadly one of the files, containing notes on motherboard got corrupted.

HP 5090B 05090-701 Rev.A PCB.pdf
HP 5090B Comparison of PSU to A version.pdf

The 5090B is quite different, it has no cable looms for the PCBs, instead the front & back panels and PCBs are all connected together using card edge connections.

David