Onboarding session for the HP-53310A Modulation Domain Analyzer

No worries!

While I don't have a special use for the Hewlett-Packard 53310A modulation domain analyzer, it hold a special place in my history. When I was 15 I was sending letters to HP (yes, the snail-mail written type) to get price information and pamphlets to complement my T&M catalogs that I was requesting every year. Yes, I had an HP T&M catalog in my school bag and was reading that during recess. How many nerd points do I get? :-) Anyway the MDA was a weird instrument and I was really curious how much it would cost. So out went a letter to HP asking for prices. Of course when HP received it they didn't know what to do: who sends price requests like that? So they went to the white pages, dug out the house phone number and called when I was at school. Not sure what my mom told them but a few days later I got some pamphlets with prices. The MDA was 414.000 BEF + VAT. BEF was the Belgian franc at the time, today it would be an almost round 10.000 euros, to which you of course have to add quite a bit of inflation. Needless to say I was speechless: I was expecting something 10 times lower. As the saying goes, if you can't guess the price, you cannot afford it / you don't need it. Ah memories... That being said, I did use it to calibrate my frequency counter and check for modulation problems in my house 10MHz reference so it's not completely useless :-)

So when this very nice example popped up on an auction site I had to give it a try: it had all the options! 010 oven time base, 001 memory extension and 031 2.5GHz extra channel (the good one, none of that 030 option rubbish!). But would it chooch?? After a quick positive self test it's time to open it. The biggest concern is the power supply: it's quite compact and its caps age thus a little too fast, meaning oozing chemicals and destroyed boards. This instrument has a little strange construction that is also present in a few instruments of the era (e.g. the HP-54542a and its siblings 54540, 54520 and 54522 plus some logic analyzers).

The case is open revealing a clean interior

The interior turned out to be very clean, with some minor dust here and there, possibly animal hairs (a cat? Those are apparently allowed in precision labs ;-) Speaking of which, you know you're in the big-boys club when you see the words "LO in" and "LO out" on your equipment!

Big Boys Club achievement unlocked!

Some fluffiness here and there but nothing that can't be swiftly cleaned up.

The CRT area in full potato-vision

The top of the dreaded PSU brick. Note the bracket for a floppy drive, which is not present in the hp53310a.

The CRT is a Toshiba jobbie.

Removing the PSU and the option 031 board reveals a few interesting chips, and quite a few of them have HP programs in them. Strange, these days you'd get only a couple of programmable logic max, plus a big-ass FPGA of course. Other times...

The edge of the option 031 board with its coax connectors

The top processor board, under the PSU

Cute flock of programmable chips on the top processor board

This DIP switch has the normal positions written on the PCB. 100% HP awesomeness, and shows the good dialog between the logic design team and the PCB team.

Battery backed-up RAM: a sign of the age of this device and something that will need some attention in the future.

The main processor (?). I think it's the same part in the HP54542A & friends

Now let's turn our attention to that potential deal breaker that is the power supply. The number of screws to open that bastardinios is impressive, but the result is worth it: no sign of caps bulging or other atrocities. Score!

PSU cover and warning labels

After much unscrewing no bad news in this PSU

The PSU part numbers

The inner construction is interesting: some heat sinks are tied together with diodes. Not sure why but I'm not opposed to it :-) On the other hand the heat sinks surrounding a cluster of caps looks like a recipe for disaster. Not here though so I'll just crawl back to my cave: all hail the HP (or in this case the subcontractors) engineers. A cap was also tied to an inductor with a zip-tie: interesting protection against vibrations...

Strange: heat sinks linked by diodes. Never seen that in 30 years of tear-downs...

Cluster of caps surrounded by heat sinks: me not likes.

Another cap shot, no leaks!

Good branded big-ass caps. No touchy: angry pixies inside!!

A cap tied to an inductor with a cable tie. Yes with an S: with 2 cable ties!

TO-3 goodness for the win!

Looks like everything is supa-gut here so let's close the PSU. No I won't recap it (for now): if it works, don't fix it! Plus it could easily add 25% to the price I paid for this beauty. One last thing to check is the oven oscillator. I'm quite curious because it's the first device I own with the venerable 10811-60111 oven oscillator. It's located with the thickness of the rear panel but protrudes a little out. It's powered by its own power supply that is on whenever the instrument is plugged (thus even when the instrument is OFF). This separate PSU even has its own fuse and voltage selector on the back panel! The caps and circuit are fine there too, so no worries: the 53310 can be closed and enjoyed after the compulsory CRT adjustments (focus.brightness etc...) as recommended in the service manual.

The HP 10811-60111 oven oscillator.

Rear panel with oven oscillator PSU fuse and voltage selector.

The oscillator oven PSU, top view

The oven oscillator PSU, side view. No bulging caps, everything is clean. Noice.

One little thing that'll be on the TODO list is to replace the NVRAM on the CPU board. It's an ST part, reference M48Z18-100PC1, which is still available from Digikey! Best of all it's one of the cheapest NVRAM in that format (not that the prices vary much though).