HP 5328A universal frequency counter teardown
Kenneth scored two HP universal frequency counters and popped them open. One generates the timebase from an uncompensated oscillator in a TO-5 package, while the other uses a large temperature controlled oven oscillator:
This entry was posted in Teardowns and tagged HP, teardown, universal counter.here is both of my counters measuring the same 1pps taken from my AdaFruit GPS receiver. The lower one with the oven-compensated oscillator (which is marked as being calibrated in 1999) is spot on, while the top counter using an unprotected crystal oscillator drifts up and down 0.1-0.2ppm, and is almost 5ppm slow…The standard time base is rated for 2.5ppm drift between 0 and 50 degrees Celsius, where the [oven oscillator] time base has the much tighter 0.007ppm. Since these counters display eight (or nine) digits, 2.5ppm is very visible, where 0.007ppm will at worst only effect the least significant digit.
Comments
Hmm, interesting. Is there such a thing as a GPS-based universal frequency counter? Might this be a project one could do at the “competent hobbyist” level?
It’s not unusual to use GPS in a control loop to discipline a Rb reference, which then generates 10MHz as the time base for a counter, so yeah, you can build a GPS-based counter. With the GPS 1pps, the challenge is that you need a really stable oscillator to multiply 1pps up to the needed 10MHz, and you need to be able to measure very small phase changes.
Interesting, thanks. Apparently I can’t post links here, but I’ve found references to a Rockwell Jupiter GPS unit (that I’ve found on eBay for $65) that has a 10kHz timebase. Presumably that’s going to be cheaper than an Rubidium reference, since I can’t find an online mention of what one would cost (read: more than you want to spend, dirty hobbyist!)
Used FE-5680 Rb references are $75 on eBay, and (supposedly) 300Hz agile. 10kHz puts a cruel upper limit on frequency range or period resolution.
Woah, cool. New toy for the wish list!
Wow, it’s $75 now? Price has been going up. If you’re interested in the FE-5680A, here is a FAQ regarding that unit: http://www.ko4bb.com/dokuwiki/doku.php?id=precision_timing:fe5680a_faq
I fairly recently purchased few 5680A and they were 40$ (few weeks ago) .. the shipment was “a lot more” but the devices themselves were only $40.
Anyhow, disciplining 5680A with GPS is going to be huge pain in the but, IMO it’s way simpler to make yourself ovenized 10MHz reference and then discipline that reference with GPS. Properly made ovenized oscillator will work incredibly stable (better then your average high end 9-10 digit counter), you just need to use GPS to get it to lock on exactly 10MHz (and not by using direct feedback every second but by fixing the cumulative error every 48 hours or so)
As for the counters, I recently took pictures of mine AGILENT 53181A counter. This is one interesting device. VERY INTERESTING device. I suggest anyone interested check out the schematic (available on HP site) … it’s super cool, it has bunch of ADC and DAC devices inside, separate interpolator .. very cool thing… schematic is a very interesting read. If you want pic’s they are available here
BTW to clarify, disciplining 5680A with GPS if you have Rx/Tx working will be fairly simple, but most 5680A I seen had Rx and Tx not connected
If you have one of the (many) variations on the 5680A without the serial port control, you can still do a simple modification allowing analog control. Check out http://www.ko4bb.com/dokuwiki/doku.php?id=precision_timing:fe5680a_faq#adding_an_analog_adjustment_pot
Yes you can do analog contro and that’s the part I said about “pain in the ..”