Cesium standards on eBay

[rubidium]

Just a discussion point here, but there are a lot of 5061A/B's and 5071A's on eBay for what I consider to be outrageous prices for "parts only" systems or systems that are apparently knocking on the doorstep of CBT end-of-life (i.e. high E-mult voltage 5071A's). Do these sellers really expect to move these things? Just curious.
Jim

[Stray Electron]

  IMO, about 97% of all of the TE sellers on E-bay are completely delusional.

[5065AGuru]

I offered one of the sellers that had quite a few of these items advice on how to do some quick tests so that they could offer the buyer at least some operating numbers to go by but never got a reply.
Say for instance they could power up their 5071A, verify it locks, give the initial parameter, (gains-EMvolts-ion pump current etc.) Let it run 24 hours and repeat giving the new values. I'd be a LOT more interested if the unit was showing a lower EM voltage, the standard gains and that it passes the profiling test. Of course some sort of warranty would be nice!

Cheers,

Corby

[Raindog]

I've noticed the same thing, most of the 5061A's I've seen are going for 3K or 1.5K for a parts unit missing the CBT with the "no way to test" disclaimer. Same goes for the 5071A's going for 20-30+K. I was able to purchase a NOS AT&T rubidium standard form a seller that was calibrated against a 5071A  prior to shipping. It has not given me any issues and have confirmed it's accuracy against my GPSDO. Trying to find a HP 5061A that is in good physical and operating condition with plenty of beam tube life for under 2K is pretty difficult, let alone a 5071A.

[rubidium]

Ya gotta love the "Pulled from a working environment" notations or, better yet. the unit with the missing CBT claiming "Acquired from a reliable source ...". 

[dietert1]

Difficult to understand this discussion. Even if one gets a working unit, it may fail at any time soon and with little chance to fix it.
Why get into this? What is the application that can't be solved with a GPSDO and/or a cheap rubidium unit? I remember buying three FE-5680A maybe 10 years ago. Two for redundancy and the third one because the second one never locked and unable to fix. Both of the remaining units still work and are stable to 10E-12 under lab conditions and accurate to 10E-9 in comparison to GPS.

Regards, DIeter

[rubidium]

Occasionally (or perhaps more accurately, rarely), one can find an old cesium unit that still has some decent life left. I acquired a PRS-50, for example, that was apparently removed from telecom service because it had 15 leaking electrolytics distributed among its various power supplies. The seller didn't routinely deal with such things and probably acquired it as part of a lot of other equipment. He originally wanted $300 for it, presumably to just get rid of it. I asked a variety of questions about it, and in the end he was kind enough to send me a short video of the unit's initialization after being plugged in. From that video I noticed dim LEDs, combined with a cyclic pattern to the LEDs. I offered $150, and was willing to take a gamble at that price that the behavior being seen in the video might be due to one or more bad power supplies that could possibly explain the unusually dim LEDs, and possibly a microcontroller experiencing brown-out and constantly rebooting to give the endless cyclic pattern. It turned out to indeed be failed electrolytics dragging down the various voltages, and $20-worth of parts and some hours of time to replace them got me to a working system that locked in 30 min. The various operating parameters (gains, etc.) indicated that there was perhaps a few years of tube life still left.

That was a $150 gamble. I've certainly "gambled" that much and more on a disappointing restaurant meal - too many times. But it appears that there's a crop of 5061A's and B's and 5071A's currently being offered for prices that are 2 orders of magnitude or more greater. Coupled with that, the sellers won't respond to any questions. Some units are even missing CBTs! So, yes, I won't gamble on any of that, and I don't know anyone who is knowledgeable about these systems who would.

Jim

[DonBox]

Has anyone purchased a new CBT to replace the 5071A? :'(

[edpalmer42]

Has anyone purchased a new CBT to replace the 5071A? :'(

Do you mean just the Cesium Tube or the entire unit?

The tube is available from MicroChip.  They bought the Agilent timing line of equipment a number of years ago.

If you're talking about a new complete unit, they sell the 5071A and the new 5071B.  I don't know if the 5071B is an upgrade or just an update of the old design.

https://www.microchip.com/en-us/products/clock-and-timing/components/atomic-clocks/atomic-system-clocks/cesium-time

If you're talking about the price for either option, well, let's just say that neither option is 'hobbyist friendly'.  The price is solidly in the range of "If you have to ask, you can't afford it"!!

[tridac]

I bought an Oscilloquartz 3210 on Ebay a few years back. On collecting it, discovered it was one of three Racal portable frequency standards, with battery backup, in short racks. Made a cash offer for all three, which was accepted. On test, two worked and locked in tens of minutes from cold, bags of beam current. The third looked ok, but zero beam current. Did a bit of work, tracing signals etc, but unable to make it work. Sold and shipped it to a collector in Japan, to recover some of the outlay. Still have the other two, ion pump and vcxo on all the time, then tube powered up once a month or so to keep the circuits and caps ok. One will be for sale (or trade) at some point, but no rush. Compared to each other with a phase meter, zero detectable drift over days, but they are good enough to detect the sawtooth like drift of the lab gpsdo. The 3210 is an all analog design, so quite early, but iirc, uses a Frequency Electronics tube. Amazing accuracy and far more than I need for work done here, but they do scratch a time nuts itch and add to the frequency standard collection...

[KE5FX]

Has anyone purchased a new CBT to replace the 5071A? :'(

It's possible to use the newer-style 5061A/B tubes in the 5071A with some hacking.  Not that they are much easier to find.

On the subject of EM voltage, I recently bought a 5071A with a standard-grade tube that was running at around 1830V.  It would lock, but would usually fail short-term ADEV.  Upon receipt, the error messages in the operating log indicated that the tube had already lost lock a couple of times.  However, the real problem turned out to be a very noisy 10811 OCXO. 

This is the second time I've seen one of the 5071A OCXOs misbehave in a way that made me think the tube was dead.  After replacing the OCXO, I found that while the EM voltage was still in the 1830-1840 range, the tube was actually within spec.  It ran for a full week with only minor variations in EM voltage (+/- 3V or so), while a dying tube will normally show a fast upward trend in EM voltage over that time span.

I think a lot of people assume that 5071As running at more than 1300V are on the way out.  But the statistics behind that assumption appear to have come from an early survey at USNO, and you don't see a lot of standard-grade tubes at USNO.  So far, my impression is that EM voltages near 1800 are not a problem for the standard tubes.

[Bicurico]

Being a TE affictionado I looked into these standards when I first read about them here on the forum. I have no real use for them, but found it fascinating.
Anyway, I did not buy any when they were available for less than hundred Euro and I noticed how their eBay price increased steadily since then.

Most TE on eBay is offered at ridiculous prices indeed.

Some deals are still there though.