I am looking forward to long term data on the Lucent unit. GPSDO's are
getting closer and closer to Cesium. Having worked for 18 month on two GPSDO
projects we find that the limiting factors are the Cesium Standards.  Working
presently on a Cesium GPSDO. Short term OCXO, medium Rb and long term
Cesium.  With Cesium may be able to use 14 day filter. Will find out. If we  do
not see an improvement we will most likely retire our Cesium units.
Bert Kehren

In a message dated 12/6/2014 10:46:57 A.M. Eastern Standard Time,
kb8tq@n1k.org writes:

Hi

magnus@rubidium.dyndns.org  wrote:

drkirkby@kirkbymicrowave.co.uk  wrote:

http://www.ebay.com/itm/HP-Agilent-5061A-Cesium-Beam-Frequency-Standard-FOR-PARTS-REPAIR-/141483787108

re-tubed with a high performance tube is > $32K.

you see are pulls from used gear. The question with them (as with any  Cs)
is just how many years (or months) is left on the tube. You physically  move
Cs from one end of the tube to the other when you operate the device. One
you have exhausted the pre-loaded stock, the tube is dead. It’s also coated
all over the inside with surplus Cs. Since signal to noise ratio is very
important, the drop in Cs at end of life and crud on the inside leads to
degradation in the performance towards the end of the tube life. Even if the
tube works, it may (or may not) be useful in a given application.

performance rivals that of most of the older Cs standards. They are way  cheaper,
and they don’t wear out. Indeed, if you have a 5071A with a high
performance tube in it, a GPSDO is not going to match it’s performance. I’ve
replaced two tubes in one of those, so they are correct when they talk about  the
projected life of the tube.

to do it “right” it’s a  major pain. Sending one back for re-tube does
require you to do all the formal  shipping nuttiness. That may or may not be an
issue on the surplus market  ….

receivers. Rubidiums do help to some degree. Comparing two  GPS clocks with
their highly systematic sources, so you can't get useful  differences that
way for the stability of the produced signal.

Unless  you are making a GPS receiver from scratch (which you might be),
there is a  certain “trust factor” that comes into using a GPS for timing.
Since you can’t  play with the firmware, you trust that the guy who wrote it
did a good  job.

In making a GPSDO, yes on a commercial basis verification against  primary
standards is likely to be required by this or that customer. In a  basement
lab, I’m not so sure that’s true. Simply comparing things against an
ensemble of “known good” designs (and cross checking the results) should be
good enough. If your design passes the performance of the ensemble, building
several of your design is likely to be cheaper than keeping a Cs running long
term. That’s even more true if you need a fully functional 5071A to do the
comparison. Let’s see .. new BMW or rebuild the 5071 … hmmm  :)

Bob

https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts

time-nuts  mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the  instructions there.

Hi

If you toss a Rb into the GPSDO “mix” things can get quite good. The Rb should be better than an OCXO in the > 1,000 second range. It’s crossover with the GPS ADEV will be further out than the OCXO’s. The gotcha with both the OCXO and Rb is their temperature dependance. Some / many / all of the lower cost Rb’s are not much better than a good double oven OCXO in terms of raw temperature performance. The approach they use to “correct” this does not help their ADEV at all. Yes, you can disable the correction and put the whole thing in a temperature controlled environment.

Lots of details to take care of. If you get them all right, you’ll beat any / all of the older Cs standards. That assumes that GPS is not deliberately lying to you :) … (off to the Conspiracy Time Nuts mailing list). One very cute addition would be to pull down the NIST GPS data and use it to correct your system on an hourly / daily basis. If you do that with common view satellites, you most certainly will beat a surplus grade Cs standard.

Bob

Well a bad tube is a bad tube and thats been my story. Though for $125 how
can I complain. But for $999 plus $79 shipping no interest at all.
When the tubes used up its used up. Generally.
Regards
Paul
WB8TSL

On Sat, Dec 6, 2014 at 12:06 PM, Bob Camp kb8tq@n1k.org wrote:

On 6 Dec 2014 17:58, "paul swed" paulswedb@gmail.com wrote:

I have never looked a tube, but what (if anything) stops them being
rebuilt? I guess someone has tried it.

They want almost $1000 to ship to me, although I suspect that they would
recalculate and get a more realistic price.

It sounds like it is not worth bothering with.

Dave

On 6 December 2014 at 17:58, paul swed paulswedb@gmail.com wrote:

I just offered them $250 for it. From what I gather, it is going to be
unfixable, but will be a bit of fun poking around inside. But I'm not
paying $1000 either.

Dave

David
I picked my unit up for $125 at a Hamfesyt and the tube was absolutely bad
as it turned out. But then what do you expect for the $. That said another
time nut gave me his dead tube from a 5060. I spent a good deal of time
getting it adapted into the 5061 now name Frankenstein. This included a new
oven controller.
But it was all about learning and boy did I. Still nothing like some of the
folks on Time nuts truly they have the expertise and feel for the beasts.

But the fumes were enough for me to get an actual feel for a CS reference
and why they were always left on. Also it drove home a lot of the physics
and math.
Now for $125, thats a pretty good education.
Good luck and may the CS fumes be with you.

Regards
Paul
WB8TSL/1

On Sat, Dec 6, 2014 at 2:47 PM, Dr. David Kirkby (Kirkby Microwave Ltd) <
drkirkby@kirkbymicrowave.co.uk> wrote:

On 6 Dec 2014 21:10, "paul swed" paulswedb@gmail.com wrote:

new

the

The fumes will do for me!

I got an HP 58503A as I need a solution, not a project.

But playing with a Cs for a bit of fun/education is something quite
different. But maybe I should wait until one comes up in the UK, shipping
is going to be $300-$500.

Due to the size of the UK, there's a lot less test equipment around here
than the US. What we do have here tends to be quite expensive.

Dave

Used to be quite a bit available at reasonable costs. But Ebay ended that
ages ago. Even for bad stuff the dollars are very high. Granted some
venders will take back items that do not work such as this one. But not
always.
And yes its not for business at all.
Regards
Paul

On Sat, Dec 6, 2014 at 4:51 PM, Dr. David Kirkby (Kirkby Microwave Ltd) <
drkirkby@kirkbymicrowave.co.uk> wrote:

$300 of which $50 was shipping. 14 day return but I pay the shipping. So
figured would only be out the $50.  Seller said would be  fipped in a
double wall box. So shipping shock was much less of a concern.

Also not for work.
On Dec 6, 2014 1:52 PM, "Dr. David Kirkby (Kirkby Microwave Ltd)" <
drkirkby@kirkbymicrowave.co.uk> wrote:

Hi

The Cs that was just bought came from the Netherlands over to the US. There is gear in Europe.

Bob

Hi

The flip side of that is being able to shop all over the globe for the stuff you are after. For the 99% of us that are not in the part of the world where things get dumped / got dumped, that’s a good thing. In 20 years of going to hamfests and sales in the areas where I lived, I never saw a single Cs source ever, let alone one for $125.

Bob

Hi Bob,

On 12/07/2014 03:23 PM, Bob Camp wrote:

Indeed. Every now and then I find a bunch of cesiums in the trunk of my
car. Then I unload them at home and give them some basic LTC and turns
out they work.

Re-located two of them to places running low on working cesiums.

So yes, there exist gear in Europe. :)

Cheers,
Magnus

Hi

The real issue is:

Do you want to spend hours a day doing searches on eBay for all sorts of weird spellings and listing conventions? Do you want to search all of the various eBay sites throughout the world? When any single piece one of a kind item comes up at a good price, it will go fast. There will not be people jumping up and down saying “look at this”. They will just buy it and move on. The only time you’ll get a hint is when somebody lists a bunch of the same thing. Then who ever finds it will buy what they want / need / can afford / crave. After that they will mention it on a list somewhere.

Bob

On 7 December 2014 at 15:55, Bob Camp kb8tq@n1k.org wrote:

There is always

http://www.geo-ship.com/

which searches all eBay sites except those that use obscure characters
like China.  The interface works well on a workstation, but is just
about unusable on my mobile phone.

Dave

HI

Right, but now you need to put in every possible spelling of anything that might reference to what you could be looking for in all the possible languages, Next you sort through a few thousand hits a day…

Consider:

You are looking for a GPSDO. The guy on the other end calls it a “frequency tamer” and never mentions GPS or OCXO’s. He only talks about the part numbers used in the device. If you are lucky he does it in English. If you want all the possible hits, the words might not be next to each other.

Lots of fun.

Bob

On 7 Dec 2014 18:59, "Bob Camp" kb8tq@n1k.org wrote:

that might reference to what you could be looking for in all the possible
languages, Next you sort through a few thousand hits a day…

“frequency tamer” and never mentions GPS or OCXO’s. He only talks about the
part numbers used in the device. If you are lucky he does it in English. If
you want all the possible hits, the words might not be next to each other.

Bob,

I think you are making it out to be much more difficult than it is.

First you say one has to search every eBay site. When I show that is not
true, you now say one needs to search every possible mis spelling of
anything of interest.

I am sure we could all find lots of GPSDO's. Finding them on eBay is the
easy bit. Deciding on which to buy is somewhat harder.

Anyway,  I might buy a Cs standard for a bit of fun, but I will not buy one
with the intention of actually using it as a serious frequency standard.

Dave.

Hi

What you showed is that you can use a wide range search tool rather than searches customized to each regional marketplace. My point is that there is a problem with that.

There are a variety of ways things like simple things like Cesium get spelled. Depending which side of the Atlantic you learned to spell, you might / might not call them mis spellings. For some reason people like to put an “a” in the perfectly good / properly spelled word Cesium.  If you use a wide range search, you need to include the whole range of terms and languages that might be used in those regions.

…. which is my main point. A search that is broad enough to find everything of interest also brings in a monstrous amount of stuff that needs to be sorted through. A wide range tool with a wide open search simply means more stuff to go through manually.

If it’s good (what ever it is), you don’t have long to decide. T Bolts sell these days for $300 or so. If one gets listed for “buy it now / free ship” at $70 it will go quickly.  Having a few thousand hits to go through each day makes that process pretty tedious. A once a week sort is unlikely to do you a lot of good.

Run a search for “frequency standard” on a search tool and see what comes up. It’s a lot of hits, the vast majority of them have no relation at all to accurate frequency. In some cases, that’s the only search that turns up this stuff.  Even the valid hits will be 99% useless because they will be repeats of stuff you have looked at before.

Without a working tube, they are not terribly interesting to play with. Reading the manual (free download) is more fun.

Finding one on eBay for $1K that is “money back guaranteed works" these days is doing very well. More typical working prices are in the > $4K range. If you go HP 5071, the prices are going to be higher than a 5061. FTS versions are likely to be lower cost than HP’s with HP tubes. The FTS / Datum boxes are less likely to have all the lab bells and whistles on them. They are more likely to be adapted to telecom service. Either way, it’s a primary frequency standard.

With a working tube, most Cs standards get sold when the beam current starts to drop / ADEV degrades. They go into yearly cal / checkup and get flagged. The company looks into the re-tube price and dumps the device. Since the “repair” is half the price of a brand new one, that’s always been a pretty common decision. Will it run for a month or a decade at "low current"? - who knows.

This is a narrow field. You will spend a lot of time searching for a deal.

Bob

On Sat, 6 Dec 2014 11:47:10 -0500, you wrote:

Hi Bert,

Out of curiosity, what Rb are you using, and how does it respond to
air pressure changes?
Combining temp control, air pressure compensation and drift
compensation can give very good results with the right Rb.

Angus.

Hi

Properly identifying / measuring pressure induced drift is not as easy as one might think. The “tweak and see” approach seems to be the best bet. Hmm … I wonder who originally suggested that …. oh, yea it was Angus.

Bob

Hi Bob, I have been having problems posting this. If other versions
show up, please disregard.

I have some questions for you.

What does the temperature controller do that degrades the ADEV? Can
anything be done to modify it to improve the performance?

Is there a list anywhere of all the details that need to be
addressed? How about words or phrases that can be used to search
google and the archives?

How can we do this? The NIST archives state

"The archive is only updated once every 24 hours, so data are not
available for today's date. Data from the previous day are added to
the archive at about 1600 UTC."

http://www.nist.gov/pml/div688/grp40/gpsarchive.cfm

Is there another page that has current data? If so, how do we
incorporate it into the GPSDO?

Thanks,

Mike