Re: PTB 100 Ball Efratom

Chuck Harris

As I recall, the short bypassed the thermal fuse.

HP has made a few bone-headed mistakes over the
years... Allowing your Rb/Cs/xtal ovens to incinerate
is very often one of them.

The battery inside of the Sulzer 2.5 and 5 references
is an accident waiting for a place and time to happen.

I removed mine because it wouldn't hold a charge.
The cells were all open circuit from the excess
trickle charging.

Blackened bulbs are almost always metal that has
been transported from where it belongs to the much
cooler bulb envelope.

In the case of neons, and nixies... the same thing
really, the black is metal that is ripped from the
glowing elements. In the process of plating itself
out on the inside of the glass, it will entrain some
of the neon, reducing the pressure inside of the
tube, and raising the striking voltage.

This is the same principle behind the ion pump.

-Chuck Harris

stevenhorii wrote:


Quite a story! I am surprised that the oven did not have an upper limit
temperature cutout.

That's like one of my Tracor/Sulzer quartz standards. Output went to zero
on all outputs, but the power lamp was also out.

I surmised it was the power supply (in a separate cylinder). Changing power
supplies fixed it.

However, I thought maybe it was just a bad capacitor or transistor in the
regulator circuit, so I opened it up. The housing will mostly filled with
black gook. It was all over everything - mostly anything dependent in the
cylinder - the gook must have flowed "down". When I looked, I was
surprised that it was not the capacitor in there (a cylindrical
electrolytic) but the transformer. The transformer has in a sealed housing
and the header had simply blown out and the electrolytic tar or whatever
they used blew out. I eventually determined that the power supply had been
running on "fast charge" mode continuously as one of the batteries in the
backup pack had failed short. The supply must have sensed low output
voltage from the batteries when I would turn off the supply when moving the
standards rack and so switched to fast charge mode. It did not blow the
fuse and the supply did not have an upper temp limit switch. I don't know
if the oil boiled or just expanded enough to rupture the header, but it
made a mess. I discarded all the stuff except the housing that was not
covered by gook. After that, I took the battery packs out of the power
supply cylinder (there were 21 of them in it) and made sure the supply
ripple was OK and the output voltage was OK with no battery pack. The
supply output was not affected by this. My application was not that
critical that I needed battery backup - I was willing to live with having
to watch the stability while the oscillator came back after a power down.

Isn't metal deposition on the inside of Nixie tubes the reason why old ones
have dark glass?


On Mon, Jul 13, 2020 at 3:00 PM Chuck Harris <cfharris@...> wrote:

I did one in a HP5065A.

I had a bad oven, and a black Rb lamp, and fixed both.
The oven had an internal short between its formvar covered
nichrome wire wound heater, and the outer mumetal shield...
This put the oven on full heat, and melted all of the solder
on the lamp's board... parts fell out of their holes.

But, curiously didn't clean the glass on the bulb. I used
a propane torch, as I recall. The glass is quartz, so heating
it is not all that risky.

Corby Dawson is where I first got the idea.

Corby has made some optical modifications to the 5065A
standards that eeek out another order of magnitude of S/N

-Chuck Harris

stevenhorii wrote:

I got my information from this article:

However, I think we are both right - see page 5 of the article. It notes
the problem is a tradeoff between too much Rb (noise) and depletion from
the Rb diffusing into the glass.

There was a guy (Corby Dawson?) who was able to restore Rb frequency
standards - maybe by doing what you suggested. Have you tried this? I've
got a non-working Efratom standard I should try it on.

Thanks for improving the answer and suggesting a fix.

Steve H.

On Mon, Jul 13, 2020 at 1:35 PM Chuck Harris <cfharris@...> wrote:

I don't think Rb depletion is quite correct.

What I, and others, have found on units that just
ran out of steam, is generally that the RF excited
Rb vapor lamp has plated out its Rb metal on the
envelope of the lamp, seriously reducing the intensity
of the emitted light.

The cure is to re-vaporize the Rb metal off of the
lamp's glass bulb.

To do this, take the Rb lamp assembly apart, and you
will generally see the lamp's envelope is black in
appearance. Heat it with a hot air gun and at a
certain temperature, you will see the black coating
simply disappear.

The Rb lamp is ready to go another decade or so.

-Chuck Harris

stevenhorii wrote:
I can’t tell you about phase noise, but a key feature of rubidium
is their operating life. The life is limited by depletion of the
in the discharge lamp. Because studies by manufacturers of the
of the standards they make, they can provide estimated lifetimes based
rubidium depletion. Because of the reliability requirements of
and time standards (for example in cell phone tower electronics) they
often swapped out for new oscillators as they get within the range of
expected end-of-life. The large number of rubidium standards that turn
on eBay is at least partly a result of this preventive maintenance
Most of the used standards likely will have limited life. However,
low cost means you might be able to afford to buy more than one so you
swap in one that works for one that has failed.

The standard I know that was designed with longer life in mind is the
Stanford Research Systems (SRS) line, Their PRS-10 model, for example,
a design life of 20 years. Just look up the model for a description of
long-term stability and low phase noise. The standard can monitor the
start voltage as it rises as the rubidium depletes. These standards
for more on eBay because of the longer life and likelihood that they
work for you even as used units. Some honest and knowledgable sellers
tell you the lamp voltage. Disclosure: no financial connection with
Ball-Efratom or PRS. I do own standards from both and I bought used
eBay and they work (and are working after a couple of years) despite
used units. On the other hand, I have some HP and Tracor/Sulzer quartz
frequency standards that are more than 20 (the HP) and 50 years old,
respectively. I’ve had to repair a couple of the Tracor units, but the
problem was almost always a transistor in the divider/amplifier stage
failed. I also have a couple of Frequency Electronics quartz standards
these are mil-spec units (URQ-10 and URQ-23) and work fine, but I’ve no
idea how long they were in service. They all have built-in battery
that are built up from “D” size nicad cells (I think - the packs are
sealed) and all have failed. They run fine without the batteries - the
power supply does not use the battery pack as a filter.

Anyone interested in time and frequency should have a look at the Time


On Mon, Jul 13, 2020 at 11:36 garp66 <@green> wrote:

Does anyone know how the PTB-100 Ball Efratom Rubidium standard
performs &
technically (data & experience), with any of the other available
frequency standards ?

(phase noise, etc...)

-- and how it ages ?

Is the PTB-100 still a useful, viable kit ?

thank you,

Join to automatically receive all group messages.