Lux, Jim via time-nuts writes:

What kind of hydrogen concentration are we talking about ?

--
Poul-Henning Kamp      | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG        | TCP/IP since RFC 956
FreeBSD committer      | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.

FWIW, hydrogen is a bigger molecule (2 atoms) than helium (1 atom, 1
molecule). That's why helium is used for leak testing. If helium won't
leak, neither will anything else.

Steve Hendrix

On 2022-12-08 04:28 PM, Lux, Jim via time-nuts wrote:

On 12/12/22 7:09 AM, Poul-Henning Kamp wrote:

Unknown - I'm looking at "if you've got a cubesat mounted on a big
rocket that uses LH2, and they go through various and sundry tanking and
detanking exercises, what might happen"   In the long term sense,
probably not a high concentration, but I could see there being times
(hours?) when there's a very high concentration.  I mean, they have
those brooms to probe for hydrogen flames for a reason.

There are cases where people think that a sensor was "poisoned" by He
(but never confirmed) because it was stored in a He-rich (but no where
near pure He) environment.  ISS has this problem - the atmosphere is
fairly rich in He, because they backfill the supply ships with He, so
when they open the hatch, a big blob of He moves into the station. (why
they don't use N2, I have no idea.. Maybe they are using He sniffers
after bolting the hatches or something)

Hi

Depends a lot on what you call a “hermetic TCXO”.

There are a lot of parts out there that have sealed crystals inside a sealed TCXO package.
The TCXO package might be resistance welded, it could be solder sealed, it might be glued
shut. The crystal likely is resistance welded or cold welded in a separate enclosure inside
the TCXO package. The TCXO package likely has an inert gas in it at around 1 atmosphere.

There are other parts that have an open blank inside a single package. The package probably
is welded shut, but these days … who can be 100% sure of anything :)

On top of this, a sealed crystal might be in a vacuum or it could be in some sort of inert gas ( often
at pretty low pressure). The same is true of OCXO crystals. The single package unit could
be evacuated or backfilled, possibly at a fairly high ( = close to an atmosphere) sort of pressure.

So lots of variables.

Worst case is a vacuum packaged high Q crystal. It gets gas instead of vacuum and the Q
goes down. If it goes down enough, the oscillator quits. Not a really high probability in a TCXO.
There are parts out there though.  If the crystal is inside it’s package and that’s inside a second
hermetic package, it’s going to take a long time for your hydrogen to get to the crystal ….

Bob

Lux, Jim writes:

Hydrogen can diffuse through any metal[1], because by and large,
metals are just nuclei in hydrogen-soup, but it takes work to do.

If the TCXO is not evacuated, I would not worry at all, because without
a pressure differential the diffusion rates nearly nothing.

If the TCXO is evacuated, and the hydrogen is present for long periods
of time, perhaps because it cannot escape from some badly ventilated
cavity, then you might see detectable diffusion, but I still cannot
imagine it having any relevant effect.

Have you checked NASA literature ?

They must have considered this for Apollo ?

Poul-Henning

[1] Really pure hydrogen is produced by pushing it through a sheet
of Palladium, which is why Fleishmann and Pons were in that business
to begin with.

--
Poul-Henning Kamp      | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG        | TCP/IP since RFC 956
FreeBSD committer      | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.

This is true and was proven by experimentation. But I believe it was a
Helium atmosphere not Hydrogen. The He poisoned the MEMS oscillator in both
phones and watches. Fault time was about 30 minutes recovery time was about
3 days. Concentrations were small enough that and employee filling party
balloons at a store had a dead I phone for a while.

Eric

-----Original Message-----
From: Poul-Henning Kamp via time-nuts time-nuts@lists.febo.com
Sent: Monday, December 12, 2022 10:10 AM
To: Discussion of precise time and frequency measurement
time-nuts@lists.febo.com
Cc: Poul-Henning Kamp phk@phk.freebsd.dk
Subject: [time-nuts] Re: hydrogen rich environment and oscillators

Lux, Jim via time-nuts writes:

What kind of hydrogen concentration are we talking about ?

--
Poul-Henning Kamp      | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG        | TCP/IP since RFC 956
FreeBSD committer      | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.

time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe send an
email to time-nuts-leave@lists.febo.com

On 12/12/22 11:21 AM, Zen via time-nuts wrote:

Yes, but H2, while a bit bigger than He, is still pretty diffusable (if
that's the right word).

And, I believe the iPhone event was a very tiny MEMS resonator in a vacuum.

I've been poking through the literature and reports, but it does not
appear to be a "well studied problem".

The challenge is that with Cubesats and the like making more use of COTS
components and techniques, there's not a lot of history to go back to. 
Space qualified parts get leak tested with He, etc. so their hermeticity
is generally good.  The same is not necessarily true of other devices.

Look for Knutdsen effect:
Knudsen diffusion, or Knudsen flux, occurs when the mean pore diameter
of the porous medium is smaller than the mean free path of the gas
particles. This type of diffusion occurs in low permeability porous
media, which have small pore radii, and at low gas pressures, when the
mean free path becomes large.
73
KJ6UHN

On 12/12/2022 2:09 PM, Lux, Jim via time-nuts wrote:

On 12/12/22 2:43 PM, Alex Pummer via time-nuts wrote:

This would be at atmospheric pressure, so MFP is on the order of 60 nm. 
I suspect that the pores in metal or silica or whatever are much smaller
than that.

Hi

If you have a welded package TCXO, it’s a reasonable guess that it may have gone
through the normal “mil spec” leak test. Mil STD-202 method 112 test condition C
gives you all sorts of fun information. It also heads off in many directions.

As normally done on a crystal oscillator, you put the parts in a pressure vessel for
about an hour. It’s pressurized at around 50 PSI with 10% helium. Yes both the time
and pressure are past what method 112 calls out as the minimums.

After being soaked like this, they go into a mass spectrometer to look for helium
coming back out of the parts.

Pretty much any mil spec or space spec welded part would go through that test or
something very much like it. That’s true regardless of who made the part.

The only relevant point here is that if this sort of exposure killed the parts, it would
show up and folks would very much know about it.

Welded crystals going into precision oscillators get the same sort of leak check or
possibly something more in depth. There’s no guarantee on this so it’s not quite as
much a sure thing as the space or mil oscillators. Still, if this killed crystals you would
know about it.

Before anybody asks, yes, some parts do leak. I don’t remember any of them being
non-functional. However, you don’t test those parts much. They are headed to the
recycle bin.

Bob

On 12/13/22 5:57 AM, Bob Camp wrote:

OK. so welded packages (particularly those that are procured for space)
aren't likely to leak, they tested, etc. I'm sort of interested, though
in COTS cheap oscillators like these tiny $1 Abracon units.  (I don't
know what they use on Arduinos and RPi, but it's something like this).

https://www.mouser.com/datasheet/2/3/ASEseries-38758.pdf

They're sort of silent on the datasheet about sealing, etc.  And at
$1.38 each, do they actually leak test them?

Hi

As price goes down, so must cost. If not, the outfit goes broke. Working
out just what gets “left out” on a $1 TCXO is almost impossible unless
you can go do an audit of the production line at the time the parts are
being made.

Next question would be: Are the parts evacuated before seal? I’d bet not.
They likely get some sort of inert gas fill and that’s about it. If so, you
already have gas in the package. There’s not a lot that a one gas
will do vs another as far as damping the crystal. Also, with a more or less
zero pressure differential, not much He or H is going to work it’s way into
the package. Yes, this assumes the parts are crystal based ….

It’s not all that hard to test this out. Get or rig a pressure container that
will take 50 to 100 PSI. Go over to the local gas store and get a small
bottle of He or H. Bomb the parts overnight at 50 to 100 PSI. Pull them
out and see if they still work …… If so, you have just done what is likely
the “100 year test” on the parts. If they don’t work, see if you have crushed
the parts ( = smashed the package flat …) and try again at a lower
pressure … :)

Bob

On 12/13/22 9:09 AM, Bob Camp wrote:

That's kind of what I thought.  I would think that if it's a MEMS
device, that the mfr would identify it as such. The Abracon part does
say "Crystal Oscillator" at the top of the data sheet.

Good idea.

An excellent project for a summer intern.

Of course, at JPL, we'd be much more elaborate, budget permitting - CAT
scan the package before and after, set up a RGA or Mass Spec, Bring in
an optical fiber(s)) from the maser to compare the frequency of the
oscillator. Certify all the high pressure gas handling equipment. The
test process alone could cost more than buying 100,000 oscillators.

In my garage - the work of a weekend.

if it's anything like where I work, If you don't test it that way, and
something happens you'll get yelled at for not doing due diligence. If you
do, and nothing happens you'll get yelled at for wasting money

eric

On Tue, Dec 13, 2022 at 10:04 AM Lux, Jim via time-nuts <
time-nuts@lists.febo.com> wrote:

--
--Eric

Eric Garner

After you stop at the industrial gas or party store to pick up bottles
of He and H, stop at a tire shop and have them remove both rear tires.
Throw half a dozen SMD oscillators inside each one, and re-fill one tire
with He and the other with H. Drive around for a day or two (or until
one or both tires go flat), remove the oscillators, and measure them.

This method not only tests He and H gas leakage but also mimics the
acceleration violence of launch and periods of gravity free-fall. No
chamber necessary. And you have to drive to pickup and return the
bottles anyway.

/tvb

On 12/13/22 10:26 AM, Tom Van Baak via time-nuts wrote:

You're assuming I don't happen to have a bottle of He or H2 at home (or
at a friend's house), and a pressure pot <grin>. These things are tiny..
I could probably electrolyze water to make enough H2 and use a spare air
cylinder with a weight to pressurize it.

But I like the idea of the field expedient vibe/shock test.

Hi

The MIL-STD 202 folks seemed to be quite happy with a 10% mix of He and whatever. My guess is
that you could fill the pressure chamber with your water sourced H2 and then blow it up to 50 PSI
with the shop compressor.

The obvious disadvantage of any process using H2 is fire / explosion. I’m not 100% sure this is the
approach I’d want to take …..

Bob

There has been a lot of work with respect to hydrogen sulfide interaction
with steel in the hydrocarbon area.

Hydrogen embrittlement might be a useful search term for your
investigation.  A believe that the American Gas Institute (AGI)  did a
couple of webinars about this topic.  You can probably find them online.
They were quite informative.

Be extremely extremely extremely cautious working around hydrogen.

On Mon, Dec 12, 2022 at 10:04, Lux, Jim via time-nuts <
time-nuts@lists.febo.com> wrote:

--
Dr. Geophysics
http://dr.geophysics.googlepages.com

On 12/13/22 11:48 AM, Bob Camp wrote:

yeah, since the flammable range for H2 is pretty wide.

So do it in the back yard, not the garage or living room.

And if it's just a few cc of H2, I'd think the pressure pot would
contain any "momentary overpressure"

This does bring up an interesting idea for time-nut-ty experimentation. 
Is there a cheap connector/feedthrough that can take 100 psi or so to be
able to make measurements "in-situ"? Hermetic RF connectors are
available (not cheap new, but perhaps surplus), but I've only used them
for 15 psi differential (e.g. atmosphere outside to vacuum inside). I've
also used non-resistor spark plugs as HV hermetic feed throughs (there's
some specific Champion number that has a particularly long reach). 
Something like a D-sub with a dozen or so pins might be convenient,
especially if it's round (since round holes are easier to drill)

Lux, Jim via time-nuts writes:

Use a window and optical transmission ?

--
Poul-Henning Kamp      | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG        | TCP/IP since RFC 956
FreeBSD committer      | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.

On 12/13/22 1:24 PM, Poul-Henning Kamp wrote:

It's hard to get power in with optical transmission.

Say you have your favorite 10811 based GPSDO and you want to see how
it's going to work in your diving bell loaded with heliox (with a coax
cable to a floating buoy with the GNSS choke ring antenna) - but not
really...

This is sort of different from the whole He, H2 question - but it raised
the interesting question of operating devices at pressures other than
room temp.  We've talked a lot on the list about schemes of varying
complexity to thermally isolate an oscillator - well, putting it in even
a partial vacuum might  be effective, one really wants to get the MFP to
"bigger" than the gap between device and wall, so convection stops being
an issue: then it's just radiation, but even if you pump down to, say,
1/100th bar, the bulk thermal conductivity will be (I think) 1/100th of
what it is in "room air".

On 2022-12-13 04:10 PM, Lux, Jim via time-nuts wrote:

I recall from a number of decades ago when I worked on aerospace systems
that used large amounts of H2, that the lower explosive limit for
hydrogen in air was 4%, so we had all our sensors set to alarm at 2%.
Keep in mind, though, that if hydrogen gets loose it can burn, but if
pure oxygen gets loose, EVERYTHING will burn. And as others said, a tiny
quantity only has so much energy available.

Steve Hendrix

Hi

Given that you can keep feeding gas to the pot ( = minor leaks in the back yard
aren’t that big a deal …), grab a tube of epoxy and attack the pressurized side of
whatever connector happens to be handy. Probably do several layers. Run
carefully spaced bare wires through the epoxy.

My choice for a connector would be something circular. I’d go that way  just to
make the mounting hole easy to drill and later maybe plug with a bolt. A 4 pin “mic
connector” comes to mind ( and there’s one sitting loose right over there ….).
Power, ground, signal out, signal in maybe.

Epoxy may be overkill in this case. RTV might do the trick. It certainly would be
easier to strip off afterwords. A lot depends on just how “valuable” the pressure
pot happens to be …...

Bob

Lux, Jim writes:

The thing about vacuum is that you have to make sure you still get
rid of enough heat to not cook your devices.

I know of at least one early cube-sat which did not deploy it's
antenna, because somebody didnt heat-sink the FET which should have
delivered power to melt the launch-restraints.

--
Poul-Henning Kamp      | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG        | TCP/IP since RFC 956
FreeBSD committer      | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.

Hi

If we are off to a stabilized / vacuum environment, it’s a pretty good bet that there
will be multiple sources of leaks / outgassing / whatever to mess things up. You
will have a pump of some sort to deal with what’s going on. The better you can make
things, the smaller a pump you would need. It’s still a pump and under some sort
of servo control ….. loop back to “pressure sensor is out of whack” a week or two
ago :)

Bob

Steve Hendrix, P.E. via time-nuts writes:

There is a very informative monograph from NASA (sorry, dont have a link
at hand) about their early experiences with cryogenic rocket fuel.

A particular interesting episode is where they have a leak in a LOX tank,
and then the firetruck of the emergency response catches fire because
they are downwind.

--
Poul-Henning Kamp      | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG        | TCP/IP since RFC 956
FreeBSD committer      | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.

Hi

I am embarking on building up a block - a common collector Butler
oscillator with an overtone  crystal at 98.304 MHz

on on FR4 , I've used the Philips NPN type  BFS17.

The results <=10 Hz phase noise were quite disappointing. (-69 @ 98 MHz)

Does anyone have any good suggestions. Maybe something slower (larger)
like a 2N5179 (or its SMT counterpart MMBT5179) will be better.

regards

glen

On 13/12/2022 11:14 am, Lux, Jim via time-nuts wrote:

Hi

If you are inside 10 ppm, it’s a good bet that you are in a region that is still influenced by the
resonator / oscillator Q. At 98 MHz this is an offset of about 1 KHz. First thing to worry about
is the Q of your resonator.

Past that, transistors vary a lot. This brand vs that brand matters. VHF transistors are rarely
well characterized for audio noise. Their performance vs spec is not a great thing to use as a
selection parameter.

Simple answer is always going to be: try it and see.

Probably better to look at 100Hz noise. At least there, you have pretty well defined limits as
to what’s ok, good, great, and impossible. As you go from -120 to -140 dbc/Hz, you pretty much
cover that entire range.

Bob

Generally speaking, the 1/f noise at, say 10 Hz, for a crystal
oscillator is determined

entirely by the intrinsic noise of the crystal.  It's really hard to
screw up an oscillator

circuit so badly that it actually contributes anything to the oscillator
1/f noise.

For instance, I'm sure we will all agree that the HP 10811 is a pretty
decent oscillator.

FYI, essentially the same crystal is used in the E1938A that I designed.
Those two

oscillators have ENTIRELY different circuits.  Yet both achieve the same
close in

phase noise as what I measured for the intrinsic noise of the crystal.
By the way,

the HP 10811 uses a selected 2N5179 transistor.  But is NOT SELECTED FOR

1/f NOISE.  Instead it is selected for good gain at high DC collector
current, because

the HP engineer who designed it had trouble with the oscillator starting
up correctly,

in conjunction with the ALC circuit.

If you use the same output circuit for your oscillator as the 10811, you
will also

have no oscillator transistor contribution to far out phase noise.
This is all

very well documented if you want to read the details.

In a previous life, I designed many many 5th overtone oscillators.  Your
crystal

like the Croven ones I used in those days is not of the same caliber as
the

10811 crystal, so therefore it is even more unlikely that the transistor

you choose is going to make any difference.  I always locked my 5th OT

crystals to something like a 10544 oscillator to clean them up close in

using PLL synthesizers.  The 10811 hadn't been invented yet.

I hope this gets you back on the right track.

Rick Karlquist
N6RK

On 2023-01-16 14:26, glen english LIST via time-nuts wrote: