Electronic pressure sensors. gauges

[George Herold]

If I make a pumped probe thing, it needs a pressure gauge.
(maybe two) How to do that on the cheap?

For the pumped chamber I'd like a range of ~10-1000 Torr,
(But I'm happy for thread drift into other ranges)
So I got out "Building Scientific apparatus"
Their list of gauges:
Mechanical (Bourdon, McLeod)
Capacitance manometer. (Star this as good candidate!)
Thermo-conductivity gauges TC, Pirani...let me add convectron
which is also interesting candidate.
Viscous drag (below 1 Torr (10^-1 - 10^-6))
Ionization (again low pressure)

So I built this sorta pirani/ convectron gauge from a thermistor.
it was mehhh... sorta ok, but was fussy about zero's. and drifty.
(perhaps due thermistor waving around on, held by it's leads.)

But I'd like to talk about a capacitance manometer.
There's a plate (of a selected thickness.. different ranges)
and two chambers with electrodes on each side.
(OK is it just me or does wikipedia stink these days...
what happened to all the articles?)
OK this looks interesting, but overly complicated.
https://www.mksinst.com/n/baratron-capacitance-manometers
back in the 80's I used a gauge that had a chamber on each
side and you set the meter to read the difference.

Later, gotta run
George H.

 

[Phil Hobbs]

On 2020-01-28 15:13, George Herold wrote:

If I make a pumped probe thing, it needs a pressure gauge.
(maybe two) How to do that on the cheap?

For the pumped chamber I'd like a range of ~10-1000 Torr,
(But I'm happy for thread drift into other ranges)
So I got out "Building Scientific apparatus"
Their list of gauges:
Mechanical (Bourdon, McLeod)
Capacitance manometer. (Star this as good candidate!)
Thermo-conductivity gauges TC, Pirani...let me add convectron
which is also interesting candidate.
Viscous drag (below 1 Torr (10^-1 - 10^-6))
Ionization (again low pressure)

So I built this sorta pirani/ convectron gauge from a thermistor.
it was mehhh... sorta ok, but was fussy about zero's. and drifty.
(perhaps due thermistor waving around on, held by it's leads.)

But I'd like to talk about a capacitance manometer.
There's a plate (of a selected thickness.. different ranges)
and two chambers with electrodes on each side.
(OK is it just me or does wikipedia stink these days...
what happened to all the articles?)
OK this looks interesting, but overly complicated.
https://www.mksinst.com/n/baratron-capacitance-manometers
back in the 80's I used a gauge that had a chamber on each
side and you set the meter to read the difference.

Later, gotta run
George H.

We use this one in our SEM cathodoluminescence system to detect when the
chamber starts to vent, so that we can turn off the MPPC bias and return
to room temperature before water starts to condense on the cold plate.
Its range is 10 to 1200 millibars, about $2 in hundreds.

<https://www.te.com/commerce/DocumentDelivery/DDEController?Action=srchrtrv&DocNm=MS5607-02BA03&DocType=Data+Sheet&DocLang=English>


Cheers

Phil

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[John Larkin]

On Tue, 28 Jan 2020 12:13:31 -0800 (PST), George Herold
<ggherold@gmail.com> wrote:

If I make a pumped probe thing, it needs a pressure gauge.
(maybe two) How to do that on the cheap?

For the pumped chamber I'd like a range of ~10-1000 Torr,
(But I'm happy for thread drift into other ranges)
So I got out "Building Scientific apparatus"
Their list of gauges:
Mechanical (Bourdon, McLeod)
Capacitance manometer. (Star this as good candidate!)
Thermo-conductivity gauges TC, Pirani...let me add convectron
which is also interesting candidate.
Viscous drag (below 1 Torr (10^-1 - 10^-6))
Ionization (again low pressure)

So I built this sorta pirani/ convectron gauge from a thermistor.
it was mehhh... sorta ok, but was fussy about zero's. and drifty.
(perhaps due thermistor waving around on, held by it's leads.)

But I'd like to talk about a capacitance manometer.
There's a plate (of a selected thickness.. different ranges)
and two chambers with electrodes on each side.
(OK is it just me or does wikipedia stink these days...
what happened to all the articles?)
OK this looks interesting, but overly complicated.
https://www.mksinst.com/n/baratron-capacitance-manometers
back in the 80's I used a gauge that had a chamber on each
side and you set the meter to read the difference.

Later, gotta run
George H.

Maybe you could pulse-heat a thermistor or thinfilm RTD or high-TC
wire and observe the cool-down curve. Something low emissivity to
reduce radiation cooling?

How about a shielded 3-terminal capacitor?

How much does a commercial vacuum gage cost?



--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

 

[George Herold]

On Tuesday, January 28, 2020 at 3:49:10 PM UTC-5, John Larkin wrote:

On Tue, 28 Jan 2020 12:13:31 -0800 (PST), George Herold
ggherold@gmail.com> wrote:

If I make a pumped probe thing, it needs a pressure gauge.
(maybe two) How to do that on the cheap?

For the pumped chamber I'd like a range of ~10-1000 Torr,
(But I'm happy for thread drift into other ranges)
So I got out "Building Scientific apparatus"
Their list of gauges:
Mechanical (Bourdon, McLeod)
Capacitance manometer. (Star this as good candidate!)
Thermo-conductivity gauges TC, Pirani...let me add convectron
which is also interesting candidate.
Viscous drag (below 1 Torr (10^-1 - 10^-6))
Ionization (again low pressure)

So I built this sorta pirani/ convectron gauge from a thermistor.
it was mehhh... sorta ok, but was fussy about zero's. and drifty.
(perhaps due thermistor waving around on, held by it's leads.)

But I'd like to talk about a capacitance manometer.
There's a plate (of a selected thickness.. different ranges)
and two chambers with electrodes on each side.
(OK is it just me or does wikipedia stink these days...
what happened to all the articles?)
OK this looks interesting, but overly complicated.
https://www.mksinst.com/n/baratron-capacitance-manometers
back in the 80's I used a gauge that had a chamber on each
side and you set the meter to read the difference.

Later, gotta run
George H.



Maybe you could pulse-heat a thermistor or thinfilm RTD or high-TC
wire and observe the cool-down curve. Something low emissivity to
reduce radiation cooling?

I copied this thermistor/ bridge 'recipe' from a vacuum guy.
it was meh. I tried wrapping a bit of Al foil around it,
made no difference AFAICT

How about a shielded 3-terminal capacitor?
Air capacitor? Don't know.

How much does a commercial vacuum gage cost?

Oh boy...IDK. I'm going to say $500 gauge and meter.
but probably cheaper somewhere...
a cheap mechanical dial.. maybe $20-50?

OK looks like a mechanical gauge costs ~$10 in one from granger.

George H.

--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

 

[George Herold]

On Tuesday, January 28, 2020 at 4:13:59 PM UTC-5, Phil Hobbs wrote:

On 2020-01-28 15:13, George Herold wrote:
If I make a pumped probe thing, it needs a pressure gauge.
(maybe two) How to do that on the cheap?

For the pumped chamber I'd like a range of ~10-1000 Torr,
(But I'm happy for thread drift into other ranges)
So I got out "Building Scientific apparatus"
Their list of gauges:
Mechanical (Bourdon, McLeod)
Capacitance manometer. (Star this as good candidate!)
Thermo-conductivity gauges TC, Pirani...let me add convectron
which is also interesting candidate.
Viscous drag (below 1 Torr (10^-1 - 10^-6))
Ionization (again low pressure)

So I built this sorta pirani/ convectron gauge from a thermistor.
it was mehhh... sorta ok, but was fussy about zero's. and drifty.
(perhaps due thermistor waving around on, held by it's leads.)

But I'd like to talk about a capacitance manometer.
There's a plate (of a selected thickness.. different ranges)
and two chambers with electrodes on each side.
(OK is it just me or does wikipedia stink these days...
what happened to all the articles?)
OK this looks interesting, but overly complicated.
https://www.mksinst.com/n/baratron-capacitance-manometers
back in the 80's I used a gauge that had a chamber on each
side and you set the meter to read the difference.

Later, gotta run
George H.



We use this one in our SEM cathodoluminescence system to detect when the
chamber starts to vent, so that we can turn off the MPPC bias and return
to room temperature before water starts to condense on the cold plate.
Its range is 10 to 1200 millibars, about $2 in hundreds.

Nice, thanks. I'll give it a whirl.
George h.

https://www.te.com/commerce/DocumentDelivery/DDEController?Action=srchrtrv&DocNm=MS5607-02BA03&DocType=Data+Sheet&DocLang=English


Cheers

Phil

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[whit3rd]

On Tuesday, January 28, 2020 at 12:49:10 PM UTC-8, John Larkin wrote:

On Tue, 28 Jan 2020 12:13:31 -0800 (PST), George Herold
ggherold@gmail.com> wrote:

If I make a pumped probe thing, it needs a pressure gauge.

Maybe you could pulse-heat a thermistor or thinfilm RTD or high-TC
wire and observe the cool-down curve.

That's the principle of the thermocouple vacuum gage, and the
thermocouple is a better choice than a thermistor, because it's tolerant of higher
temperatures, and gives (intrinsically) a differential measurement. It
takes an AC regulated current for the classical ones, and readout is just
a voltmeter. They aren't expensive to make, but the commercial ones standardized
back in the vacuum-tube era on odd connectors and awkward 'standard'
specifications.

 

[Bill Sloman]

On Wednesday, January 29, 2020 at 7:13:38 AM UTC+11, George Herold wrote:

If I make a pumped probe thing, it needs a pressure gauge.
(maybe two) How to do that on the cheap?

For the pumped chamber I'd like a range of ~10-1000 Torr,
(But I'm happy for thread drift into other ranges)
So I got out "Building Scientific apparatus"
Their list of gauges:
Mechanical (Bourdon, McLeod)
Capacitance manometer. (Star this as good candidate!)

The capacitance manometer is a nice scheme. With modern electronics setting up an AC-excited Blumlein Bridge - sort of a Wheatstone Bridge, but with a centre-tapped transformer as two of the arms, and the capacitance to the (grounded) stretched metal diaphragm as the other two - is cheap and simple..

Berylium copper shim would have been my choice of diaphragm material back in about 1965.

The ones I've seen have been expensive. For it's electron microscope chambers Cambridge Instrument went with Pirani gauges fro coarse vacuum and ion gauges for high vacuum, but that choice was made way back when capacitance manometer electronics was a lot more expensive.

Thermo-conductivity gauges TC, Pirani...let me add convection
which is also interesting candidate.
Viscous drag (below 1 Torr (10^-1 - 10^-6))

A turbo-molecular pum as a vacuum gauge?

Ionization (again low pressure)

So I built this sorta pirani/ convection gauge from a thermistor.
it was mehhh... sorta ok, but was fussy about zero's. and drifty.
(perhaps due thermistor waving around on, held by it's leads.)

You can get surface mount thermistors these days

But I'd like to talk about a capacitance manometer.
There's a plate (of a selected thickness.. different ranges)
and two chambers with electrodes on each side.
(OK is it just me or does wikipedia stink these days...
what happened to all the articles?)
OK this looks interesting, but overly complicated.
https://www.mksinst.com/n/baratron-capacitance-manometers
back in the 80's I used a gauge that had a chamber on each
side and you set the meter to read the difference.

Baratron was the best known supplier back then.

From my Ph.D. thesis

J.J Opsteleten and N. Warmholtz Applied Science Research Hague B4 page 329 (1955)

J.J Opsteleten, N. Warmholtz and J.J. Zaalberg von Zelst ibid B6 pages 129 (1956)

D.R. Lovejoy Rev. Sci. Instrum 32 page 41 (1961)

The first two were from the Philips research lab and meticulous. The third one was American, and less so.

At the time I was thinking about an electrostatic force balance version of the capacitance manometer. It wouldn't have been any good for pressure differences over a few torr, but it was an entertaining idea.

--
Bill Sloman, Sydney

 

[Bill Sloman]

On Wednesday, January 29, 2020 at 8:13:59 AM UTC+11, Phil Hobbs wrote:

On 2020-01-28 15:13, George Herold wrote:
If I make a pumped probe thing, it needs a pressure gauge.
(maybe two) How to do that on the cheap?

For the pumped chamber I'd like a range of ~10-1000 Torr,
(But I'm happy for thread drift into other ranges)
So I got out "Building Scientific apparatus"
Their list of gauges:
Mechanical (Bourdon, McLeod)
Capacitance manometer. (Star this as good candidate!)
Thermo-conductivity gauges TC, Pirani...let me add convectron
which is also interesting candidate.
Viscous drag (below 1 Torr (10^-1 - 10^-6))
Ionization (again low pressure)

So I built this sorta pirani/ convectron gauge from a thermistor.
it was mehhh... sorta ok, but was fussy about zero's. and drifty.
(perhaps due thermistor waving around on, held by it's leads.)

But I'd like to talk about a capacitance manometer.
There's a plate (of a selected thickness.. different ranges)
and two chambers with electrodes on each side.
(OK is it just me or does wikipedia stink these days...
what happened to all the articles?)
OK this looks interesting, but overly complicated.
https://www.mksinst.com/n/baratron-capacitance-manometers
back in the 80's I used a gauge that had a chamber on each
side and you set the meter to read the difference.

Later, gotta run
George H.



We use this one in our SEM cathodoluminescence system to detect when the
chamber starts to vent, so that we can turn off the MPPC bias and return
to room temperature before water starts to condense on the cold plate.
Its range is 10 to 1200 millibars, about $2 in hundreds.

https://www.te.com/commerce/DocumentDelivery/DDEController?Action=srchrtrv&DocNm=MS5607-02BA03&DocType=Data+Sheet&DocLang=English

There was a vogue for using diaphragms etched into a silicon wafer with built-in strain gauges. Motorola and Honeywell both used to sell them fairly cheaply.

--
Bill Sloman, Sydney

 

[Mike Coon]

In article <e73cfb92-293e-44b8-990d-aee924f5b626@googlegroups.com>,
ggherold@gmail.com says...

If I make a pumped probe thing, it needs a pressure gauge.
(maybe two) How to do that on the cheap?

For the pumped chamber I'd like a range of ~10-1000 Torr,
(But I'm happy for thread drift into other ranges)
So I got out "Building Scientific apparatus"
Their list of gauges:
Mechanical (Bourdon, McLeod)
Capacitance manometer. (Star this as good candidate!)
Thermo-conductivity gauges TC, Pirani...let me add convectron
which is also interesting candidate.
Viscous drag (below 1 Torr (10^-1 - 10^-6))
Ionization (again low pressure)

So I built this sorta pirani/ convectron gauge from a thermistor.
it was mehhh... sorta ok, but was fussy about zero's. and drifty.
(perhaps due thermistor waving around on, held by it's leads.)

But I'd like to talk about a capacitance manometer.
There's a plate (of a selected thickness.. different ranges)
and two chambers with electrodes on each side.
(OK is it just me or does wikipedia stink these days...
what happened to all the articles?)
OK this looks interesting, but overly complicated.
https://www.mksinst.com/n/baratron-capacitance-manometers
back in the 80's I used a gauge that had a chamber on each
side and you set the meter to read the difference.

Later, gotta run
George H.

Since you claim to be OK with thread drift, I'll throw in "Crookes
Radiometer". They are sensitive to low-range gas pressure, but I have no
idea whether there is calibration data somewhere.

Also when I was playing with pumping on liquid helium I needed a
silicone oil manometer measuring pressure relative to vacuum in the
closed-off arm. But the oil would "stick" to the closed-off end and not
measure. I solved by using liquid nitrogen (which we would always have
handy if using helium) soaked into cotton waste to temporarily pack
around the closed end. That would freeze the oil which cracked and
released the stick. My own invention; worked every time!

Mike.

 

[John Larkin]

On Tue, 28 Jan 2020 18:37:53 -0800 (PST), whit3rd <whit3rd@gmail.com>
wrote:

On Tuesday, January 28, 2020 at 12:49:10 PM UTC-8, John Larkin wrote:
On Tue, 28 Jan 2020 12:13:31 -0800 (PST), George Herold
ggherold@gmail.com> wrote:

If I make a pumped probe thing, it needs a pressure gauge.

Maybe you could pulse-heat a thermistor or thinfilm RTD or high-TC
wire and observe the cool-down curve.

That's the principle of the thermocouple vacuum gage, and the
thermocouple is a better choice than a thermistor, because it's tolerant of higher
temperatures, and gives (intrinsically) a differential measurement. It
takes an AC regulated current for the classical ones, and readout is just
a voltmeter. They aren't expensive to make, but the commercial ones standardized
back in the vacuum-tube era on odd connectors and awkward 'standard'
specifications.

The usual t/c vacuum sensor is a 4-wire device, with separate heater
and sensor elements, and works steady-state. Now that we have
microprocessors, we could have a simpler 2-wire element that's both
the heater and the sensor, probably with pulsed heating.

A thermocouple made of thin wire could be the heater and the sensor.

--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

 

[George Herold]

On Wednesday, January 29, 2020 at 12:57:59 AM UTC-5, Bill Sloman wrote:

On Wednesday, January 29, 2020 at 7:13:38 AM UTC+11, George Herold wrote:
If I make a pumped probe thing, it needs a pressure gauge.
(maybe two) How to do that on the cheap?

For the pumped chamber I'd like a range of ~10-1000 Torr,
(But I'm happy for thread drift into other ranges)
So I got out "Building Scientific apparatus"
Their list of gauges:
Mechanical (Bourdon, McLeod)
Capacitance manometer. (Star this as good candidate!)

The capacitance manometer is a nice scheme. With modern electronics setting up an AC-excited Blumlein Bridge - sort of a Wheatstone Bridge, but with a centre-tapped transformer as two of the arms, and the capacitance to the (grounded) stretched metal diaphragm as the other two - is cheap and simple.

Berylium copper shim would have been my choice of diaphragm material back in about 1965.

The ones I've seen have been expensive. For it's electron microscope chambers Cambridge Instrument went with Pirani gauges fro coarse vacuum and ion gauges for high vacuum, but that choice was made way back when capacitance manometer electronics was a lot more expensive.

Thermo-conductivity gauges TC, Pirani...let me add convection
which is also interesting candidate.
Viscous drag (below 1 Torr (10^-1 - 10^-6))

A turbo-molecular pum as a vacuum gauge?

Yeah! According to 'Building Scientific Apparatus' they
spin a magnetically suspended steel ball up to 415 Hz,
and then measure how long it takes to slow dow to 405 Hz...
(several minutes at the low pressure range.) A spendy bit of kit!

(Hey Bill thanks for the nice response.)

George H.

Ionization (again low pressure)

So I built this sorta pirani/ convection gauge from a thermistor.
it was mehhh... sorta ok, but was fussy about zero's. and drifty.
(perhaps due thermistor waving around on, held by it's leads.)

You can get surface mount thermistors these days

But I'd like to talk about a capacitance manometer.
There's a plate (of a selected thickness.. different ranges)
and two chambers with electrodes on each side.
(OK is it just me or does wikipedia stink these days...
what happened to all the articles?)
OK this looks interesting, but overly complicated.
https://www.mksinst.com/n/baratron-capacitance-manometers
back in the 80's I used a gauge that had a chamber on each
side and you set the meter to read the difference.

Baratron was the best known supplier back then.

From my Ph.D. thesis

J.J Opsteleten and N. Warmholtz Applied Science Research Hague B4 page 329 (1955)

J.J Opsteleten, N. Warmholtz and J.J. Zaalberg von Zelst ibid B6 pages 129 (1956)

D.R. Lovejoy Rev. Sci. Instrum 32 page 41 (1961)

The first two were from the Philips research lab and meticulous. The third one was American, and less so.

At the time I was thinking about an electrostatic force balance version of the capacitance manometer. It wouldn't have been any good for pressure differences over a few torr, but it was an entertaining idea.

--
Bill Sloman, Sydney

 

[Neon John]

On Wed, 29 Jan 2020 10:59:15 -0800, John Larkin
<jlarkin@highland_atwork_technology.com> wrote:

On Tue, 28 Jan 2020 18:37:53 -0800 (PST), whit3rd <whit3rd@gmail.com
wrote:

On Tuesday, January 28, 2020 at 12:49:10 PM UTC-8, John Larkin wrote:

The usual t/c vacuum sensor is a 4-wire device, with separate heater
and sensor elements, and works steady-state. Now that we have
microprocessors, we could have a simpler 2-wire element that's both
the heater and the sensor, probably with pulsed heating.

A thermocouple made of thin wire could be the heater and the sensor

John,

You're probably aware that I make neon as a hobby and part time
business.

I use a 2 wire Teledyne sensor for moderate vacuum. It comes on scale
at 300 Torr if memory serves and can be read down to about 0.5 Torr by
interpolating the pointer's position between 1 and 0 Torr.

For filling the neon tube with gas, I use a capacitance gauge with an
absolute pressure diaphragm.

For high vacuum, I have an ionization gauge. It goes down to 10e-7
Torr

The thermocouple cannot be used with any gas other than air because of
radical differences in gas thermal conductivity. I spent a lot of
time preparing translation tables for Argon and Neon. Not usable for
Xenon and Krypton because their very low thermal conductivity.

I have a few Baratron tubes for ultra-high vacuum. I don't use them
for neon work, though, because a tiny of mercury vapor will usually
kill the tube. I've tried baking out a contaminated tube at 400F and
moderate vacuum. Some will come back to life and some won't. Those
make nice shelf decorations

The only comprehensive supplier for everything vacuum is Dunaway
Stockroom. Everything from new instruments and fittings to used
equipment (refurbed or not. Very helpful technical people too.

John

John DeArmond
http://www.neon-john.com
http://www.tnduction.com
Tellico Plains, Occupied TN
See website for email address

 

[George Herold]

On Friday, January 31, 2020 at 4:33:06 PM UTC-5, Neon John wrote:

On Wed, 29 Jan 2020 10:59:15 -0800, John Larkin
jlarkin@highland_atwork_technology.com> wrote:

On Tue, 28 Jan 2020 18:37:53 -0800 (PST), whit3rd <whit3rd@gmail.com
wrote:

On Tuesday, January 28, 2020 at 12:49:10 PM UTC-8, John Larkin wrote:

The usual t/c vacuum sensor is a 4-wire device, with separate heater
and sensor elements, and works steady-state. Now that we have
microprocessors, we could have a simpler 2-wire element that's both
the heater and the sensor, probably with pulsed heating.

A thermocouple made of thin wire could be the heater and the sensor

John,

You're probably aware that I make neon as a hobby and part time
business.

I use a 2 wire Teledyne sensor for moderate vacuum. It comes on scale
at 300 Torr if memory serves and can be read down to about 0.5 Torr by
interpolating the pointer's position between 1 and 0 Torr.

For filling the neon tube with gas, I use a capacitance gauge with an
absolute pressure diaphragm.

For high vacuum, I have an ionization gauge. It goes down to 10e-7
Torr

The thermocouple cannot be used with any gas other than air because of
radical differences in gas thermal conductivity. I spent a lot of
time preparing translation tables for Argon and Neon. Not usable for
Xenon and Krypton because their very low thermal conductivity.

I have a few Baratron tubes for ultra-high vacuum. I don't use them
for neon work, though, because a tiny of mercury vapor will usually
kill the tube. I've tried baking out a contaminated tube at 400F and
moderate vacuum. Some will come back to life and some won't. Those
make nice shelf decorations

The only comprehensive supplier for everything vacuum is Dunaway
Stockroom. Everything from new instruments and fittings to used
equipment (refurbed or not. Very helpful technical people too.

John

John DeArmond
http://www.neon-john.com
http://www.tnduction.com
Tellico Plains, Occupied TN
See website for email address

Thanks John D. Google the convectron gauge. It's a Pirani gauge
that uses convective cooling to increase the pressure range to ~1 atm.
But not that accurate.
I like Duniway stock room. But all things vacuum are over priced.
(IMHO). One option is to use the cheaper 'tri-clamp' style
fittings. These are used in the food industry and seem to work fine a
moderate vacuum. (o-rings)

George H.