Helium Leak Testing

[giantsjoed]

I purchased a used Helium leak detector to perform some cryogenic testing on ball valves.
The detector did not have a sniffing device.
Do I need a specific sniffing device for a Leybold Ultratest F (expensive) or can I buy a generic device or can a device be manufactured?

 

[NGiLuzzu]

giantsjoed,
you may want to visit your leak detector Manufacurer's website, at the page

http://www.leybold.com/corporate/produkte_fr.html

and select "Leak Testing Instruments"; once you choose your model, click on "Accessories"; a list will be displayed and the last row is "Helium Sample Probes (Sniffers)".
So (considering the high cost of the detector) your local Leybold Representative should quickly provide you one of this probes... maybe ;-)

May I ask you what standards and/or specifications are you going to follow for cryogenic tests? Are your limits on external leakage (fugitive emissions) expressed in terms of flow (mbar*l/s or equivalent) or in terms of concentration (ppm or similar)? And what are the units of your detector's readout?

There's been a few discussion about this issue within this site: see Thread408-64093, Thread408-48808 and Thread124-48813, for example.

Hope this helps, 'NGL

 

[scalleke]

Having disassembled a Varian sniffer probe I believe it is definitely possible to manufacture such a device from standard parts.

A sniffer probe is no more than a small dia metallic tube attached to a hose going into the leak detector via a small needle valve.

I do not believe however if one purchases the individual parts that it is possible to make one cheaper than you can buy them.

Best regards.

scalleke

 

[giantsjoed]

Thanks 'NGL, but I have an older machine, an Ultratest F.
I did find a repair site, Tech Services in Connecticut that also sells sniffing wands for these older machines.

The specification for testing will be the Shell Spec., British Standard, a KBR spec., etc. Trying to cover as many specs as possible. The limits of units are in ml/sec, mbar.l/s, mm3/s, probably others. The unit has a readout in atm.cc/s.

Joe

 

[NGiLuzzu]

Giantsjoed,
could you please tell me the number/s of the British Standard/s applicable to helium leak detection? Are they about this issue in general or related to valves?

Many thanks in advance, 'NGL

 

[giantsjoed]

'NGL

BS 6364:1984
Specification for valves for cryogenic service

Although I will probably refer to:
EN 1626:1999
Cryogenic vessels - Valves for cryogenic service

Joe

 

[NGiLuzzu]

Thank you, giantsjoed!

As far as I can see, your limits are all in terms of flow rates. Have you ever met limits to helium concentration (ppm)? It happens sometimes in Spec's coming from U.S.A. (e.g. Bechtel ones)...

About BS 6364 (see also Thread408-81124 and Thread774-89124), in clause A.3.4(e) it says: "... check the valve gland and body/bonnet joint for leak tightness. There shall be no visible leakage".
What do you consider to be no visible leakage? A readout that remains of the same magnitude as helium background in air? In numerical terms?

Best Regards, 'NGL

 

[giantsjoed]

Fortunately, NGL, the Shell spec and KBR specification are all mass per unit of time.

No visible leakage generally applies to a hydrostatic test, in which case water is readily visible if leaking. So when the valve is full of helium, I can only assume that if I spray the joints with toluene or another low freezing solution and do not see bubbles, then I meet “No visible leakage”. No meter required.

Joe