moisture anakyzer - present status

[Bernhard Lauss <lauss@berkeley.edu>]

Dear Peter,


I think there are the following arguments to be made:


1) Any measurment of a moisture level is valuable in order
to show that the impurity yield we see is moisture.
A 50% accuracy to prove that moisture causes the effect is acceptable.



2) What accuracy do we need for our lifetime correction ?

The +- 2 degree accuracy of the NYAD and Pura device
transform into an realtive error of 30-40% on the
concentration calculated from the dew poit / saturation pressure.


Getting the correction to the lifetime from your Z>1 impurities table
(delta_r @ Y=10 - for Oxygen I calculate a concentration of 60 ppb @ Y=10)
assuming a moisture content mc in 10 bar hydrogen
I get the following table:


moisture    dew point   error        lifetime
content     C +- 2C     on mc        correction
mc in ppb               ppb from     ppm
		        dew point

9	    -90	        -3 / +4       3 +-  1

38          -82         -12/+15      10 +-  4

75          -78         -22/+26      19 +-  8

100         -76         -26/+38      27 +- 11



- Do I understand your calculation of the lifetime correction correct.

- Why is ther the factor 2 difference in Oxygen with respect to you
Note" Hydrogen impurities in MuCAP" vesion 10/3/2003


Form the table I would conclude that
i) we either have to be sure that we can do a better
cleaning than 40 ppb moisture in the TPC
or
ii) we need a more accurate dew point precision


So the statement from PURA
"If we assume a worst-case combined accuracy and resolution of +/-3C, then
-95.23 @ 10 bag = 0.003 ppmV and -89.23 @ 10 barg = 0.008 ppmV.
As such we can achieve a measurement at 0.005 ppmV to +0.003/-0.002.
We can therefore conclude that PURA will measure at 5ppbV to a
sensitivity better than 5ppbV, in fact sensitivity is encompassed
in resolution which equates to better than 1ppbV under these conditions."

Is kind of misleading for us, as this also means only a 40% accuracy on
the measured value and assumes that we get the cleaning down to 5 ppb.




The chapest more accurate dew point analyzer as far as I know
is the CR1-A from Buck Research at
http://www.hygrometers.com/CR1specs.htm
which has 0.1 degree C accuracy, a factor of
20 better than the 2 other devices.



2) How much money can we spend ?


comparing the prices

NYAD	$  2,300

PURA    $  5,000

Buck	$ 12,500



3) Difference in company conduct:

Pura -  many replies " saying sorry we do not have the answer yet "

NYAD -  very hard to reach, nobody picks up the phone which is
        only answered between 9:30 and 14:30 PMT anyway.
	1 eMail response/2 weeks
        However, I could drive over, they are in Martinez.



I think the decision path might be along:

- we (certainly Berkeley) don't have 12k$ to be spent on that problem
- it is not clear to me that there is an intrinsic advantage/disadvantage
for the PURA/NYAD model.
- we won't get all the stability data you want to have from the company
so we have to trust it in a way
- sent it to NIST for calibration is also extremely expensive
http://www.cstl.nist.gov/div836/836.05/thermometry/calibrations/fees.htm#hum

- so calibrate it before the run and after the run with the company
might be the option
- both PURA and NYAD have comparable sensitivity and accuracy

Therfore, spent 2,300 $ on a device which helps to a certain extent.


What do you think ?

				Best regards


				     Bernhard




*******************************************************************
Bernhard Lauss                 E-Mail: lauss@socrates.berkeley.edu
Physics Department
366 LeConte Hall
University of California @ Berkeley
Berkeley, CA 94720                             Tel: (510)-642 4057
United States                                  Fax: (510)-642 9811
*******************************************************************