Re[4]: Humidity sensor

[Peter Kravtsov <pkravt@gmail.com>]

Hello Peter,

Our humidity graph is corrected to 17 bar pressure of course.
I took  the saturation vapor pressure from reference book for
different temperatures, and fit this dependence.
The complete formula for calculating ppm from T_dewpoint is:
ppm = 10^(-43.42291 + 0.43575*T - 0.00149*T^2 + 0.00000188204*T^3)
This formula works for 10 bar hydrogen.
To correct the result for 17 bar, I simply divided it by 1.7.
Is it not right?

With regard to web page about CHUPS, we already have one in
our laboratory web site:
http://lkst.pnpi.nw.ru/hydrog/main.html

It is not updated by this visit results, but I'll do it as soon as
possible.

Best wishes,
Peter.


Monday, June 27, 2005, 6:41:05 PM, you wrote:

> Dear Peter,

> I just talked to Dave Alan from the PURA company and emailed 
> your graph.

> - He was surprised by the large temp sensitivity and
>   needs 24 hours to discuss with their techs. Dave said the
>   sensor is calibrated at 21 C.

> - The response to a humidiy change is usually dominated
>   by the surfaces and pipes before the sensor, not
>   by the sensor itself. He will send us an article
>   about the factors involved. I mentioned to him that
>   be bypass the volume and directly sample the input
>   flow, which is filtered and expected to be very dry.
>   He will discuss this question also with his techs.

> - You plot the graph in humidity in ppm. Is that corrected
>   for the 17 bar pressure of the system? How? Could
>   you give us the dewpoints, you were reading during
>   these measurments.

> We hope to have more information tomorrow.

> With best regards

> Peter 



> On Mon, 27 Jun 2005, Peter Kravtsov wrote:

>> Dear Peter,
>> 
>> 1. Time between the temperature points is about 2 minutes.
>>    (temperature rate is 1 degree per 2 minutes)
>> 
>> 2. We measure 4-20 mA sensor signal with 301 Ohm resistor,
>>    followed by a unity gain amplifier and 16-bit ADC with the
>>    range of [+ - 10V]. This gives about 0.3mV resolution.
>>    Total ADC error is about +- 2 LSB [+- 0.6 mV].
>> 
>> Best regards,
>> Peter-the-second.
>> 
>> 
>> 
>> Monday, June 27, 2005, 6:06:50 PM, you wrote:
>> 
>> > Dear Peter,
>> 
>> > I will try to contact the company today. 
>> 
>> > Please give me a bit more details:
>> >   - measurement time per temp point in your graph
>> >   - current measurement, which resistor did you use?
>> 
>> > Other relevant experimental details which might help.
>> 
>> > Best regards
>> 
>> > Peter
>> 



-- 
Best regards,
 Peter                            mailto:pkravt@gmail.com