Not claiming to know what causes the effects we see on the detectors, but if we use the proposed simple relationship between increasing ambient pressure = increasing detector pressure without allowance for back-streaming of atmospheric gases in to the detectors, how does this hypothesis explain the variation in detector response (e.g. bias voltage difference in Brian Joy's data set) for changes in pH2O for a constant overall atmospheric pressure (i.e moving vertically on the plot below)?
No question we've got more than one variable here, but the most striking aspect of Brian Joy's plot above are the sloping trends on bias as a function of barometric pressure. In other words, he saw that as the barometric pressure increases, one requires more bias voltage to maintain a PHA peak at 4 v. That is perfectly consistent with the hypothesis that more atoms (the higher pressure in 2 atm detectors vs 1 atm detectors) requires more bias voltage to operate.
I'm also not sure how it would explain the difference in response between the inlet and exhaust GFPC, both in terms of the scale of the vertical difference with pH20 and the overall difference in bias voltage.
Good question. I wish he had recorded the temperature of his P-10 bottle instead, but the relationship between the dew point and relative humidity is:
Td = T - ((100 - RH)/5.)
where T is temperature and RH is relative humidity. Note that the dew temperature can never be higher than the actual temperature. Given that his instrument is in Canada, maybe we can assume that low dew temperatures are due to winter (colder temperatures and drier air) and higher dew temperatures are seen in summer when the humidity is higher. In any event, for a constant relative humidity the dew temperature is essentially the actual temperature. So where is his P-10 bottle located?
The above plot seems to reveal that the higher the dew temperature (higher relative humidity), the higher the voltage necessary to maintain the PHA peak at 4v. So if H2O can indeed infiltrate the P-10 system, perhaps the water is acting as a quench gas? Or maybe the humidity is affecting the high voltage capacitors in the spectrometer pre-amps?
Though I suspect though this response is more likely some kind of plain temperature dependency in the P-10 gas or lab temperature.
As with everthing to do with physics, I suspect the mechanism that is at work on the detector response to atmospheric pressure is anything but straightforward!
Yes, indeed.