[wxsig] BP calibration (was T-238 Humidity Calibration)

[Will Beals]

Bob:

In your earlier email you actually hit upon the great debate on several
forums on how best to do BP measurements.  The problem was exactly one of
resolution versus range.  While I wasn't a direct participant in the debate,
the root issue was that in order to get the needed resolution out of the
sensor to be useful, you had to apply the offset to the analog signal before
it got digitized.  If you just fed the raw uncalibrated BP sensor into the
1-wires A/D chip, the needed range to incorporate the baseline pressures
from sea level to 8000+ feet meant that the bit left to determine the change
in resolution were not enough for meaningful BP change tracking.

At the end of the debate, the resultant sensors that came out and what I
ended up supporting assume the offset and "gain" are done in the analog
domain.

will

-----Original Message-----
From: wxsig-bounces at lists.tapr.org [mailto:wxsig-bounces at lists.tapr.org] On
Behalf Of Robert McGwier
Sent: Friday, September 02, 2005 6:30 AM
To: TAPR Weather Station SIG Mailing List
Subject: Re: [wxsig] BP calibration (was T-238 Humidity Calibration)

Ooops.   I just took the title of the note, obviously what I said below 
applies to the calibration of ALL sensors and in particular to the BP.  
If you can take measurements of known conditions, you can derive the
transfer function in software and not need to twiddle pots.

Bob
N4HY


Robert McGwier wrote:

> Will:
>
> Since you are using the 68HC908, I do not understand why serious 
> tinkering with pots are necessary.   If you measured the sensor 
> outputs with a sufficiently dense set of points (and this cannot be 
> that many),  an approximation to the exact "transfer function" from 
> humidity reading to actual humidity should be easily derived.  The set 
> of points stored and then all readings are run through this 
> "calibration engine" to produce the actual reading.
>
>
> What am I missing?
>
> Bob
> N4HY
>
>



_______________________________________________
wxsig mailing list
wxsig at lists.tapr.org
https://lists.tapr.org/cgi-bin/mailman/listinfo/wxsig