Roman Black wrote:

>Why not use one cheap thermistor attached to
>the tube,

That is also a good idea i might try.

Problem might be wind might blow, and that combined with moist dirt might=
 have as good cooling effect as the warmest water it has to detect.
Or greasy dirt isolating the thermistor from detecting water reliably. =
 Or...

Abour power need:
For any sensor i choose i may pulse it on just enougth to make it detect,=
 for one second intervals.

I guess i need *much* (thousand) longer on periods for NTC heating, than for=
 NMRI exitation pulse.

Also piezo pulsing can use very short on-times.

Another problem i am sure i wil encounter is that there might be radio=
 sources within meters, or somebody using a GSM phone...  There are also=
 often radar nearby, sometimes a gasoline engine that might have noisy=
 ignition system.  And a PIC of course and tiny switched power supply within=
 a couple centimeters.  Why do=B4nt i just drop this project...? ;)
Then, thermistor is best as i can discard any high freq. =20
I guess piezo-acoustic may give good readout, be reasonalby EMC tolerant,=
 and low drive (short pulses)

Peter:
Good idea, the microwave absorption might also be a solution.
If we could make a cheak microwave osc and reciever.
I have no RF experience so i have no ideas.

BTW, both microwave and NMRI solutions will probably make it different for=
 certification.  2,45(?) GHz is free for low power (and spread spectrum?)=
 IIRC.  But precise microwave do sound expensive.

Also moist dust with salt might damp much too.

And it might conduct too much to use a capacitive method to differ it from=
 water sucessfully.

I still believe most in acoustic method, where i can demand a large part of=
 the cavity to be filled before detection.  Water is the only matter nearby=
 that is plenty enough to fill it. (within reason)

/Morgan

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads