Thanks Peter, I'm rapidly going off the cap. idea! Actually 20cm is probably more like the actual size. I use a tank open to the water at the bottom with an air intake at the top of the tank that goes via a small compressor into a sealed tank. To dive the compressor pulls air out of the tank into the sealed tank. To surface the compressed air is released back into the tank forcing out the water (so long as you are not too deep!). So I'm currently looking at using the pressure in the sealed tank as an indication of the amount of water taken on... I could go with two float type devices but I don't think I will have the space (that was why the cap. approach looked nice not much space used). Thanks Andy ----- Original Message ----- From: "Peter L. Peres" To: Sent: Wednesday, October 18, 2000 9:37 PM Subject: [EE]: Help water level > >3. Continuous range from say 1 cm to say 30 cm deep, with 1mm accuracy. > >4. Possibly best not to use anything that depends upon pressure (since > >this may vary depending upon sub depth). > > So you want to try a wet resistance type of sensor with unknown (changing) > water of unknown temperature and make it accurate to 0.3% ? I hope that > your analog design and electrochemistry knowledge is in the range > between excellent and amazing, because you will need it. > > Most subs do not stay on even keel most of the time so you actually need > two sensors and average the read heights. Given the size it sounds more > like a 1/2 caliber torpedo than a model sub ;-) > > Now, a pair of capsulated potentiometers or torsion transducers and a pair > of small floats will probably get you near there (if 2% is near enough). > You can easily average the two readings in analog mode before converting > to digital. Don't forget to account for when the fish is standing on end > ;-). And do not rely on gravity to keep the floats on the surface of the > liquid, use springs. > > good luck, > > Peter > > -- > http://www.piclist.com#nomail Going offline? Don't AutoReply us! > use mailto:listserv@mitvma.mit.edu?body=SET%20PICList%20DIGEST > > > > -- http://www.piclist.com hint: To leave the PICList mailto:piclist-unsubscribe-request@mitvma.mit.edu