Brainstorming Time Ideas on how to measure the water level in my shuttle system, its a silent stream thing, so i have a filler tube rather than a resiviour i want to measure. Best thought i've had so far is UV led + active die, then measure where it is. only thing is that wouldn't be that easy to detect the water. any thoughts?
If you send a narrow beam through the tube, won't water's refractive index bend it more than an empty tube? Oh, and use a PIC!!!!
There are special cups for the visually impaired that emit a beep when the water level in them reaches a certain level to indicate this. Perhaps the same device could be used in this way? Just a thought
After searching about a bit I found this. This person used the same circuit. I dont really know if it would be suited to a down fill tube though.
I remember seeing something about fluid level sensing in the datasheets for the q-prox stuff, maybe you could use one of the q-slides and make a sensor for that.
http://www.performance-pcs.com/cata...=2654&osCsid=3374b5254117331bf9b92a99105e180f Use something like that?
The trouble with something like a probe in the water is that you're going to get considerable electrolysis over time. Maybe if you put a little plastic ball in your filler tube that floats. That should be more detectable than just water to some photodiodes + IR LEDs?
I was going to use a ball but what would happed you forgot to top your water up or the ball worked its way down the tube? It would go straight into the pump and either block it or brake it, then you can say bye bye to your pc :/
Could you insert a T-line at the very bottom of your water cooling system? The bottom end of the T-line could be attached to a pressure sensor that would alert you if the pressure that the water is exerting on the sensor fell below a certain level in relation to gravity. A PIC could be used to recalibrate the threshold of the sensor whenever the pump is on/off so that false triggering does not occur.
Just use a larger diameter filler tube to the rest of the system. A ball to fit the filler tube would then be too big to end up in the main loop.
Maybe a PTC thermistor in a voltage divider so it heats up a little. When there is water, it will remain cool. When there is no water, it heats up and the voltage changes. Or a pair of ultrasonic transducers. Use one to create ultrasonic waves and the other to receive them. Then let the laws of physics do the rest.
Make a capacitor: two pieces of isulated wire kept at uniform distance from one another, more water = greater capacitance. This page gives the dielectric constant of water at 80, while air is 1, should make for a very accurate scale.
i like your dye + light detector idea the most. if you have hte light detector directly under the tube, and the light source from straight above, it should work well, provided the water is the same colour when you top it up.
The 4093 is a CMOS quad NAND Schmitt Trigger gate, but only 2 are used. IC1a, R1, C1 make up a fairly high-frequency oscillator. The square-wave output passes through the caps C2,C3 to stop any DC going through the liquid. If the probe can conduct (and with CMOS all but the purest water will conduct enough) the AC signal goes through, is rectified and charges C4, so IC1b inputs are high; IC1b output goes low switching on the PNP transitor and lighting the indicator. If the probe stops conducting, C4 discharges through R2 until the gate switches the other way and the lamp goes off. You could swap the lamp for a relay with inverse parallel protection diode (1N4001), or use one of the spare gates as an inverter to turn a light on when the level drops. All the caps should be the standard 2.2n, R2 can be any value 10M-22M
Nice, really nice. : Now alas me rey has all five ports in use, but I could drill'n'fill fairly safely. Of course its not strictly necessary, but I have a thing for lights that indicate something, anything! Thanks for that one. EDIT: Just one thing, if the res itself is conductive (Annodised Aluminium) and conntected to the case (grounded, one assumes), would there be any issues with this AC signal confusing things.
i'm worried about pusling power in the filler tube, the effect it might have on the pump so i think i'll try with infra-red + ultrasound. hopefully they will work!
This idea is a good one, as someone else said, this is exactly what qprox advertise.. I bet it would be more reliable than ir. The problem I had with w/c was that I had various stuff leach into the tubes, I am sure that that had to do with me using rubbish pvc rather than silicone or tygon but whatever, this would certinally affect a uv system, less so a capacititive system. One problem you may have with the qprox chip is that it automatically recalibrates slowly over time to take acount of temperature changes so it may not actually work. If you have the inclination, create a pulse circuit and let the output of that pulse the speed sensor of the unused fan sockets. That should let you see your water level in mbm or whatever else you use. Rod
This pdf file seems to describe almost exactly what you are looking for. If you could replace the analog meter with a digital bar graph one or keep it as it is. and suspend the capacitive probe in the tube.
