Post by thread:[EE]:,[PIC]: How to measure level of liquids and s

Dan Michaels wrote: {Quote hidden}> Mark wrote: > .... > >> > >> Plus, I suspect you could use 2 or more thermistors arranged > >> vertically within the tube, plus a little math, to actually > >> determine the level, not just whether above/below one point. > > > >True on multiple height determination - though (despite my being pretty > >good in math) I don't see where the math's needed/applicable. I thought > >with 3 hot beads, you'd know one of 4 states, {"Height < A" || "Height > >>= A" || "Height >= B" || "Height >= C"}, don't see how you can determine > shades of height between say A and B (putting A lowest, C highest here, to > clarify.) Am I missing something? Has been a "Bad Brain Day" to some > degree <G> (There's not a big difference for thermistor B in temperature, > between temperature for water say 1 foot below it and 1 inch below it, > barring splashing, is there? Has been a while since I last hot-beaded <G>) > > > > Not that I've ever tried this, but I figure the temperature variation > would distribute itself along the tube in some manner [to be measured > and calibrated empirically], and the math would be some weighted > averaging routine. Just a thought. The sorta thing you try once, and > immediately get a strong hunch whether it will ever work or not. > > - Dan Aaah, you're thermally bonding the hot beads to the tube? Perhaps that'd work, might want to instead put a heater at the top of a small rod, bond the thermistors to the rod, use them to measure along the rod, maybe would work. Good news is that you can use one larger PIC and run quite a few thermistors simultaneously to get many data points simultaneously. Sigma Delta code's not too hard to find <G> Mark

Mark wrote: ..... >> Not that I've ever tried this, but I figure the temperature variation >> would distribute itself along the tube in some manner [to be measured >> and calibrated empirically], and the math would be some weighted >> averaging routine. Just a thought. The sorta thing you try once, and >> immediately get a strong hunch whether it will ever work or not. >> >> - Dan > >Aaah, you're thermally bonding the hot beads to the tube? Perhaps >that'd work, might want to instead put a heater at the top of a small >rod, bond the thermistors to the rod, use them to measure along the rod, >maybe would work. > Yeah, that's pretty much what I had in mind. Although I am not a thermal engineer, it seems to me there will generally be a temperature difference between the fluid, now matter what it is, and the air above it - due to thermal inertia - so a heater may not be required. I'd have to try this to see how it would work. It might work with 3 thermistors this way - one at the bottom [in the fluid], one at the top [out of the fluid], both of which act as references, and one in between to do the actual weighted measurement - all bonded to the same long tube. ============= >Good news is that you can use one larger PIC and run quite a few >thermistors simultaneously to get many data points simultaneously. >Sigma Delta code's not too hard to find <G> > For production stuff, I think most companies would want to go with as few parts as they could get away with --> costs, plus reliability issues. - Dan

Dan Michaels wrote: {Quote hidden}> Mark wrote: > ..... > >> Not that I've ever tried this, but I figure the temperature variation > >> would distribute itself along the tube in some manner [to be measured > >> and calibrated empirically], and the math would be some weighted > >> averaging routine. Just a thought. The sorta thing you try once, and > >> immediately get a strong hunch whether it will ever work or not. > >> > >> - Dan > > > >Aaah, you're thermally bonding the hot beads to the tube? Perhaps > >that'd work, might want to instead put a heater at the top of a small > >rod, bond the thermistors to the rod, use them to measure along the rod, > >maybe would work. > > > > Yeah, that's pretty much what I had in mind. Although I am not a > thermal engineer, it seems to me there will generally be a temperature > difference between the fluid, now matter what it is, and the air > above it - due to thermal inertia - so a heater may not be required. > I'd have to try this to see how it would work. It might work with > 3 thermistors this way - one at the bottom [in the fluid], one at the > top [out of the fluid], both of which act as references, and one in > between to do the actual weighted measurement - all bonded to the same > long tube. It'd depend on situation, around here on a cold day that wouldn't work well though - if the temp's 55F and the water pours in at 55F, you would need to rethink - in hotter climes, 55F vs. 80F, easy to see the difference. Worth a try some time when you have "spare time", that elusive thing we never have enough of <G> I wouldn't expect self-heating from the thermistors to work in this situation, bet you don't either; power resistors are cheap and plentiful though <G> (Bond a 5W or so resistor inside the top of the tube, use polyurethane power cable out the top, seal the ends with PVC cement goop after scratching the cable to give you good bonding and then using that purple MEK stuff to prep the cable? Seems to work pretty well in past, IME. Run say 1-2W into the resistor so it won't be over stressed.) > >Good news is that you can use one larger PIC and run quite a few > >thermistors simultaneously to get many data points simultaneously. > >Sigma Delta code's not too hard to find <G> > > > > For production stuff, I think most companies would want to go with > as few parts as they could get away with --> costs, plus reliability > issues. Sure - So long as it works <G> > - Dan -- I re-ship for small US & overseas businesses, world-wide. (For private individuals at cost; ask.)