Is this circuit going to be a problem?

How does that work? ...Jim Thompson

Should work, just don't let the main supply voltage get down to low near the gate voltage. Btw, when drawing out PTC/NTC or what ever the senser type is, its customary to use proper symbol so not to confuse others. Most know by the drawing that it's simply a variable R device, many love to know more about it.

Jamie

I won't say that WILL work, because there's not enough detail, but it certainly COULD work. Particularly if the design intent is to have the motor either on or off, but nothing in between.

I'd put a catch diode across the motor to absorb the inductive kick of the motor when the FET turns off.

On 2/6/2016 2:36 PM, M Philbrook wrote: > Should work, just don't let the main supply voltage get down > too low near the gate voltage.

What would happen if the motor supply was less that Vg? I.e., if its ripple was such that at times it was less.

That is the best sensor symbol, I think. The sensor is 2 electrodes immersed in water, to a varying depth. Resistance varies inversely as depth.

"will":"could" - fair enough. I was really asking "Will this circuit NOT work?" I wanted to know if there was something inherent that would prevent it.

In thinking about it more, I realize that the "large" motor current would cause a lot more ripple on the power cap and the regulator could drop out. Rather than worrying about the cap size needed to prevent that, I have isolated it with a diode. The motor works "fine" with unfiltered full wave.

But that does mean that the motor voltage will drop below Vg. I'm not clear what happens then. I don't even know what "Vds" means when it's saturated. Isn't saturated-mode Vds = Ids * Rds(on) ?

Since this is s.e.BASICS, I know that I can ask such basic stuff.

Revised, with more detail:

At the risk of provoking answers to questions that I'm not asking, here is the big picture of what I'm doing: I have a CPAP machine with a dehumidifier and that dehumidifier has a water reservoir of a cup (240ml) or so. Which means that it has to be filled EVERY night. So I intend to sense when the water is low and turn on a pump to refill it. The sensor is 2 immersed electrodes and the pump is a auto windshield-washer pump. It will probably run for a few seconds at a time, a couple of times a night.

Thanks, Bob

You need to worry about the gate voltage of the FET falling below threshold, but that's more or less independent of the motor voltage.

As long as your 12V supply doesn't drop below 10V or so the circuit should work with most power FETs out there.

Perhaps change the reg to a low dropout to be sure ??

That's a risk! Will the electrodes cover with gas bubbles and alter your resistance? Mikek

Thanks. I just got around to trying it and it works fine.

I had read that people use AC for electrode sensing, to prevent electrolysis, but I was hopeful that I had enough headroom in my ultra-low precision sensing that I could use DC.

I have done some testing and whatever bubbles there might be forming do not effect my sensing. Probably because my electrodes are about 1 sq in each and the sense current is less than a ma.

Bob

maybe you can make some browns gas!

Jamie