detecting water and probe disolving ?

[Thunderchild]

I made a simple circuit consisting of 1 BJT to pass a small current through water and the base of a BJT and a resistor so that enough current flows in the collector to light a led.

well it worked but bubbles are comming from the negative "probe" and I'm sure that the water around the positive probe is going ever so slightly white.

Am I just putting electrons into the water or am i actually disolving my probes ?

Perhaps some sort of methos using alternating current is required ? so that the prodes do not irreversably disolve ?

 

[Hero999]

Look up electrolysis on Wikipedia.

If you don't want it to happen, you should use AC and couple the probes via capacitors.

The circuit can be made using a few logic gates.

 

[Thunderchild]

ok well that starts to explain a few things, we have these pumps at work with sensing circuitry inside them, but they kind of keep going when the water has run out so I started some tests at home to see if there is a simple way of doing it. Also because he is using "AC" and there are two pumps per unit and I think the circuits are over sensitive and one picks up the signal from the other so both run when one starts.

so if I decouple the sensors with caps and say I'm running with 24 V, that means that i will have +/- 12 V and need to rectify the signal again which can be done with one diode. so maybe used the same setup with a transisor but supply it with pulses instead ?

I'll knock out a schematic and post it

 

[Hero999]

I'd recommend using a CMOS logic IC.

As you're using 24VDC, you'll need to add a MOSFET and a regulator.

 

[Thunderchild]

yep that looks sensible, just nee to figure out why the guy needed 3 probes in his sensor !

 

[chemelec]

This circuit will work, without disolving the contacts.

**broken link removed**

 

[Hero999]

I've just realised, the circuit will be easier to make and better with Schmitt triggers.

Replace the first two gates with a single Schmitt trigger oscillator and the second gate with a Schmitt trigger.

With the circuit as it is, there's a risk of the output being biased in to the linear region which won't happen if a Schmitt trigger is used because it has hysteresis.

If you don't have a Schmitt trigger logic IC handy, you can use the hex inverter: a Schmitt trigger can be made with a couple of inverters and two resistors and added to the output

If you only have quad NAND or NOR gates, it's possible to make a Schmitt trigger with an additional gate and transistor; the latter can double as the load switch.

I'll post a schematic, if you don't understand.

 

[Thunderchild]

well this is currently hypothetical I am trying to understand something of how the units we have at work function. so what difference exactly do the schmitt trigger gates make ?

 

[Hero999]

If the resistance between the probes is too high to turn the output fully off but too low to keep the output on the output will sit somewhere inbetween 0V and VDD.

In other words, if the voltage across the 1nF capacitor is in between the logic levels of the gate the output will also be neither high nor low. In the case of the CD4000 series, this will happen when the gate voltage is exactly half the power supply voltage.

I Schmitt trigger has different turn on and turn off voltages so this will never happen. For example it might turn on when the input voltage is above 2/3 of the supply voltage but it won't turn off until it drops below 1/3 of the supply voltage.

Using a Schmitt trigger also means the circuit can be built with only two gates, not that this matters as there'll be some redundant gates. As I said before, if you don't have a Schmitt trigger IC, you can easily improvise.

 

[Thunderchild]

ah yess I see your point

 

[Reloadron]

Not to deter from electronics and sensing the conductivity of water to trigger pumps I would make another suggestion.

The problems with using plates in the water and conductivity is knowing the conductivity of the water, the plates corrode depending on the true chemical composition of the water and other things haunt you. Water with solids? Water with whatever?

About ten years ago I went to work one morning following a 4 day holiday weekend. We have a huge sump in a sub basement that holds about 4,000 liters of water and pretty much deionized water with some but few impurities. We were greeted by about 4 feet (over a meter) of water with al sorts of things floating around. Obviously the system failed.

I devised a new system and after looking at sensors I kept it simple. Real simple. I used float switches. We have two pumps down there. I put the pumps on a timer so each pump had the duty for 30 min. When the sump water reached a level a pump would come on based on which pump had the duty. However a pump could fail. When the water level increased another float sensor would trip. That sensor would command both pumps to run. In the event both pumps failed as the water rose a third float switch would kick in sounding an audible loud alarm. We run 24/7 so there is always someone there and in case of a holiday weekend which started this mess the system phones home.

I had originally kicked around fancy electronic sensors but the bottom line for my application was simple float switches, using 3 as I mentioned. The only actual electronics is the timer for pump duty, the rest is plain and simple relays and pump contactors. Simple is sometimes good and that system has been running 10 years, We replaced a few pumps but it works just fine.

Just My Take
Ron

 

[KMoffett]

yep that looks sensible, just nee to figure out why the guy needed 3 probes in his sensor !

My guess is that one probe was common and the other two were high and low sense. This is so the pump doesn't cycle on and off rapidly due to wave action as it starts pumping.

Ken

 

[Thunderchild]

Not to deter from electronics and sensing the conductivity of water to trigger pumps I would make another suggestion.

The problems with using plates in the water and conductivity is knowing the conductivity of the water, the plates corrode depending on the true chemical composition of the water and other things haunt you. Water with solids? Water with whatever?

About ten years ago I went to work one morning following a 4 day holiday weekend. We have a huge sump in a sub basement that holds about 4,000 liters of water and pretty much deionized water with some but few impurities. We were greeted by about 4 feet (over a meter) of water with al sorts of things floating around. Obviously the system failed.

I devised a new system and after looking at sensors I kept it simple. Real simple. I used float switches. We have two pumps down there. I put the pumps on a timer so each pump had the duty for 30 min. When the sump water reached a level a pump would come on based on which pump had the duty. However a pump could fail. When the water level increased another float sensor would trip. That sensor would command both pumps to run. In the event both pumps failed as the water rose a third float switch would kick in sounding an audible loud alarm. We run 24/7 so there is always someone there and in case of a holiday weekend which started this mess the system phones home.

I had originally kicked around fancy electronic sensors but the bottom line for my application was simple float switches, using 3 as I mentioned. The only actual electronics is the timer for pump duty, the rest is plain and simple relays and pump contactors. Simple is sometimes good and that system has been running 10 years, We replaced a few pumps but it works just fine.

Just My Take
Ron

your right of course, I am not actually planning on building the circuit, I am more intereted in second guessing how a product we use at work works, these are small pumps that are in air conditioning evaporator units and are used to pump out condensation, so it should be pure water but I have seen sand it on units and have no idea how it got there unless picked up from the air.

I don't think there is a high and low sense, I think the designer has used 1 probe as common, one to inject the signal into the water and the other to pick it up again, small wonder that when one pump goes into action so does the other as it is so sensitive that it just picks up the signal from the other one

 

[Hero999]

What's the dielectric constant of pure water?

If AC is used and the dielectric constant is high enough, pure water will increase the capacitance between the probes which might reduce the impedance enough for the circuit to trigger.