I believe the current through the circuit is so low that even through 3Megohms most of the 9 or 5 volts will get to the gate and turn on the darlington.

If you buy the parts to this circuit and put in together it will cost very little.
Try it.
The kit will be overpriced.

 

... most of the 9 or 5 volts will get to the gate and turn on the darlington....

Darlingtons are junction transistors and have "bases" not "gates". They operate on current drive, not voltage. Otherwise you are probable right.

Ken

 

But dont you need at least 1.4V to turn on the darlington pair (according to the datasheet), but even so, I dont think you'll get enough voltage to turn on the transistor (even in active mode...) with a 3M resistance of tap water that I measured. However, I am breaking down and just buying the velleman kit from a local store and I'll just post what happens. But does anyone have an answer to my darlington pair question?

 

But dont you need at least 1.4V to turn on the darlington pair (according to the datasheet), but even so, I dont think you'll get enough voltage to turn on the transistor (even in active mode...) with a 3M resistance of tap water that I measured. However, I am breaking down and just buying the velleman kit from a local store and I'll just post what happens. But does anyone have an answer to my darlington pair question?

Well darlington pair have tons of current gain so not much inital base current is required.

As far as the effectivness of these simple water detectors, it's all about the resistance of the water. Pure lab DI water can have a resistance of 5meg/cm and is as good an insulator as there is and would probably not be detectable with these simple DC probe detectors. But any water that accumilates on a surface is bound to pick up contaiments and be able to conduct some current. So experimentation is called for.

Lefty

 

I'd use a simple conductance technique (I've actually done it). It can be a simple transistor, or an op amp, or anything.

It is true that the water requires contanaments (ie, dirt), but most floors this isn't an issue. If it is sprinkle a little salt under the detector.

 

I think you should add a current limit resistor in the base circuit.
Maybe 1800 Ohms (5ma).

KMoffett, of course you are right. If 1.4v is correct...I was thinking voltage and since the driving requirement is so low I could visualize the situation better (in my mind) by thinking of voltage. Would 1.4v get to the base. If no current is used, yes.
If extremely little current is drawn, yes. I have not consulted data sheets or done any math. So what do I know. Lets not get into that

 

You're right, might as well add another transistor for more gain. Voltage isn't really an issue I think.

 

Why not using a simple and more reliable "Float Switch" ? see some example to this link:
http://www.floatswitches.net/Miniature_Vertical_Mount_Float_Switches.html

Al.

 

That assumes there is enough water to float a switch. Most floods homeowners have to deal with is a layer. If there is enough to activate a deeper switch I'm in trouble.

My usual problem is AC condensate myself.

 

Because of the nature of water...pure or contaminated (usual), any sensor that relies on the actual presence of water is subject to contamination or error or freezing or all of the above. The only reliable detection means for detecting water is an in-direct method; ie: weight. By measuring the weight of a vessel of water, you by-pass all the usual aforementioned problems. Because water has a lot of weight for a given volume, a weight detection apparatus is easily configured.

Cheers, DPW [ Everything has limitations...and I hate limitations.]