Automatic Water Shut Off Chapter 4 ( Frankensteining a Control Unit)

RE: Automatic Water Shut Off Chapter 4 ( Frankensteining a Control Unit)

volly — Tue, 16 Aug 2016 16:08:21 GMT

[mention:f80b53cee57c44bc9d7c577d07d7c791:e9ed411860ed4f2ba0265705b8793d05]. We'll keep [mention:8fb233e8974445138ef0af2b63b49f94:e9ed411860ed4f2ba0265705b8793d05] excellent observations and points raised in mind....shall we?

RE: Automatic Water Shut Off Chapter 4 ( Frankensteining a Control Unit)

mcb1 — Tue, 12 Jul 2016 08:31:28 GMT

Nice recycling.

A question for you to contemplate.

Have you considered that the battery fuse could blow and there may be no warning.?

The "fail to safety" in me always tends to assume that the battery could go dead (they dry out) and the fuse might blow and it won't be noticed until after it's needed.

(Yes we've had exactly the same thing happen at work)

Cheers

Mark

RE: Automatic Water Shut Off Chapter 4 ( Frankensteining a Control Unit)

kulky64 — Mon, 11 Jul 2016 18:45:54 GMT

I'm little confused. Your schematic indicates that the negative input of the comparator is used as reference with one half of supply voltage on it (which is 6.8 V). In your second blog you showed that when your sensor is dry it has more than 50 Mohm resistance, so basicaly open circuit and resistor R13 should pull down the positive input of the comparator to low level. So in dry condition you have lower potential on comparator positive input and therefore comparator's output transistor should be open and pull-up resistor R14 should pull the gate of Q2 high opening the transistor. Your solenoid should be now activated, buzzer buzzing a LED lit, which is obviously not desired behavior. Or am I missing something?

RE: Automatic Water Shut Off Chapter 4 ( Frankensteining a Control Unit)

DAB — Mon, 11 Jul 2016 18:42:52 GMT

Nice build John.

Reusing older electronics makes good sense.

Most of the older gear were over engineered and thus more tolerant for repurposeing.

You did a good job with the heat shrink tubing.

DAB

RE: Automatic Water Shut Off Chapter 4 ( Frankensteining a Control Unit)

jc2048 — Mon, 11 Jul 2016 09:49:21 GMT

I think I'd want a fuse in series with the battery (most practical way would be an in-line fuse holder in the red lead that comes down from the pair of batteries). If something were to short on the board you could have the kind of currents flowing that could do a lot of damage and possibly even set light to the box.

RE: Automatic Water Shut Off Chapter 4 ( Frankensteining a Control Unit)

michaelkellett — Mon, 11 Jul 2016 08:01:23 GMT

You might like to consider adding some hystereisis around U2 - maybe 10M from output to +ve input - this will make the switching threshold of U2 change a little when its output changes state and reduce the chances of the solenoid cutting in and out if the signals are a bit noisy or the power supply dips a bit under load.

Because the humidity sensor is quite high impedance you may pick up a lot of noise and mains hum on it's leads - a capacitor to ground on U2 +ve input would help get rid of this but the hystereisis idea wouldn't work so well then - so you may need to choose. Screened or twisted cable for the sensors would help.

Ideally one side of your sensor would be connected to ground - noise pickup would be easier to deal with and a short wouldn't kill your power supply (but you won't want to change it now of course.)

I think if I'd been doing it I would have buffered the signal from the sensor with an additional op amp which could have also given me some good low pass filtering and allowed a nic elow impedance drive to the comparator so that it was easy to add hystereisis and filtering at the same time.

I'm hoping that you'll prove these additional features unnecessary when you test it.

Good luck.