How to monitor a sensor that cycles between open and closed during alarm

[signal15]

If I had a Normally Open sensor that cycled roughly every half second between open and closed during alarm, how would I monitor it on the ELK M1G?

I'm guessing the answer is that I can't.

 

[nov0798]

I think you might be able to do it with a counter in the M1G

 

[WayneW]

If I had a Normally Open sensor that cycled roughly every half second between open and closed during alarm, how would I monitor it on the ELK M1G?

What do you mean "monitor"? You could use a pulse stretcher (ie one shot timer) to make the circuit to be constantly open (or closed) while the real sensor fluctuates.

 

[signal15]

Ok, so here's the exact scenario. I want to monitor the low salt LED on my water softener. So I picked up an optoisolator for $3. As a test, I wired it in parallel with the low salt LED using test leads, and the other two pins go closed when the LED lights up. Works great in that respect.

The problem is, I didn't realize the low salt LED blinks when it's low on salt, instead of just staying on. It's 1/2 second on, 1/2 second off. Can I use a capacitor to smooth out the signal? Or is that not going to work?

 

[Sandpiper]

Ok, so here's the exact scenario. I want to monitor the low salt LED on my water softener. So I picked up an optoisolator for $3. As a test, I wired it in parallel with the low salt LED using test leads, and the other two pins go closed when the LED lights up. Works great in that respect.

The problem is, I didn't realize the low salt LED blinks when it's low on salt, instead of just staying on. It's 1/2 second on, 1/2 second off. Can I use a capacitor to smooth out the signal? Or is that not going to work?

Nice job on the electronics.....the rest can be easily achieved in the M1.

Just use an output with a timer. Write a rule like this:

Whenever ZONE X becomes not-secure
then turn OUTPUT Y on for 2 seconds

When you write the rule, mnake sure you check the box for "reset timer if already running".

Now use the OUTPUT Y signal for whatever you were going to do. This output will turn on wherenever the LED comes on and not go off until the LED stops blinking.

 

[signal15]

Nice! Now, the next question is, how do I turn off logging for a particular input. I don't want to fill my logs full of crap when the LED is blinking.

 

[WayneW]

You can use an Elk 960 or similar to convert the pulsing signal into a nearly steady signal. It looks like the Elk 960 may reset every 60 minutes, I cannot tell for sure.

http://elkproducts.com/products/elk-960.htm

 

[signal15]

You can use an Elk 960 or similar to convert the pulsing signal into a nearly steady signal. It looks like the Elk 960 may reset every 60 minutes, I cannot tell for sure.

http://elkproducts.com/products/elk-960.htm

1-60 seconds, or 1-60 minutes, set by a jumper. Looks like you need to apply at least 4.5VDC for it to trigger. I could pull power from my panel to power it, and wire the positive leg through the optoisolator output side to the trigger terminal. But now I'm spending about $30 on this instead of $3.

Can't I use a capacitor and resistor in series to smooth out the "blinking" to a nearly constant signal?

 

[Sandpiper]

Nice! Now, the next question is, how do I turn off logging for a particular input. I don't want to fill my logs full of crap when the LED is blinking.

My M1 does not log zone status changes. Does yours?
My M1 logs alarms, arming, and disarming, and a few other miscellaneous things, but not zones unless they cause an alarm. I assume your LED zone will be a non-alarm zone anyway, right?

 

[signal15]

Nice! Now, the next question is, how do I turn off logging for a particular input. I don't want to fill my logs full of crap when the LED is blinking.

My M1 does not log zone status changes. Does yours?
My M1 logs alarms, arming, and disarming, and a few other miscellaneous things, but not zones unless they cause an alarm. I assume your LED zone will be a non-alarm zone anyway, right?

I have a motion sensor set up as type Aux1, and it shows up in the logs.

 

[rbroders]

I had a gate opener at my old house that had a 30VDC light output that flashed whenever the gate was open. I attached it to a small relay with a capacitory across the coil. The capacitor held the relay open during the brief period the light was off, but shut down shortly afterwards. Worked like a charm.

I'm sure you could do the same thing. Maybe use the output from the first opto-isolator to charge a capacitor and drive a second opto-isolator which then drives the ELK input...

--Bob

 

[WayneW]

I have a motion sensor set up as type Aux1, and it shows up in the logs.

In the ELK logs? That would be quite unusual. Must be a feature of Aux1. Type 16 doesn't get logged.

My HomeSeer logs are full of motion sensor and other activity.

 

[pipeman]

I'm very rust on this, but I would try;
Ground one side of the optoisolator, drive a 220 ohm resistor in series with a 10 microfarad cap connected back to ground.
Use the common point of the R/C as your zone input.
Then, with the 2200 ohm pull-up resistor in the ELK, you would have a half cycle pulling towards 12 volts with a time constant of 22 seconds, then a half cycle pulling towards common at a time constant of 2.2 seconds.
That should stabilize somewhere below the threshold.

 

[signal15]

I'm very rust on this, but I would try;
Ground one side of the optoisolator, drive a 220 ohm resistor in series with a 10 microfarad cap connected back to ground.
Use the common point of the R/C as your zone input.
Then, with the 2200 ohm pull-up resistor in the ELK, you would have a half cycle pulling towards 12 volts with a time constant of 22 seconds, then a half cycle pulling towards common at a time constant of 2.2 seconds.
That should stabilize somewhere below the threshold.

I need to go buy myself a breadboard or something and start testing this stuff out. Using test leads in this scenario is not exactly optimal.

 

[signal15]

I nabbed myself a breadboard to set up some test circuits. I had to put a 1k ohm resistor inline with the anode on the optoisolator, or all of the current flowed through that and the LED on the water softener wouldn't light. I then put a 470uF cap in parallel with the anode/cathode on the opto and it keeps the opto in a closed state *almost* until the LED blinks back on. I think I need a larger cap, but that was the largest one I had.

I should have it working after spending a bit more time on it with a couple of beers.