Elk Keypad Wiring

[pipeman]

In my pre-wire, I ran a cat5 and an RG6 cable to my front door doorbell location.
I've been using the cat5 for a lighted doorbell.
My wife likes our present doorbell and changing that is unlikely.
Now, I have decided that location inside the door is the logical location for my Elk KP2.
Ordinarily, I would use the entire cat5, (with two unused) for connection to my hub.
I am thinking of using only 4 wires for the KP2 with termination and then would have 2wires for the low voltage lighted doorbell and 2 wires for local speakers.
Will that put interference on the data bus?


Thanks

 

[felixrosbergen]

I never understood why the ELK wiring diagram shows you have to take the Green and Organge and tie them together at the Keypad. Why can't you tie them together in the panel..isn't that the same thing but you only need 4 conductor out of the cat5?

Runnign speakers with the databus over the same cat5 sounds a bit hairy to me..but i guess just try..worst case you just have a KP2 and no speaker there..

 

[CORT]

Remember the KP2 has a local input for 1 zone. You could conceivably re-purpose the cat5 wire for the keypad and your speaker following the recommended connections. Then use the local input of the KP2 for the doorbell button. This would mean that the M1 becomes integrated with your doorbell system. You would need to put a relay in an alternate location to actuate the doorbell gong if you don't want to lose that sounder. It might be a little tricky figuring out the lighted part of the doorbell button, but that's possible too.

Finally, if you have access via an unfinised basement or crawlspace, it is not too difficult to drill up into the stud bay and pull more wires. I've done this in the past.

Good luck.

 

[mustangcoupe]

I never understood why the ELK wiring diagram shows you have to take the Green and Organge and tie them together at the Keypad. Why can't you tie them together in the panel..isn't that the same thing but you only need 4 conductor out of the cat5?

I think it is "easier" to understand the wiring this way, and if you are using a new install DBH (which uses RJ connectors) you dont have to mess around with what you are using. just crimp a connector and plug it in.

now if you are not using a hub, and just splicing wires I figure you could just use 4 wires and splice it at the control

 

[pipeman]

I understand the reason for looping the data bus wires and terminations (i think), and my home is CBS slab on grade with no attic access to this location.
Using the data bus hub normally, will still leave two wires (pins 4 & 5) unused.
As I understand, the Elk keypads have a chime only, no speaker output.
So I always figured I could send audio to keypad locations using the two unused, with Elk's 32 Ohm speaker, I've assumed the current would be low and not too much problem with length.
The doorbell circuit is 24v AC and I am concerned about crossover to the data lines.

Thanks

 

[AceCannon]

Doorbell: I hooked up my doorbell wires not realizing someone had already put a doorbell button on the wall and connected it to blue-bluewhite. Hooked it into an Elk zone and everything. Funny, it has a light in the button which is working. I have the Elk play 3 of the 800hz tones for the doorbell (for now, until CQC is deployed). I had planned to use an unlit doorbell button because I didn't want to mess with it, but it works anyway. I am not at home to check, but I think I have it as a NC zone and pushing the button opens the circuit. The current the M1G is sending through there must be lighting the button (maybe an LED?).

Keypad orange/green: this spec is to maintain the RS485 databus integrity. I posted essentially the same question around here before. If you are using the databus hub, you need to do it this way or risk problems on the bus. Here is a relevant thread. At any rate, you will still have an extra pair even doing the correct way. And apparently an illuminated doorbell is still do-able with only 2 wires, since my setup worked accidentally the first time around!

 

[pipeman]

Thanks, I'll go ahead with that

 

[AceCannon]

I double checked and my doorbell is set up as def 16=non-alarm, type 1=normally closed. Fast loop response is enabled.

I made a rule:

whenever doorbell becomes not secure
then announce miscellaneous 1

I still think it is weird that the light in the doorbell button is lit.

 

[tmbrown97]

Ace - if you figure out how you have that set up, I'd really like to know... I want the original doorbell out of the mix.

What I ended up doing to keep the light was use the Elk doorbell detector; then I broke the sound plates out of the physical doorbell so it silently triggers the plunger, but doesn't actually make a sound - then I punched through the hall wall into the coat-closet on the other side and flipped the old doorbell into the closet. That allowed me to keep the lighted button (very important to me and wife - not sure why though)... and still have the Elk play the actual doorbell. My solution works, but if there's a way to keep the light and lose the doorbell, I'll do it in a heartbeat.

Now - for the OP and Mav even... I believe the reason for the looping of the 2 pairs at the keypad has to do with the signal's bouncing and attenuation... basically it travels down one pair and back the other - and so-on through the chain. The last device has the resistor. If you drop that second pair, you'll likely kill the signal. That said, I know people have posted here before saying that they use the last remaining pair for their SP12's. I personally didn't - I ran 18/2 for them... but nobody has mentioned problems doing it that way. Using something like a doorbell detector, or figuring out what AceCannon did - you can go into the keypad's zone input and be perfectly safe with the single Cat5. To me, that sounds like your best bet.

 

[AceCannon]

Ace - if you figure out how you have that set up, I'd really like to know... I want the original doorbell out of the mix.

Now - for the OP and Mav even... I believe the reason for the looping of the 2 pairs at the keypad has to do with the signal's bouncing and attenuation... basically it travels down one pair and back the other - and so-on through the chain. The last device has the resistor. If you drop that second pair, you'll likely kill the signal. That said, I know people have posted here before saying that they use the last remaining pair for their SP12's. I personally didn't - I ran 18/2 for them... but nobody has mentioned problems doing it that way. Using something like a doorbell detector, or figuring out what AceCannon did - you can go into the keypad's zone input and be perfectly safe with the single Cat5. To me, that sounds like your best bet.

Even doing it correctly with the out-and-back loop, there is an extra pair in the cat5e. Blue and blue/white are not used in the Elk description. (1 pair for power, 2 pair for the Elk databus, 1 spare pair for doorbell or speaker or whatever.) If he used the spare pair for the doorbell, he STILL has the keypad's input empty.

 

[Lagerhead]

Even doing it correctly with the out-and-back loop, there is an extra pair in the cat5e. Blue and blue/white are not used in the Elk description. (1 pair for power, 2 pair for the Elk databus, 1 spare pair for doorbell or speaker or whatever.) If he used the spare pair for the doorbell, he STILL has the keypad's input empty.

In a pinch, can't you make the keypad last in the chain, give it the terminating resistor, and so only need one data pair?

 

[AceCannon]

Even doing it correctly with the out-and-back loop, there is an extra pair in the cat5e. Blue and blue/white are not used in the Elk description. (1 pair for power, 2 pair for the Elk databus, 1 spare pair for doorbell or speaker or whatever.) If he used the spare pair for the doorbell, he STILL has the keypad's input empty.

In a pinch, can't you make the keypad last in the chain, give it the terminating resistor, and so only need one data pair?

I am almost certain that you can. One would just have to remember that the next connector in the DBH is now not usable. If someone else needed to work on the system, it would not be obvious why the next jack in the DBH was not working.

Or worse, the next DBH jack would work but only erratically, and for non-obvious reasons.

At any rate, for the OP - connecting the keypad normally should still leave a spare 2 conductors for the doorbell, as well as a spare input at the keypad.

 

[pipeman]

Ace,
My doorbell button has an incandescent bulb and the switch (normally open) is internally wired in parallel.
I think this is the common way for typical wired doorbell systems.
If you have a standard button, Is the elk seeing a resistance (the bulb glowing), then a status change to short (when the button is pressed)?
I found an incandescent rule of thumb that says cold resistance is 1/15th of hot resistance.
My button light measures 26.5 ohms cold, that should translate to 397 ohms hot, if we gave it 16v.
Looking at the Elk book, it seems that resistance range would be too low to provide sufficient voltage, which would lower the resistance even more.
Is your button maybe an LED light?

Lagerhead,
Yeah, thats what I'm thinking, the Elk needs two bus terminations, so one would be my data bus hub terminator, the other would be the front door keypad, pull the on-board terminator and I should be OK.
That leaves me 4 wires of the Cat5E for doorbell and speaker pairs

 

[AceCannon]

Ace,
My doorbell button has an incandescent bulb and the switch (normally open) is internally wired in parallel.
I think this is the common way for typical wired doorbell systems.
If you have a standard button, Is the elk seeing a resistance (the bulb glowing), then a status change to short (when the button is pressed)?
I found an incandescent rule of thumb that says cold resistance is 1/15th of hot resistance.
My button light measures 26.5 ohms cold, that should translate to 397 ohms hot, if we gave it 16v.
Looking at the Elk book, it seems that resistance range would be too low to provide sufficient voltage, which would lower the resistance even more.
Is your button maybe an LED light?

As I mentioned in post #3 above, I suspect it is an LED. The zone is NOT setup as EOL supervised, it is a normally closed zone. So I dunno if pushing the button is shorting the conductors or opening them. The zone is set up to fire "voice misc 1" when it becomes not secure. So it is either going into trouble status or opening the connection completely.

 

[pipeman]

Thanks for the advice.

What I ended up doing is surgery on the doorbell button to replace the existing incandescent bulb with a radio shack blue led.
The LED is in parallel with the push button.
I then added a 470 ohm resistor in series and connected to the keypad input zone.
Zone voltages measure 4.8V normal and 2.5V pressed.
Used fast loop response to catch the press, EOL.
Cool blue button light is nice.
Wired the keypad normal for data bus hub, control is terminated and DBH plug termination.

Connected a SP12 speaker to the backside of the flush-mount box wired to the spare blue/blue-white pair.


Mavric, I read an earlier post where you were going to use the spare pair for speakers.
Did that work, or did you re-wire?