Replacing Existing Security System with Elk M1?

[TheDeepRough]

First of all, this board is pretty amazing, love the amount of information I have found here already!

We recently purchased a new home, that was outfit with a pretty dated 'Dictograph' security system. I have brought along, from our previous home, some Insteon capable devices (dimmers/etc). So, I would like to start down the path of some simple HA and at the same time, reviving the existing security system.

I have been looking at the Elk M1 a fair amount, and it looks like a system I can grow into, integrate existing Insteon devices as well as add additional ones down the road.

Here is my question though. The 'Dictograph' security system in place today has the following:

- Approximately 20 windows fit with magnetic contact sensors, all wires routed to the existing security planel
- 6 'plunger' sensors mounted in doors, wired to the existing security panel
- An internal siren, an exterior alarm bell
- A number of small indicator lamps around the house (presumably when the system was armed, they would be lit?)
- A couple of physical key switch panels, two indoors, one outside, that were used to arm/disarm the system

I am not so interested in the physical keys, or even the indicator lamps. But, how can I go about determining if I can re-use these magnetic contact sensors with the Elk M1? Are all sensors pretty much the same? What about the plunger sensors? Assuming it does work, does anyone have any advice on how to quickly identify which leads at the current panel go to which sensors? Or, is it just a lot of walking back and forth?


Has anyone had experience doing something similar in the past?

Thanks!

 

[apostolakisl]

Assuming they work, you can use all of the door/window sensors. Plunger contacts tend to have issues, but if yours are as old as they sound like they are and still work, they must have been pretty good.

The only issue you may have is if they used an eol (end of line resistor). Not that this will prevent you from using Elk, but rather it will require you to make some adjustments that may be quite simple, or maybe not. It just depends on what is there. It should be quite solvable regardless.

The wires going to the keyed arm/disarm may work for touchpads if there are 4 conductors.

The sirens will probably work. Just check to be certain the current draw is OK. When you say "bell" do you mean a real old fashioned electromagnetic hammer whacking bell? If so, I am quite certain that a separate source of power will be needed for it, but it still can be controlled by the Elk.

I highly suggest you get an ISY and the Elk module for it. ISY adds a ton of functionality to both Insteon and Elk. . . not to mention the seamless integration between the two systems that it provides. Plus, ISY is just about to release z-wave support so you can bring z-wave into the fold with more or less seamless integration.

 

[TheDeepRough]

Awesome, thanks for the feedback Lou

And yes, I mean an old school hammer bell. When I first got here, was messing around with the components in place, and managed to get that thing to go off. Pretty amusing to hear, if it ever went off in a real 'security' situation, I am sure my neighbors would probably think I was home schooling my kids and wanted to recreate an old recess bell

 

[jpmargis]

To determine compatibility with the ELK M1G, disconnect a pair of wires and measure the resistance with the zone secured. The resistance should be either very low, approximately 2.2K ohms, or open. Violate the zone and check for a change. If you measure less than 2.2K ohms, it is easy to add a resistor at the panel to get the ELK M1G working. If it is significantly greater than 2.2K ohms, you may need to surgically remove or replace the existing EOL resistor. This is a lot of work! Post your results for further assistance.

As far as wire tracing goes, it is probably best if you use the trial and error method. Get a friend to help you and it's not that bad. Keep in mind that you likely have several window contacts connected in series. Hopefully, they are logically grouped!

 

[TheDeepRough]

Thanks JP, I'll give it a try and let you know how it goes.

Lou, what kind of issues do plunger contacts typically have?

 

[jpmargis]

The plungers seem to collect dirt and eventually jam. They also have exposed springs which corrode and break.

 

[jpmargis]

I'm not a Dictograph expert, but it appears to be a current loop type design, so the sensors should work with an ELK M1G. The key switches and lights look very "retro". B) If you have the keys and want to keep them, we can probably get them to work with the ELK M1G as well.

 

[tmbrown97]

For the most part, you should be able to re-use the contacts and most of the wiring; I'd plan on putting in new speakers in the existing locations.

If you like jpmargis' idea to keep the keyswitches, the elk actually has zone types just for that Even the indicator lights might be reusable; back then most wiring was on/off - not a digital signal, so chances are they just run back to the panel and can be powered to indicate whichever color you want.

The wiring to the keys and keypads is important - if you have 6 wires you can do the normal databus; if you have 4 you can get a DBHR to help. You might get lucky and be able to replace the exterior one with a prox reader and the interior key switches with keypads.

 

[Del91574]

Here's my take, since I've had to work on these old panels as well as convert them. One of the companies within my area was Dictograph's largest dealer in the US.

What you most likely have is badge shaped keyswitches or pushbuttons, with a red and white lens on them. You should have at least 5 conductors behind them, sometimes more. Typically they would have a tamper circuit in series, however many times this was omitted. I doubt very highly that there is any EOLR supervision on the panel based on the information so far. Even fire alarm didn't have EOLR supervision, it was wired form A, with the circuit going back to the panel.

I'd look carefully to see if the panel is a 6V or 12V unit, that would determine what or how much of any peripherals could be recycled (like a keyswitch, bell, siren, motion, etc.) or if even desired.

As far as contacts, I agree that the plungers should be replaced. As far as magnetic contacts, I would investigate what "flavor' they may be. If you have a piece of metal installed on the window sash or if you bring a piece of ferrous metal near them and the contact actuates, replace them, they're old Ademco switches from the same vintage, the first true recessed units. If not, careful inspection as to if they are sealed reed switches or not should be done, as your system dates from the time that sealed reeds were just starting to come out.

As far as wiring and if it's viable, yes, it should be metered, however since you most likely have a couple of continous loops strung around the perimeter, it's honestly faster and easier (although at the cost of cable and labor) 9/10 times to cut and run new feeds to general areas and zone the system more "modern" rather than a single entry and perimeter zone.

 

[TheDeepRough]

I took some pictures of the various components around the house (http://cocoontech.com/forums/gallery/album/90-dictograph-system/). Seems like, as some suspected, the window contacts are all in series, terminating at one block in the security panel.

TO answer DEL - it seems to be 6V


[sharedmedia=gallery:images:513]

[sharedmedia=gallery:images:512]

 

[apostolakisl]

I took some pictures of the various components around the house (http://cocoontech.co...tograph-system/). Seems like, as some suspected, the window contacts are all in series, terminating at one block in the security panel.

TO answer DEL - it seems to be 6V

Voltage of 6 or 12v is bears no relevance as far as contacts are concerned.

If you want to keep the old bell then you will need an Elk relay to transfer power from the bell's driver to the bell since Elk can not drive something like that directly. This is no problem. Or you can just buy a new siren that is designed to work directly with Elk. If you choose to keep the old bell, and you have a power outage, the bell won't sound unless you have a separate backup power supply for it.

Your internal speaker may work as is, you will need to check the specs on it.

After seeing the photos, for as little as it costs to replace your magnetic contacts with new ones, I would just do it. Those guys are all yellowed and covered with sloppy paint. New ones are smaller and will look better. The wiring is the key. . . that is 99% of the work and your current wires will almost certainly suffice.

It sounds like you have banks of detectors wired in series. That is pretty standard. Very rarely does someone care to differentiate one window from the one right next to it and that is all you lose when several adjacent windows are wired together in one loop. I assume your doors are each on their own wire. This is important since you will be treating them differently with entry/exit delays.

If you do not have eol resistors, then you can wire directly up to the Elk and set your zones as not having them. You very well could add eol resistors. Just one per loop at the most distant (electrically speaking) point in the loop (hence the name "end of line"). If you unscrew the contactor and the wire pulls out and you have room to slip a resistor back in the hole with the wire then they can be added. EOL resistors serve to prevent an accidental or nefarious on-purpose shorting of the wires proximal to the resistor. When you have a bank of windows wired in series, as I mentioned, you only add one eol resistor to the entire bank. Ideally you do this at the contactor that is furthest out.

I don't know why you would want to save those old keyed alarm control stations. Unless you are into the retro thing, they are very limited. Since you sound like you have aspirations for a more automated system, I would suggest replacing them with keypads and getting rid of the outside one all together. 4 conductors is the minimum to put Elk keypads there and as mentioned 6 wires allows you to do it easier since the data loop can be passed onto the next keypad without needing an adapter.

 

[Del91574]

While I'm unable to view the pictures, I'll point out some points that might have been missed or IMHO emphasized when it's not an important issue to think about at this stage of planning. The goal is to cut over the new system and be able to take advantage of 20-30 years of improvements and refinements in methods of connecting a system together.

Since you believe the panel is 6V it's counterproductive or counterintuitive to keep resuscitating 6V devices on what is going to be a 12V system. You could change bulbs in the plates and salvage them, however unless you're attached to them, it's not worth keeping them living on the new system going forward. Just a disservice to yourself in the long run.

As far as the contacts go, unless they're epoxy potted, they're garbage and plan on replacing them. If you're worried about paint lines, etc. modern contacts do exist for old footprints 99% of the time.

For continous loop wiring, it's futile and a disservice again to bring that over to a new panel. Would you rather know X room, door, etc. or "perimeter" or "entry doors" caused an alarm, false or is faulted. Plan on cutting wiring back wherever possible and refeed from your panel location. I wouldn't honestly worry about EOLR's unless you physically know which is the last device on a series loop, otherwise their value is significantly diminished from really supervising the loop. They won't supervise if a contact fails, only wiring, when installed at the last detector in a loop, on the "common" side of the loop pair. Again, if you can't do it right, the value of putting them in at all starts coming into question. EOLR's are nice to have, but unless you can install them appropriately, it's just complicating and blurring the issue and you're going to gain little to none on the install and it's inherent security.

You most likely have 2 "zones" wired now, perimeter and entry/exit. At best, you might have an "interior' or "panic" zone tied in separately, but that's it, and how it was done was a big loop around a house, or as wire runs allowed. If a door was next to a window, they'd be tied together as long as a delay was not needed. There might even be timer relays or shunt switches installed in a closet. I'm very skeptical the doors were home run. If you were very lucky, you'll have a junction box or two with wiring tied through in it, but 99% of the time, the installers would not pull more cabling to the panel than what was needed to make it work. Mentality of "just makes it more difficult or cluttered".

You could salvage some keyswitches or indicator lights, that's your perogative, however installing KPAS' (if size and cost is an issue) or full blown KP's can be done, provided enough wiring is there and home-ran back to the panel, or a minimum of 2 taps per run (then adding a DBHR to the mix).

If you have junction points, it's definately worthwhile to spend time with a DMM and some help to trace down the logical wiring and then develop a plan from there.

 

[apostolakisl]

I think it is safe to say that no one here knows how your house is currently wired.

I also thing it should be your first goal to figure that out.

The simplest thing to do is simply put an ohm meter on each zone wire at the panel and run around opening/closing things to see what breaks the circuit. You'll also see if there are any resistors in the circuit based on the reading.

You can also determine the order in which the contacts are daisy chained together and thus determine where the eol should go if you choose to put one in. A rf tracer may work for that or it may not work that well. It depends on how good your tracer is and the wire characteristics. But the basic jist is you open up every door and window in your house that is alarmed. Put the rf signal source on the hot lead at the main panel. Then find the "hot" contact in the house with the rf detector. Close that door or window and then find the next "hot" one. And so forth. Where you may run into trouble is that the rf signal can leak over onto the negative lead and make the last device in the loop look "hot" as well. Or even an entirely different wire if they run parallel for a long enough distance. A sure fire way to trace the wires through your house is to set up a 12v source on the hot lead at the panel and then run around your house with a long wire going back to the neg terminal on your 12v source. With everything open, touch your multimeter hot lead to the wires at each contact and the negative lead goes back to your 12v source neg. When you find 12v, you know you have the first one. Then close that door/window and find the next one. And so on. It's the same concept as the rf tracer but induction won't confuse the issue.

 

[Del91574]

An inductive toner will bleed through an entire cable pair unless you connect one end to ground and likewise on the cable that is being toned. They're designed for cable pairs. For inductance in a cable mass, that's up to judgement as to which cable it might be and then being able to short a pair to verify the pair. If you split the toner between a conductor and EG you would be able to meter between a suspected "hot" conductor and EG to see voltage (the audio peaks) using a DMM relatively easily, and without dragging wire around the house, downside is you need a toner.

 

[apostolakisl]

Alternative is to use house ground. Connect the hot lead at the panel to house ground and then check your ohm reading between house ground and the wires at each contact. This assumes a single cohesive house ground. It is possible in very old homes to not have that, but you probably do.