Yet another HAI vs Elk Threas

[fxs158]

Personally having installed my HAI system about 5 years ago never installed EOLR until this past year. Always had smoke detectors but never had them connected to the panel. I decided this year to add both smokes and CO detectors to the panel. Prior to doing this I decided also to put EOLR on all the zones and I probably should have done it 5 years ago...

I want to thank everyone for the feedback. I placed my order today and ended up going with HAI, I have not wired my house yet so maybe I will start a home automation thread!

Thank you all again!

 

[icellama21]

If you want your security system to be UL approved, then EOL resistors are not optional, and many insurance companies require a UL approved security system for coverage. UL approval is also required for most government and high-security installs. Its your house and you can do what you like, just know that if you DON'T use EOL resisters, and a zone wire should get pinched and short-out, you will never know it. Use them and your panel will tell you within seconds when it occurs. There is no argument not to use them, except for laziness.

I'll throw out a reason. Transient resistance. If you have frequent lightning or voltage spike issues, sometimes the voltage on your zones can change by a few volts. As Spanky pointed out for an ELK system, when using an EOL on a NC loop, going from secure to insecure only requires a swing of a few volts. On a NC loop with no EOL, it requires 7-8 volts or more. Significant difference.

From my testing:

EOL: 7V secure, at ~9V transitions to insecure
no EOL: 0V secure, at ~9V transitions to insecure

I believe the insecure/open loop thresholds are low to allow the system to operate on partially discharged battery voltages. If there are any alarm engineers out there they can comment, but it seems removing EOL and using NC loops does provide a bit of false-alarm immunity.

The cost of removing EOLs is possibly not detecting an alarm. My 2 cents, use EOLs for life-safety(smoke detectors), and use layered sensors for security so you don't rely on a single sensor. Given typical police response times and the ease of jamming cell-phone signals(the kids in my town do this when looking for prescription drugs) and cutting phone lines, I am not sure the difference in security is significant.

 

[fcwilt]

Do you take into account that there is a minimum loop response time? Did you test with DC levels or truly induce a transient?

 

[icellama21]

Do you take into account that there is a minimum loop response time? Did you test with DC levels or truly induce a transient?

I have not fully modeled this for a real world scenario. With a typically long loop, small gauge wire and typical resistance to ground in a security panel, it is very conceivable that a lightning transient would greatly exceed the loop response time. Nearby lightning strikes can induce tremendous voltage in any wire, and the built-in inductance and resistance will cause the EMF to take some time to dissipate.

Is there really a scenario where a nearby lightning strike will induce enough voltage to trigger a circuit with an EOL, but not one without, I don't know. I do know that I was having false alarms in electrical storms and removing EOLs and making all loops NC so they are always tied to ground helped significantly. It is also likely that reducing the resistance in the loop(removing the EOL) and tying it to ground(NC) is a bigger factor than the "trigger" voltages.

 

[icellama21]

Here is another forum:

http://www.cocoontech.com/forums/index.php?showtopic=9730
http://www.cocoontech.com/forums/index.php?showtopic=7775

It looks like Spanky and ELK products may have some more scientific experience with this. I'm interested if anyone else has noticed a difference with/without EOLs.

 

[cornutt]

With a typically long loop, small gauge wire and typical resistance to ground in a security panel, it is very conceivable that a lightning transient would greatly exceed the loop response time.

You think so? Don't forget that these are digital sampling systems. I don't think they sample all that fast. 5-10 Hz is plenty fast enough to catch all of the events that a security system cares about. In fact, sampling any faster is probably counter-productive because, as you say, you just wind up seeing a bunch of transients and line noise.

 

[Digger]

Not every situation is the same. I can say from an experience earlier this year where an AHJ rejected an install because the installer did not follow the mfg instructions. The installer had to go back and correct a few things that were specified in the installation instructions.

So if the installation instructions state to use an EOL Resistor and the AHJ cares to check it could be an issue (as would any other requirement in the instructions).

Personal opinion is that in a residential installation it is not that critical to have an EOL especially if you check your system somewhat regularly (UL states once a week but that is not practical for non life safety devices etc).