Possible to calculate speed with ELK M1 and 2 Sensors?

[Big517]

Hello All,
 
I have a 2 Optex Laser break sensors spaced roughly 10 feet apart from each other and shooting across my driveway to detect someone or something approaching the house.
 
Also using Vera Alerts (independent of my Vera System I want to note) to have my ELK Talk through a phone that is plugged into the paging system of my home's audio distribution.
 
So right now it will announce "There is a (Vehicle/Person) (Entering/Leaving) the driveway" throughout the home (4 different conditions).
 
I've attached my rules hoping someone can benefit from them at this stage, and also offer advice on improvements and understand how to integrate my request if possible.
RULES ATTACHED BELOW
 
This works extremely well, and without error for a year now.
 
I want to take it one step further and determine the speed at which a vehicle would be approaching.
 
This may be beyond the capabilities of the ELK M1 Gold, but there are a lot of things in there that i'm still discovering so I figured I would post here.
 
Ideally if we could output the speed as a number in MPH, and set a counter to that number.  If we can get a number output I could probably take care of the rest, and I would be able to have my system alert me at what speed the vehicle is traveling.  (and ultimately play a warning to slow down or flash lights etc...)  

Is there a way to do this calculation on the ELK?  

 

 

[Del91574]

I'm a little tired to go through the rules, but I'm guessing you have PE beams and not lasers...the Optex lasers are about $8K a piece.
 
That said, we've done this many times using cartels or their contemporaries, with some relay logic, but nothing too painful. Usually we save the rules for the heavy hitting and announcing, but it can be done either way. In the case here, the rules allow a more fine toothed adjustment for timing relating to the location of the detectors.
 
Speed is going to be a hard item to judge with the equipment installed in your case, unless you add some more logic.

 

[gatchel]

Since Speed=Distance / Time and you only have 10 feet between the beams you are going to need a pretty good timing resolution to get this to work and be accurate.
 
I ran a calculator for 3 yards at 20 MPH and that is about 3/10 of a second. 30 MPH is 2/10... I think you are going to be hard pressed to get the Elk hardware alone to give you an accurate MPH estimate. Just a guess.
 
If anything you would have to spread out the distance between the beams 2x or 3x if it is even doable.

 

[BraveSirRobbin]

You would need to know the scan time of the PLC in the Elk, then the acquisition time of the digital input (signal to registering an address change in the PLC) to see if you will get the resolution Gatchel mentions above.

 

[Big517]

It's impossible to get above 15mph. These are right after a 90deg turn so I don't think anything will be above that.

@DELInstallations, What is this "Cartels or their contemporaries" you reference?

So the concept is there but does the elk have abilities to calculate fine numbers, or am I limited to creating a series of timers that indicate mph thresholds and base it off which timer was on when the 2nd beam tripped?

 

[Del91574]

I was referring to other components instead of PE beams...driveway loops and related technology, so when it's installed it's invisible.
 
I'd have to sit down and draw it out then factor in the logic, however you're going to be limited by the Elk's  zone response time, which IMHO, should not be "fast" but around the 400 mS mark for the majority of systems out there on the input, then the time of the relay opening/closing at the Optex, which is usually a fixed value.
 
Irregardless of what is done at the panel, your fault times are still going to be limited by the Optex units....they have a relay which faults and then restores, so I don't believe there's much adjustment or ability to do much in that regards, so now you're trying to work within the confines of the zone response time and the PE beam relays, which is going to be tough only having 2 detectors and located where it sounds like they are.
 
The ugly and down and dirty way to do this, with a lot of experimentation would be to install some capacitors that hold the loop voltage present in relation to longer than the actual physical state of the zone, assuming the Optex faults for long enough and the capacitor bleeds faster than the Optex relay, then testing to make sure it doesn't affect the inbound/outbound timing or fault timing....not fun.

 

[Big517]

Ahh, the sweet sound of a deflating balloon...
 
I appreciate the details and you are correct, that's pretty ugly.   Hopefully someone will find use in my rules at least to determine if it's a vehicle or a human approaching and respectively leaving.  
 
If anyone has any other suggestions or inexpensive hardware solution (plug and play preferably) i'm open to ideas.

 

[Del91574]

It could be done other ways, but I'm guessing the location of the PE beams is going to be the issue. A couple of timer relays and relay logic, then the M1...well it's doable, but the timing involved, you might end up having the car almost at the house before it triggers.
 
I'm just picturing a not fun project in my head that would requre a 3rd sensor input to provide another "AND" to the boolean, so the panel could then determine the timing (which the 3rd "AND" sensor should drop out prior to the directional portion of the boolean)