Yeah, there are all kinds of missing pulse detector algorithms that could be
used.
If you did it on the current input, it could even be done with split core
current transformers.
Simple hardware. Painless install. Pain is in the software.
From: Lewis Bergman
Sent: Tuesday, November 28, 2017 8:13 AM
To: af@afmug.com
Subject: Re: [AFMUG] Teach me about tower lights
I would think consumed current would work as long as the time resolution is
fine enough, which I am sure yours is. Those LED flashes don't last long. I
have asked one of our suppliers we buy the most from to have an engineer give
me the specs so you can see what you are dealing with. In addition, many use
some very large caps on the older strobe models so you don't see any difference
on the incoming side. I haven't seen any of the LED models with big caps. I
wouldn't think you would need to monitor before the controller as your draw
should definitely be different than normal, at least at night, when the
controller isn't working correctly.
The biggest issue I would think would be how to get that current initially.
With all the different configs out there and different manufacturers the number
of combinations become enormous. Some designs actually have the controller
inside the top beacon (for some ungodly reason). There are a lot of different
LED designs with different numbers, quality, and lumen of LED's.
Maybe some kind of learning routine where you set it to learn for 1 minute, it
watches the pattern of draw, then sets that as normal for that time period,
then it does the same thing at midnight. I guess midnight would work except
nearer the poles. After that I would give your day/night timings a good bit of
slack as some are not the best at mounting their photocells in good locations.
Maybe 30 minutes to an hour each way of sun set/rise.
On Mon, Nov 27, 2017 at 10:21 PM Forrest Christian (List Account)
<li...@packetflux.com> wrote:
Can you think of any reason why simply measuring consumed current and making
sure it matches an expected pattern (I.E. toggling between all-on and "blink")
with the correct frequency and during the right time of day wouldn't work with
any of the systems you've put in?
Of course with a contact or two input as an additional input to catch the
failures the controller knows about?
On Mon, Nov 27, 2017 at 7:22 PM, Lewis Bergman <lewis.berg...@gmail.com>
wrote:
All the new controllers that I have seen have a dry contract for failures.
Of course the controller has to be working. I am afraid there are about a dozen
manufacturers each with several models. That's a pretty big matrix.
We just installed one last week with steady side lights because the
customer didn't want to apply for the new E class system without sidelights so
they have an all LED system with solid side lights and 3 flashing beacons.
It has a dry contract off the controller.
On Mon, Nov 27, 2017, 7:23 PM Chuck McCown <ch...@wbmfg.com> wrote:
I have thought about doing something similar many times over the years.
I always figured I would have it learn the healthy current of the fully
functional system with all lamps on and send an alarm if the current changed.
From: Forrest Christian (List Account)
Sent: Monday, November 27, 2017 6:19 PM
To: af
Subject: [AFMUG] Teach me about tower lights
One of the most popular questions/product requests I get is about tower
light monitoring.
It used to be that I sent people toward one of the off-the shelf tower
light monitors with a contact output and then monitored that with a base unit.
Unfortunately it seems that most of these only work with incandescent, and most
people are moving toward LED lights. So I'm getting a lot of 'I haven't found
one which will work with led lights'.
I know that many lighting controllers do some monitoring and have relay
outputs for failures. Recent feedback from customers seem to indicate that
this is a lot like the fox watching the henhouse, in that often the lighting
controller is the problem, especially with newer led lights which don't fail as
much.
As a result of the above, I'm looking at doing a tower light monitor.
But I need some help understanding the reality of lighting towers, especially
since things have changed over the years, and some older towers I'm sure are
grandfathered under an older lighting/painting scheme.
The first question is related to flashing lights on a tower: The FAA
circular seems to state that all lights on a tower should flash together. Are
all tower lights all designed so that all of the flashing lights on a tower
blink at the same time? If so, it makes monitoring them a lot easier, since
there's only two lighting levels to monitoring.
I'm also wondering about control systems, and how they are wired. My
google searches seem to indicate that most towers with blinking lights have a
control box at the bottom which controls the lights on the tower (and sometimes
provide a indicator of a certain type of failure). Is this common? What
other systems are out there?
I am also intrigued by the daytime lighting systems, and probably need to
think about monitoring these as well. I'm assuming these are strobes. Anyone
install one of these?
I'm also curious about what type of power consumption the newer LED
lights have, and if measuring AC current is really even practical on these.
Any other information anyone can point me toward would be appreciated.
--
Forrest Christian CEO, PacketFlux Technologies, Inc.
Tel: 406-449-3345 | Address: 3577 Countryside Road, Helena, MT 59602
forre...@imach.com | http://www.packetflux.com
--
Forrest Christian CEO, PacketFlux Technologies, Inc.
Tel: 406-449-3345 | Address: 3577 Countryside Road, Helena, MT 59602
forre...@imach.com | http://www.packetflux.com
