🚀 go-pugleaf

In sci.electronics.design Dale <ewgeneral@yahoo.com> wrote:
> Hi everyone out there.
> 
> Im looking at detecting failure of HPS street lamps where say you have
> up to 20 lamps on a particular circuit. Im only interested in knowing
> that one or more lamps are in failure state (i.e. from the moment they
> start to go through their on-off cycle)
> 
> (If anyone has got ideas on when a particular lamp has failed, via a
> single feeder circuit, of course I would be very interested, even
> maybe have a job for you!!! ;)

Finding a particular failed lamp is trivial.

PIC, and a powerline modemm.

Getting the cost down is the interesting bit.

Measuring the average power ripple in the 0.5Hz (or whatever) band may 
work for 20 lamps, how many lamps it will work for without false triggering
will depend on the normal amount of noise in the power draw at those
frequencies.
This would need little more than a couple of A/Ds, some software, and a
PIC, along with a current transformer to measure the current.

Then again, if you'r putting up a lot of lights, there may be better ways.
An little mains driven oscillator that squeels at a few Khz at a well 
defined frequency when the bulb is not lit would probably work well for
lots more lamps.


-- 
http://inquisitor.i.am/    |  mailto:inquisitor@i.am |             Ian Stirling.
---------------------------+-------------------------+--------------------------
He had been eight years upon a project for extracting sunbeams out of cucumbers,
which were to be put in vials hermetically sealed, and let out to warm the air
in raw inclement summers.  -- Jonathan Swift, "Gulliver's Travels" (1726)

> Getting the cost down is the interesting bit.

Use a photodiode to detect the light. Can't be to hard to do, they even come
in those 150 in 1 electronics project kits. Couple it to a small signal
transmitter that will send a code or serial number down the power line every
few minutes that it can't see the light. Could even be included in the
electronics to turn on future light fittings at night.

Also, it's not restricted to one lamp type, but will work for a wide variety
of lights. Maybe the traffic lights could be done....

Hope this helps,
Peter

Dale wrote:

> Hi everyone out there.
> 
> Im looking at detecting failure of HPS street lamps where say you have
> up to 20 lamps on a particular circuit. Im only interested in knowing
> that one or more lamps are in failure state (i.e. from the moment they
> start to go through their on-off cycle)
> 
> (If anyone has got ideas on when a particular lamp has failed, via a
> single feeder circuit, of course I would be very interested, even
> maybe have a job for you!!! ;)
> 
> Failing a method for above, even a pointer or two to any links that
> may characterise the failure mode i.e. current/voltage characteristics
> of such would be a start.
> 
> Regards,
> 
> Dale


What about RF interference caused by the faulty lamp and detectable via the 
a power cable pickup

In sci.electronics.design Bushy <please@reply.to.group> wrote:
>> Getting the cost down is the interesting bit.
> 
> Use a photodiode to detect the light. Can't be to hard to do, they even come
> in those 150 in 1 electronics project kits. Couple it to a small signal
> transmitter that will send a code or serial number down the power line every
> few minutes that it can't see the light. Could even be included in the
> electronics to turn on future light fittings at night.
> 
> Also, it's not restricted to one lamp type, but will work for a wide variety
> of lights. Maybe the traffic lights could be done....
> 
> Hope this helps,

The detection circuit is trivial, and inexpensive.
The power supply for it, the coupling and bypass arrangements, confirming
it doesn't do nasty EMC things and has a MTBF greatly exceeding that
of the lamp housing (say 20-50 years) and installation issues could all make 
it more expensive than a first look would make it seem.

Then you've got to look at if as you'r already putting one circuit board
into the light, connected to the mains, might it be a good idea to 
integrate an electronic ballast?

-- 
http://inquisitor.i.am/    |  mailto:inquisitor@i.am |             Ian Stirling.
---------------------------+-------------------------+--------------------------
"I meant, have you ploughed the ocean waves at all?" Colon gave him a cunning
look. 'Ah, you can't catch me with that one, sir' he said 'Everyone knows
horses sink'                                       --  Terry Pratchett - Jingo

In article <65050f09.0301282112.294b0685@posting.google.com>, Dale wrote:
>Hi everyone out there.
>
>Im looking at detecting failure of HPS street lamps where say you have
>up to 20 lamps on a particular circuit. Im only interested in knowing
>that one or more lamps are in failure state (i.e. from the moment they
>start to go through their on-off cycle)
>
>(If anyone has got ideas on when a particular lamp has failed, via a
>single feeder circuit, of course I would be very interested, even
>maybe have a job for you!!! ;)

  How about detecting voltage applied to the ballast and lack of current 
flowing through the lamp?
  If the lamp is not the pulse start kind (although most HPS lamps are) 
then detecting full line voltage (or ballast open circuit output voltage) 
across the socket may be easy and a sign that the lamp is not working.

  NOTE - allowances must be made somewhere for lamps requiring time to 
cool down after a brief power outage before they are able to restart.

 - Don Klipstein (don@misty.com)

Thanks to everyone for help on this.

Don, your suggestion is ok if we can get access to the internals of
the lamps, but we have to do this remotely from the 'feeder pillar'.
The idea is to retro fit networks which have up to 20 or so lamps on
the same 'feed'.

Cheers,

Dale

don@manx.misty.com (Don Klipstein) wrote in message news:<slrnb3rkjf.o7s.don@manx.misty.com>...
> In article <65050f09.0301282112.294b0685@posting.google.com>, Dale wrote:
> >Hi everyone out there.
> >
> >Im looking at detecting failure of HPS street lamps where say you have
> >up to 20 lamps on a particular circuit. Im only interested in knowing
> >that one or more lamps are in failure state (i.e. from the moment they
> >start to go through their on-off cycle)
> >
> >(If anyone has got ideas on when a particular lamp has failed, via a
> >single feeder circuit, of course I would be very interested, even
> >maybe have a job for you!!! ;)
> 
>   How about detecting voltage applied to the ballast and lack of current 
> flowing through the lamp?
>   If the lamp is not the pulse start kind (although most HPS lamps are) 
> then detecting full line voltage (or ballast open circuit output voltage) 
> across the socket may be easy and a sign that the lamp is not working.
> 
>   NOTE - allowances must be made somewhere for lamps requiring time to 
> cool down after a brief power outage before they are able to restart.
> 
>  - Don Klipstein (don@misty.com)

I read in sci.electronics.design that Dale <ewgeneral@yahoo.com> wrote
(in <65050f09.0301282112.294b0685@posting.google.com>) about 'High
Pressure Sodium Lamp Failure Characteristics', on Tue, 28 Jan 2003:
>Im looking at detecting failure of HPS street lamps where say you have
>up to 20 lamps on a particular circuit. Im only interested in knowing
>that one or more lamps are in failure state (i.e. from the moment they
>start to go through their on-off cycle)
>
>(If anyone has got ideas on when a particular lamp has failed, via a
>single feeder circuit, of course I would be very interested, even
>maybe have a job for you!!! ;)
>
>Failing a method for above, even a pointer or two to any links that
>may characterise the failure mode i.e. current/voltage characteristics
>of such would be a start.

When they fail, they draw almost no current, don't they? So obviously,
there is no difficulty in detecting that 'a' lamp has failed. But
detecting which one, without installing some electronics in each lamp,
is unlikely to be feasible. Time-domain reflectometry? But you would get
lots of 'clutter' from permanent discontinuities in the cables.
-- 
Regards, John Woodgate, OOO - Own Opinions Only. http://www.jmwa.demon.co.uk 
Interested in professional sound reinforcement and distribution? Then go to 
http://www.isce.org.uk
PLEASE do NOT copy news posts to me by E-MAIL!

ewgeneral@yahoo.com (Dale) wrote in message news:<65050f09.0301282112.294b0685@posting.google.com>...
> Hi everyone out there.
> 
> Im looking at detecting failure of HPS street lamps where say you have
> up to 20 lamps on a particular circuit. Im only interested in knowing
> that one or more lamps are in failure state (i.e. from the moment they
> start to go through their on-off cycle)
> 
> (If anyone has got ideas on when a particular lamp has failed, via a
> single feeder circuit, of course I would be very interested, even
> maybe have a job for you!!! ;)
> 
> Failing a method for above, even a pointer or two to any links that
> may characterise the failure mode i.e. current/voltage characteristics
> of such would be a start.
> 
> Regards,
> 
> Dale

Will probably get flamed for this, but whats wrong with someone
driving around and looking to see which ones dont work - or is that
not cost effective?

Andrew VK3BFA

ewgeneral@yahoo.com (Dale) wrote in message news:<65050f09.0301282112.294b0685@posting.google.com>...
> Hi everyone out there.
> 
> Im looking at detecting failure of HPS street lamps where say you have
> up to 20 lamps on a particular circuit. Im only interested in knowing
> that one or more lamps are in failure state (i.e. from the moment they
> start to go through their on-off cycle)
> 
> (If anyone has got ideas on when a particular lamp has failed, via a
> single feeder circuit, of course I would be very interested, even
> maybe have a job for you!!! ;)
> 
> Failing a method for above, even a pointer or two to any links that
> may characterise the failure mode i.e. current/voltage characteristics
> of such would be a start.
> 
> Regards,
> 
> Dale

Hi, just give the residents a phone number to ring when the lamp
fails, just costs the price of a stamp.

On 5 Feb 2003 05:18:25 -0800, chrishelpfl@aol.com (chris) wrote:

>ewgeneral@yahoo.com (Dale) wrote in message news:<65050f09.0301282112.294b0685@posting.google.com>...
>> Hi everyone out there.
>> 
>> 
>
>Hi, just give the residents a phone number to ring when the lamp
>fails, just costs the price of a stamp.


Yes, and as an added incentive, give the "failed-lamp-reporter" a
5-10% reduction in their power bill for the month.

Barry Lennox