I would expect long delays to be built in, since high power gas lamps can be damaged by quick power on/off cycles.
I would expect long delays to be built in, since high power gas lamps can be damaged by quick power on/off cycles.
How did you reset them ?
Were they actually photo detectors, or just simple timers ? [ie does your photo detector comment, relate to the timers ? ]
Seems the ideal solution is a combination of both.
One of those applications that is 'so simple' no one bothers to do it properly...
-jg
Did you do this from above (ie emulate the sun ) ?
-jg
Maintenance people would go around to the various locations, and reset the clocks. The point (which I didn't make clear) was that this was a terrible design. Even if it weren't for the problem with loss of power, they would still have to go around and reset them as the nights got longer in winter, etc.
Almost all of the lighting control I've seen has been with photo detectors (the round thing with the twist-lock connector). I've only seen timers used on this one project, and it seemed like a pretty poor idea. I don't have any direct experience with street lighting, but had assumed that they were usually controlled by photo detectors of some sort.
I'm pretty sure that the lights here in Australia are controlled by photosensors - at least that's one input to the equation. On particularly dark afternoons (storm brewing etc) some lights definitely come on before they 'normally' do. And they do appear to be switched on in banks rather than individually.
Regards,
-- Mark McDougall, Engineer Virtual Logic Pty Ltd,
Now you've taken *all* the fun out if it!! ;)
Regards,
-- Mark McDougall, Engineer Virtual Logic Pty Ltd,
The photoresistor that goes from +V to the comparator has shorted to ground. It now pulls down during the day instead of up.
-- David M. Palmer dmpalmer@email.com (formerly @clark.net, @ematic.com)
It depends on how you go about it. Do you build a pole-climbing robot with a camera to read the model number, or do you chop down the pole so you can read it from the ground? Do you phone the manufacturer or hack into their complaints department database?
-- David M. Palmer dmpalmer@email.com (formerly @clark.net, @ematic.com)
I say build your pole-climbing robot to read the model number, then hack into the company's computer but download the schematics. Diagnose the likely fault and design a retro-fit solution. Upload the solution back into the computer. Then hack the complaints database, flagging that particular lamp as being scheduled for the retrofit.
Then camp out next to the streetlight with aforementioned lemonade.
If the retrofit works, then you've correctly identified the problem!
Update this thread!
Regards,
-- Mark McDougall, Engineer Virtual Logic Pty Ltd,
Okay, let's see a schematic where a single point failure will cause a phase reversal..
Best regards, Spehro Pefhany
-- "it's the network..." "The Journey is the reward" speff@interlog.com Info for manufacturers: http://www.trexon.com
View in monospaced font. Sorry I don't have a good ASCII schematic editor
V2 V1 +---R2----+-----R3----+ | | | V+ -+--R1----+ PR +---- Ground | | | | B +---R4----+-----R5----+ | | V3 | | Q------------------+Vgate | Ground
B is the bulb for the streetlamp Q is a transistor such that a high Vgate causes a current through B PR = photoresistor is infinite resistance in the dark, 0 ohm in daylight
(R2/R3) < (R4/R5)
If all the resistors are in spec so the circuit is working: In the dark, Vgate > V1 In the light Vgate gets pulled _down_ by the short to V1
Now, if R3 fails open so the circuit is broken: in the dark Vgate < V1 (and also V1 == V2) In the light, Vgate gets pulled _up_ by the short to V1
Choose the resistors so that, e.g. if you are using a FET with a 10 V gate voltage for the transistor:
Vgate_night_working ~ Vgate_day_broken ~ 10 V Vgate_day_working ~ Vgate_night_broken ~ 1 V
(The voltage drop through R1 gives the extra degree of freedom to allow the FET gate to get appropriate voltages in all 4 cases)
This is probably overly contrived as a circuit that shows the observed behavior. (Although maybe not; if you want to add additional components for hysteresis, this circuit geometry may be appropriate as a starting point.)
-- David M. Palmer dmpalmer@email.com (formerly @clark.net, @ematic.com)
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.