New Photocell Cycling Issue

[Lightall66]

I have volunteered some time to my church and am fixing some lighting and electrical issues they've been having. Everything was going just fine until I went to repair one of their HID (Metal Halide) wall fixtures. They told me they had just put in a new kit and lamp but the fixture still wouldn't light. The fixture also has a photocell.

The first thing I noticed was the core and coil (4 tap) was hooked to the wrong voltage and the photocell was also incorrectly hooked up. I resolved both issue's and reinstalled the lamp. The lamp fired and I started to close the fixture when the lamp went off. First thought was bad lamp. Replaced lamp but this didn't resolve the issue. Next I measured voltage on the line side and found good constant voltage. Then measured the load side of the photocell (photocell well taped) and found the cycling issue. While it didn't make sense as usually PC's usually work or they don't, I replaced the PC with a new one and still had the same issue.

The fixture works just fine with the PC bypassed.

I have been pushing a desk for the better part of the last 10 years but had about 15 years field experience prior to my management position. Now while my tools maybe a bit rusty, I didn't figure my troubleshooting skills were that bad. What am I missing? New lamp, ballast and second new photocell?

 

[dicklaxt]

Sounds like the cell is getting a reflection shadow bright to dark or vice versa,re-aim the eye to another direction and see what happens.
dick

 

[Lightall66]

That was one of my initial thoughts as well. But the photocell was well taped when i was testing it.

 

[dicklaxt]

That was one of my initial thoughts as well. But the photocell was well taped when i was testing it.

I have had photo cells that would function even when taped,how I don't know but it drove me nuts before I found out what was happening.I thought the eyes were bad,this was on a wild game feeder and I had to put a test button in the circuit to the eye,then it worked fine in the test mode.

Re aim the eye and see what happens with natural light.

dick

 

[gar]

100814-0913 EST

I am not familiar with photocell modules for light control.

Is this a two wire or a three wire device. Two wire means that all energy to operate the electronics must come from current thru the device, either leakage or load current. Three wire means full line voltage is available to power the electronics whether the load side is on or off.

A three wire device should have no problems if the photocell is shielded from the controlled light.

A two wire device has the potential for a problem if the leakage current when the load is off is insufficient. Not your problem.

Or if when on there is not enough load current to maintain the relay on. This will very much depend upon the design of the device. For example it might not be current, but a Triac might be the relay and there could be a deliberate delay to turn on to provide a sufficient pulse of voltage to a storage capacitor to supply power to the electronics for a half cycle. Other possibilities exist for powering the electronics. Fundamentally one wants to design a control device with low voltage drop when on to avoid wasted power.

You need to understand how your device works to make troubleshooting easy. With a suitable load in your shop try a unit that failed. Lean how it operates relative to a load. I do not know your lamp voltage and current ratings, but suppose 120 V and 5 A. In this case I would try a 750 W heater as a test load. This is a moderately constant load resistor. Then maybe a 50 or 60 W light bulb. These are a low resistance when cold and about 288 or 240 ohms when hot.

If the photocell device worked as expected with loads like these, then in the shop I would experiment with the actual load and probe further into why the strange result.

.

 

[ptonsparky]

Mixing the line load on a photocell will cause them to cycle.

 

[gar]

100814-2024 EST

ptonsparky:

Do you know if the type you are referring to is a two wire or three wire unit? Then if it appears to be a two wire is it actually a three wire where the third wire is the EGC and the EGC is actually used to provide power?

.

 

[ptonsparky]

3 wire units. Hot, switch leg, neutral.

 

[hurk27]

100814-2024 EST

ptonsparky:

Do you know if the type you are referring to is a two wire or three wire unit? Then if it appears to be a two wire is it actually a three wire where the third wire is the EGC and the EGC is actually used to provide power?

.

Gar here is one of the most common type photo cells in use:

http://forums.mikeholt.com/showpost.php?p=1186111&postcount=18

There is a heater next to a bi-metal strip contact, which is parallel with the LDR (Light dependent resistor) so if the load is removed then the heater cools off and closes the contact turning the fixture back on, if the hot (black wire) and load (red wire) is reversed, it will cause cycling to occur, as the load is removed and replaced. and yes most PC's are 3-wire, Hot, load, neutral. there are some very small two wire units made with a photo transistor, but they will have an instant dimming effect as the ambient light falls off the load lamp is turned on very slowly from full dim to bright.

 

[gar]

100814-2137 EST

Thanks.

The bi-metal type clearly requires a continuous power source and if wired backwards makes a nice low frequency oscillator when the photocell is illuminated.

Next I measured voltage on the line side and found good constant voltage. Then measured the load side of the photocell (photocell well taped) and found the cycling issue.

Connected backwards and with the photocell covered, per the original post, the normally closed switch should remain closed and the light would be on. This would not oscillate until sufficient light was on the photocell.

Something does not seem consistent here. The description of the problem conditions are not correct, or some other circuit than a CdS in series with a resistor is being used.

.

.

 

[ptonsparky]

The typical PC controlled fixture will come on briefly when power is first applied, even in broad daylight. You should expect it. Switch line to load and they blink.

 

[kwired]

The typical PC controlled fixture will come on briefly when power is first applied, even in broad daylight. You should expect it. Switch line to load and they blink.

And with the OP's HID lamp the lamp will go out and he will need to wait for restrike of the arc. He may be losing power someplace besides the photocell. He should run tests for voltage stability with a resistive load - and preferably with a load that is reasonably heavier than the luminaire this will heat up a bad connection if he has one and cause it to fail quickly.