0-10 volt LED dimming

[Canton]

I am posting this in this forum to get feedback more from an Engineering stand point.

The issues on this job grow day to day. Here is the latest issue.

We have 0-10 volt DC dimming with LED lights. The wiring method that was chosen was 12/4 MC. Two conductors carry the 277 volt power and neutral and the other two are the 0-10 volt DC.

These drivers in the LED lights source the 0-10 volt DC. The dimmers in each room are just an on/off button with a slider that inserts resistance in the 0-10vdc line. No resistance (10vdc) is full brightness, increase the resistance till you get 0 volts or short the wires together and you have full dimming.

I am getting a vibrating/pulsing in the lights in certain rooms. But in certain rooms with the same amount of lights i am not getting this effect...? I have been able to correct it by running low voltage wire for the 0-10vdc and abandoning the two extra conductors in the MC for the 0-10vdc.

What I am trying to figure out is why this is happening? Is the AC side inducing and interfering with the 10 Vdc side? Why only some rooms and not all with the same exact set up? It also intensifies with longer runs and more lights.

One other interesting thing. I disconnected the 0-10 volt DC from all three lights in one of the rooms and just left the two conductors in the MC used for the dimming circuit connected in and out of each light. I then removed those same two wires from the dimmer so they were connected to nothing at each end, at this point they were just two extra wires through the MC about 30' long. I took some "AC" voltage readings with a High and Low impedance meter and had 30-60 Volts AC between them and 50 - 100 Volts AC to ground....! Regardless of which meter i used i definitely had significant voltages on those wires.

Also, some of the 4 wire MC that was used was MC light (aluminum jacket), some of it was a heavier steal jacket. When the MC sheath was stripped back some of the brands twist their conductors inside and some are straight. Could any of this affect why it is happening in some of the rooms?

 

[gar]

150228-0950 EST

I suggest your problem possibly can be solved by about a 25 ufd 25 V electrolytic capacitor at the light fixture across the 0-10 terminals with proper polarity observed. What size capacitor is sufficient is determined by the external capacitance in your power cable, and the internal impedance (resistance) at he 0-10 input.

Your 0-10 V input is a 10 V source with an internal impedance of about 10 k or possibly 1 k. Determine this by measuring the current when the 0-10 terminals are shorted. Using a linear curve variable resistance as a load across the 0-10 terminals gives you a non-linear change in voltage across the terminals vs angular position of the variable resistor.

When the 0-10 terminals are open or the variable control resistor is at maximum resistance the voltage across the 0-10 terminals is a maximum and the brightness is maximum. When the variable control resistor is 0 or at a minimum, then the brightness is minimum.

I will guess that internally the dimmer is measuring the voltage across the 0-10 terminals, and not the current thru the terminals. Also I guess that the 0-10 terminals are isolated electrically from the 277 V power.

Knowing the internal impedance of the 0-10 input may provide a better idea of why your long cable run is causing problems. You can guess at 20 to 100 pfd for the capacitance per foot of your cable run. I don't believe it will be 100, but assume it is, then 100 ft is 10,000 pfd or 0.01 mfd. At 60 Hz 0.01 mfd is about 230,000 ohms. A divider of 23 from 277 is about 12 V of ripple added to the control signal at full brightness and assuming a 10 k input impedance. This is just a ball park idea because I have taken a lot of liberties in making these estimates.

The idea of a shunt capacitor at the 0-10 input is to greatly reduce the coupled noise ripple.

.

 

[Canton]

150228-0950 EST

I suggest your problem possibly can be solved by about a 25 ufd 25 V electrolytic capacitor at the light fixture across the 0-10 terminals with proper polarity observed. What size capacitor is sufficient is determined by the external capacitance in your power cable, and the internal impedance (resistance) at he 0-10 input.

Your 0-10 V input is a 10 V source with an internal impedance of about 10 k or possibly 1 k. Determine this by measuring the current when the 0-10 terminals are shorted. Using a linear curve variable resistance as a load across the 0-10 terminals gives you a non-linear change in voltage across the terminals vs angular position of the variable resistor.

When the 0-10 terminals are open or the variable control resistor is at maximum resistance the voltage across the 0-10 terminals is a maximum and the brightness is maximum. When the variable control resistor is 0 or at a minimum, then the brightness is minimum.

I will guess that internally the dimmer is measuring the voltage across the 0-10 terminals, and not the current thru the terminals. Also I guess that the 0-10 terminals are isolated electrically from the 277 V power.

Knowing the internal impedance of the 0-10 input may provide a better idea of why your long cable run is causing problems. You can guess at 20 to 100 pfd for the capacitance per foot of your cable run. I don't believe it will be 100, but assume it is, then 100 ft is 10,000 pfd or 0.01 mfd. At 60 Hz 0.01 mfd is about 230,000 ohms. A divider of 23 from 277 is about 12 V of ripple added to the control signal at full brightness and assuming a 10 k input impedance. This is just a ball park idea because I have taken a lot of liberties in making these estimates.

The idea of a shunt capacitor at the 0-10 input is to greatly reduce the coupled noise ripple.

.

Gar, thanks for the info. We decided to break up the runs halfway through. Where we broke them we changed over to a low voltage two conductor cable. This seems to have broken up the interference onto the dimming circuit.