181201-1934 EST
sosuaveone:
An LED semiconductor chip is a diode that emits light. Thus, it conducts in one direction only. So long as you do not exceed the PIV rating (Peak Inverse Voltage or breakdown) you can apply a reverse voltage and not damage the LED. No light is emitted in the reverse direction. In the forward direction a lot of current flows at a low voltage, somewhat above 1 V. So long as you do not exceed a reasonable current the chip will not be burned out. Current limiting is required. In the forward direction an LED emits light in contrast to an ordinary diode.
What is an LED puck light? By that I mean how does it work? Is it just an LED chip or chips? Does it use resistive current limiting or something more complex?
Note: strip LED lights are strictly DC. What does that mean? It means that the strip consists of LED chips in series with a series current limiting resistor, and many of these in parallel as the strip gets longer. A strip LED assembly may light on AC, actually it should, but will only conduct on 1/2 cycle. If AC is applied it must not exceed the PIV rating. In the forward direction the internal current limiting resistor, and the number of LED chips in series determines the design rated voltage.
On DC a strip LED assembly is polarity sensitive? Meaning works with on polarity and not with the other.
An LED chip with applied DC does not emit noise.
How does the puck light compare with the strip light. I don't know. Possibly some literature will tell you. Not likely, but possibly indirectly.
An ordinary CREE 9.5 W screw in bulb has lots of internal electronics that allows it to run from a 120 V 60 Hz sine wave, a phase shifted dimmer output, or a reduced voltage sine wave. How does the puck light compare with the CREE? I don't know. Does the CREE emit RFI? Some but not a severe amount in the AM broadcast band. It is more in the 50 kHz range. If fed from a phase shift dimmer it will make more noise in the AM range, but this is a result of the dimmer and not the LED bulb. From the dimmer noise results because of the rapid change of voltage at the turn on in the cycle. This is also true of an incandescent bulb on a phase shift dimmer.
How can you study your LED puck problem. I suspect a battery powered AM radio can be a good detector. Pick a frequency with only a weak AM signal. Apply power to the LEDs, does the background noise increase? If so, then a useful detector.
Apply a DC input to the puck light. Do you get noise? If so, then you have puck problems. You could use a filter at each puck, and probably solve the problem, or select a different puck.
If no RFI with DC excitation, then your problem originates with the dimmer power supply. This requires changing to a different supply.
If you need dimming, then there are ways that would minimize noise.
.
