LED lamp QRM - solved!

Last week I became really scared because I turned on 40 meters and there was AC buzz all over. I have a relatively quiet noise floor which is advantageous for contesting and DXing, and my fear was that my neighbors had bought some cheap device with a switching power supply that would spew noise all over the HF bands. Or was it the new fridge we bought? Or was it something else? Thankfully cooler heads prevailed and I went about finding the noise source methodically. Watch how I got it done!

Troubleshooting

The very first thing I did was to get a shortwave radio and verify that the noise was there. Indeed, a portable shortwave radio confirmed that it was there. Actiually, I used my Kenwood TH-D74A in general coverage shortwave mode. It works well for this purpose.

The next step was to shut off the main breaker in the house. We did that, and the noise was gone. This alone was a huge sigh of relief! The noise was inside this house, and this means I had full control over it, at least.

Then, I took the shortwave radio and began to walk throughout the house, bringing it close to things. This is a classic noise hunting technique that previously netted me a computer switching supply that was wiping out 80 meters.

I noticed that the living room produced more noise - aha! It's in here! So I began pointing to more and more stuff, switching off things one by one - TV, UPS, blu-ray (you still have one of those, right?) Then, almost quite by accident, I shut off the lights. The noise fell silent. What??? Yes, I had found the source! It turned out to be the living room lights, track lighting connected to a dimmer switch. I knew one of them had to be the culprit, but which one was it?

LED lights

LED and CFL lights get a bad rap for interference, and rightfully so. A number of tests by ARRL labs and others have shown them to be prolific noise generators. However, I had tested the bulbs that were installed here for ham radio friendliness before the return window had expired, and I returned any that were bad. What I was left with was a mix of GE, Cree and Kichler (Lowe's house brand it seems). 

The culprit was a  Cree BR30 LED floodlight. I noticed that one of them was taking 1/3 of a second longer to start. Removed it, and *poof* the noise was gone. 

Happy ending!

This lamp all of a sudden began to spew noise when it was fine for well over a year. I am thinking that a capacitor became defective over time. This would also explain the delayed starting. 

I'm going to contact Cree and see what they say.

See you on the low bands.
88
Ria, N2RJ

Comments

  1. Interesting. I've checked them when new, like you did, but haven't had any problems, knock on wood
    73 de kg2v

    ReplyDelete
    Replies
    1. I had no problems with these either until last week, but I suspect the RFI may be a warning sign of impending failure.

      Delete
    2. Every LED in my home that failed is a Cree. Just sayin'

      Mike - W2CHN

      Delete
  2. I've heard reports of light dimmer switches themselves being notorius noise generators.

    ReplyDelete
  3. I had a bad problem on 80m. Noise floor was S9+5dB. At first I thought it was my neighbor's solar farm on his roof or maybe the new LED street lights, but it was always there. So I went around the house unplugging anything I could see. Turned out to be a Chinese build battery charger that had a microprocessor in it. Unplugged and *poof* ! Noise floor went down to S2. "FCC Type Accepted my ass!" LOL

    ReplyDelete
  4. I have had no luck with Cree lamps either. So far two of 8 failed in less than a year!

    ReplyDelete
  5. If only mine was so easy. I have 40/9 on 80 and 9's on 40. Forget 160. Every time I think I find it, I find done check and it is no longer there but elsewhere. So frustrating

    ReplyDelete
  6. I suspect a lot of devices from China are "FCC type accepted" but use cheap parts wherever possible. So while they may meet the type acceptance, they may not stay that way for very long. If you have other such lamps in your house, It would be very interesting to see if they fail after a certain time in service.

    I'm sure the factory probably doesn't think anyone will notice the difference and realize what the cause is. Except maybe a few hams...

    ReplyDelete

Post a Comment

Popular posts from this blog

New video on YouTube - connector wrapup (for now)

Welcome, Hackaday readers!

Ria blogs anew!