Z-Wave devices - Can be serious generators of RF noise

K7JV

My amateur radio station has suffered from high ambient noise in our home from day one, nearly six years ago. Z–wave has been in service in our home, also from day one. In the past few weeks, I finally put all else aside and went on a mission to find my problem, and thanks to the awesome information available on the panadapter and the waterfall of my FLEX 6400, I have tumbled to the fact that the Leviton DZ15S in-wall switches are extremely high generators of RF interference. Has ANYONE researched and learned any useful lessons on this problem? I did not find where any of the other 23 Z-wave devices are interference generators, which is good. It is my hope to share my experience with others that may have the same problem and maybe not even know it, as well as to hopefully gain from the experience of others.

On the FLEX, with virtually all of the rest of our home powered down except for the breaker that serves our great room and back porch, I found that the above switch device would raise the noise floor on the radio by a whopping 17 dB!!! Furthermore, it is interesting that the interference is far greater when the switch is OFF than when it is ON. Seventeen dB is BIG! In addition, I learned that the DSL line coming into our home is a huge interference generator! Our refrigerator is no slouch at creating highly visible noise and interference, particularly when the door is opened or closed.

I seriously wonder how compatible Z-Wave technology is with amateur radio, and I'm so thankful that the FLEX 6400 could be such an incredible tool for ferreting out my problems. I have a 3+ page writeup on what all I found and would share it if it would be of value to anyone.

Beyond that, I wish you all a very Merry Christmas and a far better and Happy New Year!

PS: All of my testing so far has been in the 80M amateur radio band, which is from 3.5 MHz to 4.0 MHz.

VE7ATJ_Don

I can commiserate... in my old home, the brand new LG fridge we installed completely wiped out the 40m band whenever it was on. I tried power line filters, RF filters at the fridge, etc. The only way I could use 40m was to switch the fridge off at the breaker -- which wasn't a great solution, but at least during cooler months, I could use 40m for a couple of hours.

I'm in a new home now and still setting up HF antennae. Hopefully the noise won't be as bad (we bought a Bosch fridge :-))

Jeff W7NEE

Thank you for sharing this! I too have a high noise level after locating to a New residence. I would be most interested in your 3 page document for ideas of what you have discovered.

K7JV

RFI and NOISE FLOOR ISSUES.docx

RFI and NOISE FLOOR ISSUES - ZWave Related.docx

I don't know if this will work, but I'll give it a try. The larger document is looking at the various breakers that have Z-wave devices on them with the pic on the left being the baseline with the entire home "dead". The one on the right in each case is with only that breaker energized. Note the high noise floor that starts at 3.8 MHz. I'm 90% sure that is from the incoming DSL line. The only question I have is why it did not appear in the baseline picture. More testing to do on that issue.

K7JV

The smaller document, if it makes it to where it will be visible to you, is looking at each Z-wave device. I meant to say that in my reply a few minutes ago. As for the DSL noise present, I found that disconnecting the DSL line at the modem in our laundry room cabinet does not affect it. But when I disconnect it at the phone company service entrance box, it goes away. But when I took the baseline photo in the other document, I had NOT disconnected it at the phone company box. So I'm not sure why it wasn't present. Like I said, more testing to do on that issue.

I really am considering changing from using our traditional land-line DSL provider and going to a cable internet provider. But I want to confirm that the DSL is the absolute source of that plateau of noise above 3.8 MHz before I make that plunge.

VE7ATJ and W7NEE, thank you for your input. Cheers and best 73, and more to come when I learn more.

K7JV

More information - The reason the left pictures in the larger files were showing the noise floor to not have the risen portion from 3.8 MHz and up was because that image was showing my receiver to be on my ANT 2, which is a Grayline flagpole antenna that is fed from about 30' of coax, none of which is close to any radiating systems or devices to speak of. The ANT 1 that I've been working with, and that is the right picture in each pair, is quite the opposite in that the antenna is in my attic along the entire length of our home and is exposed to a number of systems including several Ethernet cameras, most of the homes service wiring, and the incoming phone and satellite cables and conductors.

BUT, I have learned more!!! The "table" on the noise floor that begins at 3.8 MHz and goes up from there has been positively identified. It is directly from the local telephone company's DSL line coming into our home. I learned that disconnecting the "twisted pair" at either the DSL modem, or the DSL filter that feeds the land line phones withersoever dispersed around our home, or both, the effect on the noise floor "table" is only minor. BUT, when I disconnect the "twisted pair" at the service entrance on the side of the home, the noise floor "table" goes away, entirely! So it is definitely being caused by the e-field from the "twisted pair" between the service entrance and the cabinet where all of our internal phone lines and TV coax lines are terminated. That incoming line parallels my folded dipole in the attic, perhaps 30" away from the lower antenna wire for a distance of about 20'.

We have now just made the decision to change our Internet service provider from the local phone company to the local cable company. That should solve what is likely my worst noise nightmare of the bunch. I am of the strong belief that by virtue of the design of coaxial conductors such as those used by TV cable companies (and amateur radio stations as well) concentrate nearly 100% of the e-field to be within the dielectric between the center conductor and the shield. Both are current carrying conductors, but they should be minimum radiators of errant RF due to that design. My belief and hope is that by using the cable Internet source, my RFI to the folded dipole should be negligible, if not non-existent. And it should actually save us around $60 per month, so its a win-win.

Questions? More to follow. I'm probably a couple of weeks away, at least, from having the cable source installed and put into service.

Cheers to all, and a very Merry Christmas!!