Air Conditioner RFI
|Jul 24th 2011, 19:58|
Total Topics: 0
Total Posts: 0
A new AC/Heating central air unit has just been installed in my house. It is generating spikes approximately every 18 khz and steady noise at a lower level between spikes.
The greatest interference is on 17, 15, 12 and 10 meters. The spikes ride upward of 30 dB above ambient noise and the steady noise appears to be at least 10 dB.There is also interference on 40 and 20 to a much lesser degree.
The pickup is via close by antennas/transmisssion lines, not through the AC power. The central system fan appears to be the source of the spikes and then when the cooling system activates the steady noise appears.
I have found unused wires in the thermostat cable and grounded them at one end - the AC unit. This includes cabling to the outside heat pump. It did nothing.
Any thoughts? Help....
|Jul 25th 2011, 11:12|
Super ModeratorJoined: Apr 4th 1998, 00:00
Total Topics: 0
Total Posts: 0
Let me start by explaining what ARRL knows about these systems and what it doesn't (yet) know.
With today's concerns about energy efficiency, technology is changing to make better use of energy. Most power supplies today use digital switching technology because it is more efficient than older analog power supplies (and a LOT lighter).
Switching supplies typically make noise every "N" kHz, where N can range from 5 kHz to as high as a few hundred kHz. Other switching supplies appear to create a broadband hash that extends across many MHz.
Electirc motors can also be noisy. Older motors could use brushes and armatures or commutators, and sparking could cause noise over many MHz. Motor speed could be controlled, but the process to do so was somewhat inefficient.
Many newer motors more efficiently control motor speed with pulse-width modulation. These are used in equipment such as pool pump motors and HVAC systems, among others. If the pulses are not well filtered, these motors can make horrible noise, and ARRL does have a few reports of interference from systems in neighboring homes. In most of those cases, the neighbor was not cooperative and those neighbors now have an autograph from Laura Smith of the FCC. The cases are still being resolved.
For the most part, PWM motor systems are controlled by the FCC Part 15 rules on conducted and radiated emissions. There are exceptions. If the frequency of the pulse rate is below 9 kHz, the device is not considered to be using RF energy, so it is classified as an incidental emitter, subject only to the rules that require good engineering practice in the design and that require that the operator of the device control harmful interference.
The bad news is that there is another exemption to the rules that control emissions: residential appliances.These, too, are subject only to the rules that require that actual intererence be addressed.
You are correct that the problem is caused by conducted signals, but when those signals get on to wiring connected to the unit, that wiring radiates the signal. Most Amateur receivers have very well-filtered ac lines, so it is unlikely that the problem is caused by conducted emisisons being conducted into your radio. This can be demonstrated quickly by disconnecting the antenna from your radio. If the noise goes away, it is a radiated phenomenon.
The good news is that based on the signal levels you are reporting, it appears that the system does meet the rules and, at the levels reported, it is probable that interference from neighboring homes would be much lower than you are reporting. The 18-kHz spaced signals could be audible, but not loud and more than likely, the broadband noise would not be present if this were being operated in a neighbor's house. This means that your unit does contain more internal filtering than the present FCC rules would require.
However, in cases of operation close to the unit, ie it is operating in your own home, additional filtering may be necessary.
The noise coming from the unit could be either common-mode or differential mode in nature (see http://www.arrl.org/files/file/Technology/RFI%2520Main%2520Page/Hallas.pdf).
More than likely, the noise is being conducted onto the ac mains. It could also be conducted onto other wiring such as the thermostat wiring although this would be unlikely, especially since it didn't change with grounding of the unused pairs, which would probably would have made a difference (for better of for worse) if the noise were on the thermostat wiring.
If the noise is common-mode in nature, you may obtain some relief with a common-mode choke. Obtain an FT-240-31 ferrite core (Amidon, Dan's Small Parts, Palomar) and wrap about 10 turns (if possible) of the ac wiring supplying the unit onto the core. You will need to include hot and neutral leads, plus the ground wire. You can also use -43 ferrite mateiral.
It will probably be necessary to cut the wirng and install the common-mode choke using wire nuts. This should be done by a qualified electrical contractor or installer. A serious shock hazard exists on the electrical wiring, and incorrectly done, the system could pose a serious safety hazard!
Considering the frequencies involved, you may be able to use clamp-on ferrite chokes and obtain some relief. Make sure they are either -31 or -43 ferrite material. You would need five to 10 of the larger clamp on ferrites to provide much effectiveness at upper HF. At lower frequencies, even more would be required. Still, they may be a good troubleshooting tool.
However, having said all that, it is more likely that your interference is caused by a differential-mode signal that would not be much impacted by the common-mode chokes. In this case, a differential-mode filter may be required. Corcom makes a series of ac line filters that may be of help, although you will have to carefully match the configuration and current ratings.
Now, for the part we don't know -- the ARRL Lab is in the process of testing a number of these filters, to ensure that they will be effective. Especially if your system operates on 240 volts at fairly high current, the filters will be expensive. That's the bad news.
The good news is that this case of interference to your own home gives ARRL a unique opportunity to have the ham have better access to the system than would occur in a neighbor's home, so if you are willing to work with our staff by telephone, in consutation with the manufacturer as needed, ARRL will obtain what appears to be a suitable filter and we can get some specific reports on its effectiveness. When all is said and done, we won't ask you to incuor the additional expense of removing the tilter.
If you are game, contact Mike Gruber, W1MG@arrl.org and ask him how to proceed.
When ARRL completes its testing of the ac-line filters, we'll post the results on the RFI area of the Technology pages on the ARRL web site.
That's where we stand today. We'll try to help get this resolved for you, as this will be the easiest PWM case we have to date, because it's yours instead of a neighbor/s. (Hopefully you are a homeowner, not a renter!)
Ed Hare, W1RFI
Technology forums moderator