Linksys (Cisco) SPA921 VOIP phones producing noise in 70cm?

[yahya]

I would like to share an odd discovery. I have wireless weather stations by Oregon Scientific and Hama operating at 433.92 MHz, which have had problems with synchronization with the outside temperature transmitter. I finally realized that the problem could have been caused by RFI and tuned my venerable Baofeng and IC-R2 to the above mentioned frequency to find that there was a heavy noise present within ca. 5 to 7 yd. from the operational SPA921 phone. When the mains power supply plug was disconnected from the phone, the noise disappeared in the same moment. When the VOIP phone was unplugged and off, the weather station received the data from the wireless outside temperature probe without hassle. Note the power supply for the above mentioned phone does not produce RFI when plugged in to the mains.

I have two such old phones at two different locations, which were purchased a long time ago from different outlets. They have different firmware and mfg. dates, and surprisingly they have a large sticker claiming that they were FCC rules compliant, which have been know to be very stringent. While these phones are no longer supported my the manufacturer, I would like to ask if someone also has such a device and a radio scanner and could kindly tune to 433.92 MHz to check if any noise is present in vicinity of an operating VOIP phone of the SPA series?

 

[dkcorlfla]

One thing to do for sure is to try and rule out the switching mode power supply. Maybe try putting the phone in a mircowave oven (make sure it is off!)

Then move the Baofeng close to the PS outside of the microwave oven. Some of the RFI may come down the power corded if the problem is in the phone itself but you might get some clues where the problem is.

Perhaps you have another PS from something else and swapping them could help find the cause.

 

[yahya]

I already tried the SPA921 with a different brand 5V power supply. The effect was similar like with the original power supply. When I plugged it in, the RF noise at 433.92 MHz immediately appeared. The strength was at full scale on Baofeng and Icom R2 in vicinity, and the noise started fading at the distance of ca. 7 yd. from an operational SPA921. So, I do think the phone itself is the source of the noise. I have two different SPA921s of different mfg. years and with various firmware. Both behave the same and produce RFI at 433.92 MHz while operational. And that frequency is used by a number of radio-controlled items, like wireless weather stations, wireless key fobs, etc. No wonder such items' signals are "snowed" by the SPA921 while in vicinity. I really wonder if other owners of the SPA series VOIP phones encounter similar problems.

 

[RFI-EMI-GUY]

Since 433.920 is an ISM frequency, it is possible that 13.560 which is also an ISM frequency, is used to clock the CPU etc within the phone. The harmonics are what result. In any event, open the phones up to see what kind of shielding exists and clean them up or add foil tape to the exterior case and ground it to the other shielding ground plane. Putting ferrite chokes on the power and other leads may also reduce the radiation.

 

[KevinC]

Is the noise present with the phone powered but the ethernet cable not plugged in?

 

[yahya]

Is the noise present with the phone powered but the ethernet cable not plugged in?

Yes, it is.

 

[yahya]

Since 433.920 is an ISM frequency, it is possible that 13.560 which is also an ISM frequency, is used to clock the CPU etc within the phone. The harmonics are what result. In any event, open the phones up to see what kind of shielding exists and clean them up or add foil tape to the exterior case and ground it to the other shielding ground plane. Putting ferrite chokes on the power and other leads may also reduce the radiation.

Thank you for the valuable suggestions. Well, I also had an SPA901, which was a simple VOIP phone and when I wanted to disassemble it to repair the power socket, I accidentally tore off the RJ45 socket. So, this time, I give up.

I would be grateful if someone who also has a VOIP phone of the SPA series could check on a scanner if the noise is present at 433.92 MHz when the phone is powered on.

 

[dkcorlfla]

Ok, you made me curious. I have a SPA 303 so I hooked up an indoor antenna near the computer and fired up GQRX.

Not a trace of RFI from the SPA 303 but the monitor on the computer was a total mess. Check out the screen shots. One shows the waterfall with the SDR set to 434Mhz and the antenna about 2 feet from the monitor. I moved it as far away as I could to check the SPA 303 and there was still a trace from the monitor but nothing from the SPA 303. Perhaps it has something to so with " Phone Series, the SPA921 IP Phone includes a high-resolution display," I took a look at the SPA 921 and it does not look much different then my SPA 303 but the SPA 303 just has a plain monotone LCD display.

The second picture shows 434Mhz with the SDR hooked up to my outdoor ham radio VHF/UHF vertical.

Hope this helps, Dale

 

[yahya]

Thank you for the measurement tests that you made. The SPA303 is actually a 3-line phone, while the SPA921 supports only one line. I presume these sets must differ in terms of hardware.

I also checked a third SPA921 that I have and which is actually bricked. Upon plugging in the power supply, the noise at 433.92 MHz commences, but since the bricked SPA921 does not boot further, the noise goes off after a few seconds. I think that RFI-EMI-GUY is right that the noise originates from the CPU.

 

[yahya]

A small update. I disassembled the bricked SPA921 that I have. The CPUs are not shielded at all. I removed a jumper from the port marked by the yellow rectangle, and the phone entered into the SOS Phone in recovery mode. As you can see in the picture below, the noise at 433.92 MHz is also present, and this is the third SPA921 that I tested. The CPU marked ESS Visba3 is hot.

 

[dkcorlfla]

Looks like it might be time to consider a replacement. Back when I used to work (retired now) we had the Polycom VOIP phones and I was impressed with the HD voice. The audio is not bad at all on the SPA 303 but I always wanted to get in under the hood of one of the Polycom phones. The phones we had at work were provisioned by the provider so I never got to see what I could break. I was kicking around the idea of picking up one of the Polycom VVX 311 as they are very low cost (less then $20 on Amazon)

Maybe try one of these or a different model or make and if it makes RFI at 434Mhz send it back and try something else.

 

[kruser]

I have a Linksys SPA2102-R that also emits a fairly strong emission but centered a little higher on 433.9335 MHz.
I placed an RF probe over the board and the strong noise source is from the CPU chip which is also not shielded in my 2102.
My 2102 is no longer provisioned but the RFI still appears a few seconds after applying power.

I tried placing the 2102 in a steel box but it still radiated a strong signal on the above freq.
It seemed that the power wire is radiating the noise from the 2102 when inside the steel box. The power supply cord from my 2102's SMPS is shielded but it looks like the RF is riding on the cables shield. I didn't try to tame it with ferrites due to the strength of the signal. If I ever use the thing again, I'd try experimenting with ferrites and different power supplies. I know I have a regulated linear supply here I can use if I ever mess with the VoIP adapter again.

The 2102-R is a VoIP adapter only. You must plug in your own analog phone set but the 2102 emits the noise with no phone or ethernet cable attached. Plugging in an ethernet cable does raise the noise signal strength some but sniffing the cable with an RF probe does not really show that the ethernet cable is radiating the noise source.

 

[yahya]

@kruser
Thank you for the reply. So, my SPA921s were not alone as far as RFI is concerned. I can't believe how these sets were certified by the FCC.
Obviously, it's time to retire my SP921s and use something more contemporary.