Optus box (CAU) RF Shortwave HF Interference Ham radio QRM - Solved
Introduction
Shortly after getting into shortwave listening (SWLing), I began to notice a peculiar pulsating noise scattered throughout the shortwave (HF) bands. It was especially noticeable on 40m (the 7 MHz Ham band) and a few other higher frequency clusters.
I knew this was interference from within my QTH, because, on a portable shortwave receiver, the interference did not occur outside in the open. Having switched off the power to everything, the QRM still remained.
With the portable receiver, the QRM was traced to the phone lines. Upon further investigation, it was found the Optus CAU (grey Motorola box), mounted on the exterior, was the culprit.
The Optus CAU
Now, more information about the CAU. The purpose of the CAU is to provide telephony services to a property via the Optus coaxial network, officially known as the HFC (hybrid fibre coax) network. These telephony services are in the form of a standard copper pair, providing POTS or PSTN. In order to do so, the CAU mimics a telephone exchange by providing a POTS interface that communicates back to the central telephone exchange via a built-in coax converter.
How do the phones get their line voltage from a coax cable? Simple - the cable that runs from the power pole to the property is known as a siamesed cable - one that has both the coax plus a copper pair that provides 97 volts of power. This supplies power to the CAU, which is used for its circuitry to interface the POTS service onto the coax. The good thing with this service (as opposed to new Optus cable installations) is that Optus provide all the power required to run the telephones even during a mains power outage.
The coax also provides several well known services - namely cable TV and cable internet. The external power is not required for cable internet (the cable modem has its own power supply). So, for some cable TV or cable internet only installations, the CAU is sometimes omitted. If the customer requires a telephony service, the CAU is required to provide the POTS interface.
Interestingly, newer Optus cable installations do away with the CAU alltogether, but instead places an integrated box in the home that provides both an ethernet interface for the internet and a POTS interface for telephones. Only a coax connection from the street is required. The box requires mains power but has a backup battery, ensuring phone communications in the event of a power outage. This takes away the power supply responsibility from Optus and makes it necessary for the customer to provide power for the box.
Initial investigations
getting up a ladder and opening the CAU's access panel with a security hex bit reveals a connection bay that has provisions for several PSTN lines, power input, earth and the coax in/out.
Most early CAU installations did not have a telecommunications earth connected. Optus subsequently revisited all older installations and rectrospectively installed a telecoms earth to those that did not have one. There were upsides and downsides to this. It makes the system safer from lighting strikes, surges and electrical faults and lessens Optus' liability for damage in such situations. But when the earth is connected to the home's electical earthing system, it allows the RF noise to travel thoughout the home's electrical wiring. Temporarily disconnecting the earth does seem to reduce the QRM, but not by much. It should be noted that disconnecting a telecommunications earth can potentially be both illegal and dangerous.
There was something else that was radiating. As I had cable internet, the coax running to the cable modem was the next thing to be looked at. The cable modem and also coax outer shield were grounded. This did seem to reduce the radiation from the coax, but did not entirely eliminate it.
Next was the phone service. The phone sockets were wired back to the CAU's PSTN ports. The copper pair cable is a balanced pair. However, it does make a really good antenna for radiating interference. I then proceeded to disconnect power from the CAU by removing a jumper block. The interference stopped immediately. However, that left me with no telephone service. After reinstating the jumper, the CAU took about a minute to reboot, and afterwards the interference returned.
I then disconnected the phone lines from the CAU, and the noise was drastically reduced but not eliminated. But again, it left me with no phone service so that was no good either. The copper pair was reconnected. However in this case, with the lines disconnected, the level was so low, the receiver had to be within metres of the CAU. So, some sort of filtering could possibly be added to stop the noise from travelling down the copper phone lines.
Ferrite beads to the rescue
Having established that it was the CAU that was the source of the noise with the phone lines helping to radiate the noise, I decided that filtering was the solution. The noise radiated from the CAU itself is low enough to be tolerated, but the radiation from the phone line was not.
Because the phone line is a balanced pair, the common mode radiation can be nulled by placing the pair inside a ferrite bead. This, in theory, should stop the radiation, but still permit the phone line to operate properly.
So, I went out to the local electronics parts supplier and purchased a large ferrite RF supression bead. The bead has an outer diameter of about 25mm and an inner diameter of about 12mm. More than enough to place a phone line through.
I disconnected the phone line from the CAU, placed it though the bead and reconnected. I then checked the noise by surfing the bands on the receiver, and to my disappointment, the noise did not drop by much. The phone lines were still radiating QRM!
So, what next? I did recall reading somewhere that multiple turns though a ferrite bead will increase its supression capability, so I got back up, unhooked the line again and looped the line through the bead another two times. In fact, for best effect, you should loop it through as many times as you can. But in this case, there was only enough slack to put it though three times. So I left it at that, and went to test the signal again.
What a difference it made! By having the cable loop though the ferrite bead at least two times, the common-mode interference drastically dropped. I could still hear it faintly, but it no longer had such a strong presence that made listening to DX hams impossibly.
So to all those who say "Ferrites make no difference", they are probably making one mistake: threading it through only once. Ferrites work best when the cable is looped multiple times, as this multiplies the suppression effect. They also work best close to the source of the radiation to be supressed. This means placing the bead as close to the CAU as possible. And it won't hurt to place more beads along the length of the phone line if circumstances allow.
The ferrite works by being more permeable than just air, so the field around the cable finds the ferrite an easier path. The ferrite, essentially a 'shorted' magnetic loop, will simply turn the absorbed RF into heat. However this heat is tiny, and becomes dissipated as quickly as the RF is absorbed. But the levels are high enough to be picked up and amplified by your receiver to cause annoyance.
Conclusion
Motorola seems not to have not designed these CAUs to be RF compliant. It is surprising they are even classified as FCC Class B. They emit noisy trash and hash. Fortunately there is a way to mitigate the effects. One, by removing power (bad because it cuts off phone services), and two by filtering (a compromise between completely removing power and having some residual QRM). Obviously, filtering is the solution, although motorola should have spent a few dollars investing in adding ferrites to their CAUs. Or even Optus recognising that this is a problem and offering to fit ferrites for customers.