CAPITAL CURRENTS: Witnessing the FCC’s white space tests
‘White spaces’interfere with TVs hooked up to cable.
In July, I had the opportunity to observe the FCC’s “white space” tests and the severe direct pickup interference to cable TV reception that occurred. This creates a new challenge for the 120 million TVs that are directly connected to cable with a set-top box.
Companies like Google, Dell and Microsoft hope the FCC will allow the use of unlicensed transmitters on frequencies that are allocated for TV broadcasting. Since only a few TV channels are actually used for broadcasting in most cities, the conjecture is that the remaining TV spectrum, so-called “white space,” can be used for broadband local area networks. The problem for cable subscribers is that cable systems use all of the TV spectrum, and there are no white spaces on cable. That’s not a problem when using well-shielded cable set-top boxes, but the FCC tests showed that TVs are not well shielded, and TVs directly connected to the cable network are very likely to suffer interference from white space transmitters.
I witnessed tests conducted by FCC staff engineers on July 22 and July 31 at two private homes in Maryland.
One test on July 22 was interference from a white space transmitter into two TVs that were directly connected to an analog cable channel – a large (40-50 inches), flat-screen Toshiba and a 19-inch Samsung CRT. The Toshiba was in the same room as the transmitter, while the Samsung was in the next room.
Cable channel 73 was selected (visual carrier at 517.25 MHz), and the transmitter was set to 512-518 MHz (TV channel 21). The separation distance was about 1 meter. Initially, the transmitter was set to 100 mW output power – the maximum power allowed under the Commission’s proposal – into an omnidirectional antenna, and there was severe interference that looked like granular noise on the Toshiba. The power was decreased by 3 dB steps to 50 mW, then 25 mW, then 12.5 mW, then 6 mW, and the interference was still perceptible. Even at a 3 meter separation distance and 6 mW output power, the interference was still perceptible.
On the Samsung TV in the next room, the interference was worse. Even with the signal going through a wall, there was significant interference. Interference was also seen on cable channel 72, because the transmitter frequency overlapped channels 72 and 73. The interfering signal was leaking into both TVs’ tuners – direct pickup interference.
Two other tests were done at this house. In one, the cable was connected directly to a set-top box, and then directly into the TV. There was no interference until the 100 mW transmitter was within 1 foot of the box. The tuner in the cable box had better shielding than the tuners in the TVs.
The other test used a set-top box but reconnected the “rat’s nest” of subscriber-installed cables, switches, splitters and other devices that had previously been connected to the cable. The results were similar to the first test, with interference showing up from transmitted signals as low as 10 mW. In this case, the interference was leaking into the cables, switches and splitters.
On July 31, two interference tests were done: First, the TV was connected to a cable set-top box tuned to a digital channel; and second, the TV was directly connected to cable and tuned to an analog channel.
For the first test, the cable box was set to channel 6, and the Pioneer TV displayed a WETA (Washington, D.C., PBS station) digital program that was being carried in the cable channel 81 multiplex (564-570 MHz). The white space transmitter was set to off-air channel 30 (566-572 MHz). There was a considerable amount of subscriber-installed house wiring that could not be disconnected.
At a separation distance of 4 meters, with the output power set to 100 mW, there was some blocking and freezing of the picture. When the separation was reduced to 2 meters, and then to 1 meter, the blocking and freezing became more frequent. Since the cable box was probably well shielded, this showed the effect of signals leaking into the customer-installed wiring.
Next was a test of direct pickup by the Pioneer TV’s analog tuner. The cable was connected directly from the wall to the Pioneer and tuned to cable channel 72 (510-516 MHz), and the transmitter was tuned to off-air channel 21 (512-518 MHz). At a distance of 1.5 meters, and with output power of 100 mW, there was noise evident on the display, but not as bad as on the Toshiba TV tested the previous week. The power was decreased to 50 mW (still interference), and then to 25 mW (barely noticeable).
These results confirm tests done at the FCC Labs last year. Interfering signals from these unlicensed transmitters can easily leak both into your own TVs and into your neighbors’ TVs. FCC immunity rules require “cable-ready” TVs to work without interference in fields as high as 100 mV/m, but I calculate that the white space transmitter created fields much higher than that.
If the FCC approves unlicensed transmitters on TV white space frequencies without first addressing cable’s interference concerns, it creates a new challenge for those 120 million analog TV sets now connected to cable without a set-top box. Even “cable-ready” TVs – both analog and digital – will suffer.