colleague Linley Gumm and I recently made an accidental discovery. When we
turned our RF test bed on for a second day of testing of inter-band
interference, all seven of our DTV receivers failed to come on. Some had much
greater jamming than in our tests two days earlier.
previously wrote about inter-band interference in which two or more undesired
signals on the high VHF band channels generate third-order distortion products,
which fall in the UHF band where they raise the noise floor under the desired
ATSC signal. (See “Interband RFI: Should
If the noise floor under the desired
signal comes within 15 dB below the desired signal, reception fails due to this
IBI, which has seldom been considered because the RF selectivity of receivers
in the past helped protect against it. This may not be the case with the
integrated circuit tuners that dominate DTV receivers today.
When Linley and I first discovered the anomaly, we had no idea why the
change happened or which set of data were incorrect. We moved the spectrum
analyzer to an output of the eight-way signal splitter, which feeds our seven
receivers and to a HP 436 Power Meter. Wow!
third-order IM products across much of the UHF band as shown in Fig. 1. Note
that the center frequency of the spectrum analyzer in Fig. 1 is 587 MHz (Ch.
3). Third-order IM “Bee-Hives” generated by Chs. 9 and 11 are centered at 579
and 591 MHz. Their third harmonics are centered at 567 and 603 MHz. That proved
these spurious responses are third-order intermodulation products. So we
tracked the source of the third-order distortion, which turned out to be the
DTV receiver we failed to turn on.
|Fig. 1: Third-order distortion products generated in a modern DTV receiver
when not operating.
|Fig. 2: Third-order distortion products generated in the input circuitry
(yellow trace) of modern receivers vanish when they are not in operation
vanish (blue trace) when the receiver is operational.
When all seven receivers are
turned on, there was nothing in the UHF band and all seven receivers displayed
Fig. 2 shows a blue trace, which is the
baseline noise floor in the UHF band when all receivers were on.
When at least one receiver was off, the distortion products
appeared (yellow trace). Somehow, a receiver, when off, can generate
third-order distortion products, which travel back through the signal splitter
to the other receivers also connected to this signal splitter. Those receivers
were jammed by these distortion products. When all receivers were on, no
third-order distortion products were generated. All receivers worked.
When a modern receiver is off, its RF input impedance is not
75 ohms, so the coax from the signal splitter is not properly terminated. We
suspect the third-order distortion products were generated in the front end of
the off receiver. With the receiver “dead,” there would be no back bias voltage
applied to junctions of diodes or the RF amplifier transistor, so our VHF
signals drive the unbiased junction in conduction.
is true, then the “off” receiver looked like a highly nonlinear load (the
conducting junction) to the coax cable. Of course, the “on” receivers received
the third-order distortion products (IM3, third harmonics and Triple Beats)
being generated in the “off” receiver. These fell in Chs. 32 and 34. We had
tuned all seven receivers earlier to Ch. 32.
or more strong high VHF band signals are received, there will also be Triple
Beats (TB) generated in an off receiver. There are 31 Triple Beats generated by
all seven high VHF signals. The lowest TB is 177 MHz + 183 MHz + 189 MHz = 549
MHz, Ch. 25.
The highest TB is generated by signals on Chs.
11, 12 and 13. The sum of their center frequencies is 621 MHz. The highest
frequency in this spectrum is 627 MHz in Ch. 40. Third harmonics of high VHF
band signals are 3* 177 MHz – 9 MHz = 522 MHz. The highest third harmonic is 3*
213 MHz + 9 MHz = 648 MHz Ch. 43.
products are generated in a receiver when not operating. Which UHF TV channels
could be jammed by two high VHF band signals?
channel pair is Chs. 7 and 8. The center frequency of Ch. 7 = 177 MHz, and for Ch.
8; 183 MHz. The center frequencies of the IM3 for this pair is 2* 177 = 354 +
183 = 537 MHz. This is in Ch. 25 (536 – 542 MHz. The lowest frequency component
of this IM3 is 9 MHz lower: 528 MHz, which is in Ch. 23.
Likewise, the highest frequency of this IM3 is 2*213 + 207 = 633 MHz
(center frequency). Adding 9 MHz to get the highest frequency component, we get
642 MHz, which lies in Ch. 42. Most of these channels are those most likely to
remain TV channels after re-packing next summer.
this accident worth writing about? Consider that many homes having a rooftop
antenna are receiving signals from that antenna via a signal splitter. Now
consider that one receiver is tuned to a UHF channel, say 32 or 34, while a
pair of strong VHF signals, say on Chs. 9 and 11, are also feeding both
receivers. If one of these receivers is “off” the other may not be able to
receive the desired UHF channel (32 in this example). It is possible that
someone sooner or later would stumble across the fact that one receiver was
“off” when the other receiver could not receive Ch. 32 or 34.
In a second scenario, the rooftop antenna is feeding multiple apartments
in the same building. In this scenario it would be nearly impossible to
determine the cause of erratic reception of the UHF channel.
Many, if not most, homes today receiving TV signals with a rooftop antenna
have at least two receivers connected to the antenna by means of a signal
splitter. If one of these receivers is off, and the other is on, and tuned to a
weak UHF signal, it may not work reliably. That is, it may be getting interference
generated by the off receiver’s front end. Apartment houses, hotels etc. have a
master antenna and distribution amplifiers to compensate for the attenuation of
the signals by signal splitters.
Very strong high VHF
band signals can cause interference to receivers tuned to many weak UHF
signals. So this form of Inter-Band Interference is indeed real. I deemed this
hitherto little known interference mechanism worth writing about because I
expect it will be an annoyance to OTA viewers after channel re-packing.
Stay tuned for more good news.
Charles Rhodes is a consultant in the
field of television broadcast technologies
and planning. He can be reached via
e-mail at [email protected]