On 25/01/2020 13:37, tony sayer wrote:
In article , The Natural Philosopher
scribeth thus
On 24/01/2020 20:34, Bill Wright wrote:
On Friday, 24 January 2020 02:37:09 UTC, The Natural Philosopher wrote:
I fly model planes. The barbed wire fence glitch is well known

I haven't heard of this and I'm very interested. Please tell us more.

back in the days of 35MHz NBFM it was well known that flying at
reasonable range over rusty fences would cause momentary loss of or
corruption of signal. Generally seen as a massive twitch on all flight
surfaces and a loss of throttle.

35MHz was never the best of mediums anyway but club fliers got to know
that certain places were prone to this effect.

Whether it was multipath, or diodic mixing causing intermodulation I
cannot say - or both.

For more information, the signal modulation on FM RC is/was a series of
frequency excursions marking the edges of a series of channel frames
about 1ms long with the lengh of a frame being the position of the
associated servo - in the range of 0.5ms to 1.5ms. A very long frame -
10ms or so - was used as a synchronisation frame.,

This form of modulation has only one benefit. A simple CMOS shift
regsiter could decode the pulse train to the servos.

But as an interference rejecting schema it was pants. Loss of a frame
edge meant one channel got two channels worth of info and the subsequent
channels got te wrong info.

Hence the massive glitch/twitch.


Bill



Interesting!

Time for OFDM perhaps;?..

we have now gone 2.4GHz. Essentailly the wifi band. This uses a
collision detection algorithm and spread spectrum. Also some frequency
hopping. Since the frame rate is low - a complete model only needs 64
bits every 20 ms - a massive data rate of 3200 bits per second - the way
the systems work is to transmit a frame with error correction and a
transmitter ID, and back off if there is collision. The comms is bi
directional so the receivers request retransmits if a frame is corrupt.

Loss of a frame simply means that the model carries on doing whatever it
was doing - the actual perception is that it's sluggish and
unresponsive. Ther is usally a watchdog tomer that will move the
controls to user configurable preset positions - usully cut throttle and
mild level turn - in the event of total signal loss.

Since every transitter has a unique MAC code, operation of dozens of
models simultaneously is possible - all too many transmitters does, is
slow the data rate down a bit. Its like having a shared wifi network.

I used to get some issues flying through a tight beam from the local
radio mast that you know of all too well. Lets hope te 2.4GHz stuff is
more resilient


--
"Strange as it seems, no amount of learning can cure stupidity, and
higher education positively fortifies it."

- Stephen Vizinczey