From: Paul Keinanen on 10 May 2005 09:11:19 -0700
While a backplane would not be suitable for running the RF signals,
it
would be a good idea to have a common control interface standard.
This
might be some sort of serial interface or perhaps a CANbus interface
as used on some AMSAT satellites.
Who says a "backplane would not be suitable?" :-)
Those PC backplanes carry terribly broad spectra of
RF...from (literally) DC on up to the low microwaves.
No "perhaps" about it. Thing is, the layout can NOT
be done as if it were wire-wrap; i.e., in random
order of wire placement.
The PCI signals must run on transmission lines, since the receiver is
not activated by the for forward wave, which is reflected by the
mismatched end of transmission line and the receiver is only activated
by the combination of the forward and reflected wave. So indeed, the
layout is critical to get the signal through, even if no crosstalk
problems would exist.
But...there CAN be COUPLING there and that is, very
definitely, part of the layout. When mixed with
analog signals - as would be the case in a "radio" -
the layout can be critical.
In a PC, the signals are all around a few volts, thus the crosstalk
problems are not so bad.
Look again at the 3.3 V logic thresholds. :-)
In a radio receivers, the signal levels vary
from less than a microvolt to several volts, so the crosstalk issues
are much more demanding.
I will disagree on radio receivers on such wide dynamic
ranges. "Several volts" INTO a receiver front end?
No. Such levels aren't encountered in practical
locations and would, definitely, cause enough IM
that would create much distortion and spur products.
In radio transmitters, YES, but those stages can be
individually shielded and thus isolated...do NOT
need to be close to the control lines...or even need
control lines (in the case of an amplifier block).
Microstrip transmission lines would hardly be
enough, at least striplanes with grounded traces between the signal
conductors in the middle layer would be required, so the minimum would
be a 3 layer PCB.
Not the case in practical RF structures done in
the last three decades. [been there, done that,
got lots of T-shirts] It is BETTER to have good
stripline and microstrip as opposed to "ordinary"
PC layout, but that isn't an absolute necessity.
The IEEE-488 requires a lot of signals and a complex handshaking, so
in practice, you would need an interface chip anyway.
That was cited solely as an example of something
that IS mature and used daily in radio-electronics
testing.
The CANbus has been used in the automobile industry for more than a
decade. The CANbus has a nondestructive collision system, so this
makes it possible to have a true peer-to-peer communication system,
without complex protocols (such as token passing).
IF and only if this SDR of the future NEEDS micro-
computer control...or even modular microcontroller
sub-systems.
Trying to use an EXISTING computer interface system
isn't always good because that system has worked for
a decade-plus. While automotive computer interface
system speeds are increasing with increasing control
demands, radios aren't quite vehicles. The control
needs aren't quite the same.
The AMSAT thing I was referring to is a standard PCB, with a size
about a D connector, with an interface chip on it and it has a few
digital signals. It is included in every module on the bigger AMSAT
birds. This bus structure greatly simplifies the wiring between
modules.
I've had hands-in on earlier unmanned spacecraft but
understand the principles...which are similar to the
interface chips for things like USB adapters to work
with Serial or Parallel port peripherals with PCs.
One SOC (System On a Chip) that is essentially a
dedicated mircocontroler is all that is needed.
[FTDI makes those chips, Mouser sells them]
What you describe is more like an outgrowth of the
existing microcontroller adaptation to amateur
radio (and, more, to commercial radio) equipments.
The front panel controls are coupled (mostly) via
DC lines to the actual signal controls on PC boards
to reduce the mechanical complexity...which allows
greater freedom of layout and compactness.
[positive attributes for spacecraft as well]
My "ancient" Icom R-70 receiver has a central
microprocessor doing a great number of control
tasks...and does have some external control
capability through a rear connector. At about
two decades old, that's just one example of what
already existed - in radios - some time ago and
still does. Modern amateur transceivers usually
have two microcontrollers. Some of those allow
external control and a few are entirely controlled
externally. The basics have already been laid
down for the SDR system on what CAN work.
What is lacking is STANDARDIZATION. That can't be
worked out in newsgroups, but requires much more
organization...and willingness to compromise
(almost impossible in newsgroups, heh heh). See
any of the industrial standards (EIA, AES, etc.
in the USA) which are the first steps towards
making ANYTHING "plug and play." Example: The
Cannon "D" connector was on the market in the
early 1950s. A combination of factors made it a
practical connector line used in many electronic
things. Eventually, it became so common in the
USA that it was Standardized in shape, materials,
dimensions, etc., despite the original company
changing in corporate evolution. Wide use made
it "standard." The 25-pin and 9-pin D connectors
are on practically every PC today...as they were
in the beginning of the PC in 1981.
Standardization isn't anywhere close to reality
for SDR now. Nobody can seem to agree on WHAT
range of control is needed, let alone details of
the controlling interface signals. :-) That
might work itself out later.