[Milsurplus] Radar simulator

[J. Forster]

Last fall, I was designing exactly what you seek for use in the CIC of the
Battleship Massachusetts. Unfortunately, the project had to be postponed
for personal reasons, but I did develop an architecture which should work
very well.

Essentially, it was done with an EPROM and counter and video D/A and
worked like this:

The address lines of the EPROM were divided into two blocks, one for
Azimuth and one for range.

The Azimuth address bits were devived in one of two ways, because the
design was for two different radars.

In one case, the radar antenna was motor driven at a constant rate by a
constant speed motor and the position monitored by a synchro. In this
case, I designed a counter to drive a digital to synchro converter,
simulating the antenna's rotation by incrementing the counter every so
many sweeps. This simulated a rotating antenna, and the simulated input
was sent to the PPI and PPI Repeaters and the servos therein slaved the
yokes to the digitally commanded position. The design was for 10 bits of
Azimuth data. The same 10 address bits went to the EPROM.

In the other case, the yoke was driven by a constant speed motor directly.
The yoke gearbox also drove a synchro, which commanded the antenna
position through a Ward-Leonard and big servo motor. In this case, the
synchro output was digitized with a 10 bit synchro to digital converter
and those address bits used for the address bits of the EPROM.

The scan was handled by a cloch and synchronous digital counter. We
decided how many pixels were needed to reasonably represent a scan and
calculated the correct clock rate. The zero count state of this counter
was used to trigger the display sweep and counts above the end of the
active sweep allowed for retrace time. This is a bit wasteful of EPROM,
but saves other timing logic.

The output of the EPROM went to a simple 2 bit D/A converter and thence to
one input of a MCL mixer. The other input to the mixer came from a small
crystal oscillator at the radar's IF frequency. Essentially, the simulator
output 0, 1/3, 2/3, full and simulated 4 return strengths. By selecting
other output bits from the EPROM in pairs, additional scenes are possible.
More intensity resolution is possible, of course.

Refinements include replacing the EPROM by RAM, so slowly moving scenes
can be displayed, and provisions for triggering the IFF capabilities.

The hardest part, IMO, is generating the EPROM loads and I've not got a
good solution for that as yet.

FWIW,
-John


> I have a Radar PPI indicator, which works, scan goes around tube etc  etc.
>
> I want to supply the video i/p with simulated returns. The thing
revolves at
> a given period, stepper motor driven. Each scan from centre to edge take so
> many microsecs.
>
> So I need to store the appropriate number of scans per revolution, each
scan
> length can be held in a memory, at some point during each scan a value
needs
> storing to indicate the trace to bright up and paint an image.
>
> Fairly straight forward I would have thought but before I reinvent the
wheel
> I was wondering if anyone else has tackled this type of project?
>
> Ben G4BXD.
>
> ______________________________________________________________
> Milsurplus mailing list
> Home: http://mailman.qth.net/mailman/listinfo/milsurplus
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:Milsurplus at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html