[AMPS] Microprocessor-Controlled Amps

[Dave Kirkby]

Jim Benson wrote:

> Wondered if anyone has experience with microprocessor-controlled
> amplifiers.  eg. SGC SmartPower Cube) ......

Sorry I dont know the amps you talk of.  If you do find anything useful
on microprocessor control of amps, I would be very interested. I have
bought a Tern V104 controller board (uses NEC 25 CPU) and has AD
converter,  DA converter, real time clock, watchdog timer,  etc on the
board. I'm in the process of building this into a 2m amp using the
3CX5000A7 or YC156 tube. The software was written in C and has been
tested on a simulation of the amp, that uses leds to replace the power
relays, pots to simulate the analogue inputs etc.

I'm in the process of packaging the processor board into a very good
quality RF tight box (machined from a sold block of aluminum) that has
64  Pi section  feedthru filters with two 2.5nF caps and an inductor -
that should keep the RF out and the CPU clock away from the antenna! All
signals entering or leaving the CPU box are buffered, as the processor
cant be expected to drive the 5nf feed through filters. Other I've
communicated with, who have experience of CPU control of amps, suggest
that keeping the RF from the processor and the clock away from the
antenna are major challenges. Hence I've taken this quite seriously.
Obviously if ones measures anode current, and an arc occurs, there is a
risk of damage to the AD converter or even the processor. I've yet to
find out who well it handles deliberate arcs.

My software/hardware controls the amp, implements timing for the
heaters, relays, keeps the fan on for a couple of minutes when the amp
is switched off etc, It measures the important parameters - anode
current, anode voltage, grid current, fan temp (fan has an internal
thermistor), fan air pressure, input and output powers, temperatuers,
swrs etc etc etc, handles errors in an appropiate mannner and prints any
errors to a 16x2 character LCD display. Thats the theory anyway. As I
say, so far it has not been tested in a complete amp.

For anyone contemplating microprocessor control of an amp, I suggest
strongly you look at the Motorola Application note, with a title
something like - Designing for EMC compatibility with single chip
microcontrollers (you will find it on the Motorola web site)  I'll dig
out the full reference if anyone wants it. That covers some rather
interesting (and not so obvious) ways of improving the reliability of
CPU products from the EMC point of view. One idea I liked was to program
any unused memory with jumps to a error handling subroutine. If the
processor stars executing code at memory that is not used (as it might
if there is an EMC problem) instead of crashing, the program gets
directed to where you want. There are some obvious ones too - supply
decoupling, watchdog timers, but some of the software ideas were new to
me - but then I am not a computer scientist.

--
Dave Kirkby G8WRB
Department of Medical Physics,
University College London,
11-20 Capper Street,
London,
WC1E 6JA.

Tel: 0171 209 6406 (International +44 171 209 6406)
Fax: 0171 209 6269 (International +44 171 209 6269)



--
FAQ on WWW:               http://www.contesting.com/ampfaq.html
Submissions:              amps@contesting.com
Administrative requests:  amps-REQUEST@contesting.com
Problems:                 owner-amps@contesting.com
Search:                   http://www.contesting.com/km9p/search.htm