[ARDF] Re: Montreal fox controllers

[Gerald Boyd]

Jennifer,

Also don't use the spring type fuse holders between the controller and the
battery. 

Use the small automotive blade type fuse holders (can get them at radio
shack). 

The spring type holders can break contact during transport into the field
and reset the controller! Not a good thing to have happen on event day when
people are ready to start. 

We don't have an ON/OFF switch. We use power pole connectors to power
everything up and issue a master reset with a sync cable (a box with a
pushbutton and a cable to each controller). 

Our controllers have been very stable. As on both event days they came up on
time and stayed in sync.
 
Jerry
-----Original Message-----
From: ardf-bounces at kkn.net [mailto:ardf-bounces at kkn.net] On Behalf Of
Charles E. Scharlau
Sent: Friday, February 17, 2006 6:27 AM
To: ardf at kkn.net
Subject: [ARDF] Re: Montreal fox controllers

Hi Jennifer,

I've built (or re-built) about ten of the old version of Jacques & Francois'
Montreal fox controller. Based on my experience, and what I've heard about
the alternatives, the Montreal controllers have proved to be the most
reliable and accurate controller available; and the least expensive to boot!
I am not sure what improvements are in the newer version, but whatever they
are, they are not mandatory for having good reliable controllers.

There are a few things you can do to ensure that your Montreal controllers
work reliably and keep their timing relative to one another:

1. Use a high-quality IC socket for the PIC. You want to be able to remove
the PIC for re-programming, and a quality socket will ensure that the
processor stays put when shaken, and stays reliably installed through
changes in temperature and humidity.

2. Solder each part carefully. Soldering is the number 1 source of problems
in my experience. Only for that reason would I recommend a PCB instead of
hand-wiring on a perf-board. Also, if you really want to ensure that each
controller has an identical clock frequency, a PCB helps ensure identical
parts layout, and prevents random stray capacitance from affecting clock
synchronization. Otherwise, perf-board wiring should work just fine.

3. Program your PIC with software that stores ALL controller parameters in
non-volatile memory. Earlier versions of software (and perhaps the more
recent ones, too) read data (such as fox number) from the DIP switch at
power up. The DIP switch is a definite weak link, and is best used only for
initial programming, not for setting controller parameters on the fly. I
prefer to never use the DIP switch at all, and simply reprogram each PIC
before installing it in the controller. I have software available should you
need it.

4. Use identical, high-quality, crystals and oscillator circuit capacitors
in all the controllers you build. The capacitors (C1 and C2 in Jacques'
schematic diagram, I believe) are especially key. Do not use standard
ceramic caps having a tolerance of +/- 20% and which are not stable over
temperature. Instead, use the "COG" ceramics (+/- 5%) available from
Digi-key: 399-1890-ND. (Also, I don't recommend trying to use a trimmer
capacitor to tweak the oscillator frequency. If you carefully adjust each
oscillator circuit you will have them "perfectly" aligned only so long as
they are at the same temperature and they don't get jostled. You are better
off just using quality fixed-value caps and identical crystals. An added
plus: no alignment or frequency counter required!)

5. Use a minimum number of connectors, and where you must use a power or
audio connector, use a high quality type, and solder carefully.

All of the Montreal controllers I've built have been 100% reliable, and keep
their timing (relative to one another) within ~3 seconds for more than a
week of continuous use. They will be used for the US Champs in April... and
they are the least of my worries!

What a great V-day gift!

73,
Charles
NZ0I


_______________________________________________
ARDF mailing list
ARDF at kkn.net
http://www.kkn.net/mailman/listinfo/ardf