FW: [Elecraft] Auxbus driver

[Jack Brindle]

OK, so how about this. More than a few microseconds delay could cause 
major problems, and most likely will.

10mS delay will cause a complete failure, and much confusion in the 
system. Anything more than 5 microseconds will probably get you there 
also. In the implementations, we are concerned with as little as 1 
microsecond changes. Why? Because the delay cascades through the 
protocol data bits, eventually throwing off the bit timing.

Of just as much concern is the likelihood of RFI getting into the 
system with cables of just about any length. Cables longer than a few 
feet tend to pick up RF and inject it into the system, again causing 
bus confusion and failure. Hit a resonant cable length for the 
frequency of K2/100 transmission, and you will see great problems.

I'm not sure I could be any more explicit than this. Extending the 
AuxBus cable beyond the (short) desktop lengths is a recipe for system 
failure. Wayne, Eric, Gary, Scott and Lyle will tell you the exact same 
thing!

On Wednesday, September 10, 2003, at 04:03  AM, Dan Barker wrote:

> That's just the point. We need the Powers-That-Be to define 
> "significant"
> distances. If significance is 5 nS, then 1 meter is the limit. If
> significance is 10 mS, then the limit is a couple KM.
>
> Dan / WG4S / K2 #2456 / 10cm Auxbus path<g> (Builtin KAT2)
>
> <snip>
> In short, officially, it is a bad idea to extend the bus for 
> significant
> distances.
> </snip>

-Jack Brindle, WA4FIB
=======================================================================
MacDobs - helping to shift the paradigm for low-cost amateur astronomy.