[time-nuts] GPS DO Alternatives

Jim Lux jimlux at earthlink.net
Sat Dec 8 22:52:46 UTC 2012


On 12/8/12 9:30 AM, johncroos at aol.com wrote:
>
>
> Instead the discussion has centered on what microprocessor (of a hundred
> that would work)
> and how to eliminate "glue" logic and and a few analog parts to save
> money. This is silly - silicon is
> CHEAP.

Silicon is cheap, but for one-off fabrication by a hobbyist that isn't 
always the case.  As was mentioned in a couple of the mails in the long 
discussion, for a single person to build something like this requires a 
combination of skills and materials. Someone may be fine at software, 
but doesn't want to fabricate circuitry, or vice versa.

So, there was discussion of what could you do that would literally be 
"plug and play" with minimal hardware design and assembly required (so 
the playing would be with software).


This isn't an unusual scenario.. The AMSAT folks have run into it vis a 
vis ground stations. So have others (APRS).  A colleague of mine (N5BF) 
comments that what you really need is something where someone can 
"impulse buy" enough to do something useful fairly quickly.  The "kit 
idea": buying $100 worth of parts and then having to spend 6 weeks 
assembling and testing means that lots of people will have $100 parts 
bags sitting on a shelf, unused.  You'd be better off selling a $200 
assembled and tested widget.  Yes, you won't sell quite as many, but a 
LOT more of them will be actually used than those bags o'parts.


A particularly attractive model is where you have a hardware component 
that is delivered pretty much ready to go, with basic software, and the 
"fooling around" is with changes in the software or parameters. For the 
GPSDO world, this might be experimenting with different filters and 
holdover strategies, or maybe tuning it to work with your particular OCXO.

This is why the Arduino is so popular.  No or minimal soldering 
required, a wealth of simple software that almost does what you need it 
to, be it monitoring the temperature of your beer fermentation, turning 
on and off sprinklers or whatever.

Anything where the software is quite complex, that will inhibit 
experimentation, unless there's a lot of documentation of the theory of 
operation and software design, and the software has to be written to 
facilitate modification.   For the Arduino, the limited amount of 
storage sort of self limits the complexity of applications.  Once you 
move into the PC world it gets a lot harder.  And realistically, a lot 
of hobby written software doesn't have a good architecture or underlying 
design.  It sort of just growed in place with successive modifications 
to add features, etc.  And it works, but it's not very easy to figure 
out how to modify it.






More information about the time-nuts mailing list