DSPRSSTC?

["Tesla list" <tesla@xxxxxxxxxx>]

Original poster: "Steve Conner" <steve.conner@xxxxxxxxxxx>

>About drawbacks: instead of a PLL you could plug in a small DSP like free
>samples of TI's TMS320LF2401A (32-pin, 40mips) or Motorola/Freescale
>56f8122 etc series (48-pin, 40mips, free devel tools).


It is quite tempting. My day job is designing and programming DSP and uC
based scientific instruments so I could probably manage it ok. But I can
think of drawbacks to DSPs too.


Pros of using a DSP-

All the things Jan said, plus a few other neat things I can think of like
built-in bang energy and RMS/peak current measurements, remote control and
data logging to your laptop, and last but not least a "singing arc" MP3
player :D


Cons-

Latency. All that signal processing takes time and it may turn out that the
DSP could not respond as quickly as an analog PLL controller.

Hassle. Learning another DSP instruction set and development toolset (I only
know the Analog Devices and PIC ones) Getting PCBs made and soldering tiny
surface mount things to them. Coming home from a day of programming to...
more programming...

Noise immunity. Most new devices run at 2.5 or 3.3v, I use 15v CMOS logic at
the moment.

Difficulty of debugging. What happens if your JTAG debugging dongle gets hit
by a ground arc :-0


Overall I think I will stick to 1970s logic for my SSTCs and save the DSP
stuff for my audio and radio projects. Unless anyone who is up to speed on a
suitable toolset wants to get together with me and develop an open-source
DSPRSSTC.

Steve C.