Content-Transfer-Encoding: 8bit
Message-ID: <3617AC3245C2D1118A840000F805359C017528D0@pdw-mercury-1.dera.gov.uk>
From: "Talbot Andrew" <ACTALBOT@dera.gov.uk>
To: rsgb_lf_group@blacksheep.org
Subject: LF: Re: DDS Source board,
Date: Thu, 11 May 2000 09:52:51 +0100
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1; format=flowed
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

Having a board with an uncommitted PIC means that the same hardware can
be used in any way that the user wants just by re-writing the firmware.
I was going to use a normal 18 pin DIP package rather than SMT PIC so it
can be used in a socket for maximum flexibility.    

I do realise PC timing is appalling in any of the Win operating systems
- that's why I still use DOS as much as possible and have a second
computer - an ancient 486 - dedicated to radio purposes with most ham
progs using EVM varients rather than Soundcard.  Even so, Win operating
systems can generate real timings to 55ms accuracies (can't they ?).  So
real time rates based on 1 second or longer schemes shouldn't be too far
out.   The 'PLX software for PSK31 with the EVM uses the serial port in
near real time and is quite happy with Win 95 / 98 etc - 

For high prescision work timing, sync pulses from a GPS receiver could
go to a port on the PIC, then use the PC via a handshake to set
parameters.   However, I'm uneasy about using the DDS control for
anything other than frequency setting for FSK and other INCOHERENT modes
anyway.  I believe PSK is better generated in dedicated hardware (or by
DSP and mixing, but that's another story).  That way, the frequency
source stays inviolate and fixed and modulation is applied via proper
waveform shaping to a modulator that can be as simple as an EX-OR gate;
or better, a mixer.

Andy   G4JNT


>From Larry Kayser :

>I offer that to have maximum efficiency when using a serial line into a
PIC
>to control the AD9850 it will be necessary to have your board OR an
external
>drive signal with the signal from the 16F84 for the W_CLK signal so
that
>accurate control of the FSK or PSK transition events can occur.  The
control
>could occur from either the 16F84 or from an external signal source.

>The first need is that an external drive signal can control the timing
of
>the BPSK or the FSK change / shift event.  If I have it right, I
suggest
>that the on board 16F84  needs to also sense when the external control
>signal on W_CLK has occurred so that the 16F84 can under program
control
>allow the next event to be uploaded and prepared to shifting into the
DDS
>portion of the chip.  IF I HAVE THIS RIGHT, this is the only way that

>accurate timing of events into the DDS for FSK and PSK can occur.  I
can
>absolutely assure you that the ASYNC serial port output of any Win95,
Win98,
>NT-4, and NT-2000 WILL NOT achieve the timing accuracy needed for FSK
or
>BPSK state changes.


-- 
The Information contained in this E-Mail and any subsequent correspondence
is private and is intended solely for the intended recipient(s).
For those other than the recipient any disclosure, copying, distribution, 
or any action taken or omitted to be taken in reliance on such information is
prohibited and may be unlawful.