Date: Tue, 31 Mar 2015 20:12:37 -0230 (NDT)
From: jcraig@mun.ca
To: rsgb_lf_group@blacksheep.org
In-Reply-To: <551B07AA.5010900@gmail.com>
Message-ID: <Pine.LNX.4.64.1503312003210.9373@plato.ucs.mun.ca>
References: <4698D0522AFD45F896A41F2D16A2D831@AGB> <BC1061758A4D4230AA9719C22735EBF4@gnat>
 <19EEF7A6DD30462FA6ACC0F015DC9F52@AGB> <5509E3D5.90900@gmail.com>
 <55102D0C.8040405@gmail.com>,<55148587.3090101@gmail.com>
 <7E7DFBB4D102A04DB5ADC88D66628A4A234F1A25@ICTS-S-MBX1.luna.kuleuven.be>
 <520562D5C7694FF69C20E595E00B793D@F6CNIToshiba> <55155AD6.1090102@gmail.com>
 <551B07AA.5010900@gmail.com>
MIME-Version: 1.0
Subject: Re: LF: WSPR at 4xf from a Raspberry Pi
Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Status: O

Hi Stefan,

That's very clever.  It seems the code is quite complex; can you explain
how it generates a carrier to someone who is an ungifted programmer?
My level is not much better than this:

def beep13(t2):
   i=0
   while i<500*t2:
     i+=1
     GPIO.output(13,1) #GPIO.HIGH)
     time.sleep(0.001)
     GPIO.output(13,0) #GPIO.LOW)
     time.sleep(0.001)
   return

I used this piece of python as a 500 Hz side tone for a pi keyer but
doubt it will work at 479 kHz.

73 and thanks for the reports on your progress.

Joe

On Tue, 31 Mar 2015, DK7FC wrote:

> Hi MF,
>
> Got no answer from PE1NNZ so i spent a bit more time and effort to 
> 'understand' the C-code, i.e. find the correct line where to put a '4*' to 
> get what i want. I found it (it is: */double freq = 4*centerFreq + 
> 4*((double)(-511 + i))*symOffset/3.0;/* ).
> I also compiled it successfully. Now i can externally set the same frequency 
> (e.g. 475685) but the output and frequency shift is 4x higher. This can be 
> passed through my /4 CD4013 stage, which is used to generate a perfect 
> rectangular, opposite phase signal, after the input signal has been distorted 
> a bit while passing an optocoupler (EMC stuff...).
> I know several OMs are using such divide-by-4 stages, which, so far, didn't 
> allow to transmit WSPR. Now, with the Raspy these days are gone, at least for 
> me.
> If some one is interested, i would of course share the code (you already see 
> it and just need to find the line in your wspr.c file).
> I'm now trying to run a bit more power as usual since the Raspi is directly 
> connected to the (real) PA: 
> https://dl.dropboxusercontent.com/u/19882028/MF/Raspi_to_QRO_link.jpg
>
> 73, Stefan
>
> PS: Ooooh, of course it could be that someone did the same before me, maybe 
> even better and so on...  so it is maybe not a 'first' whatever!!!!
>
> Am 27.03.2015 14:27, schrieb DK7FC:
>> [...]And i wrote an email to PE1NNZ and asked to a modification of the 
>> C-program to generate a signal with 4x higher FSK shift. Then it it would 
>> be possible to use that version for OMs having a divide-by-4 stage in their 
>> PA (to get a perfect 50% duty cycle and rectangular shape (after passing 
>> optocouplers etc)).
>
>