KitsAndParts.Com Forum

Well, I made my first homebrew SSB contact (ever) just a few minutes ago (19:45Z) on 20 meters with this setup



I'm using the K2 for RX. T/R is done with an old DowKey relay (to the left of the K2) which is triggered with a toggle switch. The relay also supplies +12V to the *JBOT linear amplifier during TX. RX muting was done by turning down the volume control on the K2 during TX. In between the power meter and the DDS-VFO is the WB2EBY amplifier running about 15 watts output with a watt or so drive from the JBOT. I'm only using the amplifier on the BPF board (lower right) since the extra Power from Diz's RF amp wasn't necessary. Obviously this setup is subject to change. I want to try Diz's linear but took the easy route with the JBOT since it was easier to set up. I also want to try my DC receiver.

The QSO was with AC0BU, Carson who was in a bucket truck with a Yaesu and a Hamstick during the contact.  Carson, in Corydon, Iowa said the signal was weak and that I sounded fine as near as he could tell - under the circumstances.

Ok, that was fun - about as much fun as ham radio gets.

Tom, ak2b

*JBOT by Ashar Fahhrhan http://www.phonestack.com/farhan/jbot.html
My JBOT Youtube Video at http://www.youtube.com/watch?v=rID2-uZrCgE

w4gns wrote:

You worked a mobile with 15 watts SSB ?

Hi Gary,
He kept signing mobile but somehow I had in my mind that he was in the bucket of the bucket truck during the QSO .
I guess I was excited. Thanks for the comments.

Tom, ak2b

wb2me wrote:

While I was looking for a microcontroller prototype board I forgot about these:
http://www.arduino.cc/

Funny you should mention this device. I heard of it yesterday on the QRP-L list, looked it up and ended up ordering the Duemilanove board from Spark Fun http://www.sparkfun.com/commerce/produc … cts_id=666. I paid $30 for it. It looks like you can use it on any platform and the best thing about it is all the support. I'm not that good a C programmer but  I think I can muddle by. Looks like a lot of fun.

Also, saw this on the QRP-L reflector this morning http://www.kkn.net/archives/html/QRP-L/ … 00011.html
which looks REAL interesting. The good stuff is at http://openqrp.org/.

Tom, ak2b

The picaxe looks like the basic stamp grown up. 
Not a bad way to prototype things, but the performance won't be up to the arduino.

W4GNS wrote:

Good Point, actually I bought an Aduino Duemilanove Board several months ago, and also a book, played with it awhile , but I tend to go back to the Picaxe for whatever reason. Maybe I had better finish reading the book and play around with it a tad more. C does not really appeal to me, but who knows. and I found this PDF file a few days ago, I found it to be good reading, the subject is making PCB's for SMT's and its super easy, for anyone interest   http://www.vakits.com/datasheets/MakingSMTPCB.pdf

Looked at the pdf (yuck!).  Well there are much better ways to make a PCB than using a sharpie for resist.  I like the iron on method, using the toner from a photo copy machine or laser printer.  There are special transfer sheets for this but many use glossy photo paper with good results (but lots more clean up after peeling off the paper).  Also the common ferric chloride etchent is VERY messy.  I now use a mixture of 66% Hydrogen-peroxide (drugstore strength) + 33% Hydrochloric Acid (that's common swimming pool muriatic acid).  Pour the Acid into the H2O2 when mixing!   This etchant is quite fast.  Also it can be reused as the copper does not dilute the strength of the solution very fast.

How much power were you running during the 'test?  Sounds like an after burner might be in order (unless your goal IS QRP that is).
I found an article in Feb QST for 1963 that looks like what I might eventually end up with here, driven by something like the JBOT.
I have a suitable power transformer and about a dozen good 1625's plus just about all the other required parts (think I might have to
dig up a pair of 0D3's).  I still like messing with glass fets.

I dug out the spec sheet on the Max6954 so I now know what to stuff into the registers.  Going to try to light up the seven segment readouts
next.
  The Miami-Dade county hamfest was yesterday.  It was so poorly attended that it wasn't worth the gas, tolls, and admission price.
But I did manage to find a nice brand new chassis, a PC board full of some nice slug tuned coils in cans (tunable BP filters?), and some
breadboard sockets.  The former Motorola 'free-flea' is next month, hope it will be better attended than usual.  Where have all the hamfest
attendees gone ... fleabay?  (Or are they SK's?)   At least Dayton was well attended this year (or so I heard from a few hams at work that were there).

Well, here it is: SSB transmitter, the Video
http://www.youtube.com/watch?v=njo4Z_8cdWs

I got my Arduino board yesterday and have been playing around with it. Setting up the software and hardware was painless. The website seems to have anticipated every idiot question I could ask before I asked it. Nicely done.
I went through the brief tutorials and really like what I see. On one hand the concept lends itself very well to the beginner but doesn't limit you if you happen to grow beyond the simpler concepts. It's easy to see why this devolpement package is so popular. Support seems to be everywhere.

Tom, ak2b

Tom, I'd like to post a zip file with the source code I'm developing for the mega16/32 driving the DDS-60.
I don't see how to post an attachment to this site though.  I guess you post pictures by pasting them in, but I'm not
sure of that either.   

The link you showed to the Arduino controller is a bit basic.  I'm using the AVR spi hardware to talk to the DDS-60.
The Arduino library does have functions for the LCD driver with examples so it shouldn't be hard for you to get
an LCD up and running.

Here is my rotary encoder code, it is in C not wiring as in the Arduino, but I think it would work in the Arduino.

// encoder clocks are portb 2&3
#define phaseA PIND2
#define phaseB PIND3
//encoder states as seen at portb inputs
#define STATE_MASK 0x06

#define STATE0 0x00
#define STATE1 0x02
#define STATE2 0x06
#define STATE3 0x04

void init_decoder(void)
{
  DDRB &= ~(_BV(PIND2)|_BV(PIND3)); //input pins
  last_state = curr_state = 0;
  dial = 0;
}

void decode(void)
{
  curr_state = PIND & STATE_MASK; //read encoder

  switch(curr_state){
  case STATE0:
    switch(last_state){
    case STATE1:
      if(dial < 127)
    dial++; //increment dial
      last_state = curr_state;
      break;
    case STATE3:
      if(dial > (char)-128 )
    dial--;
      last_state = curr_state;
    }
    break;
  case STATE1:
    switch(last_state){
    case STATE2:
      if(dial < 127)
    dial++; //increment dial
      last_state = curr_state;
      break;
    case STATE0:
      if(dial > (char)-128)
    dial--;
      last_state = curr_state;
    }
    break;
  case STATE2:
    switch(last_state){
    case STATE3:
      if(dial < 127)
    dial++; //increment dial
      last_state = curr_state;
      break;
    case STATE1:
      if(dial > (char)-128)
    dial--;
      last_state = curr_state;
    }
    break;
  case STATE3:
    switch(last_state){
    case STATE0:
      if(dial < 127)
    dial++; //increment dial
      last_state = curr_state;
      break;
    case STATE2:
      if(dial > (char)-128)
    dial--;
      last_state = curr_state;
    }
    break;
  }
}


//interrupt on signal from encoder.
//call decode to increment/decrement dial
ISR(INT0_vect)
{
  decode();
}

ISR(INT1_vect)
{
  decode();
}


I've written code to receive input from a standard 38khz IR remote control.  I'll post that later once I clean it up.  It will
need to be made a bit more generic as  it's currently coded to accept IR codes from only one make/model IR remote control.
It will work with any IR remote control based on an industry standard NEC controller chip, but it expects a specific vendor code.
(I work for Niles Audio, so I'm using a Niles R8 remote).  But many vendors used this IR standard.


Here is the DDS code:
The frequency to DDS word calculation is done in integer math, I hand coded the DDS constant
as a binary fraction.

#include <avr/io.h>
#include <avr/interrupt.h>
#include <stdio.h>
#include <stdlib.h>
#include <avr/pgmspace.h>
#include <string.h>
#include "uart.h"


#define DDS_LOAD _BV(PB4)
#define DDS_CLK  _BV(PB7)
#define DDS_DATA _BV(PB5)
#define DDS_DDR  DDRB
#define PORT_DDS PORTB

#ifdef DEBUGDDS
static char buffer[40];
#endif


// (2**32 / 180000000) + (rem of 2**32/180000000)*(2**24)
// generates a binary floating point number as an integer
#define Konst (unsigned long long)0x17dc65de

// perform a floating point mult of the frequency by the
// dds constant for 180mhz clock and a 32 bit accumulator.
// Integer math is performed and the fractional portion
// of the result is shifted away.
unsigned long CalculateDDS(unsigned long freq)
{
  unsigned long long K = Konst; //(unsigned long long)0x17dc65de;

  union dds{
    unsigned long long Product; //64 bits
    unsigned long Part[2]; //32 bits * 2
    unsigned int i[4]; //16 bits * 4
  }dds;


  dds.Part[0] = freq;
  dds.Part[1] = 0;

#ifdef DEBUGDDS
  sprintf_P(buffer,PSTR("freq m->l %x,%x,%x,%x\r\n"),
        dds.i[3],dds.i[2],dds.i[1],dds.i[0]);
  uart_write(buffer,strlen(buffer));
#endif

  dds.Product *= K; //multply freq by const

#ifdef DEBUGDDS
  sprintf_P(buffer,PSTR("prod m->l %x,%x,%x,%x\r\n"),
        dds.i[3],dds.i[2],dds.i[1],dds.i[0]);
  uart_write(buffer,strlen(buffer));
#endif

  dds.Product = dds.Product << 8; //shift left one byte

#ifdef DEBUGDDS
  sprintf_P(buffer,PSTR("ret m->l %x,%x,%x,%x\r\n"),
        dds.i[3],dds.i[2],dds.i[1],dds.i[0]);
  uart_write(buffer,strlen(buffer));
#endif

  return dds.Part[1]; //return upper long int as result
}


void spi_byte(char ch)
{
#ifdef DEBUGDDS
  sprintf_P(buffer,PSTR("ddsbyte = %x\r\n"),ch);
  uart_write(buffer,strlen(buffer));
#endif

  SPDR = ch; //load spi to send a byte
  while( (SPSR & _BV(SPIF))==0){} //wait for done
}

void send_dds_freq(char *freq)
{
  //enable the spi as master, send LSB first, clock active high
  //using F_CPU/16 (500khz), can go faster
  SPCR = (_BV(SPE)|_BV(DORD)|_BV(MSTR)|_BV(SPR0));

  PORT_DDS &= ~(DDS_LOAD); //bring load (cs) low
  //send bytes 0-4 of the long word
  spi_byte(freq[0]);
  spi_byte(freq[1]);
  spi_byte(freq[2]);
  spi_byte(freq[3]);
  //send the control byte
  spi_byte(1);

  //latch the data
  PORT_DDS |= DDS_LOAD; //raise the load flag
}
 


void init_dds_spi(void)
{
  //clear bits going to DDS
  PORT_DDS &= ~(DDS_LOAD|DDS_CLK|DDS_DATA);

  SPSR = 0; //clear SPI2X
   
  //DDS bits are all outputs
  DDS_DDR = (DDS_LOAD | DDS_CLK | DDS_DATA);

  //enable the spi as master, send LSB first, clock active high
  //using F_CPU/16 (500khz), can go faster
  SPCR = (_BV(SPE)|_BV(DORD)|_BV(MSTR)|_BV(SPR0));

  PORT_DDS &= ~(DDS_LOAD); //bring load (cs) low

  //send 5 bytes of zero to DDS
  spi_byte(0);
  spi_byte(0);
  spi_byte(0);
  spi_byte(0);
  spi_byte(0);
  PORT_DDS |= DDS_LOAD; //raise the load flag

  //do it again to make sure!
  PORT_DDS &= ~(DDS_LOAD); //bring load (cs) low
  spi_byte(0);
  spi_byte(0);
  spi_byte(0);
  spi_byte(0);
  spi_byte(0);
  PORT_DDS |= DDS_LOAD; //raise the load flag
}

Hi Ken,
There is no way to post files to the forum directly. I do it by running a personal web server (IIS) on my XP Pro box and then run  an app that reports my IP to no-ip.org. It used to be free but now I pay a few dollars a year. I used to have ak2b.com registered but keeping up a website is a lot of work. In any case my web server still runs locally and I just keep all the pictures and sound files in virtual directories and point to them from the FluxBB.
I looked at your code and want to give it a shot. I have a 9851 hanging around so I ordered another DDS-60 board to try out your code with. I'm also interested in the K1EL project with the Arduino. It seems Steve just got the transceiver boards in and is trying them out.

Ken, I think it is time to start another 'Topic' under 'RF Toolkits'.  I'll let you have the honors since it is your code The DDS-60 is a excellant board and as far as I can tell there are no other controllers for it.

Tom, ak2b

W4DNQ wrote:

Tom,
Got the VFO built and MAN what a diff.
That N3ZI DDS was only sending a 300mv signal to the mixer and Diz's VFO was over 1.5v.
It seems that the low IF I had was from the small signal from the DDS.

I ordered a bunch of Copper Clad board today as I am going to take somewhat the
same approach you did. It is getting to the point now that that there are just to many
wires all over the place. UUUHHHH !

Will keep all informed,
Ron / W4DNQ

If diz's diode mixer is typical it should need 7dbm input for the LO.  At 50 ohms 7dbm is about 0.5 volts so that DDS is down
a bit less than half of what is required.  1.5v would be about 16.5dbm which would be over drive for the mixer, though
some mixers do require more than 7 dbm.  Mini circuits makes mixers in the 7 - 17 (or more) dbm range.
To convert volts to dbm look here: http://www.sengpielaudio.com/calculator-volt.htm

wa2me wrote:

1.5v would be about 16.5dbm

Hi Ken,
I think Ron was talking p-p not rms since this is what Diz specifies in the last instruction for building the VFO.

Diz wrote:

Finally, adjust R5 for your output. 1.4V P-P = 7dBm

Tom, ak2b

More and more I've started using the "Universal Generic PCB" for building projects. I would make one suggestion and that is to make the input/output 3 pin instead of 2 pin connectors. And along the side I would like to see at least one 3 pin connector for power.
I have no problem building with smd parts. The pad widths are just about right for a size 1206 cap or resistor. If you need to add two together you can mount an adjacent smd on the opposite side.
Anyway, this board is W7ZOI's bi-directional amplifier in response to the ones in the BitX20. If you read the article it pretty much sums up what this little amp will do. It is virtually identical in specifications to the aritlcle. After having built and tested this amp I will  build up two more to use with my SSB transmitter to make an SSB transceiver

Tom, ak2b