Topics

QCX-SSB: SSB with your QCX transceiver


Dean Smith
 

OMG! I get to choose  which oled to run with, 90% through the build now.. Making a cnc'd case over the weekend, pics to follow, thanks Guido, Manuel (great kit instructions, easy to follow!)
Exciting times :)


Ryan Flowers
 

Hmm, I wonder: Will running a 20x4 I2C LCD display be just as simple as replacing those few lines? I'm getting quite excited about this build now :D 

Ryan Flowers



On Thu, May 7, 2020 at 10:31 AM Dean Smith <bardezbiker@...> wrote:
OMG! I get to choose  which oled to run with, 90% through the build now.. Making a cnc'd case over the weekend, pics to follow, thanks Guido, Manuel (great kit instructions, easy to follow!)
Exciting times :)


--
Ryan Flowers - W7RLF
MiscDotGeek - QRP and More


ajparent1/KB1GMX
 

Running a larger LCD is a code change even with Guido's code.
They use the same hook up but the code has to know there are
4 lines and all.  Not a difficult change but then the user interface 
code need to change to take advantage of it so it more than a line
of code.

Allison
-------------------------------
Please reply on list so we can share.
No direct email, it goes to bit bucket due address harvesting in groups.IO


Ryan Flowers
 

Yes of course, I was thinking more of the I2C aspect of it :) 



On Thu, May 7, 2020 at 10:44 AM ajparent1/KB1GMX <kb1gmx@...> wrote:
Running a larger LCD is a code change even with Guido's code.
They use the same hook up but the code has to know there are
4 lines and all.  Not a difficult change but then the user interface 
code need to change to take advantage of it so it more than a line
of code.

Allison
-------------------------------
Please reply on list so we can share.
No direct email, it goes to bit bucket due address harvesting in groups.IO


--
Ryan Flowers - W7RLF
MiscDotGeek - QRP and More


ajparent1/KB1GMX
 

Ryan,

Well you can get the 20x4 LCDs with I2C adapters on the back
or with the parallel interface like the 20x2 and 16x2.

However even with I2C there is configuration code that has to
match the display.

The Oleds I've played with tend to be noisy and eat power,
like the look, but batteries.  Back lit LCDs run the led current
down and they stay readable at far less power.

Allison
-------------------------------
Please reply on list so we can share.
No direct email, it goes to bit bucket due address harvesting in groups.IO


Ryan Flowers
 

Yes, I have one of those and have used 16x2's on I2C before as well. I do realize that more configuration will be required, and that won't be a problem once I get it working. 

Thanks for the input :) 



On Thu, May 7, 2020 at 11:33 AM ajparent1/KB1GMX <kb1gmx@...> wrote:
Ryan,

Well you can get the 20x4 LCDs with I2C adapters on the back
or with the parallel interface like the 20x2 and 16x2.

However even with I2C there is configuration code that has to
match the display.

The Oleds I've played with tend to be noisy and eat power,
like the look, but batteries.  Back lit LCDs run the led current
down and they stay readable at far less power.

Allison
-------------------------------
Please reply on list so we can share.
No direct email, it goes to bit bucket due address harvesting in groups.IO


--
Ryan Flowers - W7RLF
MiscDotGeek - QRP and More


Skip Davis
 

Guido,
Thanks for all the updates I’m also very interested in this and have been following the github up. I have QCX-40 that will be used for the conversation to SSB.
This looks like a fun project to experiment with once I finish all the noodling over your’s and Manuel’s information.

Skip Davis, NC9O


Guido PE1NNZ
 

Many thanks Skip!

But actually this thanks should go to ALL of you, because without you this project would have simply not possible to bring it to this stage. So thank you!

And.. as we are not there yet... please keep experimenting, thinking, brainstorming, commenting, testing, helping, coding, building, ...    if you feel too it.

73, Guido


Colin Weaving
 

Nice job on the QCX-SSB manual. I am almost convinced I can handle that, albeit my programming knowledge is non existent.

Is there a schematic that details the modified QCX. I am afraid I am old school, and it would be almost unique if I did not have some trouble shooting to do. Then I would find the schematic indispensable.

Initially I wondered what the fuss was about, but It has developed into an intriguing project.

Colin M3WCK

On 7 May 2020, at 20:12, Skip Davis via groups.io <skipnc9o=yahoo.com@groups.io> wrote:

Guido,
Thanks for all the updates I’m also very interested in this and have been following the github up. I have QCX-40 that will be used for the conversation to SSB.
This looks like a fun project to experiment with once I finish all the noodling over your’s and Manuel’s information.

Skip Davis, NC9O



Guido PE1NNZ
 

On Thu, May 7, 2020 at 11:29 PM Colin Weaving <cwweaving@...> wrote:
Nice job on the QCX-SSB manual. I am almost convinced I can handle that, albeit my programming knowledge is non existent.

Is there a schematic that details the modified QCX.  I am afraid I am old school, and it would be almost unique if I did not have some trouble shooting to do. Then I would find the schematic indispensable.

Initially I wondered what the fuss was about, but It has developed into an intriguing project.

Colin M3WCK
> On 7 May 2020, at 20:12, Skip Davis via groups.io <skipnc9o=yahoo.com@groups.io> wrote:
>
> Guido,
> Thanks for all the updates I’m also very interested in this and have been following the github up. I have QCX-40 that will be used for the conversation to SSB.
> This looks like a fun project to experiment with once I finish all the noodling over your’s and Manuel’s information.
>
> Skip Davis, NC9O
>
>
>




David Goodrow
 

Red outlined sections of the schematic are components not needed?


Manuel; DL2MAN
 

Hello David,

Red outlined sections are changed in comparison to original schematic.
The missing sections are..... Missing ;)

Please read Guidos info from the project page.
(His link named other Infos) and you should be good to go.

73 Manuel DL2MAN


Dean Smith
 

Hi, Manuel, all,
Thought I'd add a couple of pointers to documentation for rev5 pcb, that people might find helpful when following the instructions.
Firstly the value (fat if changed) word fat would be better as bold.
C35 should be soldered slightly higher off the pcb to allow you to solder the pin nearest IC2. Otherwise it will be uber tricky!
C32 the ground pad on the Pcb looks as if it's floating. I'll assume the minus lead of the capacitor needs a board ground.
L5, 6 100uh. 47uh are supplied. I take it that's a typo.
Best
Dean (G7EOB)


Dean Smith
 

Ok scrub C32 comment its grounded on topside of pcb! Just looked.


Christiaan PA3FUN
 

Hi All.

I did my best to build QCX-SSB as neatly as possible on a unused R4 board.

I loaded R1.02a.

Display showed 1 line of white blocks, other line blank.

Following the (excellent) "troubleshooting" section on qrp-labs website I ended up suspecting the (i2c-communication with the) Si5351a.

I stripped Guido's code to the bare minimun, just enough to initialize the LCD and write the version-info to it, see below.

Do basically ALL interaction with the Si5351a eliminated from the code.

Now the LCD awaked, see photo attached. So, processor obviously okay & up-and-running, and wiring processor<->LCD okay.

I checked I2C-wires, SDA and SCK. No short-circuit, perfect connection from the Si5153a-chip to the processor. 1K pull-ups in place.

The Si5351a -power measures 3.5V. The 27MHz oscillator works (as observed thru freq.counter &oscilloscope).

No short-circuits between any other pin of the Si5351a.

However, no output. And, after, reloading the full code-base, hang-up again. Tried the same with Hans' rev 1.5 (yes I know, not intended for this modified board) with same result. Display top-row white blocks, bottom-row blanks.

 

HELP......

 

 

void setup()
{
  digitalWrite(KEY_OUT, LOW);  // for safety: to prevent exploding PA MOSFETs, in case there was something still biasing them.


  initPins();

  lcd.begin(16, 2);  // Init LCD
  for(i = 0; i != N_FONTS; i++){  // Init fonts
    pgm_cache_item(fonts[i], 8);
    lcd.createChar(0x01 + i, /*fonts[i]*/_item);
  }

 
  show_banner();
  lcd.setCursor(7, 0); lcd.print(F(" R")); lcd.print(F(VERSION)); lcd_blanks();

}

void loop()
{
}


Manuel; DL2MAN
 

@Dean:
Thanks for your feedback. As you might have figured out already, english is not my native language. So it is of great value for me, if a native speaker gives such a feedback.
In german this would be "fett", so I thoght "fat" would be apropriate :) But now, as you mention it, I remember the term "bold". Thanks again.

Regarding C35: Valueful input. Will mention it.

L5/6 are 100µH in Guidos Schematic, and I didn´t re check those. Thanks. Will change this also.

In the next revision of manual, there will be also Mic and Key Wiring, as Guido thought, it might be helpful.
If someone else realized somthing missing or wrong, please give me a feedback.

@Christiaan:
I asked Guido to have a look at it.
As far as I can see from your Picture, you did not do the full modification since T1 is in Place and C1 is also in Place. So you did not do the SDR Mod ?
I can´t give you much input, since I went all the way (Multi Band, SDR RX and SSB TX).

73 Manuel; DL2MAN


Kelly Jack
 

Step 4 should read pins 6 and 8.

Great job on the manual.

73

Simon
VK3ELH


Guido PE1NNZ
 

Hi Christiaan,

Can we debug this offline? I have whatsapp, can you give me a call?

On Fri, May 8, 2020 at 12:30 PM Christiaan PA3FUN <christiaan@...> wrote:

Hi All.

I did my best to build QCX-SSB as neatly as possible on a unused R4 board.

I loaded R1.02a.

Display showed 1 line of white blocks, other line blank.

Following the (excellent) "troubleshooting" section on qrp-labs website I ended up suspecting the (i2c-communication with the) Si5351a.

I stripped Guido's code to the bare minimun, just enough to initialize the LCD and write the version-info to it, see below.

Do basically ALL interaction with the Si5351a eliminated from the code.

Now the LCD awaked, see photo attached. So, processor obviously okay & up-and-running, and wiring processor<->LCD okay.

I checked I2C-wires, SDA and SCK. No short-circuit, perfect connection from the Si5153a-chip to the processor. 1K pull-ups in place.

The Si5351a -power measures 3.5V. The 27MHz oscillator works (as observed thru freq.counter &oscilloscope).

No short-circuits between any other pin of the Si5351a.

However, no output. And, after, reloading the full code-base, hang-up again. Tried the same with Hans' rev 1.5 (yes I know, not intended for this modified board) with same result. Display top-row white blocks, bottom-row blanks.

 

HELP......

 

 

void setup()
{
  digitalWrite(KEY_OUT, LOW);  // for safety: to prevent exploding PA MOSFETs, in case there was something still biasing them.


  initPins();

  lcd.begin(16, 2);  // Init LCD
  for(i = 0; i != N_FONTS; i++){  // Init fonts
    pgm_cache_item(fonts[i], 8);
    lcd.createChar(0x01 + i, /*fonts[i]*/_item);
  }

 
  show_banner();
  lcd.setCursor(7, 0); lcd.print(F(" R")); lcd.print(F(VERSION)); lcd_blanks();

}

void loop()
{
}


Dean Smith
 

Microphone and key schematics would be great. As I'm sure this will be my next block to overcome.
Looks like Hans summers has truly sorted Cat function for qcx, i think this would be a great idea with these mods for ft8/Js8 it would take considerable guess work out of setting up and moving frequency, modes ect.  But I bet memory on the 328 is tight, dream on :)


Guido PE1NNZ
 

Just had contact with Christiaan, here is a follow-up:

So, Christiaan's QCX-SSB kept hanging due to wrong fuse settings. It is the best to use the default Arduino fuse settings (you can burn them as states in the assembly guide by selecting Tools > Burn Bootloader). This will burn fuses: E=FD H=DE L=FF

Once programming is done, it is also important to disconnect the ISP cable; otherwise the QCX also keeps dead.

I have also uploaded a test version R1.02b that should fix an LCD issue (please report if you have it with 1.02a and is it fixed in R1.02b?)

73, Guido

On Fri, May 8, 2020 at 3:16 PM Guido PE1NNZ via groups.io <threeme3=gmail.com@groups.io> wrote:
Hi Christiaan,

Can we debug this offline? I have whatsapp, can you give me a call?

On Fri, May 8, 2020 at 12:30 PM Christiaan PA3FUN <christiaan@...> wrote:

Hi All.

I did my best to build QCX-SSB as neatly as possible on a unused R4 board.

I loaded R1.02a.

Display showed 1 line of white blocks, other line blank.

Following the (excellent) "troubleshooting" section on qrp-labs website I ended up suspecting the (i2c-communication with the) Si5351a.

I stripped Guido's code to the bare minimun, just enough to initialize the LCD and write the version-info to it, see below.

Do basically ALL interaction with the Si5351a eliminated from the code.

Now the LCD awaked, see photo attached. So, processor obviously okay & up-and-running, and wiring processor<->LCD okay.

I checked I2C-wires, SDA and SCK. No short-circuit, perfect connection from the Si5153a-chip to the processor. 1K pull-ups in place.

The Si5351a -power measures 3.5V. The 27MHz oscillator works (as observed thru freq.counter &oscilloscope).

No short-circuits between any other pin of the Si5351a.

However, no output. And, after, reloading the full code-base, hang-up again. Tried the same with Hans' rev 1.5 (yes I know, not intended for this modified board) with same result. Display top-row white blocks, bottom-row blanks.

 

HELP......

 

 

void setup()
{
  digitalWrite(KEY_OUT, LOW);  // for safety: to prevent exploding PA MOSFETs, in case there was something still biasing them.


  initPins();

  lcd.begin(16, 2);  // Init LCD
  for(i = 0; i != N_FONTS; i++){  // Init fonts
    pgm_cache_item(fonts[i], 8);
    lcd.createChar(0x01 + i, /*fonts[i]*/_item);
  }

 
  show_banner();
  lcd.setCursor(7, 0); lcd.print(F(" R")); lcd.print(F(VERSION)); lcd_blanks();

}

void loop()
{
}