Re: zBITX CW Mode - Proof of Concept

#118115

Note - This is a follow up to my 29APR25 Posting, which almost got me there.? Some of the background that is that original posting may be of interest as it is not repeated here.

Goal

Make my zBITX usable for manual Straight Key CW Mode and manual Iambic Paddle CW Mode.? At any speed.

Approach

Fix it in software.?? Work within the confines of the existing zBITX and sBitx software.? In other words, apply band-aids, not do a major rewrite.

Solution

The solution was achieved by modifying just the sbitx_gtk.c file, which is a bridge between the underlying sBitx software and the zBITX software/hardware.? My solution was to change the two CW input pins (Dit & Dah) to use Interrupts (IRQs), add IRQ handlers, and then modify the ui_tick() function.

The ui_tick() function is a very busy function that, among many other things, calls the existing sBitx function that looks at the Ptt(Dit) and Dah hardware inputs. The ui_tick() function is triggered periodically by a hardware clock.? The ui_tick() function runs nearly every millisecond.? But, if the ui_tick() function has not completed all of the processing from the previous clocked trigger, that new trigger is not acted on.? So, it is not deterministic.

What was happening was that the ui_tick() process was frequently busy doing something else that took some time to complete whenever it needed to be processing a manual CW key-down occurrences in real time.?? Thus, it had a very poor and unpredictable response, and would not support rhythmic manual CW keying ?

My solution was to:

1.????? Have my Ptt(Dit) and Dah IRQ handlers immediately make the initial call to the existing sBitx function that looks at the Ptt(Dit) and Dah hardware inputs. ?If that existing function (and it’s cast of supporting functions) finds a key down situation, the device is then put into Transmit Mode (the? in_tx? variable is set to True).

2.????? Move the call to the to the existing sBitx function that looks at the Ptt(Dit) and Dah hardware inputs to the very top of the ui_tick() function.? Right after making that call to check for CW activity, I then test the state of the ??in_tx ??variable.? If the? in_tx ?variable is True, then I exit out of the ui_tick() function right away so that I am not getting bogged down by processing anything else in that function.

3.????? Modify the ui_tick() function to test the state of the ?in_tx ?variable after each of the other various tasks that it does. ?If the? in_tx ?variable is True, then I exit out of the ui_tick() function right away so that I am not getting bogged down by processing anything else in that function.? (These additional exits were added because ui_tick() could already be doing something whenever the key down IRQ occurs. ?My multiple test and exit strategy gets me out of the ui_tick() function cleanly, as soon as it is practical to do so.

Result

My zBITX is now quite responsive to manual CW.? With a Straight Key and with Iambic Paddles.? It is Semi-QSK, not Full QSK.? I use a 750ms Semi-QSK Tail-Timer (this is the ‘Delay’, which is user adjustable from the zBITX GUI ‘Set’ selection). ?This tail timing suits my operating style.? Whenever the device is sending CW, the GUIs are no longer being updated. ?And the knob is disabled.

Giving up the GUI activity was the price I paid to make the device responsive to manual CW without doing a total software rewrite.? Because ?ui_tick() ?may already be executing some task that happens to be updating the screen during each initial Receive to Transmit switchover, you will occasionally see something ‘blip’ on the display during the Receive to Transmit switchovers.? Then the display will become static.

Implementing This Change

This is not an official zBITX release.? It is not in the zBITX github repository.? It is only here, in the sbitx_gtk.c file that I have attached to this posting.? ??If you are willing to rebuild your sbitx executable, which lives in your zBITX ??/home/pi/sbitx ??subdirectory, you can try out my solution:

1.????? Save the existing /home/pi/sbitx/sbitx_gtk.c file to a safe place.

2.????? Put my sbitx_gtk.c? file into your /home/pi/sbitx subdirectory.

3.????? From the command line. Execute ??./build??? ?Expect no errors.

4.????? Verify that you have a sbitx executable in your/home/pi/sbitx subdirectory.? It should be around 1.1 gb in size (I would occasionally get a zero length sbitx file).

5.????? Reboot your zBITX?? Plug in your Straight Key or Paddles.? You should now be good to go with manual CW.

Caveat

This is not an official zBITX release.? I am not setup for using SSB, AM, FT8, or any other mode with my zBITX.? ?While I do not see where my changes to the sbitx_gtk.c file to make manual CW usable will have any impact on these other modes, I have not tested any of them.

Going Forward

Perhaps Ashhar Farhan, zBITX developer, will consider my changes to allow manual CW at any speed to be usable on his zBITX, and perhaps will implement them (or some other solution) in a future official zBITX firmware release.? All of the changes that I made to my sbitx_gtk.c? file are identified with a ‘k3jzd’ comment.

?

I am considering this resolved.? It meets my needs.? Look for you on the air (on CW). ???

?

72,

Jody – K3JZD

sbitx_gtk.c

Re: zBitX not reporting on pskreporter.info

#118114

That bothers me too, but it's not the issue here.

?

Vy73 Juergen DF7II?

FS zBitx :: EU Only.

#118113

Hello!

?

Im selling my zBitx from the first batch and delivered last April.

Besides the transceiver and supplied accessories (OTG, usb cable, stylus) im also adding:
HDMI to mini HDMI cable?
BT keyboard
4 x 18650 batteries
USB charger for 2 18650 batteries

located in Portugal and able to ship within the EU (shipping cost to be added)

Asking 230€.

Re: SBit V4 Thoughts

#118112

I think offering various RPi models directly from HF Signals would be a bit of a challenge and isn't really necessary.

The sBitx is sold as a havkable radio, with limited support after sale, the owner takes responsibility for the hardware/software with minimal support from HF Signals. That's my characterization, the actual terms are listed on the HFSignals.com website.

To stock multiple RPi models to include in the radio would require a more involved order/build process, I think the more practical answer is to make the RPi board easily replaceable by a ham with a screwdriver - as it currently is.

If all you want is Firefox and to store some PDFs, there's no reason to increase the RAM in the sBitx as currently supplied, at least that's my opinion.

The higher-res display is the big improvement, the radio itself doesn't really need more RAM.

Ken, N2VIP

Show quoted text

On May 11, 2025, at 00:30, Gerald Sherman via groups.io <ve4gks@...> wrote:

One thing that might be worthwhile is to offer the V4 with different Pis, depending on the amount of RAM (2, 4, or 8 Gb) individual users desire.  I realize this option would have to be reflected in the pricing.  I would be willing to pay more for more memory.

Re: Mouse & Monitor Help

#118111

Mouse wheel doesn't work for me, sometimes a value will change by one when clicked on, not always

Re: Mouse & Monitor Help

#118110

So I checked it now - all conected to zBitz.

Adjusting for example the Audio level works somehow.

Pushing down the mousewheel makes reacting the settings, but it needs a hughe movement of the wheel to make a jump and that not smoothly. It is sometimes jumping in a more wild way. Not useable with my mouse. A Question of resolution settings of the mouse? At windows and linux and rpi4 it works correct.

Would be interesting to know what mouse other users are using and how it is connected to zBitx.

Through Bluetooth or USB or USB-Hub?

73 Michael

Re: Mouse & Monitor Help

#118109

Interesting, I have to try it too. At the moment I'm connected via VNC to a headless zBitx - at this configuration it doesn't work correct.

73, Michael

Re: Splatter-proof radio?

#118108

Isn't splatter in CW mode caused by the rapid rise of the signal on key down or rapid fall on key up?? How is that handled in the sBitx?? Is there a level control that uses a raised cosign or similar algorithm to prevent key clicks?

?

I must look into the code to see how CW is being processed.

?

73

Evan

AC9TU

Re: Splatter-proof radio?

#118107

Thanks, Evan!? ?I agree that if you overdrive the finals you can produce IMD.? ?But I think the modulation is by Fourier Transform and band-limited in the Fourier Transform, so unless the transform errs, how can it produce a time-domain output that has signal strength outside of the limited set bandpass of the creation?? ? ?

That intrigues me.? ? ?

Also, I watch a number of my friend's signals on the spectral display and I see a lot of wasted energy in their lower-frequency parts of their audio signal.? ? When it is the higher frequency parts that are so important for speech recognition -- especially with older ears with possibly lower response?in the mid and higher range.? ? Before this radio I didn't really ever see that stuff.? ??

Interesting.

Gordon KX4Z

Show quoted text

On Mon, May 12, 2025 at 7:00?AM Evan Hand via <elhandjr=[email protected]> wrote:

Gordon,

?

I believe that this radio can produce "splatter" if the level is set too high.? I have not looked into the code, so I do not know if any key shaping is done in CW mode, for example.? Overdriving the audio can also produce splatter.? The PA gain (NOT Mic gain) is post DSP filters, so splatter is still possible.?

?

Splatter is one reason the QRP-Labs QDX is not suitable for CW operation.? The QDX lacks key shaping, even though it is a single-frequency signal generated directly from the Si5351.

?

I could be wrong so that I will look for others' comments.

?

73

Evan

AC9TU

Re: Splatter-proof radio?

#118106

Gordon,

?

I believe that this radio can produce "splatter" if the level is set too high.? I have not looked into the code, so I do not know if any key shaping is done in CW mode, for example.? Overdriving the audio can also produce splatter.? The PA gain (NOT Mic gain) is post DSP filters, so splatter is still possible.?

?

Splatter is one reason the QRP-Labs QDX is not suitable for CW operation.? The QDX lacks key shaping, even though it is a single-frequency signal generated directly from the Si5351.

?

I could be wrong so that I will look for others' comments.

?

73

Evan

AC9TU

Re: zBitX not reporting on pskreporter.info

#118105

Is this thing bothering you? Doesn't it bother you that you can't make a QSO with the /p or /prefix callsign?

--

Gyula HA3HZ

Re: Splatter-proof radio?

#118104

There is some kind of control on the alsa mixer, because earlier, when the topic of discussion was why the output signal is smaller during digital transmission, I saw this. Call the alsa mixer and try to adjust the volume, the slider automatically jumps back to a level when transmitting. I didn't look any further for where this is set, but the regulation works.

--

Gyula HA3HZ

Re: zBitX not reporting on pskreporter.info

#118103

I only see who hears me. The display of the stations I can hear is missing.

Splatter-proof radio?

#118102

I have a question for all you extremely BRIGHT people out there.

?

Watching how the sBitx works, and thinking through how it mathematically computes the single side band modulation, based on a transform of the analog audio input, I am thinking that it is impossible to "splatter" outside the normal bandwidth.? ?That would be highly desirable, because it basically means you don't need ALC (automatic level control) and that you sorta have a "built in compression" if your audio gets louder.

?

Am I right in these thoughts?? ? I've studied the ALC circuitry by putting some debugging print statements in it, and I don't think I've ever seen it do much at all and this was part of what got me thinking that this radio may simply NOT produce splatter.

?

Your input welcomed,

?

73

Gordon Kx4Z

?

Re: SBit V4 Thoughts

#118101

Many thanks to all that contributed!!?

?

73'

?

Dick

Re: Mouse & Monitor Help

#118100

开云体育

With a similar setup....

My mouse has a center 'wheel' that allows me to 'scroll' the settings' up or down.

My mouse also has a 'click' input on the center wheel.

Works very well.

Don in Austin, N5CID

===========================================================

On 5/10/2025 3:32 PM, cam1aw via groups.io wrote:

Show quoted text

So I finally was able to chase down enough cables and a micro USB hub to hook up to a HDMI monitor, mouse and keyboard and got it all working along with connection to my laptop over wifi. My question is when I mouse click settings boxes with the remote mouse and monitor, the item high lights, but can't see to be able to change the value like say the drive setting. What am I missing?

_._,_._,_

Re: Mouse & Monitor Help

#118099

in reply to cam1aw... on the SbitX, once you click on a setting to highlight it, you use the small knob to adjust the value.? I suspect it's the same on the ZbitX.? Is that what you were looking for?

73 de W0ZF

Show quoted text

On Sat, May 10, 2025 at 3:32?PM cam1aw via <plcs=[email protected]> wrote:

So I finally was able to chase down enough cables and a micro USB hub to hook up to a HDMI monitor, mouse and keyboard and got it all working along with connection to my laptop over wifi. My question is when I mouse click settings boxes with the remote mouse and monitor, the item high lights, but can't see to be able to change the value like say the drive setting. What am I missing?

Re: zbitx, compiler optimizations #zbitx

#118098

Peter thanks for sharing interesting performance data!  

You may be right that delayTicks could be shortened.  There is a note in the i2cbb.c source code that says 

    delayTicks = 400;       // Delay value empirically chosen to be twice the value that just start to cause I2C NACKs - N3SB 

That note was in the sbitx code for machines with the RPI4 processor.  I don't know what it should be for the Raspberry Pi Zero 2W processor in the zbitx - maybe cut it down proportionate to processor speed difference?
-- 
Mike KB2ML

Show quoted text

On Fri, May  9, 2025 at 12:48 PM, Peter, DL4PIT wrote:


Just for test I changed delayTicks to 1 (so i2c_delay counts to 1 instead of
400), and this seems to remove the clogging of the ui_tick completely.

Anyone knowing the code much better than I do:
I guess the burning of CPU time in i2c_delay has some reason, can anyone
enlighten me why this is done?

Re: Z bitx poor sensitivity

#118097

开云体育

It’s not an 817?

the sensitivity will never be as good as an 817 put a signal generator on it.?

You’ll probably find a one micro volt sensitivity.

Sincerely, ?Frank Krozel t:?1-630-924-1600?frank@...

BSEE, ? AASEET, ? FCC Lic.# PG-18-19178, ? KG9H

Electronic Instrument Associates?- Central, Inc. ...since 1971

w:???d/l our linecard at???

Show quoted text

On May 11, 2025, at 6:45?AM, g0uvm via groups.io <g0uvm@...> wrote:

?
Hi all, I received my Zbitx a little over a week ago and have not had time to really try it out.

I have noticed it seems to be very insensitive to any signals.

When I switch over to my ft 817 signals are strong then immediately back to my zbitx they are either not heard or barely audible.

I have varied the if control between 0 and 100?

mans have gone the age options to no avail?

Any help would be greatly appreciated.

Thank you -Sean de g0uvm

--

?

de KG9H

Life member, ARRL

Re: zBitx power draw

#118096

A poor quality switching power supply may significantly impair your RX sensitivity. I have experienced that with both sBitx and zBitx. Finally for sBitx (and possibly other future equipment) I have bought a dedicated power supply for HF transceivers (namely it was: ).

I still had to run the DC cables through the ferrite ring to reduce the RX noise floor.

?

73,

Wojtek - SP5DAA