Re: Icom microphone on the sBitx?
that's great!? ?I was doing experiments with the digital signal processing tonight and strangely, I got poor results from a homebrew mic that I think I had working very well with my DE -- when tried on the V2.....? mysteries never cease.
That is fantastic?that you have the icom mic working! I have the parts to try the same thing, will need to do that. THANKS! Gordon
On Tue, Jan 16, 2024 at 12:26?AM John Terrell, N6LN < N6LN@...> wrote: Gordon,
Thanks again for your wonderful relay mod for keying an external amplifier. By using the sBitx V2/V3 to drive my amplifier to almost a kilowatt I was finally able to get reliable signal reports on SSB tonight with the Icom handheld microphone. I am happy to report that the setup has excellent audio quality. Now when I operate remotely by QRP I¡¯ll know that the SSB doesn¡¯t need any further modification.
73,
Jack, N6LN
|
Re: Icom microphone on the sBitx?
Gordon,
Thanks again for your wonderful relay mod for keying an external amplifier. By using the sBitx V2/V3 to drive my amplifier to almost a kilowatt I was finally able to get reliable signal reports on SSB tonight with the Icom handheld microphone. I am happy to report that the setup has excellent audio quality. Now when I operate remotely by QRP I¡¯ll know that the SSB doesn¡¯t need any further modification.
73,
Jack, N6LN
|
Success ...
I just watched Ashhar's video showing how to use CW mode, where he shows using the keyboard to send a CW message.? Obviously, his didn't crash.
What was different?
Ashhar's desktop taskbar was "hidden".? So, I have just tried mine in CW mode with the taskbar hidden, and the "sBitx" program functions normally when pressing the [KBD OFF] button !!
Strange behavior, but it works none the less.? I will leave it to the "coders" to work out why, as I am no coder.
--
Pete VK3PYE
|
Re: Icom microphone on the sBitx?
// TBD: Something strange is going on, this should have been the otherway
if (r->mode == MODE_LSB || r->mode == MODE_CWR)
// zero out the LSB
for (i = 0; i < MAX_BINS/2; i++){
__real__ fft_out[i] = 0;
__imag__ fft_out[i] = 0;
}
else
{ // zero out the USB
for (i = MAX_BINS/2; i < MAX_BINS; i++){
__real__ fft_out[i] = 0;
__imag__ fft_out[i] = 0;
}
// GLG: double the eventual USB signal to make up for unknown losses
for (i = 0; i < MAX_BINS/2; i++){
__real__ fft_out[i] = __real__ fft_out[i] * 1.4;
__imag__ fft_out[i] = __imag__ fft_out[i] * 1.4;
}
}
//now rotate to the tx_bin
for (i = 0; i < MAX_BINS; i++){
int b = i + tx_shift;
if (b >= MAX_BINS)
b = b - MAX_BINS;
if (b < 0)
b = b + MAX_BINS;
r->fft_freq[b] = fft_out[i];
/////////////////
But which file is this?
Jack, N6LN
|
Re: Icom microphone on the sBitx?
Here is the code snippet where I changed the USB filtering:? ?The coefficient may well be different for every different radio.? ?I twiddled this on the real radio after starting on my computer.? ?I think I ended up with between 1.4 and 1.7 for the coefficient.? ?Here, for safety, I have put in 1.4? ?You will need to measure your own radio.? Remember these are VOLTAGES, power differences will be squared.
// TBD: Something strange is going on, this should have been the otherway
if (r->mode == MODE_LSB || r->mode == MODE_CWR)
// zero out the LSB
for (i = 0; i < MAX_BINS/2; i++){
__real__ fft_out[i] = 0;
__imag__ fft_out[i] = 0;
}
else
{ // zero out the USB
for (i = MAX_BINS/2; i < MAX_BINS; i++){
__real__ fft_out[i] = 0;
__imag__ fft_out[i] = 0;
}
// GLG: double the eventual USB signal to make up for unknown losses
for (i = 0; i < MAX_BINS/2; i++){
__real__ fft_out[i] = __real__ fft_out[i] * 1.4;
__imag__ fft_out[i] = __imag__ fft_out[i] * 1.4;
}
}
//now rotate to the tx_bin
for (i = 0; i < MAX_BINS; i++){
int b = i + tx_shift;
if (b >= MAX_BINS)
b = b - MAX_BINS;
if (b < 0)
b = b + MAX_BINS;
r->fft_freq[b] = fft_out[i];
On Mon, Jan 15, 2024 at 10:00?PM Gordon Gibby via <docvacuumtubes= [email protected]> wrote: Yup.? ?Big difference.? ?Congrats!? Remember we can adjust ANY microphone through simple math now in this radio.? ?I'm getting ready to do a science fair experiment with my precalculus students where we emphasize the highs and de-emphasize 200 Hz and below to see if it increases intelligibiliity.? ?Will putt out some code snippets soon on this presuming it works.? ?It worked perfectly to fix the power differences?between lower and upper sideband....
Gordon
On Mon, Jan 15, 2024 at 9:55?PM John Terrell, N6LN < N6LN@...> wrote: HUGE difference! Check it out for yourself with my audio files.So here's what both sound like. I transmitted my voice through each microphone and recorded the audio received on my Icom-7600 directly through the USB connection to the Voice Recorder accessory on my connected Windows 10 PC. So this is exactly what it sounds like on the receiving end.
Amazing.
73,
Jack, N6LN
|
Re: Icom microphone on the sBitx?
Yup.? ?Big difference.? ?Congrats!? Remember we can adjust ANY microphone through simple math now in this radio.? ?I'm getting ready to do a science fair experiment with my precalculus students where we emphasize the highs and de-emphasize 200 Hz and below to see if it increases intelligibiliity.? ?Will putt out some code snippets soon on this presuming it works.? ?It worked perfectly to fix the power differences?between lower and upper sideband....
Gordon
On Mon, Jan 15, 2024 at 9:55?PM John Terrell, N6LN < N6LN@...> wrote: HUGE difference! Check it out for yourself with my audio files.So here's what both sound like. I transmitted my voice through each microphone and recorded the audio received on my Icom-7600 directly through the USB connection to the Voice Recorder accessory on my connected Windows 10 PC. So this is exactly what it sounds like on the receiving end.
Amazing.
73,
Jack, N6LN
|
Re: Icom microphone on the sBitx?
HUGE difference! Check it out for yourself with my audio files.So here's what both sound like. I transmitted my voice through each microphone and recorded the audio received on my Icom-7600 directly through the USB connection to the Voice Recorder accessory on my connected Windows 10 PC. So this is exactly what it sounds like on the receiving end.
Amazing.
73,
Jack, N6LN
|
The init_waterfall is called just once at the beginning of the
program. it doesn't get called repeatedly at all. the rtty, psk31
modes are dud in v3, instead, you should use DIGITAL and switch to the
fldigit screen and use RTT/OLIVIA/et al from the fldigi screen. I will
remove the PSK31/RTTY on gtk from the next release.
- f
toggle quoted message
Show quoted text
On Tue, Jan 16, 2024 at 8:07?AM Peter VK3PYE <vk3pye@...> wrote:
On Tue, Jan 16, 2024 at 11:02 AM, Dave, N1AI wrote:
I am running the text-based UI (GTK) and it has the KBD button on the bottom right corner
Hi Dave, just to confirm, mine is crashing on the "sBitx" program as you show (above), and with the same modes -- 2TONE, DIGITAL, RTTY, PSK31, CWR and CW.
I have tried it using both the "mouse" and touch screen.
The "Web" (in browser) program seems to be OK, as it already has the keyboard running "down the page".
*** Thanks for everyone's input on this *** What a great little rig :-)
Pete VK3PYE
|
On Tue, Jan 16, 2024 at 11:02 AM, Dave, N1AI wrote:
I am running the text-based UI (GTK) and it has the KBD button on the bottom right corner
Hi Dave, just to confirm, mine is crashing on the "sBitx" program as you show (above), and with the same modes -- 2TONE, DIGITAL, RTTY, PSK31, CWR and CW. I have tried it using both the "mouse" and touch screen. The "Web" (in browser) program seems to be OK, as it already has the keyboard running "down the page". *** Thanks for everyone's input on this ***? ?What a great little rig? ?:-) Pete VK3PYE
|
barry,
Run the free -h command to see how much free memory you have. If I'm not mistaken the the OS only has 100MB of swap. You can increase it to 2GB by editing the /etc/dphys-swapfile setting. See this site for how-to.
On Mon, Jan 15, 2024 at 7:43?PM Dave, N1AI < n1ai@...> wrote: On Mon, Jan 15, 2024 at 08:02 PM, barry halterman wrote:
Dave, this PI4 ?only 2GB. That might explain the problem.
Yes, and debugging via this channel is better than nothing, but quite slow.
I think I'll have to dig out my V2 which still has a 2GB Pi in it and see what happens.
That won't happen tonight, maybe tomorrow, or maybe someone else will pick up the ball and run with it.
That reminds me, time to watch some football!
?
--
Regards,
Dave, N1AI
Jerry, AC9NM ÊÖÖеÄÄñÔÚ¹àľ´ÔÖÐÖµÁ½¸ö
--
Jerry Ponko, AC9NM
|
On Mon, Jan 15, 2024 at 08:02 PM, barry halterman wrote:
Dave, this PI4 ?only 2GB. That might explain the problem.
Yes, and debugging via this channel is better than nothing, but quite slow. I think I'll have to dig out my V2 which still has a 2GB Pi in it and see what happens. That won't happen tonight, maybe tomorrow, or maybe someone else will pick up the ball and run with it. That reminds me, time to watch some football! ? -- Regards, Dave, N1AI
|
Re: Icom microphone on the sBitx?
Thanks, john, I actually haven¡¯t used the built-in software very much yet. I do a great deal of Winlink and so I have to have an external sound interface in order to do that because there is very limited support for Linux in WINLINK.?
?It works just great once a bunch of things are figured out, and part of it was the fact that it really needs a high audio input, which I sure has confirmed incidentally and gave us a number for which I forgot. Additionally I had to adjust the digital filtering to handle the discrepancy between lower side band and upper side band which I think Ashhar has also confirmed. ? ? I simply multiplied the coefficient on the side that needed more power by about the ?square root of two and all was much better. ? ?At some point I¡¯m going to map it better and make a better inverse filter but for the moment that got me working, with full or close to full power on upper side band?
Getting ready to use this radio for a really neat science project with my pre-calculus students and I¡¯ll be writing more about it here soon, I¡¯m really amazed at what we have access to?
toggle quoted message
Show quoted text
On Jan 15, 2024, at 19:55, John Terrell, N6LN <N6LN@...> wrote:
?Gordon, I have a signalink that I use for FT8 with my FT-857D, and then I run WSJT-X on my laptop connected to the Signalink and the Yaesu, on nice days in the patio. Do you use the signalink with your sBitx? I run the WSJT-X program within the sBitx and with a larger monitor in the shack. So I don't need to use the Signalink. I use the 7-inch screen for convenience when I want to quickly change bands or modes and if I don't have the webUI window at the front. But I suppose if I go mobile with the sBitx I'll just use a laptop for the webUI.
Jack, N6LN
|
|
|
Re: Icom microphone on the sBitx?
Gordon, I have a signalink that I use for FT8 with my FT-857D, and then I run WSJT-X on my laptop connected to the Signalink and the Yaesu, on nice days in the patio. Do you use the signalink with your sBitx? I run the WSJT-X program within the sBitx and with a larger monitor in the shack. So I don't need to use the Signalink. I use the 7-inch screen for convenience when I want to quickly change bands or modes and if I don't have the webUI window at the front. But I suppose if I go mobile with the sBitx I'll just use a laptop for the webUI.
Jack, N6LN
|
PS: You can type:
q
at the (gdb) prompt to exit the debugger.
--
Regards,
Dave, N1AI
|
On Mon, Jan 15, 2024 at 07:24 PM, barry halterman wrote:
Dave, what I ended up with.
That makes it clear where the failure is, but not why.
?is the spot (in the current code, which may or may not be 100% the same as what is in the image):
This code looks like:
So the sbitx code is asking for memory, the request fails, so the sbitx application prints that message then exits.
The request fails either because either:
(a) the calculation for how much memory to ask for is in error (maybe f->height is absurdly big?), or
(b) the system really doesn't have enough memory to satisfy the request
?
Maybe mine is working because my sbitx v3 is using a Pi 4 with 4GB, I don't know.?
Or maybe some other code path is tying down a lot of memory when you happen to press that button, that would explain it working in some cases and not in others.
I'll try to poke around some more, but hopefully this clue gives others some ideas as well.
But thanks for providing that big clue!
--
Regards,
Dave, N1AI
|
Dave, I ran it again and it said ¡°No Stack¡±. Same info as the picture but I did not do the bt in gdb before. Barry ? Sent from for Windows ?
toggle quoted message
Show quoted text
From: Dave, N1AISent: Monday, January 15, 2024 7:03 PM To: [email protected]Subject: Re: [BITX20] KBD on v3 screen ? On Mon, Jan 15, 2024 at 05:26 PM, Peter VK3PYE wrote: I have just received my sBitx v3 complete kit last week, and same issue with the on-screen keyboard with mine.? No updates or additions to my software as yet, so it is "as received".?
I have a V3 board-only kit ( no touch screen ) and have booted the sdcard it came with.
Unfortunately I can't reproduce this bug using the mouse, and my V2 touch screen is not accessible right now.
I am running the text-based UI (GTK) and it has the KBD button on the bottom right corner.
I have tried turning KBD on and off when using CW, RTTY and DIGITAL modes using the mouse and have no crash or lockup.?
So either I am not doing the right thing to reproduce the problem, or the bug has to do with the interaction of the touch screen and the sbitx program.
If anyone is willing/able to run the debugger, maybe we can see where it is crashing.
The basic steps are to close the sbitx window, open a Terminal window, and do the following:
$ cd $HOME/sbitx
$ gdb ./sbitx
[ lots of output ]
(gdb) run
[ sbitx starts, then you put it into the foreground if it is not already, make it full screen if it is not already, then go into the right mode (CW, DIGITAL, etc) and hit the KBD button]
[ Terminal window should show a message from gdb about a signal (?!?) ]
(gdb) bt
[ Several lines of output, CAPTURE THIS via screen shot or camera, post it here ]
(gdb) quit
$?
?
--
Regards,
Dave, N1AI
|
On Mon, Jan 15, 2024 at 06:45 PM, Doc Sarvis wrote:
I was able to use another linux box to image the 64gb sd card. Thanx for all the help and your patience. It works. WooHoo.
Glad you are up and running!?? ? -- Regards, Dave, N1AI
|
Dave, what I ended up with. 
? Sent from for Windows ?
toggle quoted message
Show quoted text
From: Dave, N1AISent: Monday, January 15, 2024 7:03 PM To: [email protected]Subject: Re: [BITX20] KBD on v3 screen ? On Mon, Jan 15, 2024 at 05:26 PM, Peter VK3PYE wrote: I have just received my sBitx v3 complete kit last week, and same issue with the on-screen keyboard with mine.? No updates or additions to my software as yet, so it is "as received".?
I have a V3 board-only kit ( no touch screen ) and have booted the sdcard it came with.
Unfortunately I can't reproduce this bug using the mouse, and my V2 touch screen is not accessible right now.
I am running the text-based UI (GTK) and it has the KBD button on the bottom right corner.
I have tried turning KBD on and off when using CW, RTTY and DIGITAL modes using the mouse and have no crash or lockup.?
So either I am not doing the right thing to reproduce the problem, or the bug has to do with the interaction of the touch screen and the sbitx program.
If anyone is willing/able to run the debugger, maybe we can see where it is crashing.
The basic steps are to close the sbitx window, open a Terminal window, and do the following:
$ cd $HOME/sbitx
$ gdb ./sbitx
[ lots of output ]
(gdb) run
[ sbitx starts, then you put it into the foreground if it is not already, make it full screen if it is not already, then go into the right mode (CW, DIGITAL, etc) and hit the KBD button]
[ Terminal window should show a message from gdb about a signal (?!?) ]
(gdb) bt
[ Several lines of output, CAPTURE THIS via screen shot or camera, post it here ]
(gdb) quit
$?
?
--
Regards,
Dave, N1AI
|
On Mon, Jan 15, 2024 at 05:26 PM, Peter VK3PYE wrote:
I have just received my sBitx v3 complete kit last week, and same issue with the on-screen keyboard with mine.? No updates or additions to my software as yet, so it is "as received".?
I have a V3 board-only kit ( no touch screen ) and have booted the sdcard it came with. Unfortunately I can't reproduce this bug using the mouse, and my V2 touch screen is not accessible right now. I am running the text-based UI (GTK) and it has the KBD button on the bottom right corner. I have tried turning KBD on and off when using CW, RTTY and DIGITAL modes using the mouse and have no crash or lockup.? So either I am not doing the right thing to reproduce the problem, or the bug has to do with the interaction of the touch screen and the sbitx program. If anyone is willing/able to run the debugger, maybe we can see where it is crashing. The basic steps are to close the sbitx window, open a Terminal window, and do the following: $ cd $HOME/sbitx$ gdb ./sbitx[ lots of output ](gdb) run[ sbitx starts, then you put it into the foreground if it is not already, make it full screen if it is not already, then go into the right mode (CW, DIGITAL, etc) and hit the KBD button][ Terminal window should show a message from gdb about a signal (?!?) ](gdb) bt[ Several lines of output, CAPTURE THIS via screen shot or camera, post it here ](gdb) quit$?? -- Regards, Dave, N1AI
|
John Terrell, N6LN
Peter VK3PYE
JerryP
Dave, N1AI
