Date   
Re: V4 TX problems and new finals #ubitx-help

Don - KM4UDX
 

Evan, Mark, Raj, Anthony, & Co.

What is marveling about this lovely radio is its resilienance to stupidity.

What is marveling about this community is how unbelievably kind, helpful, generous and talented you people are.

How many times did I turn the bias pots the wrong way? Counter what? And the driver adjustment does what which way?  Why does the uBITX go dark and reset when my DVM probe slips and shorts the  IRF510 pins?  Which pin is the gate (again?)?  Why does my uBITX go all Chernobyl when I put 2Ov on the PA brown wire and fail to notice that I'm stuck in TX  WSPR digital mode 100% duty cycle for 10 minutes with way too much drive?

Why didn't I study electronics in school? Okay there were way more girls in English class, but still...it would be sort of useful around now...

I have to draw little cute diagrams of transistors and label them because I can never remember what lead is what. I feel like a first grader learning how to tie their shoes. 

If I set the bias current drain, then the voltage on the gate can't be zero.  Ah....so there is an error here...wait...is this that well meaning dope Don?  Good grief, he  can't get him out the door if you pushed him -- electrically. Hahah.

And I always feared that my problem (s?) are due to "operator error". I seem to have lots of that. Hahaha. 
 
uBITX  is just humility spelled backwards.  It is sooooo much better and smarter than I am.  But, that makes it really fruitful to study and poke and add to. 

I'll get back to Lab Class and report out on the "Rs" and the "TPs" and the counter clock wise what? Hahah.

Thanks a hundred million.

Don


Re: Let's Minimize the Library problems.

Ashhar Farhan
 

Tom,
I have moved on to using nothing except the standard arduino libraries. The current build is dependent only in the i2c and spi libraries. I am going bare metal on the touch library as well. I will oush the touch and display code into one file that you can rewrite for every display.
- f

On Sun 15 Dec, 2019, 10:14 AM Tom, wb6b, <wb6b@...> wrote:
Perhaps the biggest problem with STM32F103 Blue Pill development on the Arduino platform is there are two competing tool chains, the original "Paul Stoffregen" toolchain and the "official" STM32 tool chain endorsed by STM. I would prefer the official STM32 toolchain, but most Arduino Blue Pill projects seem to have been developed with the Paul Stoffregen version, so I use it. 

Another big issue I've found, for example with the graphics/touchscreen libraries, is several libraries such as the Adafruit graphic libraries need to work together with the lower level display module drivers for the TFT and touchscreen hardware. 

Library writers upgrade and enhance their libraries and no regression tests are (or can be) done on other libraries compatibilities, especially the non-mainstream libraries like those for the Blue Pill. And writers of the STM32 libraries may not be maintaining their libraries anymore, so they don't get updated. 

For what ever reason, the ESP32 toolchain developers have been doing a much better job in keeping their libraries updated to reflect changes in other libraries they are dependent on.

In an example of trying to get a touchscreen display working on a Blue Pill, AdaFruit updated one of the screen attributes from one parameters to two. (I've forgotten which one, but they changed the X and Y parameters [possibility a dot size value] to be separately specified, rather than the X and Y values being equal.) There was no backwards compatibility so the STM32 touchscreen libraries no longer worked. The change was done in a way that adding simple wrappers, still required culling through the AdaFruit code to correct numerous instances in the code. (There were still other issues with the hardware assumptions around interrupts and SPI ports, that clashed in the latest libraries, but may not have in older libraries when the STM32 toolchains were first introduced, years earlier)

Unlike the ESP32 toolchain, the STM32 toolchain(s) seemed to get off to a bad start, along with libraries breaking over time as other libraries are updated, generating confusion and trouble it so far has not recovered from. Then adding two toolchains to top the confusion off. 

For many simpler projects, these issues are less troublesome and the Arduino platform is still simple fun to use.

Finally, I don't bother with putting boot loaders on the Blue Pills. I either flash the program with the serial boot loader built into the chip, with a USB to serial adaptor, or use one of the inexpensive ST-Link adaptor clones from eBay. Although I might try the various boot loaders out at some future point. I can see for something like the uBitx installing a USB boot loader would be the best way to go for usability. 

Whichever toolchain you use, as Jack pointed out, let people know where the actual source location of library you used can be found.  As, many libraries have more than one version and authors.

Tom, wb6b

Re: V4 TX problems and new finals #ubitx-help

Evan Hand
 

Don,

The initial post states that you were able to follow the bias adjustment procedure correct?  And that you were able to adjust the bias resistors to get 100 ma increase for each IRF510.  That should pretty much eliminate them, the bias and connections through to the PA power supply.  

You have checked the other driver transistors and they seem to checkout as well.

At this point I would remove power from the rig
Look at R97 and R98 to verify that they are not open. 
Verify that TP7 is at ground to check the output coil on the output tranformere. 
Then without any power, verify that there is continuity from TP7 to TP8, followed by applying power to ONLY the main board (Red wire) and not the PA (Brown wire). 
Then select each band and verify connection from TP7 to TP8 with the key or PTT pressed. 

This is to verify that all of the relays are working, and that the inductors are not open.

That is the extent of what I can think of to test without an RF probe or oscilloscope.

There are RF probes that can be made with a few components.  Check out this web link:
http://n5ese.com/rfprobe1.htm

Let us know if the continuity checks play out.

73
Evan
AC9TU

Re: bitx v6 is here

Buddy Brannan
 

Oh wow…thanks Farhan! That’s pretty useful information to have. I think that putting keypad control together will be useful still, especially for direct frequency entry, but glad to hear that all functions can still be activated without the touch screen. With morse output, it really could be used eyes-free. Great news, and I’ll pass it on.

Vy 73,


Buddy Brannan, KB5ELV - Erie, PA
Email: buddy@...
Mobile: (814) 431-0962

On Dec 14, 2019, at 9:44 PM, Ashhar Farhan <farhanbox@...> wrote:

Buddy,
All the user interface is usable without needing touch. If you tap on the tuning knob for a second, you can move the focus from one button to the next and then click in the button to select again. Each button has text associated with it that can be read out. I have taken care to see to it that all features are accessible with and without touch. For example, the ability to enter a new frequency using the touch pad is replicated in fast tune where if you hold down the button while on one of the vfos , you can now quickly move across the bands in 50 khz steps.

Jack,
The function names are confusing as well..many variables are declared and not used any more. There is a large amount of code that is commented out as well. I have to clean that up, refactor reassemble. I will hopefully do this after Christmas.
The graphics library should really be called just a user interface library. The ILI9341 uses a SPI interface where it can fill a rectangle with a 16-bit colour. I used this to draw vertical lines, horizontal lines and pixels. No other graphical primitives have been used. The text characters are plotted pixel by pixel. The entire graphics routines are less than 100 lines. An often reused routine is the drawText routine that draws text centered inside a rectangle with a border. That makes all the buttons trivial. Almost all the user interface is just buttons, very much inspired by the jackal.
While tuning, i observed that only one digit if the frequency read out changes most of the time. So, the displayVFO routines figures which digits need to be repainted and updates only those. This routine is probably what makes the interface usable with the 8 bit power of the Arduino Nano.

- f

On Sun 15 Dec, 2019, 5:43 AM Buddy Brannan, <buddy@...> wrote:
Ooh…a thing that reads the screen in morse? That’s half the battle toward eyes-free operation. The other half, of course, is some additional way to control besides the built-in touch screen interface, like a programmable keypad. Seems to me there were a couple of implementations of this floating around. Do please elaborate.


Buddy Brannan, KB5ELV - Erie, PA
Email: buddy@...
Mobile: (814) 431-0962



On Dec 14, 2019, at 12:38 PM, Ashhar Farhan <farhanbox@...> wrote:

By morse reader, I actually meant software that reads out the screen in morse code.
I had fooled around with many morse decoders. Nothing really works on the bands. CW skimmer is the only thing that works under contest conditions.
So, if you want a morse code decoder that really works well, can adapt to any radio and needs no special hardware, then, spend a month, 20 minutes a day and practice at 30 wpm.
- f

On Sat 14 Dec, 2019, 9:56 PM splat1956, <dnrabin7@...> wrote:
Hi
This is very exciting. I have been thinking a long time but have not purchased one because I was afraid that the soldering and wiring instructions would be confusing. There have been many posts by confused builders.

This sounds great. Is there a morse code decoder built in or coming soon ? Your earlier post said " I have added a morse reader already. I shall finish integration to the user interface soon."
Please let us know when that is incorporated. I am unlikely to upgrade the software-firmware once it is working and that would be a very nice feature.

Great idea on the calibration with a known station signal for people without test equipment.

Thank you very much for this great upgrade !

David W9PH





Re: V4 TX problems and new finals #ubitx-help

 

Hmm,

A counterfeit could be 0V at its "gate" though. Did you ever track down the source of the new finals?

73


Mark

Re: Let's Minimize the Library problems.

Tom, wb6b
 

Perhaps the biggest problem with STM32F103 Blue Pill development on the Arduino platform is there are two competing tool chains, the original "Paul Stoffregen" toolchain and the "official" STM32 tool chain endorsed by STM. I would prefer the official STM32 toolchain, but most Arduino Blue Pill projects seem to have been developed with the Paul Stoffregen version, so I use it. 

Another big issue I've found, for example with the graphics/touchscreen libraries, is several libraries such as the Adafruit graphic libraries need to work together with the lower level display module drivers for the TFT and touchscreen hardware. 

Library writers upgrade and enhance their libraries and no regression tests are (or can be) done on other libraries compatibilities, especially the non-mainstream libraries like those for the Blue Pill. And writers of the STM32 libraries may not be maintaining their libraries anymore, so they don't get updated. 

For what ever reason, the ESP32 toolchain developers have been doing a much better job in keeping their libraries updated to reflect changes in other libraries they are dependent on.

In an example of trying to get a touchscreen display working on a Blue Pill, AdaFruit updated one of the screen attributes from one parameters to two. (I've forgotten which one, but they changed the X and Y parameters [possibility a dot size value] to be separately specified, rather than the X and Y values being equal.) There was no backwards compatibility so the STM32 touchscreen libraries no longer worked. The change was done in a way that adding simple wrappers, still required culling through the AdaFruit code to correct numerous instances in the code. (There were still other issues with the hardware assumptions around interrupts and SPI ports, that clashed in the latest libraries, but may not have in older libraries when the STM32 toolchains were first introduced, years earlier)

Unlike the ESP32 toolchain, the STM32 toolchain(s) seemed to get off to a bad start, along with libraries breaking over time as other libraries are updated, generating confusion and trouble it so far has not recovered from. Then adding two toolchains to top the confusion off. 

For many simpler projects, these issues are less troublesome and the Arduino platform is still simple fun to use.

Finally, I don't bother with putting boot loaders on the Blue Pills. I either flash the program with the serial boot loader built into the chip, with a USB to serial adaptor, or use one of the inexpensive ST-Link adaptor clones from eBay. Although I might try the various boot loaders out at some future point. I can see for something like the uBitx installing a USB boot loader would be the best way to go for usability. 

Whichever toolchain you use, as Jack pointed out, let people know where the actual source location of library you used can be found.  As, many libraries have more than one version and authors.

Tom, wb6b

Re: How to get rid of Local FM broadcast stations #ubitx

Gavin
 

I stripped all the paint from the connections to the chassis, and twisted the wires up a bit more. There is continuity between the coax ground and the other grounds. Any other ideas why its picking up these local stations?

Re: Let's Minimize the Library problems.

Ashhar Farhan
 

Cory, jack,
Great ideas. I am travelling and I will try implementing these.
One of my key constraints is to keep things working with what others find easy to use. I would switch from Arduino to the Blue pill in a second except that programming it from Arduino or Platform I/O isn't so simple. You have burn in the boot loader that needs you to hook it up with a serail to usb cable, then switch over to usb programming.
I just give this as an example. Platform io isn't as familiar to many as the horribly kuldgy Arduino IDE. I aleast wish they had implemented a goto line feature!

73, f


On Sun 15 Dec, 2019, 2:16 AM Cory King, <cory@...> wrote:
And sorry to double reply, “good ide support” I mean something other than Arduino IDE.  VSCode or JetBrains CLion.  Both have much better syntax highlighting, auto-completion, git integration and oh so much more.

Manual to set up uBitx Ver 6

wnpauls
 

Is there a manual, or pictures, or help to set up the new uBitx Ver 6? 

If so, where can I find it?

thanks,  Paul K0ZYV

Re: bitx v6 is here

Ashhar Farhan
 

Buddy,
All the user interface is usable without needing touch. If you tap on the tuning knob for a second, you can move the focus from one button to the next and then click in the button to select again. Each button has text associated with it that can be read out. I have taken care to see to it that all features are accessible with and without touch. For example, the ability to enter a new frequency using the touch pad is replicated in fast tune where if you hold down the button while on one of the vfos , you can now quickly move across the bands in 50 khz steps.

Jack,
The function names are confusing as well..many variables are declared and not used any more. There is a large amount of code that is commented out as well. I have to clean that up, refactor reassemble. I will hopefully do this after Christmas.
The graphics library should really be called just a user interface library. The ILI9341 uses a SPI interface where it can fill a rectangle with a 16-bit colour. I used this to draw vertical lines, horizontal lines and pixels. No other graphical primitives have been used. The text characters are plotted pixel by pixel. The entire graphics routines are less than 100 lines. An often reused routine is the drawText routine that draws text centered inside a rectangle with a border. That makes all the buttons trivial. Almost all the user interface is just buttons, very much inspired by the jackal.
While tuning, i observed that only one digit if the frequency read out changes most of the time. So, the displayVFO routines figures which digits need to be repainted and updates only those. This routine is probably what makes the interface usable with the 8 bit power of the Arduino Nano.

- f

On Sun 15 Dec, 2019, 5:43 AM Buddy Brannan, <buddy@...> wrote:
Ooh…a thing that reads the screen in morse? That’s half the battle toward eyes-free operation. The other half, of course, is some additional way to control besides the built-in touch screen interface, like a programmable keypad. Seems to me there were a couple of implementations of this floating around. Do please elaborate.


Buddy Brannan, KB5ELV - Erie, PA
Email: buddy@...
Mobile: (814) 431-0962



> On Dec 14, 2019, at 12:38 PM, Ashhar Farhan <farhanbox@...> wrote:
>
> By morse reader, I actually meant software that reads out the screen in morse code.
> I had fooled around with many morse decoders. Nothing really works on the bands. CW skimmer is the only thing that works under contest conditions.
> So, if you want a morse code decoder that really works well, can adapt to any radio and needs no special hardware, then, spend a month, 20 minutes a day and practice at 30 wpm.
> - f
>
> On Sat 14 Dec, 2019, 9:56 PM splat1956, <dnrabin7@...> wrote:
> Hi
> This is very exciting. I have been thinking a long time but have not purchased one because I was afraid that the soldering and wiring instructions would be confusing. There have been many posts by confused builders.
>
> This sounds great. Is there a morse code decoder built in or coming soon ? Your earlier post said " I have added a morse reader already. I shall finish integration to the user interface soon."
> Please let us know when that is incorporated. I am unlikely to upgrade the software-firmware once it is working and that would be a very nice feature.
>
> Great idea on the calibration with a known station signal for people without test equipment.
>
> Thank you very much for this great upgrade !
>
> David W9PH
>
>
>




Re: bitx v6 is here

Buddy Brannan
 

Ooh…a thing that reads the screen in morse? That’s half the battle toward eyes-free operation. The other half, of course, is some additional way to control besides the built-in touch screen interface, like a programmable keypad. Seems to me there were a couple of implementations of this floating around. Do please elaborate.


Buddy Brannan, KB5ELV - Erie, PA
Email: buddy@...
Mobile: (814) 431-0962

On Dec 14, 2019, at 12:38 PM, Ashhar Farhan <farhanbox@...> wrote:

By morse reader, I actually meant software that reads out the screen in morse code.
I had fooled around with many morse decoders. Nothing really works on the bands. CW skimmer is the only thing that works under contest conditions.
So, if you want a morse code decoder that really works well, can adapt to any radio and needs no special hardware, then, spend a month, 20 minutes a day and practice at 30 wpm.
- f

On Sat 14 Dec, 2019, 9:56 PM splat1956, <dnrabin7@...> wrote:
Hi
This is very exciting. I have been thinking a long time but have not purchased one because I was afraid that the soldering and wiring instructions would be confusing. There have been many posts by confused builders.

This sounds great. Is there a morse code decoder built in or coming soon ? Your earlier post said " I have added a morse reader already. I shall finish integration to the user interface soon."
Please let us know when that is incorporated. I am unlikely to upgrade the software-firmware once it is working and that would be a very nice feature.

Great idea on the calibration with a known station signal for people without test equipment.

Thank you very much for this great upgrade !

David W9PH


Re: Digital mode transmit via CAT but MIC-connector switches to Transmit directly - #cat #digital #v5 #ubitx-help - SOLVED #digital #v5 #ubitx-help #cat

Erwin - PE3ES - F4VTQ
 

Solved by inserting the small PTT switch into the orange Raduino line. Now I can switch to Tx by PTT via CAT with or without plug in the mic jack. And can activate Tx via the little PTT switch manually when a plug is in the mic jack.


--
PE3ES - F4VTQ - Erwin
73

Re: PTT with digital modes #bitx20 #cat #digital

Erwin - PE3ES - F4VTQ
 

Just had the same since yesterday. I installed the small PTT push button in the orange cable coming from Raduino board and going to the microphone key jack. Now th problem is over.

You can now do a CAT PTT with the CEC firmware (with or without a plug in the key jack) or a manual PTT when the plug is in the mic key jack.
I have used JS8CALL and WSJT-x to test PTT.
I have not transmitted so far.
--
PE3ES - F4VTQ - Erwin
73

Re: uBITX CW offset woes

Gordon Gibby
 

You are very gracious.

I think the real issue is that my other rigs know if you are in CW mode or in single side band mode, and so they can’t offset the beat frequency oscillator appropriately

The MicroBid X with Stock software does not know; if you push the mic button you are single side band if you push the key you are CW., and until you do that it doesn’t know. By merely adding a way for it to know which technique (SSB vs cw) you prefer this could be solved.

On Dec 14, 2019, at 09:34, Bill Cromwell <wrcromwell@...> wrote:

Hi Gordon,

Either way we setup the dial is right. Either way is wrong. If we use it the wrong way. As we have both said, we have to know how the radio in front of us works if we are going to have good results. That means we don't have any part of our emission outside the band AND our signal is where the other ham can hear it. For SSB we must present the right sideband, too.

Once upon a time I got a headache sorting it all out. At that time the headache medicines had a commercial showing Thor's hammer inside a human head striking an anvil complete with lightning strikes. If you remember those you will know it was at least a week ago:)

73,

Bill KU8H


On 12/13/19 8:35 PM, Gordon Gibby wrote:
Aargh. I may be wrong here! If I am, my apologies, it makes my head spin a bit. I’ll have to go fiddle with the radio and try it out.
--
bark less - wag more


PTT with digital modes #bitx20 #cat #digital

V Zecchinelli
 

Good Afternoon,

I don't know if this has already been discussed.  I found one message with same issues but no answers.  Could someone point me in the right direction?

When I plug 3.5mm cable into PTT/Mic my V5 uBitx immediately goes into transmit.  I tried stereo cable, mono cable and even just a bare plug with no cable attached.  CAT works fine and so does audio but I can't get past the radio going into TX.

Using Windows 10 on Acer PC,  WSJTX V2.12.  CEC V1.08.

Thank you,
Vince N1VIN

Re: Let's Minimize the Library problems.

Cory King
 

And sorry to double reply, “good ide support” I mean something other than Arduino IDE.  VSCode or JetBrains CLion.  Both have much better syntax highlighting, auto-completion, git integration and oh so much more.

Re: bitx v6 is here

 

Rick,

 We have one on order for testing and fitting of the AGC just to be sure it fits and works as it should. We will keep the group updated with our findings.

73
--
David

 N8DAH
Kit-Projects.com

Shop is open!

Re: Let's Minimize the Library problems.

Cory King
 

In my personal projects that use Arduino stuff (or in my case ESP32/8266) I have been using platform.io.  In addition to really good IDE integration, it has a way to express library dependencies.  


Its way isn’t as fancy as something you’d see with a “real” package manager like NPM but it is good enough that anybody pulling down your source code will know what libraries to install—in fact platform.io will do it for them.

If it was me, I’d switch all the code to use platform.io.  It could also help with porting to ESP and teensy as the system helps manage any compiler flags you’d need in order to add platform-specific code.

Re: Sidebands reversed after Memory Manager Used

Anthony Gomes
 

Bill, I had the same issue with SSB (LSB & USB getting inverted) with my scratch build. I don't like any software which does code changes magically, I like "Breaking the Magician's Code" maybe it just me. Don't get wrong the Memory Manager has it's own benefits, I would have taken a backup of the config before using the Memory Manager and doing any changes. I use the Memory manager to backup and restore my best tuned config and make another Arduino nano clone.

I think the best option will be to start with the CEC recommend factory reset and calibration process.

73
Anthony
VU3JVX 

Re: Frequency calibration V5

Cory King
 

“Replacement frequency source”.

So i picked up one of those NanoVNA’s on amazon and it can generate a constant signal on any given frequency.  I’m wondering how I can use that to set the BFO / calibration.

I have yet to find a concise YouTube video or guide on how one actually knows when the BFO and calibration are both set correctly.

my guess is for BFO I’m looking for the setting where the “tone” generated by VNA goes away— meaning the frequency on the dial is now sitting right on top of what would be the carrier.  But I could be completely wrong.


but seriously, the descriptions of “zero beat” I read online don’t click with me...   this stuff is all just math... surely somebody can describe it that way.