Topics

uBiTX #v6 - First steps after assembly #v6


Urs, dl1kdx
 

Hi guys,

I am new to the forum, received my uBiTX v6 yesterday, assembled the kit same night and encounter some issues. I would like to share with you my issues, kindly seeking for your expertise and advice how to fix them. Here are my setup, observations and issues:
1. Power supply is 13.8V out of a regulated 20A-capable device.
2. Current in Rx mode with running TFT display is approx. 350mA.
3. After a couple of minutes the 5V regulator mounted on the Raduino board gets boiling hot.
4. The TFT is "not touchable", neither pushed with the fingers nor the pen, it doesn't re-act.
5. The displayed menus doesn't re-act on spinning or pushing the rotary encoder.
6. The rotary encoder appears to work well, verified with my Digital Multimeter.
7. Current in Rx mode without TFT display is approx 150mA and the 5V regulator remains at moderate temperature.
8. I re-installed the latest version of Ashhar's master code to the Arduino Nano successfully, to verify that it works properly. It does and none of the above points change.
9. Finally I installed Reed N's modification of the master code with minor operational improvements:
10. After pushing the rotary encoder several times the uBiTX starts sending the VFO-A frequency in morse code through the speaker and volume control works well.
11. Occasionally the Split button illuminates.
12. However, I am still not able to browse through the menus or to change VFO-A frequency.
13. Powering up the uBiTX while pushing the rotary encoder, brings up the "Click on the cross" window, but still the screen is "untouchable" and doesn't accept input through the rotary encoder.

I never had the uBiTX switched on longer than 5 minutes, because of point 3. Enclosed some pics illustrating my observations.

Any kind of comments, ideas etc is highly appreciated.

73 de Urs, du1/dl1kdx


Reed N
 

1. Sounds fine
2. Sounds fine
3. Yep. If you search this group, you'll find lots of reports of hot 5V regulators. It's normal.
4. Did you calibrate the screen? I just explained how to do this earlier today: https://groups.io/g/BITX20/message/78865
5. When you first boot (in stock or my version), the knob adjusts the VFO A frequency - it doesn't select.
6. Sounds good
7. Sounds fine
8. Kinda expected
9. N/A
10. If you press and hold the knob in while in tuning mode, that activates the morse menu feature, which "plays" menus to you, so that you can navigate the radio blind. You can turn it back off by long pressing the knob again.
11. Might be a screen calibration thing. See 4.
12. Power radio off. Power radio on. Press knob button momentarily (briefly). At that point, you should see a small white border around VFOA that moves when you turn the encoder.
13. Did you click on the 4 crosses and then set the BFO and oscillator cals? If you didn't, good. I'd recommend against it, since you'll lose the factory BFO and oscillator cals, and have to re-cal from "scratch". If you did, not the end of the world, but also strange that the calibration wouldn't make your touch screen work.

It strikes me as odd that it sounds like the knob button only works as a long press for you. I would expect it either to fully work, or not work at all. I know some folks have accidentally plugged the encoder into the audio jack on the motherboard instead of the raduino jack, but if the button works somewhat, that sounds unlikely to be your issue. If you power the radio on, then after the screen is drawn, press the knob briefly (short press), do you not see the white border around VFO A? It's a thin border, so perhaps you just didn't see it?

Another thing from the earlier thread today that may be worth trying is loading the test branch: https://groups.io/g/BITX20/message/78792. It will let you test the inputs and outputs of the radio, by providing audio and on-screen feedback (see youtube link in that message for demonstration).


Reed


Urs, dl1kdx
 

Hi Reed,

I verified the input-output functionality using your sketch. Touchscreen and rotary encoder seem to work well, apart from a huge delay of tuning knob input: https://groups.io/g/BITX20/message/78885
Same time I can't follow your calibration instructions (https://groups.io/g/BITX20/message/78865): entering selection mode works, VFO-A field gets framed but the frame doesn't move when spinning the tuning knob.

vy 73 de Urs, du1/dl1kdx


Urs, dl1kdx
 

In addition: when I touch the screen now the SPL button highlights in a toggling manner with a few secs delay, regardless where I press. So it looks like the whole display is layout as SPL button initially.
I read the advice to very long press the tuning knob to enter setup menu, but it is not working for me.

My thoughts:
1. Is there an Arduino sketch available which starts in setup mode straight after re-boot?
2. What exactly is the purpose of the PDQ_GFX_UPDATE? It seems to aim on improving the performance of display and encoder?

Any idea how to proceed?


Reed N
 

Hi Urs,

On my software, if you hold the tuner button down when you power on, it goes into calibration mode, as I think you already encountered.

The PDQ GFX branch is outdated vs the master branch. All of that functionality is already merged. I left it there since there are some links directly to it. You're welcome to use it, and it should work, but it's less featured, and has some of the stock software quirks.

I'm rather perplexed by the knob issues you're having. I've never encountered latency on the buttons and knob encoder, and am unsure where or how that would happen. It's directly connected, and is interrupt-driven, so it should be one of the fastest parts. I wonder if there's something up with the nano itself. I don't suppose you have another nano to test with? I know you said you checked the encoder, but perhaps consider trying a different encoder as a sanity check too?


Reed


Urs, dl1kdx
 

Hi Reed,

many thanks again for your feedback.
I just watched your clip on YT (https://www.youtube.com/watch?v=Wbpqw5-YeXc) and wow(!) the feedback of the touches and encoder tweaks is amazing compared to the performance of my Nano while running the same sketch. Mine is like running in very slow motion closed to be frozen. I will check on my Nano, unfortunately I don't have a spare one.

73 de Urs


Reed N
 

The first thing I'd recommend checking on the nano would be the oscillator Y1 on pins 7/8 of the ATMega (if you have access to an oscilloscope?), since if that's not working right, resonating at 16MHz, it could make the uC do all sorts of wonky slow stuff.


Reed


Urs, dl1kdx
 

Hi Reed,

unfortunately I don't have an oscilloscope at hand right now. I ordered a few Nanos as replacement instead.

I wonder a bit about the numbers in the wide bar separated by comma in your YT clip: https://www.youtube.com/watch?v=Wbpqw5-YeXc
I bet these are the selected x-y-coordinates on the display? Mine are in the range 179-344 and 41-126 only.

73 de Urs


Reed N
 

Hi Urs,

Yeah, they are the touch input coordinates. They're dependant on the touch calibration, though, so if you can't do a good touch cal, the numbers will be off from what they should be.

With any luck, a new nano will magically make everything work as intended.


Reed


Urs, dl1kdx
 

In the course of troubleshooting the uBiTX v6, I measured the voltages on the plugbar connecting the raduino to the mainboard with a Digital-Multimeter.
Same time Reed's "Input-Output-Debug" sketch is running in the Arduino Nano.

Here are the results in Volts referring to the uBiTX v6 schematics:
Pins 1, 2: 5.0
Pins 3, 4: 13.5
Pin 17: 2.5
Pins 5..16, 18: 0.0

Pins 5, 8 & 11 count a frequency of 16 kHz, probably the real frequency is out of range of my multi-meters range.
Pin 17 changes frequency between 440 and 1300 Hz by pushing the button of the encoder like it is heard over the speaker.

Are these voltages as expected? I suspect that the delayed processing of the inputs made on the rotary are caused by issues on the mainboard.


Urs, dl1kdx
 

My problem isn't related to the mainboard. I powered up the Raduino board bypassing the mainboard, s. pic below.
Running the Raduino separately leads to the same huge delay of processing inputs from the encoder and the range of x-y-coordinates which can be addressed is very limited.


Reed N
 

Hi Urs,

5V on the key and PTT inputs makes sense, as they have internal pull ups.

2.5V on the tone pin makes sense, because that's half of 5V and it's a 50% duty cycle PWM producing the audio.

0V on the clock pins make sense, because the test build you're using doesn't turn on the frequency generator.

0V on the other control pins makes sense, as the test build doesn't try and control any of those.


Reed


Urs, dl1kdx
 

Hi Reed,

many thanks for your evaluation. So the diagnosis is narrowed down to a malfunctioning Nano. I am waiting for the replacement Nano now ..

73 de Urs


Dave Dixon
 

Hi Reed,
              How you bearing up with the virus,hope all families well over there,here been self isolated since middle of Feb when it was first announced having been brought over to our country.Gardens never looked so good and all those projects and jobs on radios and on my motorbike a honda vfr 800 fi in the garage have been sorted out,our youngest son just bought a Yamaha R 1 sounds awesome and looks the business looking forward to donning my leathers and getting back out on the roads over here at 71 im gonna be careful and steer clear of cities til ive got my riding legs back.I got rid of my V6 ubitx changed the screen ran kd8cec firmware but my 2 v3s seem better so im sticking with them.My 1983 transceiver produced by 2 uk amatuers in 1983 in Ham Radio Today as a kit called OMEGA a 9 band ssb/cw 10 watts is on the operating table having a variable 100 watt pa fitted plus updated transisters etc (photos) on my qrz page its been my main station until the v3 ubitx arrived,so now the omegas having a long earned rest until alls well..Been really nice over here in the uk for the past 3 weeks NO RAIN frosty nights but cracking warm sunny days.Stay well safe Reed and all of you that read this.Dave G0AYD.QRZ.COM 


On Sun, 17 May 2020 at 10:27, Reed N <greenkid336600+groupsio@...> wrote:
Hi Urs,

5V on the key and PTT inputs makes sense, as they have internal pull ups.

2.5V on the tone pin makes sense, because that's half of 5V and it's a 50% duty cycle PWM producing the audio.

0V on the clock pins make sense, because the test build you're using doesn't turn on the frequency generator.

0V on the other control pins makes sense, as the test build doesn't try and control any of those.


Reed


Urs, dl1kdx
 

I replaced the Arduino Nano today: I notice a different layout of the stock Nano vs. the substitute.

1. Reed's Input-Output-Debug sketch looks and feels now like in his YT clip, s. first screenshot below. The display is still following the encoder's inputs a bit slow, but compared to the stock Nano a remarkable improvement. The x-y-coordinates are now in the expected range.

2. Running reed's Master-Code sketch, s. second screenshot below. Long pressing the encoder leads to enable morse code representation of the VFO-A frequency and presents the white selection fame. Nevertheless, rotating the encoder doesn't move the frame to the adjacent button. Short pressing the encoder has no effect, same like pushing buttons on the display.

3. I suspect now issue with the Raduino board or wrong wiring of the encoder.

3. It's time to hit hay now. I will resume tomorrow.




Reed N
 

Hi Urs,

Great to hear that the nano replacement was a step in the right direction!

I thought you had said you tested the encoder. I guess I should have asked how you tested it.

The picture you posted of the test screen surprises me a bit. It looks like the right hand square is red, which means the code thinks that the screen is being long-pressed, but clearly you are not touching it. Does the right square ever turn black or white?


Reed


Urs, dl1kdx
 

Hi Reed,

I tested the encoder with my multi-meter in continuity mode. When I rotated the encoder I could hear short beeps on the different for clockwise and anti-clockwise turn directions, same with button pressing. Below photos of the encoder wiring.

The screen just reacts on long pressing the encoder, screen touches are recognized with your test SW and not in the master code sketch. Only the top left square changes color accompanied with change of low and high audio tone. Long press of encoder enable or disables audio presentation of the displayed frequency, which can't be changed.

So the main issue remains, I can't access the setup menu. This is valid for the original master code as well.

73 de Urs


Urs, dl1kdx
 

Summary of my testing like described above:

1. I am not able to enter setup menu, neither with the origin master code nor Reed's version of master code nor a replaced Nano.
2. While Nano is under master code: short pressing encoder button, rotating and touch screen are not recognized.
3. Nano with Input-Output-Debug sketch shows that encoder button pushes, rotating and touch screen are recognized.
4. Swapping out the Nano leads to a wider range of x-y-coordinates when I-O-Debug sketch is running.
5. Raduino board being not plugged into the main board behaves like described under points 1.-3.

My conclusion is that the Raduino board malfunctions, probably the Si5351A chip as the heart of the board fails.


Reed N
 

Hi Urs,

You could be right about the raduino board being bad, but I'm not totally sure. One more thing you could try as a sanity check would be to directly connect the nano's D6 and ground to the speaker (for feedback) and A0, A1, A2 to encoder's A, B, and button connectors (and also connect the encoder to ground). Running the input debug sketch with JUST the encoder and speaker should eliminate any issues with the raduino board, and with the speaker, you should be able to immediately hear stuff when you turn the knob. You could also try this with another encoder, if you happen to have one in your junk box.


Reed


Urs, dl1kdx
 

Hi Reed,

thanks for your great advice. I will figure it out and give update later.

To clarify the connectivity:
1. Speaker connects straight to D6 and GND.
2. Encoder's A & B connects to A0 & A1.
3. Encoder's button to A2.
4. Encoder to GND.

Could you please name the encoder's colors according to the pics above?

73 de Urs