Topics

U3S WSPR on 30 m Error after transmission #u3s #30m


jakob@...
 

I ran a U3S for several weeks on WSPR in 30 meters, 0.6 watts into a magnetic loop antenna. Since two days, all of a sudden, the transmitter does not stop sending a carrier after the 161 elements. It will just leave the transmitter on with the same power. If I cycle the power everything is normal until element 161. It leaves the power on again after that. What could be the reason?


Shane Justice
 

Hi Jakob,

Is your U3S set up for a single band, or do you have the filter relay board in your kit?

If the former, how are you getting the RF from the lpf output to the back panel?

If the latter, what band is next in line to be output? Do you have it set to transmit after the calibration period (assuming you have the QLG-1)? How are you getting the RF from the filter relay board to the rear panel? 

Can you supply photos and more detail of your configuration?

73,
Shane
KE7TR


On Oct 13, 2020 at 16:28, jakob <jakob@...> wrote:

I ran a U3S for several weeks on WSPR in 30 meters, 0.6 watts into a magnetic loop antenna. Since two days, all of a sudden, the transmitter does not stop sending a carrier after the 161 elements. It will just leave the transmitter on with the same power. If I cycle the power everything is normal until element 161. It leaves the power on again after that. What could be the reason?


jakob@...
 

Thanks, Shane, for coming back on this issue. I made it 6 bands and ran it on all bands during testing. The problem is not the 30 meter LPF, since the system worked well until a few days ago, and even now with the problem, it emits 0.6 Watts. The problem is that it does not stop transmitting after it finished the count 161. I have not done any changes to the settings of the U3S lately, and these settings worked OK for many weeks. Is it possible that this effect (not stopping transmitting) can be related to the GPS board or the cable to it? I have re-arranged some components in my shack, including the antenna cable, but the GPS still blinks normally and transmits the time properly to the U3S.
73 de AA0JR, Jakob


Shane Justice
 

Hi Jacob,

The reason I'm asking about the configuration of your U3S, is that you may be getting RF into the PIC processor, and it is getting into some sort of locked-up state.

How you get the RF out of the U3S may be allowing some rf to travel on the power cable, either inside, or outside of the U3S.

Think about how you rearranged your shack - it may lead you to what is happening. 

You might try connecting a dummy load directly to the RF out of the U3S, and see if it operates correctly. If so, try connecting the dummy load through a length of coax to the U3S and repeat the experiment. If the dummy load on coax causes the lockup, then it is probably leakage from the coax getting into the power cable. If the U3S hanks up both ways, then check how you are getting RF off the filter relay board to the back panel- poor ground, too much exposed lead at either end, etc.

Those are the issues that  quickly come to mind.

73,
Shane





On Oct 14, 2020 at 16:58, jakob <jakob@...> wrote:

Thanks, Shane, for coming back on this issue. I made it 6 bands and ran it on all bands during testing. The problem is not the 30 meter LPF, since the system worked well until a few days ago, and even now with the problem, it emits 0.6 Watts. The problem is that it does not stop transmitting after it finished the count 161. I have not done any changes to the settings of the U3S lately, and these settings worked OK for many weeks. Is it possible that this effect (not stopping transmitting) can be related to the GPS board or the cable to it? I have re-arranged some components in my shack, including the antenna cable, but the GPS still blinks normally and transmits the time properly to the U3S.
73 de AA0JR, Jakob


geoff M0ORE
 

I think I have seen somewhere that you can reset the software to factory default. Just had a skim through the operating manual and assembly manual but can't find it so maybe in the FAQ. I am sure someone else will know. It is possible that you have inadvertent changed something without realising it. You will have to re-enter all your parameters again so do that in small steps, try just one band at a time.

Also, try a manual transmit using the right button to see if that causes a lockup.

On 15/10/2020 04:22, Shane Justice wrote:
Hi Jacob,

The reason I'm asking about the configuration of your U3S, is that you may be getting RF into the PIC processor, and it is getting into some sort of locked-up state.

How you get the RF out of the U3S may be allowing some rf to travel on the power cable, either inside, or outside of the U3S.

Think about how you rearranged your shack - it may lead you to what is happening. 

You might try connecting a dummy load directly to the RF out of the U3S, and see if it operates correctly. If so, try connecting the dummy load through a length of coax to the U3S and repeat the experiment. If the dummy load on coax causes the lockup, then it is probably leakage from the coax getting into the power cable. If the U3S hanks up both ways, then check how you are getting RF off the filter relay board to the back panel- poor ground, too much exposed lead at either end, etc.

Those are the issues that  quickly come to mind.

73,
Shane





On Oct 14, 2020 at 16:58, jakob <jakob@...> wrote:

Thanks, Shane, for coming back on this issue. I made it 6 bands and ran it on all bands during testing. The problem is not the 30 meter LPF, since the system worked well until a few days ago, and even now with the problem, it emits 0.6 Watts. The problem is that it does not stop transmitting after it finished the count 161. I have not done any changes to the settings of the U3S lately, and these settings worked OK for many weeks. Is it possible that this effect (not stopping transmitting) can be related to the GPS board or the cable to it? I have re-arranged some components in my shack, including the antenna cable, but the GPS still blinks normally and transmits the time properly to the U3S.
73 de AA0JR, Jakob


jakob@...
 

Hi Shane, leaking RF is a very good point. I did not think of that. I have a bunch of RF chokes on order with Amazon and should get them in before the weekend.I will experiment with them and keep you posted.
The cabling inside the box is coax. But I did not use the 4 screws to secure the upper half of the box. Maybe this has to do with the problem.
73 de AA0JR


Shane Justice
 

Hi Jacob,

A test with the dummy load may help you sort this out. 

Geoff raises a good point, as well. I just received a new Wattmeter last night, so I can check my U3S for power level and get the bias set correctly now, so perhaps I'll know more quantitatively what it is doing. I previously set bias by watching current draw on my lab power supply. I'm waiting for a couple of components to finish a QCX+ build for 40m, which should be here in the next few days, then I have to get antennas up! So far all experiments are with a dummy load...

73,
Shane
KE7TR


On Oct 15, 2020 at 08:48, jakob <jakob@...> wrote:

Hi Shane, leaking RF is a very good point. I did not think of that. I have a bunch of RF chokes on order with Amazon and should get them in before the weekend.I will experiment with them and keep you posted.
The cabling inside the box is coax. But I did not use the 4 screws to secure the upper half of the box. Maybe this has to do with the problem.
73 de AA0JR


jakob@...
 

I have further tested the U3S. The problem occurs on all 6 bands, with antenna or dummy load. I have re-arranged external cables too. No difference. It is now very clear to me that the program is corrupted and needs to be re-flashed. Maybe I even need to replace the MCU chip ATmega328P. I will study how I can reflash that thing. I probably buy some replacement chips, since they come for less than $2.00.
Can anybody point me to the best instructions on how to re-flash? Is in-circuit reflashing possible, or do I need additional circuitry? I have Windows 10 PCs.
Thanks for all the help.
73 de AA0JR Jakob


Alan G4ZFQ
 

Can anybody point me to the best instructions on how to re-flash? Is in-circuit reflashing possible, or do I need additional circuitry? I have Windows 10 PCs.
Jakob,

This is my way https://sites.google.com/site/g4zfqradio/qrplabs_program_chip_with_USBasp
Look on the QRP Labs site and in this groups Files. There are lots of ways, lots of pages, there must be something to suit you.

73 Alan G4ZFQ


Shane Justice
 

Hi Jakob,

Before you go to the trouble ofreflashing the PIC, let's examine how you get the RF from the filter relay board to the BNC on the back panel- how are you attaching coax to the RF out part of the relay board? How does the coax attach to the BNC?

What type of coax are you using?

Have you verified the configuration of the jumpers on filter 0 on the transmitter board, and the diodes and jumpers on filter relays 1 through 5? Do you have the highest frequency filter in position 5, as noted in the instructions?

These checks are worth your time to make sure you have all the RF ducks in a row before looking at the PIC as a corrupted device.

I hope this is helpful for you.

Best Regards,
Shane




On Oct 16, 2020 at 08:19, jakob <jakob@...> wrote:

I have further tested the U3S. The problem occurs on all 6 bands, with antenna or dummy load. I have re-arranged external cables too. No difference. It is now very clear to me that the program is corrupted and needs to be re-flashed. Maybe I even need to replace the MCU chip ATmega328P. I will study how I can reflash that thing. I probably buy some replacement chips, since they come for less than $2.00.
Can anybody point me to the best instructions on how to re-flash? Is in-circuit reflashing possible, or do I need additional circuitry? I have Windows 10 PCs.
Thanks for all the help.
73 de AA0JR Jakob


Arv Evans
 

Which PIC device are you talking about?
_._


On Fri, Oct 16, 2020 at 1:07 PM Shane Justice <justshane@...> wrote:
Hi Jakob,

Before you go to the trouble ofreflashing the PIC, let's examine how you get the RF from the filter relay board to the BNC on the back panel- how are you attaching coax to the RF out part of the relay board? How does the coax attach to the BNC?

What type of coax are you using?

Have you verified the configuration of the jumpers on filter 0 on the transmitter board, and the diodes and jumpers on filter relays 1 through 5? Do you have the highest frequency filter in position 5, as noted in the instructions?

These checks are worth your time to make sure you have all the RF ducks in a row before looking at the PIC as a corrupted device.

I hope this is helpful for you.

Best Regards,
Shane




On Oct 16, 2020 at 08:19, jakob <jakob@...> wrote:

I have further tested the U3S. The problem occurs on all 6 bands, with antenna or dummy load. I have re-arranged external cables too. No difference. It is now very clear to me that the program is corrupted and needs to be re-flashed. Maybe I even need to replace the MCU chip ATmega328P. I will study how I can reflash that thing. I probably buy some replacement chips, since they come for less than $2.00.
Can anybody point me to the best instructions on how to re-flash? Is in-circuit reflashing possible, or do I need additional circuitry? I have Windows 10 PCs.
Thanks for all the help.
73 de AA0JR Jakob


Shane Justice
 

Hi Arv,
The PIC is the microcontroller on the U3S, unless I'm sadly misinformed...

Shane


On Oct 16, 2020 at 12:28, Arv Evans <arvid.evans@...> wrote:

Which PIC device are you talking about?
_._


On Fri, Oct 16, 2020 at 1:07 PM Shane Justice <justshane@...> wrote:
Hi Jakob,

Before you go to the trouble ofreflashing the PIC, let's examine how you get the RF from the filter relay board to the BNC on the back panel- how are you attaching coax to the RF out part of the relay board? How does the coax attach to the BNC?

What type of coax are you using?

Have you verified the configuration of the jumpers on filter 0 on the transmitter board, and the diodes and jumpers on filter relays 1 through 5? Do you have the highest frequency filter in position 5, as noted in the instructions?

These checks are worth your time to make sure you have all the RF ducks in a row before looking at the PIC as a corrupted device.

I hope this is helpful for you.

Best Regards,
Shane




On Oct 16, 2020 at 08:19, jakob <jakob@...> wrote:

I have further tested the U3S. The problem occurs on all 6 bands, with antenna or dummy load. I have re-arranged external cables too. No difference. It is now very clear to me that the program is corrupted and needs to be re-flashed. Maybe I even need to replace the MCU chip ATmega328P. I will study how I can reflash that thing. I probably buy some replacement chips, since they come for less than $2.00.
Can anybody point me to the best instructions on how to re-flash? Is in-circuit reflashing possible, or do I need additional circuitry? I have Windows 10 PCs.
Thanks for all the help.
73 de AA0JR Jakob


Shane Justice
 

Arv,
I just looked- I was sadly misinformed. It's an Atmega328P!

Sorry about that, chief!


On Oct 16, 2020 at 12:33, Shane Justice via groups.io <justshane@...> wrote:

Hi Arv,
The PIC is the microcontroller on the U3S, unless I'm sadly misinformed...

Shane


On Oct 16, 2020 at 12:28, Arv Evans <arvid.evans@...> wrote:

Which PIC device are you talking about?
_._


On Fri, Oct 16, 2020 at 1:07 PM Shane Justice <justshane@...> wrote:
Hi Jakob,

Before you go to the trouble ofreflashing the PIC, let's examine how you get the RF from the filter relay board to the BNC on the back panel- how are you attaching coax to the RF out part of the relay board? How does the coax attach to the BNC?

What type of coax are you using?

Have you verified the configuration of the jumpers on filter 0 on the transmitter board, and the diodes and jumpers on filter relays 1 through 5? Do you have the highest frequency filter in position 5, as noted in the instructions?

These checks are worth your time to make sure you have all the RF ducks in a row before looking at the PIC as a corrupted device.

I hope this is helpful for you.

Best Regards,
Shane




On Oct 16, 2020 at 08:19, jakob <jakob@...> wrote:

I have further tested the U3S. The problem occurs on all 6 bands, with antenna or dummy load. I have re-arranged external cables too. No difference. It is now very clear to me that the program is corrupted and needs to be re-flashed. Maybe I even need to replace the MCU chip ATmega328P. I will study how I can reflash that thing. I probably buy some replacement chips, since they come for less than $2.00.
Can anybody point me to the best instructions on how to re-flash? Is in-circuit reflashing possible, or do I need additional circuitry? I have Windows 10 PCs.
Thanks for all the help.
73 de AA0JR Jakob


Arv Evans
 

Shane

I don't think Hans has ever used a PIC controller in any of his designs.

The ATmega328 AVR microcontroller (IC1) is pre-programmed with firmware to control the LCD, buttons, and GPS interface. The Si5351A Synthesiser module kit includes its own on-board 27MHz crystal reference oscillator. On/off keying is handled in software by switching on/off the output stage of the Si5351A synthesiser IC. The Q1 BS170 MOSFET is the power amplifier, producing about 250mW from a 5V supply on 30m (power output varies on different bands). Finally standard 7-element low pass filter plugin modules remove harmonics of the transmission frequency. 

Not sure why Hans has avoided the PIC (you may have to ask him) but 
it may be because we started using AVR's way back when Methuselah was 
just a child!   ;-)

Arv
_._


On Fri, Oct 16, 2020 at 1:33 PM Shane Justice <justshane@...> wrote:
Hi Arv,
The PIC is the microcontroller on the U3S, unless I'm sadly misinformed...

Shane


On Oct 16, 2020 at 12:28, Arv Evans <arvid.evans@...> wrote:

Which PIC device are you talking about?
_._


On Fri, Oct 16, 2020 at 1:07 PM Shane Justice <justshane@...> wrote:
Hi Jakob,

Before you go to the trouble ofreflashing the PIC, let's examine how you get the RF from the filter relay board to the BNC on the back panel- how are you attaching coax to the RF out part of the relay board? How does the coax attach to the BNC?

What type of coax are you using?

Have you verified the configuration of the jumpers on filter 0 on the transmitter board, and the diodes and jumpers on filter relays 1 through 5? Do you have the highest frequency filter in position 5, as noted in the instructions?

These checks are worth your time to make sure you have all the RF ducks in a row before looking at the PIC as a corrupted device.

I hope this is helpful for you.

Best Regards,
Shane




On Oct 16, 2020 at 08:19, jakob <jakob@...> wrote:

I have further tested the U3S. The problem occurs on all 6 bands, with antenna or dummy load. I have re-arranged external cables too. No difference. It is now very clear to me that the program is corrupted and needs to be re-flashed. Maybe I even need to replace the MCU chip ATmega328P. I will study how I can reflash that thing. I probably buy some replacement chips, since they come for less than $2.00.
Can anybody point me to the best instructions on how to re-flash? Is in-circuit reflashing possible, or do I need additional circuitry? I have Windows 10 PCs.
Thanks for all the help.
73 de AA0JR Jakob


Arv Evans
 

Shane

No problem.  We all get confused at times.  Wait until you are my 
age (79).  It just gets worse as one gets older!

My intent was just to avoid confusing others who might think the 
U3S was PIC based.  Trying to fit a PIC into the U3S schematic 
could be a bit confusing.

Arv
_._


On Fri, Oct 16, 2020 at 1:37 PM Shane Justice <justshane@...> wrote:
Arv,
I just looked- I was sadly misinformed. It's an Atmega328P!

Sorry about that, chief!


On Oct 16, 2020 at 12:33, Shane Justice via groups.io <justshane@...> wrote:

Hi Arv,
The PIC is the microcontroller on the U3S, unless I'm sadly misinformed...

Shane


On Oct 16, 2020 at 12:28, Arv Evans <arvid.evans@...> wrote:

Which PIC device are you talking about?
_._


On Fri, Oct 16, 2020 at 1:07 PM Shane Justice <justshane@...> wrote:
Hi Jakob,

Before you go to the trouble ofreflashing the PIC, let's examine how you get the RF from the filter relay board to the BNC on the back panel- how are you attaching coax to the RF out part of the relay board? How does the coax attach to the BNC?

What type of coax are you using?

Have you verified the configuration of the jumpers on filter 0 on the transmitter board, and the diodes and jumpers on filter relays 1 through 5? Do you have the highest frequency filter in position 5, as noted in the instructions?

These checks are worth your time to make sure you have all the RF ducks in a row before looking at the PIC as a corrupted device.

I hope this is helpful for you.

Best Regards,
Shane




On Oct 16, 2020 at 08:19, jakob <jakob@...> wrote:

I have further tested the U3S. The problem occurs on all 6 bands, with antenna or dummy load. I have re-arranged external cables too. No difference. It is now very clear to me that the program is corrupted and needs to be re-flashed. Maybe I even need to replace the MCU chip ATmega328P. I will study how I can reflash that thing. I probably buy some replacement chips, since they come for less than $2.00.
Can anybody point me to the best instructions on how to re-flash? Is in-circuit reflashing possible, or do I need additional circuitry? I have Windows 10 PCs.
Thanks for all the help.
73 de AA0JR Jakob


Shane Justice
 

Yep- brain fart on my part!


On Oct 16, 2020 at 13:05, Arv Evans <arvid.evans@...> wrote:

Shane

I don't think Hans has ever used a PIC controller in any of his designs.

The ATmega328 AVR microcontroller (IC1) is pre-programmed with firmware to control the LCD, buttons, and GPS interface. The Si5351A Synthesiser module kit includes its own on-board 27MHz crystal reference oscillator. On/off keying is handled in software by switching on/off the output stage of the Si5351A synthesiser IC. The Q1 BS170 MOSFET is the power amplifier, producing about 250mW from a 5V supply on 30m (power output varies on different bands). Finally standard 7-element low pass filter plugin modules remove harmonics of the transmission frequency. 

Not sure why Hans has avoided the PIC (you may have to ask him) but 
it may be because we started using AVR's way back when Methuselah was 
just a child!   ;-)

Arv
_._


On Fri, Oct 16, 2020 at 1:33 PM Shane Justice <justshane@...> wrote:
Hi Arv,
The PIC is the microcontroller on the U3S, unless I'm sadly misinformed...

Shane


On Oct 16, 2020 at 12:28, Arv Evans <arvid.evans@...> wrote:

Which PIC device are you talking about?
_._


On Fri, Oct 16, 2020 at 1:07 PM Shane Justice <justshane@...> wrote:
Hi Jakob,

Before you go to the trouble ofreflashing the PIC, let's examine how you get the RF from the filter relay board to the BNC on the back panel- how are you attaching coax to the RF out part of the relay board? How does the coax attach to the BNC?

What type of coax are you using?

Have you verified the configuration of the jumpers on filter 0 on the transmitter board, and the diodes and jumpers on filter relays 1 through 5? Do you have the highest frequency filter in position 5, as noted in the instructions?

These checks are worth your time to make sure you have all the RF ducks in a row before looking at the PIC as a corrupted device.

I hope this is helpful for you.

Best Regards,
Shane




On Oct 16, 2020 at 08:19, jakob <jakob@...> wrote:

I have further tested the U3S. The problem occurs on all 6 bands, with antenna or dummy load. I have re-arranged external cables too. No difference. It is now very clear to me that the program is corrupted and needs to be re-flashed. Maybe I even need to replace the MCU chip ATmega328P. I will study how I can reflash that thing. I probably buy some replacement chips, since they come for less than $2.00.
Can anybody point me to the best instructions on how to re-flash? Is in-circuit reflashing possible, or do I need additional circuitry? I have Windows 10 PCs.
Thanks for all the help.
73 de AA0JR Jakob


geoff M0ORE
 

A few days ago, I suggested that you re-set the U3S to default settings as you may have accidentally entered a rogue setting and not noticed it. Very easy to do with the method of programming. You should only need to re-flash the CPU for a change of software.

I did look for the method of resetting the CPU ( this is not the same as re-flashing the CPU ) and looked through the user manual and assembly manuals but could find no reference to how to do it. The U3S  faulting FAQ says it is in the user manual but I can't find it. The FAQ also says that a noisy PSU can cause spikes which may upset the processor but you will have checked that and be using a good PSU won't you.

Bearing in mind that the shack clock and the VFO/Sig gen also use the same PCB and processor, you could try using the same method as these for a factory reset.

This is very easy and much easier than re-flashing the CPU.

When powering on the U3S, you see a splash screen showing you what unit it is, a software version number and the QRPLabs logo. Whilst this screen is showing, press the right button. If the factory reset option exists, you will see the message " Reset? Sure? Press left btn".

Please let us know what happens. My U3S is undergoing mods at the moment so I can't try it here and I suspect that others with working units won't want to have to go through a complete programming procedure on a working unit.

Be brave, try a factory reset. You can't kill a CPU by pressing any buttons.


On 16/10/2020 16:18, jakob@... wrote:
I have further tested the U3S. The problem occurs on all 6 bands, with antenna or dummy load. I have re-arranged external cables too. No difference. It is now very clear to me that the program is corrupted and needs to be re-flashed. Maybe I even need to replace the MCU chip ATmega328P. I will study how I can reflash that thing. I probably buy some replacement chips, since they come for less than $2.00.
Can anybody point me to the best instructions on how to re-flash? Is in-circuit reflashing possible, or do I need additional circuitry? I have Windows 10 PCs.
Thanks for all the help.
73 de AA0JR Jakob


jakob@...
 

I spent a few hours yesterday trying to understand the problem better. I have no new information, other than 

1. Since about a week, after the transmitter worked properly on the 30-meter band for several months, all of a sudden and without accessing the system, the WSPR cycle stopped after the element 161 (the last element) and the transmitter did not shut down. The system is hanging and only a power cycle brings it back to life. 
2. This last transmission was properly propagated and the usual receivers reported it.
3. Power reset and switching the band to any of the other 5 frequencies did not show any difference, so the problem is easily reproducible on all frequencies (80-40-30-20-15-10 Meters) with antenna or dummy load.
4. The U3S unit seems not to have a 'factory reset' feature. I could not find it in the documentation. 

I conclude it is a program execution failure after the program code got corrupted. I will buy a few ATmega328P chips and a programmer to flash a new chip.
73 de AA0JR Jakob


geoff M0ORE
 

Hi Jakob,

My U3S has software version 3.12a which  believe is the current version. You can confirm this by the splash screen which appears when you switch on. If you have an earlier version, then it may be worthwhile programming the current version.

When the splash screen appears, press the right button. A message appears " Reset...? press left btn." Press the left button . The unit will now revert to factory settings.

I have just confirmed this procedure on my U3S so I know it exists.

I spent a few hours yesterday trying to understand the problem better. I have no new information, other than 

1. Since about a week, after the transmitter worked properly on the 30-meter band for several months, all of a sudden and without accessing the system, the WSPR cycle stopped after the element 161 (the last element) and the transmitter did not shut down. The system is hanging and only a power cycle brings it back to life. 
2. This last transmission was properly propagated and the usual receivers reported it.
3. Power reset and switching the band to any of the other 5 frequencies did not show any difference, so the problem is easily reproducible on all frequencies (80-40-30-20-15-10 Meters) with antenna or dummy load.
4. The U3S unit seems not to have a 'factory reset' feature. I could not find it in the documentation. 

I conclude it is a program execution failure after the program code got corrupted. I will buy a few ATmega328P chips and a programmer to flash a new chip.
73 de AA0JR Jakob


Alan de G1FXB
 

RE:- U3S reset.
assuming you are using the latest firmware and working buttons.
According to Hans own great book of words.....

" On powering up the kit, a message is shown for a few seconds on the LCD,
which indicates the version number.
If the Right button is pressed during the display of this splash screen message,
then a confirmation message appears saying “Reset? Sure? Press left btn”.
Now if you press the left button,
the chip EEPROM contents are erased and returned to the original factory settings.

The effect of this feature is to entirely return the microcontroller chip to its original settings,
it is the same as buying a new chip and inserting it.
All calibration settings, message, frequency, etc. are erased
and returned to the default settings.
"

Alan

On 17/10/2020 22:47, jakob@... wrote:

I spent a few hours yesterday trying to understand the problem better. I have no new information, other than 

1. Since about a week, after the transmitter worked properly on the 30-meter band for several months, all of a sudden and without accessing the system, the WSPR cycle stopped after the element 161 (the last element) and the transmitter did not shut down. The system is hanging and only a power cycle brings it back to life. 
2. This last transmission was properly propagated and the usual receivers reported it.
3. Power reset and switching the band to any of the other 5 frequencies did not show any difference, so the problem is easily reproducible on all frequencies (80-40-30-20-15-10 Meters) with antenna or dummy load.
4. The U3S unit seems not to have a 'factory reset' feature. I could not find it in the documentation. 

I conclude it is a program execution failure after the program code got corrupted. I will buy a few ATmega328P chips and a programmer to flash a new chip.
73 de AA0JR Jakob