ATS-MINI FEATURE REQUESTS

[G8PTN]

Please see attached a log showing bugs and feature requests for the ATS-Mini Firmware.

Please use this section for any future requests.

73, Dave (G8PTN)

ATS_MINI_LOG_1.pdf

[pablowoiz]

thanks for your excellent job!
The calibration is limited to 2khz +- . My Radio is unfortunately almost 3khz away from the real frequency....
Would be posiible to extend it to 4 or 5 Khz up and down?

73 de LU1AGH

[G8PTN]

Hi,
It is not practical extending the calibration offset further, since it already compromises the “chuff” free tuning range.

The VFO/BFO tuning mechanism has a maximum span of 32kHz (+/- 16kHz).
With the calibration range set limited to +/- 2kHz, this reduces the chuff free span to 28kHz (+/- 14kHz).

I am afraid if there is an excessive offset due to the reference oscillator, you will need to tune the dial frequency to compensate for this.

73, Dave

[pablowoiz]

Oh... ok

one more question: If I turn the encoder too fast, it doesn´t respond properly. And would be nice a faster response for scanning the band.
Maybe is a problem with my encoder, but I suspect it´s about some "refresh time" or internal clock..

Thanks Again! LU1AGH

[G8PTN]

Hi,
Yes, the “chuff” free tuning and compensation do not really work very well together. On the ATS20_EX, the range is actually limited to +/- 600Hz. I was originally going to remove the compensation option completely but decided to leave it. Provided the oscillator is within +/- 66ppm the 2000Hz was adequate at the maximum frequency of 30MHz. I did look at the option of moving the displayed frequency rather than the BFO, but then all the band limits and stored frequencies do not work correctly.

With respect to the tuning speed, yes, it is sort of limited by the display refresh at the moment. I plan to have a look at this to see if I can improve the way it works. The screen update takes between 50~75ms. I could give priority to the tuning, but then there will be a lag for the display updating. I will add this to the log of things to look at.

I hope that covers your questions.
73, Dave

[G8PTN]

Hi,
With respect to the question on faster tuning, do you think if I can provide higher steps sizes on SSB it would help with the issue?

Although it sounds odd, I think I can allow all steps sizes.
Obviously at the higher step sizes it will not be "chuff" free and in some cases it would just be toggling between the band limits.

It would also mean that if you are using the calibration option (which is per band), the value would need to be readjusted if there has been a broad frequency change, since the required offset is frequency dependant. If for example the "ALL" band is being used, if the offset at 10M is 500Hz, then the same ppm offset at 1MHz would be 50Hz and 30MHz 1500Hz etc...

73, Dave

[copernicweb]

Hello,
Thanks for the great job!
I was wondering if the ATS_MINI with si4732 would be capable of receiving AM air band (108-136Mhz) or higher VHF band (144-146Mhz)?
If so, would it be hard to add them to this firmware?

73,
Jeremy

[G8PTN]

Hi Jeremy,
I am not familiar with the SI4782.
I know you can use a simple external down converter for AM. They are about £25 on the market place sites.

As far as I am aware, the SI4732/5 does not currently support NBFM, although there is a placeholder in the PU2CLR library.

73, Dave

[copernicweb]

Sorry, I meant SI4732.
I read that it does not natively support the 118-136Mhz band but it might be possible to patch the firmware to acces it.
73,
Jeremy

[pablowoiz]

Hi, With respect to the question on faster tuning, do you think if I can provide higher steps sizes on SSB it would help with the issue?

Although it sounds odd, I think I can allow all steps sizes. Obviously at the higher step sizes it will not be "chuff" free and in some cases it would just be toggling between the band limits.

It would also mean that if you are using the calibration option (which is per band), the value would need to be readjusted if there has been a broad frequency change, since the required offset is frequency dependant. If for example the "ALL" band is being used, if the offset at 10M is 500Hz, then the same ppm offset at 1MHz would be 50Hz and 30MHz 1500Hz etc...

73, Dave

Hey Dave!

No, I don´t think bigger steps will help. For the ham bands 1khz or 0.5khz would be the most usable step. I don´t see any problem if the display refresh not so fast as the frequency. Anyway when numbers are changing fast, we can´t read them.. but we can hear some station. I think to give priority to the frequency knob over the display refresh could be a very good option.
Another idea is to add an assignable option of the "turn while pressing" like Volume, or different steps (0,1khz, 1khz and 10khz) for example. It coud works as fast accesible fine tuning (like rit in transceivers) or fast moving around the band.
I also find usefull if the MODE stores in each band. then we can move from 80mts (LSB) to 10mts(USB) with out correcting the mode every time.

Anyway, just ideas. And one more time Thanks! you are converting a toy in a great Radio for the Ham Bands!
put a paypal for donations! I will contribute!!

73, Pablo

[yarda]

As far as I am aware, the SI4732/5 does not currently support NBFM, although there is a placeholder in the PU2CLR library.

AFAIK nobody released the NBFM patch yet, but according to some traces it might exist. The si4732 internals is a black box, so the patches are binary blobs. I wish somebody would reverse the chip or leak the internal documentation, then the chip capabilities could be highly extended by custom patches. But this apply to all projects using si4732.

[yarda]

@G8PTN could you please in the future release new firmware versions as a github releases? In such case people could watch just for new releases and needn't watch all project activity. AFAIK to the new github release you can also upload the custom zip files or other assets.

[G8PTN]

Hi Pablo,
All understood, If I can add the bigger steps, I might do it anyway.

The problem with adding button press options is that is sometimes becomes difficult to differentiate actions. At the moment the button press is actually implemented as de-bounce followed by edge detect.

The rotary encoder Up/Dn takes priority and triggers an interrupt. There is a state machine for the rotary encoder (Rotary.cpp) which I have set for full step decoding at the moment to avoid erroneous actions. The encoders used on the ATS-Mini are typical very low-cost mechanical encoders and generate lots of glitches when rotating. If you rotate them too fast, Up/Dn actions will not be decoded.

I did a few timings checks in the code.
The fastest rate that an up/down is decoded is approximately 10ms.
If I enable the SI4732 actions, approximately 1 in 3 encoder triggers are actioned.
If I enable the SI4732 + Display actions, approximately 1 in every 7~10 encoder triggers are actioned.

I will investigate further, but that is why I was thinking of adding the wider steps, since it would be quicker to move across the band in SSB. At the moment you need to switch to AM, set the step to a wider option, tune and then switch back to SSB.

With respect to the MODE per band, I do plan to have a look at this.

The stock firmware does not appear to have the option to manually reset the EEPROM at boot up, so I am trying to maintain memory map compatibility in case people want to revert to the stock firmware. If the stock firm has implemented the app_id check, I may be able to change that in the ATS-Mini code and then it would force the stock firmware to reset. I have not tried this yet.

PS: Thank you for the offer, but I do not expect or want any payment for the firmware. I actually only developed it as an exercise to see if I could provide an alternative UI for the ATS-Mini, and Paul OM0ET asked if I could release the binary files for users to try. As stated in the Q&A I will try to fix bugs and add features where possible.

73, Dave

[G8PTN]

Hi,
I do not plan to maintain the code as GitHub release at the moment.
I will release the full package of code + binary in the FIRMWARE directory.
The front page readme will show the latest version and the history information is in the document.
I was only working on this for personal use and do diff checking in Notepad++
73, Dave

[oleksii-b-username]

It would be great to implement display auto off, like on start - full ledcWrite(0, 256); after few seconds ledcWrite(0, 10); then 0 - off. With toggle menu settings saved in eeprom.

[G8PTN]

Hi Pablo,
For completeness, I have checked a few points on the stock firmware.

• I updated the app_id on ATS_MINI to 65 from the original value of 47. Then loading the stock firmware, it does cause an EEPROM reset. This means that it may not be necessary to maintain memory map compatibility. If users revert back to the stock firmware it should automatically reset.

• I checked the rotary encoder behaviour. On the stock firmware the Rotary.h enables half-step mode decoding, which is not a good idea. On my hardware with the stock firmware, on rotating the encoder in one direction, I get both Up and Down actions, which is to be expected in half-step mode.

• I measured the nominal supply current on the stock firmware, which is approximately 83mA. On ATS-Mini with the CPU clock set at 80MHz, the nominal current is around 72mA (with brightness = 32). If I increase the clock speed to 240MHz the current increases to more or less the same as the stock firmware 83mA. I may increase the clock speed to 240MHz on future builds, which may make a slight improvement on the overall processing time, although generally the screen update dominates everything.

73, Dave