Re: Bug fix for Mobo 4.3 software
- I was working on building a atmega8 (I had one lying around) version of your controller but the bumble-b arrived quicker from the US than the local electronics retailer could ship me a 16 MHz xtall and zener diodes. Certainly plenty of options available now thanks to your work.
I've got some work to do on my usbsoftrock program to support all the new commands. I still haven't got around to writing the code to set BPF limits yet, with the at90usb devices it's almost easier just to re-flash the firmware for those changes :-)
I'm about to release further changes to quisk under linux to enable smooth tuning, iq balancing etc, and some initial tx support. Slowly working on improving sdr in linux, bit by bit.
--- In firstname.lastname@example.org, Loftur Jï¿½nasson <lofturj@...> wrote:
> Hi Andrew,
> I'm happy that you find the AT90USB162 firmware useful. It has provided me
> with limitless amounts of fun to work with it. Your bug report has been
> noted, thanks. I never actually did test the BandSub/BandMul code, so I'm
> not surprised :)
> Next update of the Mobo 4.3 code will include a few other bugfixes (the peak
> envelope power func is slightly broken, and power bargraph breaks at values
> above 15W). Rotary Encoder Resolution will be moved to an "Out" type
> command, numbered 0x36 and use a signed 32bit integer to allow for a larger
> step resolution (use the Encoder as a bandswitcher). This is as per
> suggestion from Fred PE0FKO. I will also look at making the number of BPF
> and LPFs configurable, should be very easy.
> I plan to do the next Mobo 4.3 firmware update in about one months time.
> For the last few days I have been working on doing a quick and dirty update
> of the ATmega168 firmware to incorporate most of the functionality in the
> AT90USB code, the reason being that I wanted to add a LCD display and a VFO
> knob to my already built SR6.3 + Mobo 3.6. This task is not fully complete
> yet, but the current code already includes I2C driven LCD, bargraph display
> of power/swr and a Rotary Encoder VFO, and it seems to work fairly well.
> The Readme.txt file included in the ATmega zip file includes a description
> of the new Cmd 0x36 which will be included in the Mobo 4.3.
> I have promised myself that I will start playing with SDR on Linux soon,
> played quite a bit with Linux a few years ago, before I became a victim of
> Windoze again :))
> Thanks again,
> 2009/9/28 andrew.nilsson <andrew.nilsson@...>
> > I've uncovered two small bugs with 16.10 1.02 firmware from the Mobo 4.3
> > project.
> > Both are to do with command 0x39 in AVR-Mobo.c with CALC_BAND_MUL_ADD:
> > 1) The values are copied from Em.BandSub and Em.BandMul which would be
> > eeprom and I believe they should be Rm.BandSub and Rm.BandMul.
> > 2) replyBuf is a sint16_t so the offset for the second memcpy should be 2,
> > not 4.
> > I'm using the firmware with a bumble-b from fletchtronics and a RXTX 6.3
> > together with the Tony's ABPF board. I've also made a change to provide the
> > 2 bit BPF addressing directly instead of the PC8574 usage. I can supply this
> > if wanted.
> > I built the BPF with 80-6m hence the need for per band multiply/subtract
> > values.
> > The whole combination works really well, and the move to the at90usb162 was
> > well worth it. I'm impressed so few changes were required to support my
> > non-standard configuration - the configurability is impressive and I look
> > forward to slowly expanding the device support.
> > The only other change I might make is to make the number of BPF and LPFs
> > configurable through another 2 macros - would improve it for non-Mobo users
> > of the firmware.
> > Well done guys (and for atmel for the at90usb line)
> > Andrew vk6jbl