ICOM Rc-28

[djacobow]

This is great work, and if you publish your work or make a product, you will be a hero to many. It's the essence of "amateur" radio.

Don't worry about the haters who seem to think tinkering with their gear is some sin against Lord Icom.

[TheBay]

Well done! I've been subscribed to this thread and reading every update. Incredible bit of work you have done there.

Myself and I'm sure a few others would love to build one.
Look forward to reading further updates as you progress.

[eb4eqa]

Wow, great achievement. It would be nice to have free info available to build one. Even if you decide to keep it for yourself, still amazing job 

Roberto EB4EQA
www.rbarrios.com

[cncjerry]

here's an update:

I had a heck of a time getting my code to work in two-way mode.  Remember my base code was originally designed to be one-way as a mouse and keyboard act.  There is a slight difference in a USB keyboard as the caps lock and other lights are set by the host PC.  But that is done with the control endpoint (I'm told) and I am using the bulk in/out endpoints.   Sending my frequency updates to the host code were being paced at 1.2sec vs the radio was just ripping right along.  This drove me crazy as I just couldn't figure it out.  Finally I found a packet coming from the RS-BA1 code that asked for the device firmware level.  I was replying but about 50ms too slowly.  So this meant I had to re-architect the code to be two-way (interrupt responding) to this packet as well as pacing my code against the host application ack's coming in.  Then I had to program the ramping to take advantage of the ramp feature on the radio.  This requires an internal timer.  Anyway, I got it working so now I can move ahead.  I must have recompiled, tested and traced my code using Wireshark (thanks again for the that advice) 100 times.  The radio would spin the frequency really fast, hard to even see the numbers, but the RS-BA1 software would just click along in 1.2sec jumps.  Now both spin like crazy.

Anyway, I realize there is interest in this as us amateurs like to tinker so I am leaning more towards offering this as a QST or QEX project rather than trying to make it commercial.  The hassles around a commercial product plus fighting with Icom as I reverse engineered their device would take the fun out.  I'm sure I can get over that hurdle but it is their design so unless they publish the USB encoder design requirements in the future, which I doubt, I feel like I would be stealing.   But this is looking pretty cool.   A slanted 3x5 LCD touch screen with an encoder and knob.  There's enough horsepower in this board to also embed a CW reader, decoder for digital modes, etc.  Way beyond my scope, I just want a couple more encoder dials (to hang my overweight knob on  )

So now onto the LCD design.  Big bug there as I can't get the USB and LCD code running at the same time, I mean WTF??.  I am hoping it is just related to stack and heap sizes and not some crazy interrupt problem.  Could also be the clock or just be the board I am using as many of the signal lines are dual function.  The LCQ goes white when I initialize the USB device.  But I have a ton of experience in LCD and TFT screens so this should work out after about 50 compiles or so.

Jerry

[cncjerry]

here's the STM32F469 development board I am thinking of using. It presents equivalent to an early iphone, has the same look and feel with 680x480 resolution.  Only problem with the board is the one I have has developed a crack in the cable or maybe it is just loose.  The screen is sort of flickering that goes away when I press on the right side where the cable is captured under the display.  The display itself is no more than 1/32nd thick with a backlight about 3/32nds.  The display was rubber-glued down and I guess it dried out over the year or so since I bought it. 

The board itself has all the features we need including 5 leds, SDCard, audio, mics, adc, dac, etc, etc.  Don't think I'll be using much of that, though.  The encoder will be mounted to the right with three buttons under it.  I'll use the display for the 10 buttons and wire that into the accessory jack.  If I can figure out how to enumerate two devices (encoder and keyboard) then I can send the recorded CW and voice buttons directly instead of the extra wire into the accessory jack.  That is much cleaner, I'll admit, but haven't figured it out.  I am getting pretty good with all this USB stuff though.

One thing nice about this board is that it presents to a connected PC as a USB drive so for firmware updates, you just take the executable and copy it over.  Don't have to mess with DFU, etc.

Here's the board with an early keypad overlay from another project.  I am using this one for testing and will hope to design some cooler looking buttons consistent with the IC-7610 display.

Overall, fun project now that I have it all working. 

Jerry

by the way, the second USB is for the integrated debugging.  The white cable goes to either a PC running RS-BA1 software or the supported Icom radio.

[cncjerry]

Now I'm rethinking this board as the flat cable that wraps around underneath is twitchy.  I've had the board around 2yrs and I guess the flat cable has a crack.  So I ordered another and will try to get ST to replace this one.  Wish me luck.  I have about every flavor of their STM32F4 and F7 board and this form factor seems to fit my plans.

The only encoder I have around here is a good-quality Grayhill with a detent.  It goes for around $25, not the best but it works for testing.  I coded an interrupt driven routine to process the encoder.  Works so quickly that you can see the detents.  The interrupt routine processes the transitions from either pin and depending on if it is a high going or low going pulse, and the status of the other line, you get the step and direction of rotation.  It wasn't easy but neither was all the USB code.  I looked all over and couldn't find encoder code with step and direction for the HAL library instead of the standard peripherals.

The touch screen is working fine and I'm going to add a keyboard enumeration to this device so I don't have to connect to the accessory port.  I should have that working by the time the new board gets here.  Then I'll mount the parts on a frame.  I have a heavy duty encoder with bearings on its way from eBay.  I was thinking of just ordering a US Digital or Bourns encoder but thought I would try these optical encoders off eBay for this and other projects.  They cost $18 and look well made.

So this project will end up costing about $100 but could have been zero if I was happy with the encoder and my STM32F469I board didn't decide to break a cable I can't even see.  It looks like that flat cable has about 50 wires in it too.  When I torque the board a little, it starts jumping all over the place so it must have something to do with refresh, no idea.  I remember when that board came out people were wondering on the ST forum if the screen would last.

Anyway, this has been fun, very satisfying project.  Just enough difficulty to keep my interest.

Jerry

[cncjerry]

I have everything working now and I am going back and cleaning up the code.  Since the board has a mini SD card, I added an INI file that allows you to program the functions behind each of the 3 sets(or more) of buttons.  Instead of emulating a keyboard on USB I now have a cable that plugs into the remote jack on my 7610 so I can send pre-programmed CI-V commands.  The large center soft button in the upper middle is where you would press for transmit.  I also put the operating frequency there now and mode.  Since the buttons are now CI-V programmed you can send up to 72 characters of CW behind any button or better yet, send any CI-V function.  This leaves the scope up on the screen while having all the CW functions handy.  This along with the encoder was my original objective for the device.

The photo doesn't do it justice at all.  I also still have the screen saver on.

The other picture is the Grayhill encoder I'm using.  Only issue with it is the freedom of movement is really high so the knob has to be balanced or maybe a friction drag with a lock as it will settle if the knob is out of balance.

I have an acceleration ramp that along with the loose encoder and high pules per rotation, allows me to give it a spin and go across the entire band in 1hz tuning mode or zero in on a signal.  This makes it many times more pleasant to use than the RC-28 which I also have.  I'm going to post a video of the operations.

I started designing a box with my CAD system.  I'm going to cut the prototype out of billet aluminum.  It's about 7" wide and 2.65" deep, slanting back to front with the back about 1.75".

The board also has an ADC and DAC and I was thinking about adding some DSP tone functions for SSB two-tone testing but that will be down the road. I still haven't decided where the project is going but I've had a number of inquiries.

More to come.

Jerry

[strimarc]

Any more news on this?  I would love to build one...  much nicer and more functional than the rc-28....

[cncjerry]

i've been traveling and not able to finish the case.  I'm also not wild about the case I've designed thinking more like the ICOM detachable front panel displays now vs a simple slanted box.  I have all the code running and have been tweaking the acceleration but otherwise I'm happy with the performance.  I find it much nicer to use than the RC-28 as I can give it a flick and scan 50k in the 1hz tuning mode.  I have the microSD card working to load CW programs for the buttons.  I tried writing some PC VB code to talk to OEM logging and control programs like Rumlog and others like it works with RS-BA1 but decided to set this aside. Everything I do to it though, has to be created as there is a severe lack of available code for the STM32F4 platform using HAL.  I also wrote about 10 pages of doc, mostly specs and background, but I haven't decided what to do with it yet.  I find it to be really cool and handy, especially with the CI-V macros behind the buttons.  It makes the 7610 more SDR without having to have a PC attached...

I have the week off between Christmas and New Years so I'll make a lot of progress then.

Jerry

[strimarc]

Make any progress?  Any chance you'd publish what you've got?  I'm sure there a few people that would love to run with it... 

[gi1mic]

I know this is an old topic but it was never resolved - so, if anyone is intrested I have released an arduino emulation of the RC-28 remote controller which can be used to control the RS-BA1 software.

While it is fully working I still class it as "work in progress". The code and instructions are available at https://github.com/gi1mic/rc28_emulator.

The code only works on an Arduino Pro Mini "Leanardo / 32u4" based Arduinos and requires some knowledge of working with the Arduino IDE in order to use it.

The RC-28 is the ICOM USB jog wheel for use with the RS-BA1 remote control software. The build cost, excluding case is around $40 including optical encoder, switches, LED's and tuning knob.

Gareth

[TheSteve]

Nicely done! de VE7FM

[cdev]

There is a Linux library to use Hercules DJ CD mixers (a popular brand) as SDR controllers

You're not alone in wanting to clone the RC-28. I've always figured it was nothing more then a mouse with a unique VID and PID. I was surprised it needed a firmware update to work properly with the IC-7610. The Griffin Powermate seems like a cheap alternative if it could be made to work.

[kc9umr]

Gareth-

I've been tinkering for a while on something similar - have you ever considered a branch of this project that would communicate over CI-V?

this would (as long as the CI-V address of the radio could be modified easily) make the project easily adaptable for any Icom radio

The only gotchya is that it would tie up the CI-V port, which isn't a killer for some of the newer radios (7300, 7610), but for older radios (737, 735, 706) might require addition of a CI-V bus controller to avoid collisions. 

Now that I'm putting my station back together, I'm hoping to get back to this project in the near future.... unless Win4Icom adds support for some of the fancier shuttlepro type devices on the market...

[bufffler]

Has anyone tried this with a 7610?
Anything further from cncjerry?

[Sidlinger]

I have the opposite desire. I bought an RC-28 for my IC-705 only to discover that RS-BA1 does not expose all functionality (no WFM iirc.) So now I would like to repurpose my RC-28 for use with Linux radio management or other software. I haven't started this project yet but I'm posting here in case anyone else has a similar interest. This would be a software-only effort, I don't want to modify my fancy and overpriced device. I guess a driver or something although I don't know how things are abstracted in Linux.