After much soldering and consternation, I managed to get this combination to work reliably. I can now utilize all the cool digital modes on my mac via FLDigi and my ProIII. The equipment is
- MacBookPro 17" running OSX 10.5.7
- XCode Installed
- IC756 PROIII
- FlDigi v3.11.x (got it to work with x=2,4,5)
- Hamlib 1.2.9 & 1.2.10(dev version)
- FT232R USB breakout board + other components wired as described below
Issues I encountered:
Hardware: (I had to build my own very simple USB/CI-V interface)
- Why not use one of those $10 integrated USB/CI-V cables for ICOM? Simple - couldn't get it to work. After staring the interchange on the scope for a couple hours and twiddling settings, I suspect the USB interface in the cable can't sink enough current to allow the ICOM to drive the interface low.
- Why not use the DLP 232M interface described elsewhere here in eHam? Doesn't work. Between the Mac and the ProIII. Again, I suspect the little interface board can't sink enough current to allow the ProIII to drive the line low. Also, the 2-wire CI-V connection is bidirectional, and I suspect this isn't handled well by the chip directly.
Software: (I recompiled/built everything locally on my machine. Can't you just use fldigi out of "the box")
- Yes, downloading the fldigi .dmg and running it worked quite fine. Though there were some subtlties in setting up the IC756ProIII interface in via hamlib. Turned out this was my own user problem - but by the time I had realized that, I had already gone through days of strife trying to build fldigi natively from the source. This required me to build hamlib locally, which required me to install swig, python, perl, tcl, etc. I must be out of my mind. If anyone actually cares to try this, I'm happy to let you know how I did it, though I suspect this will be different for each machine/software combination. Help in this area is truly sparse. In many ways you're on your own trying this - the net at large is of minimal assistance. Mostly because almost nobody cares about the MacOsX/hamlib/application combination. All I can say is, "Have hope. This can be done."
The hardware I set up is very simple. I used a FT232R breakout board I got from www.sparkfun.com
. The key things to remember
- do not expect the USB in the mac to adequately power the connection between the computer and the rig. Assistance is needed. After experimenting with various configurations, I settled on this one:
- let the USB bus power the FT232R, but then power the ICOM as suggested in the ARRL Handbook - through a couple buffers on an SN7417 TTL. The key here is that the single twisted pair between the ICOM and the FL232R is bidirectional.
- Power the SN7417 via a separate regulated +5V supply. I used a 7805 in one implementation and a 78L05 in another. I drove them from a variety of unregulated DC sources, ranging from +6 to +12V. I rapidly blew up the smaller 78L05 regulator. So I stuck with the 7805. Of course, the drop out voltage is higher on the bigger regulator, so you have to run it from a minimum of +7V. I just peeled off +12V from the rig supply itself.
- YOu connect two buffers, head to tail. These are pins 8 and 11 on the chip. The free ends are connected to the TX & RX on the FT232R board. To be precise, RX goes to pin 10 and TX to pin 9. Output to the rig is taken off the shorted pin 8 and 11 combination. Run the output through a 10uh choke in series, and slap on a 10nf cap to ground on the output as well - filter, condition, keep out rf.
- the SN7417, separately powered, has enough drive/sink capability to make the ICOM happy.
Once adequately hooked up, one starts FLDIGI, configure using HAMLIB, select the IC756ProIII profile and the appropriate serial port (in this case, it will be something like /dev/cu.usbserialXXXX_yyy where XXXX_yyy depend on the interface chip you are using) and hit initialize. If it works, you will be able to slew the passband either via the rig VFO knob or the computer. If it doesn't, youll see a message from hamlib saying there's no communication to the rig.
In this case, one thing you could try, had you installed hamlib, is to directly control the rig through the application "rigctl" which would be located in the directory /opt/local/bin had you installed hamlib.
Doing this stuff - installing hamlib, building hamlib, running rigctl, all requires knowledge of how to bring up a terminal window and running MacOSX as if it was UNIX.
Anyway - after a couple weeks struggling with this, I just had to tell someone I got it working.
73 to all