开云体育

Hi Phil,

Inrad sells a 6th-order 47.055MHz crystal roofing filter meant for the FT-980, but I don't know anyone who has installed it (and certainly no one who has done pre/post testing to see effects on sensitivity). Cost is $125 (plus S&H), and it is currently in-stock.

I will include some details in the "mod" files folder. Give me a couple hours ...

Note: You will almost certainly limit FM bandwidth to 5kHz (vs the 12kHz native bandwidth). While the initial tolerance of the matched crystals will likely be good, I'm not so sure how a VHF filter is going to behave over time as the crystals age. At 47MHz holding a 5kHz band requires VERY closely matched crystals.

--
-Mat Breton, N8TW

Thanks Fred.

I created a Dropbox folder for "work in progress" type material, and uploaded my current hamlib alpha-testing directory. Let me know if you don't receive and email invitation ...

fldigi has a number of different CAT method options built-in supposedly, including hamlib. However, I was unable to get it to connect via hamlib to the FT-980. I'll try again.

flrig appears to be closed-source, which is a potential issue (have to convince the developers to assist if the base code needs to be modified. The rig description files are .xml-based, which helps ... but the FT-980 has a particular quirk that could be a problem. The FT-980 CAT system uses an echo back PLUS an "execute" command, in addition to the command. While a few other rigs seem to have the "echo-back", the "execute" command seems to be unique.

1. PC -> Rig: Command
2. Rig-> PC: Echo-back
3. PC-> Rig: Execute

If the flrig base code doesn't support both the echo-back and execute quirks (so we can just create the .xml file), then we would need the developer support.

But first things first for me: I'm going to concentrate on hamlib (fix bugs, add functionality). What is really intriguing is the rigctlcom program (TS-2000 translator) that might even help workaround fldigi/flrig.
--
-Mat Breton, N8TW

Phil,

It's been a while since I had my FTV-901R/FT-980 connected together, but I did a test with one of my '980s last night (w/o transverter). The V/U button doesn't have an effect, and the Operating Manual (which describes its function) seemingly bears this out. You must have a transverter connected to the radio and the appropriate lines on the ACC connector toggled before the button will change the display.

73 - Fred, N8YX

I'd love to see this fully completed and can help debug, develop, what have you. See my comments in the WSJT-X thread regarding flrig - that's another package in need of update.

73 - Fred, N8YX

Ubuntu or Red Hat are what I've used the most - various flavors thereof - though you can install anything you want as long as you can get to the repos for the tools and support libraries. Most mirrors will have everything you need.

Slightly off-topic: We need to fix flrig. It doesn't support the '980 either.

73 - Fred, N8YX

This is how I set it up as well (except I'm using a W64 system).? I think I've figured out how hamlib works, next step it to set up a toolchain.

I may have to set up a linux machine since it will be a lot easier to get Git and GCC running on it than to do the same in Windows. I have an old desktop laying around I could use.

Any recommendations from anyone on a good Linux distro for programming??


--
-Mat Breton, N8TW

I've gone through "Alpha" testing of the opensource hamlib interface library code (v4.0). The ft-980 code (ft-980.c) is listed at "Alpha", or untested.

? ? ?The Alpha code only has 6 functions listed: get_frequ, set freq, get_mem {number}, set_mem {number}, get_mode, set_mode.?

? ? ?I found only one error: the set_mem error has a bug in a line that limits the ability to set the active memory number from 1 - 15, not 1-16. I found the bug in the 'C' code.

My next step is to report this to the developers ... although I plan to set up a toolchain and finish the development myself.

If anyone is interested, let me know: I'll set up a Dropbox with my documentation and progress.

--
-Mat Breton, N8TW

Well, that just proves that you never stop learning!

I now have the FT-980 functioning in WSJT-X, which is running on Mint Tricia 19.3 (see above). My PTT setup in WSJT-X is VOX and this is handled through a SignaLink USB interface. Once I set audio and power output levels, the only thing I need to do is to use the program to QSY when the various bands open and close.

Also installed MCS 'FT-980 CAT' (from the group's Files area) via Wine and got it to work after checking symlinks, verifying device entries in the WINE config files and so forth. IF Shift, Width and most other controls seem to function.

Refer to Wald's setup in to original message in thread for setup details. FWIW, I'm using the OEM Yaesu CAT interface here.

73 - Fred, N8YX

What are you doing with setup parms in WSJT-X for the FIF-232/'980, Mat?

I have 5 serial ports on this particular system - /ttys0 and /ttys4-7. I -think- my CAT connection is currently on /ttys0 but want to make sure of software setup before I go chasing more hardware stuff.

Settings are 4800, 8 data bits, two stop bits, no handshake (also tried Xon/Xoff) and cannot seem to get the radio to be recognized.

73 - Fred, N8YX

Phil, You were right on about the RSGB RadCom review. I just uploaded the article to our Files Folder 08.

73
Roger VE7LB

Phil,

You can do this from the front panel by selecting the "Key -> V/U" buttons. I have an FTV-901R that's been modified by Yaesu to work with the '980 and IIRC it toggles those lines as its band switch is moved. (Would hook them up and try an experiment but the transverter is stored away at the moment).

The FT-980 Technical Supplement provides info on which lines are used with a transverter for frequency display. I'd given some thought to rewriting the firmware and adding our 222 and 902MHz allocations to the available selections if space permits.

73 - Fred, N8YX

Hi All.

I'm new here, so be kind.......? ? 8-)

I've got two 980s - one of them was my first solid state rig, bought new in the mid-80s and the other is a donor rig I got hold of for a song some 20 years ago.

I seem to remember that when the rig is set to the 10m band, and certain pins are taken to GND on the ACC1 socket, the frequency display could be made to display 50 / 144 / 432 etc as the 10m band was traversed in 500KHz jumps. The rig stayed on 10m, but displayed the desired transverter frequency 'band'.

I cannot find this info documented anywhere however. I seem to remember reading about it in an RSGB RADCOM magazine (probably within the review of the 980 by G3SXJ I think his call was) but even then I don't recall a source for the info being stated. I can't see anything mentioned in the 980 manuals......

Does this ring any bells with anyone or have I been sniffing the PCB degreaser a bit too much.....?

73,

Phil Smeaton
G0HSS / VK4BAA / VK4KW

Your logic makes sense: if the blocking diode was defective-shorted (vs open), it could expose the MCU pin to raw IO voltage (vs emulating a low-side driver). If the FC-757 is like the 980, Yaesu didn't seem to provide a huge amount of protection to the IO. The FILT pins on the ACC1 plug go right into CMOS logic gate inputs for example.

I haven't reverse engineered this aspect of the radio (as I only wish I owned the 757 tuner), so I'm going to have to wish you luck!
--
-Mat Breton, N8TW

Did some digging into the tuner this morning. First order of business was to verify correct voltage on each of the four Band Data lines as the various bands were selected by the '980. This checked out.

Next, that data was verified on D03-D06 at the cathodes first (nearest the radio - consult the Instruction Manual schematic) then at the anodes (nearest the uPD7507 CPU). All check except for the "B" line. This line makes up part of the control nibble for 3.5, 7, 18 and 21MHz. If the right voltage isn't present, the tuner will merely go to the nearest band which matches the nibble value (e.g., 14MHz in binary is 0101 and 21MHz is 0111 - MSD first.) A check of diode D04 showed it to be bad; it was replaced but the problem persists.

All pull-up resistors and other components in the area check out, though if you examine the "B" data line of the CPU at the CPU itself with an ohmmeter, it reads 110KOhms - while the other (properly working) data pins read 101K or thereabouts.

This leads me to believe the CPU is partially bad. Note that if you select Manual mode, ALL data lines are pulled low at the CPU and band switching is handled via a different means - the tuner will still work properly in this fashion so in Manual the defective pin isn't an issue. Unfortunately, the device is a factory-programmed part. I'm half tempted to redesign the tuner to use a BASIC Stamp or similar embedded micro that has a lot of I/O pins, barring me being able to find another tuner or CPU Board.

73 - Fred, N8YX?

The next time someone takes off the IF PCB, could you snap a high-res picture for me? I'm trying to check to see if a jumper on the back was put back on the correct pads ... doesn't seem to match the schematic. No rush ... just keep me in mind ...

--
-Mat Breton, N8TW

My rig actually came with the MH-1B8 hand mic (the FT-980 standard 600 ohm hand mic) ... I quickly retired it somewhere. I'll try to dig it up. I would not be shocked if the mic element was not in as good a shape as it used to be. I a number of older mics: many have aged well ... but a few have not.

Note: if I set my mic gain at max (or anywhere near it) on my FT-980, I will overload the mic pre-amp resulting in poor audio. Any chance this could be it?

--
-Mat Breton, N8TW

Mark,

Do you have another mic to try with your rig?

I get more than adequate drive with an MD1- series desk mic and don't have to open up the Mic Gain very far to obtain an ALC reading.

73 - Fred, N8YX

Anyone else in the group using this combo?

I dragged my tuner and FIF-232 out of storage today and along with one of my '980s, set it up for FT8. That rig and tuner ensemble work MUCH better than other contemporaries of the period (and even a few later rigs such as Kenwood's 9x0"AT" equipped models) .

One fly in the ointment: With the tuner set to "auto", it appears that some bands won't auto-select under control of the radio. Putting it in "manual" mode, band switching works properly - so this indicates the CPU and switching logic local to the tuner is in order. I tried two different interface cables and verified connections for each.

This leaves the isolation diodes between the remote connector and the CPU. I'll post a reply with results of testing/replacement when I dive back into the unit in case someone encounters the same problem.

73 - Fred, N8YX