Adding CAT to 2nd "Standard" COM port in FT-991A

Discussion in 'Radio Circuits, Repair & Performance' started by PY2RAF, Apr 12, 2020.

ad: L-HROutlet
ad: l-rl
ad: L-MFJ
ad: abrind-2
ad: Left-3
ad: Left-2
ad: QSOToday-1
ad: Subscribe
  1. PY2RAF

    PY2RAF XML Subscriber QRZ Page

    Hi Forum,

    In my quest abusing my FT-991A, I found that I can actually have two simultaneous CAT ports (already documented in this thread), with the catch that besides the Enhanced COM port from USB bus, you will need to use the RS-232 DB-9 port. I've been using happily this scenario at home.

    However, due to COVID-19, I'm staying at the country QTH, and I brought the 991A but forgot (ehr, sort of, it was just too complicated to remove) to bring the serial cable + Serial-to-USB adapter. And dual CAT is a must to me.

    Well, that got me thinking - I have already studied that part of the equipment when trying to grok the usage of the "Standard COM port" of my 991A and the conclusion is that Yaesu just didn't wire the TX/RX leads (Blue and Red arrows). As they do with every other product in their line-up, including the flagship FT-DX101D.

    Pin it.

    Now... what if I wired the Standard COM port by jumping the CMOS-to-RS232 driver TX/RX feeding lines?

    See the below schematic (click to enlarge):

    [​IMG]

    Here, I'm tapping the CMOS lines coming from the CPU that enters/leaves the Q1052, Analog Digital's ADM3202 CMOS-to-TTL driver in CMOS levels (3.3V) and jumping to the unused "Standard" RX/TX pins in Q1008 CP2105 Serial-to-USB bridge.

    There are two lines, a blue (TXD) and yellow (RXD). Fortunately, Q1008 provides a test pad plus a required 100R resistor in the TX line. Easy peasy, just land the TX line in the TP1001 pad. See the below blue arrow:

    [​IMG]

    However, the RXD pin (yellow arrow) does not provide any landing pad, connection or whatsoever. To give it a yay and complicate further things, Q1008 is a QFN chip.

    So in the previous schematic, I have added a 100R resistor (probably embedded on the wire) that would land on RXD data.

    This circuitry is located on the flip side of the Main Unit. But they are really close, and shouldn't be an hassle (fortunately there's a great technician in São Paulo):

    [​IMG]

    I don't intend to be working simultaneously with port DB-9 and USB; I will use USB permanently.

    So, do you envision any potential problem on carrying such mod?

    My only "concern" (notice the quote) is sharing the same TTL line between the Q1052 and 1008. And I'm sort of unsure/unconvinced why would this be a problem.

    ESD, shielding etc shouldn't be a problem, since this is all dealt internally in the radio; actually reducing the attack/problem surface by not using the DB-9 connector.

    Do you see other problems that I'm failing to see / perceive / think of?

    I'd appreciate a lot to learn about.

    73, stay safe;

    - RF.
     
  2. KI5BPK

    KI5BPK Premium Subscriber QRZ Page

    Rodrigo-

    Interesting idea (but see below) ... thanks also for the information on the second CAT port being active simultaneously with the first one. I'm going to need to check that out.

    Here are a couple of small problems with your idea:
    The first is that you have the data I/O pins reversed in your schematic. These would be correct if you were trying to connect the serial port in the back to the "standard" serial port from the CP2105, but that isn't what you're trying to do.

    The second is that the Tx pin from the ADM3202 (pin 12) and the Tx pin from the CP2105 (pin 21 and the 100 ohm resistor) will be in contention with each other, I think. You normally wouldn't be able to connect both together without one loading down the other, making the serial data from your computer to the CAT interface unreliable.

    One solution to this problem, if you are absolutely not going to use the CAT interface on the rear of the radio, would be to lift pin 12 of Q1052 (ADM3202) and connect your jumper wire to the PC board (not the pin of the ADM3202, it would just be left floating).

    Another solution would be to create a "diode-OR" type configuration, since I believe that the data pins are "active low" (confirm this before going forward with this suggestion). You'd first lift pin 12 of Q1052 like above. Then install one signal-type diode (think 1N4148) with the cathode on pin 12 of Q1052 and the anode on the PC board pad under pin 12. Another one of these diodes would have the cathode on the resistor at TP1001 (for the output pin 21 on the Q1008/CP2105) and the anode also on the pad under Q1052 pin 12. Then put a mild pull-up resistor (10K ohms would probably be OK) from the pad under Q1052 pin 12 where the diode's anodes are to pin 16 of Q1052 (Vcc/3.3v). Again, this will only work if the data is active low (inverted) because the diode configuration I've described assumes that the outputs of the CP2105 and the ADM3202 are at 3.3v when idle. Once more, please confirm my belief here before you take this approach.

    This is definitely an interesting idea, and I'm definitely interested in looking into your original discovery as well using a separate USB-to-serial adapter to see how that works. I don't think I will be attaching a GPS to the radio, and having the second CAT interface would be very interesting.

    73
    -- Gil (KI5BPK).
     
    PY2RAF likes this.
  3. KI5BPK

    KI5BPK Premium Subscriber QRZ Page

    I just discovered something else that's interesting about the RS232-based CAT port: It does not support the AI (auto-information) and PS (power switch) CAT commands. There may be others as well, so be aware of this. It will affect your use of this mod as well if you decide to go forward with it.

    This restriction is not mentioned in the CAT reference manual or in the manual.

    -- Gil (KI5BPK).
     
    PY2RAF likes this.
  4. PY2RAF

    PY2RAF XML Subscriber QRZ Page

    Hey Gil -

    First of all; HUGE THANK YOU VERY MUCH for your well thought post and insights - It is really appreciated!

    So, for the first finding - Damn, nice catch - you are right! I have indeed mixed up the pins, thank you very much! We would have found it during tests, but you already spared a lot of hassle with this finding - tips hat, awesome!

    I tend to agree with your second finding too. The diode solution is very elegant, but is adding a lot to complexity - This is something that Yaesu engineering could come out with - Or something done with a switch, relay, etc - It could be such a product differential offering... People that uses N1MM plus other software and have to resort to HRD to glue/bridge stuff... Go figure.

    And also - wow, thanks for the AI CAT command tip! I never played with this command and it is of great help; there was a time when I was wanting to monitor radio operational data and the AI is just. Perfect. Thanks for the heads up!
    I can't test it right now (well, since I don't have conditions to use the DB9 port), but as soon I get home in São Paulo will give it a try and will let you know.

    I will be resorting to lift up the pins and isolate it with kapton and route the leads to the CP2015. But first, I need to get off this quarantine.

    Thanks again for your comments and findings, Gil! And stay safe.

    73;

    - RF.
     

Share This Page