If you avoid USB control of the rig through CAT commands, you would completely avoid the problem. Sounds like the front panel control avoids this issue and then you would be OK. Note that I have not tried this avenue so these are assumptions. Of course, if the rig is polling the radio to see what its frequency is through the Signalink (I do not own one so cannot comment further on whether this is wrong), then we're back to the same problem.
Just thinking out of the box here, but have you tried a wireless Bluetooth CAT controller instead of using a hard wired USB cable connected to the computer? My thinking here is perhaps the USB cable is injecting some kind of data signal noise from the computer USB port directly into the radio CAT port, and perhaps when using a wireless Bluetooth CAT adapter it wont do this anymore because it eliminates the wire? I am using one on my Yaesu FT-817 to control N1MM logger. It works with any kind of software really. Basically the Bluetooth connection just looks like any regular serial comport on the PC end of things. You just point your software to whatever comport the Bluetooth connection is assigned to and you're good to go. Except there's no longer any cable connected between the rig and PC anymore. If you don't have Bluetooth capability on your PC, they make a little inexpensive dongle ($5 or something) that plugs into your USB port and looks like this: This is just a short video showing you what these wireless Bluetooth CAT adapters are exactly, (not my video) but these adapters for Yaesu rigs are all over eBay. I think the Yaesu CAT control port is standard and function the way same on all the FT- "8" series radios. But confirm that your FT-891 is with Yaesu to be sure. I'm not 100% sure about that model. These wireless Bluetooth CAT adapters wont break the bank or anything and it might be worth a try. Here's one for 10 bucks but there's many others.
Thanks for the detailed suggestion. Unfortunately, the 891 only has a USB control port and not the DIN plug style port you're referring to with the other "8" series radios, so this approach would not work. It is possible I suppose to use the USB-to-bluetooth dongle you show but I have not tried that and I don't know whether it would cure the problem or not. Perhaps someone else out there can give it a try and report back if it helps. I can say that my educated guess is that the difficulty is directly inside the radio and not with the cable - i.e. they have coupled digital transitions on the internal PCB traces to the audio or RF sections of the radio. But it's a guess. This comes from the fact that I tried multiple USB cables at different lengths with and without toroids and none of it helped.
I wasn't aware CAT was a USB port on this rig, but perhaps converting the CAT port to Bluetooth may be a work around to the problem. Good luck.
Yes, that would work if you manually tune the radio to your favorite FT8 frequency, seeing as CAT control is separate from audio in/out. (It wouldn't work for any contest situation or other things where you want the computer to log frequency, of course.)
In case I was reading manuals wrong, I just double-checked and the FT-891 appears to be different from the e.g. FT-857D. That radio has the same 8 pin mini-DIN plug on the back that would work with the Bluetooth adapter you originally suggested, but it has a menu item 020 that explicitly switches the port between CAT / LINEAR / TUNER mode. However, for the FT-891, there is no such menu item and the only thing you can do is switch between the port's LINEAR and TUNER functions (Menu item 16-15). The only mention of CAT is with the USB port. So I think the CAT control can only work with the USB port. If someone figures out this is wrong, please let us know!
I wonder if there's a way to cobble up a USB to mini-din female adapter (converter cable) for the back of the rig. There might already be something like that commercially available. I know older computer keyboards which use mini-din connectors could be adapted to work with USB ports. With such an adapter cable, you could plug the Bluetooth CAT adapter into it.
The key question for that approach is whether the USB serial lines (D+, D-) are equivalent to the RXD / TXD lines for the older radios with the mini-DIN plug. I think the answer is "NO" because in other contexts, USB to serial requires an active device (such as the FTDI series) rather than a simple wiring job, as the USB wire protocol is not the same as a regular serial line protocol. This is all more trouble than I have time for, regrettably, but maybe these ideas will spark someone else to try a workaround. PS: the older keyboards with both PS/2 (DIN) and USB plugs have an active device in them and a corresponding driver on the PC, so it's not a straight wire through.
Yeah I thought about that after making my post. Another random thought I had is the guys who are making these Bluetooth CAT adapters for Yaesu rigs and selling them on EBay are actually installing mini- din connectors on these Bluetooth circuit boards. However, I think these Bluetooth circuit boards are actually originally intended to work with USB ports. Perhaps they could customize one for you with a USB plug on the end instead of a mini-din? Heck, they might even be buying the Bluetooth circuit board configured that way in the first place.
I think it's worth a try. The worse case scenario is the noise is still there, but you upgraded to a wireless CAT control setup (this works with your smartphones and tablets too) so nothing is lost anyways. But who knows, it might actually work. Good Luck and keep us posted.
I've simplified the idea considerably for the test. I'm waiting for a USB type A to type B adapter to arrive which is straight through (no cable; < 1 inch). My computer will look funny with a radio sticking out of the back, but this will be faster. If the problem still exists then I will have verified it's inside the radio to my satisfaction. By the way, the radio arrived back here and I had the covers off briefly yesterday. Troubleshooting the radio at the PCB is not for the faint of heart. As one would expect, tons of SMD 03 size parts in there and the digital lines on the top board are not accessible without removing the top PCB. That requires significant disconnection and lifting of some regulators/transistors attached to the case heat sink, etc. and I stopped when I saw that. [WARNING: do not go even as far as my simple visual inspection without proper antistatic gear!] Even with microscope station equipment, I would do more damage getting in there and rooting around, especially since no schematic PDF appears to be available anywhere. It would also be a huge job requiring a lot of time, which regrettably I don't have. So it got buttoned up and is still "meeting factory specifications" (sic). Of course, sending it to Yaesu was the whole point of not having to consider that level of debug, but you all know how that worked out. Sigh.
This evening, I tried the final test: I used a straight through USB type A to type B connector, pictured below. This is approximately 2.5 cm long and has no cabling at all. It was a bit of a physical gyration to get it plugged into the computer and the radio, but I did, fired up the CAT logger, and... No change at all. Screamingly loud "click" city. As I expected: the problem is definitely in the radio. (I note with some irony that the borrowed 20+ year old Kenwood TS-450SAT sitting on this desk is totally silent under CAT control with a USB to serial /FTDI chip cable.) Alas, that's all the time I'm going to waste beating my head against a wall with an unfixable problem arising from very poor mixed signal design. My very limited hobby hours are much more well spent in constructive activities like home brewing or working with the BiTX40 or QCX kit staring at me from the shelf. So you will see this radio appear on the for sale section of this website soon, with a link to this thread for honesty. If someone doesn't care about CAT control or polling, this is still a sensitive RX and TX unit and you'll be getting it almost brand new. Or maybe someone with a full SMD capable station plus a whole lot of time can wade into the PCB and try and fix the issue with decoupling/bypass caps or the like (good luck, especially without a schematic, but some people like a challenge). For my part, I'm pretty ticked off, and I'm done with Yaesu and their shoddy quality control on this radio plus, more importantly, the complete unwillingness of Yaesu HQ in Japan to assist this customer despite my going out of the way to give them multiple chances to make things right. I do appreciate everyone's help here. The community has been very supportive. For those buying this radio - I hope you are free of this problem and can enjoy the rig. But I can't guarantee you won't end up with this issue either. And I certainly hope you never have to send it to repair. Caveat emptor. 73 and good DX to all.
