[Trlog] YCCC SO2R box?

[w9cf at arrl.net]

Let me start with: I am not an expert of serial ports.
My understanding of stop bits is that it tells the sending side how
much idle time to wait before sending the next character. So Having the
sending side have 2 stop bits and the receiving side 1 stop bit just
makes the connection a little slower than it would be with 1 stop bit
on both sides. If the receiving side is expecting 2 stop bits but only
gets 1 in principle it could be a problem if the chip/operating system
needs that time for something, but for a modern system it's hard to
imagine this being true. In addition, with USB serial, and in particular,
the Arduino, I don't think this setting even matters, but again I'm not
an expert and may be wrong.

I wrote the code for the Winkeyer by reading K1EL's API in conjunction
with buying a Winkeyer chip from K1EL (he selss them for about $10)
and throwing together a Winkeyer with an $2 Ebay FTDI USB to TTL serial
converter a few resistors, capacitors and 2n7000 FETs to test on real
hardware.

When George asked me about the K3NG clone not working, I eventually bought
an Arduino clone and with a few LEDs was able to test.

It turned out that there was a bug in my Winkeyer code that sent an
eroneous command that the real Winkeyer ignored. In addition, looking at
the K3NG code, his handling of buffered speed changes did not seem consistent
to me with the Winkeyer API. I changed my code to use a subset of the Winkeyer
commands that looked to be handled the same way for the real Winkeyer and
for the K3NG code.

However, like George, I use a YCCC SO2R box for my shack. The YCCC SO2R
box does not use the Winkeyer code at all. Instead it enumerates as a
human interface device. Trlinux uses the libusb-1.0 usb library to talk
to it.  That code is in the keyeryccc.pas file along with some glue code
in hid.c hidapi.h and hidp.pas.

I started fooling around with supporting the SO2R mini with NN1C's firmware,
but I stopped when Tree introduced his own firmware. It would not be hard
to have support for the original NN1C firmware, but I thought it would
be too confusing and didn't pursue it, since Tree's version will
eventually have more features.

73 Kevin w9cf

On Fri, May 27, 2022 at 12:36:54AM +0000, George Fremin III wrote:
> On Fri, May 27, 2022 at 02:26:26AM +0200, Martin Kratoska wrote:
> 
> > Now my question:
> > The Winkeyer USB documentation says there is 1200 Baud, 8 bits, no
> > parity and 2 stopbits. The same is used in TR log for linux, the
> > keyer works perfectly, I was unable to reproduce the error similar
> > to  TR4W.
> > 
> > TR4W uses only one stop bit. What impact can have to use one stop
> > bit instead two?
> 
> Recently W9CF worked on the code to allow for a K3NG type keyer to
> work wiht trlinux.  And for me it is working the same as the winkeyer
> does.
> 
> Perhaps he can give some of the more techincal insights of what he is
> doing or you can take a look at the source code.
> 
> The keyer that he make work most recently for me is this one:
> 
> https://hamprojects.info/mortty/
> 
> I have not used it for a long enough time to tell you if it has any
> connection issues over time with trlinux.
> 
> I have used a real K1EL Winkeyer with trlinux for contests with no issues.
> 
> -- 
> George Fremin III - K5TR
> geoiii at kkn.net
> http://www.kkn.net/~k5tr