[WT-support] Using microHAM MK2R or MK2R+, Elecraft K3s, and Win-Test
Joe Subich, W4TV
lists at subich.com
Thu Nov 15 00:57:52 CET 2012
> I have considerable experience trying to get two K3s to work reliably
> at W7RN using an MK2R with current firmware (at the time, about a
> year ago). No settings worked reliably with Win-Test, even setting
> polling interval to 500 or 1000 ms. The COM port connections to the
> radio simply would not stay up. microHAM blamed Elecraft's serial
> port implementation, since it can fail to respond to serial commands
> for too long a period under certain conditions, such as changing
> bands.
I do not have the old (two to three years old) e-mails but as I recall
you refused to use the settings provided by the factory that were known
to work reliably for others. The Elecraft K3 *does* ignore polls when
changing bands *as documented by several independent developers* and
contrary to Elecraft's documentation is does not respond with a "?"
when busy.
> Using the same setup at home with two K3s, MK2R+, but with very old
> Router version 5 and corresponding older firmware (supporting only a
> single-com port per radio), the system is very reliable with two K3s,
> even when changing bands.
You were told how to make the then current (and current today) version
of microHAM Router's CAT port behave exactly like Router 5.x but did
not choose to do so. Simply set 2nd CAT to None and clear the "forward
autoinformation to CAT Port" boxes for *both* ports. With the 2nd CAT
port disabled and "forwarding" disabled, no data routing is performed
resulting in single port operation identical to the older software.
> In sum, the older Router works well, the newer Router does not, at
> least not with two K3s and Win-Test in Auto poll mode, though it's
> possible it may work better now than when I last tried it.
If you over poll there will be problems. Since the K3 goes dumb when
changing bands, it may be necessary to slow down polling to accommodate
the random response delays and/or lack of response. Alternatively,
the software could be coded to suspend polling for 1 second after any
command that would result in a band changed or wait until receiving
the automatic "band change" data packet from the K3 before polling
resumes.
> microHAM's documentation shows selecting a polling rate of 100 ms
> with the Elecraft K3 in the screen shots:
>
> http://www.microham.com/Downloads/MK2R_Win-Test_USB_Setup.pdf (Page 3)
> http://www.microham.com/Downloads/MK2R_Win-Test_LPT_Setup.pdf (Page 3)
>
> Is 100ms the recommended polling rate for K3s?
100ms was the default when selecting K3 in Win-Test at the time those
screen shots were taken. I would personally not poll any more rapidly
than 500 msec without extensive testing.
> A few weeks ago, I could I could not get SO2R operation to behave
> properly using the microHAM control protocol and Win-Test with a
> micro2R. I selected "Last One Wins," but WinKey messages would always
> complete on Radio 1 before the "interrupting" message was sent on
> Radio 2.
Does Win-Test abort the message on Radio 1 before switching to Radio 2?
What is the timing between the WinKey abort command and the Change
focus commands? What you describe sounds very much like an application
issue.
> I then switched the SO2R method from microHAM Control Protocol to
> Classic Auto Control, using virtual COM port pins, and everything
> started working perfectly, with Radio 1 / Radio 2 CW message
> switching occurring immediately, one message interrupting the other,
> as desired.
Again, control timing is an application issue and I have no doubt
that the internal timing can be different for flipping a bit/pin
vs. sending commands.
microHAM Router and the various interfaces respond to the inputs
as presented. "Classic Control" is not intended for any use
other than software that does not support the microHAM protocol
(typically DOS or *NIX based applications or non-SO2R aware
software).
73,
... Joe, W4TV
On 11/14/2012 6:02 PM, Bob Wilson, N6TV wrote:
> On Wed, Nov 14, 2012 at 1:57 PM, Joe Subich, W4TV <lists at subich.com> wrote:
>
>> First the WinKey v10 is the oldest version of the WinKey chip, and it
>>> has bugs. The current release is 2.3.
>>>
>>
>> First, WinKey version 10 is the current level for the original WinKey.
>> *There are no bugs* although some features operate differently than
>> the current Winkey 2 chip.
>
>
> I stand corrected.
>
>
>> If you are using an MK2R or MK2R+ with an Elecraft K3, communications
>>
>>> will be unreliable and the MK2R will often hang.
>>>
>>
>> *This is complete and utter nonsense!* N6TV has no status regarding
>> "bugs" in MK2R/MK2R+ as he consistently refuses to operate use the
>> software hardware *as designed*.
>
>
> I have considerable experience trying to get two K3s to work reliably at
> W7RN using an MK2R with current firmware (at the time, about a year ago).
> No settings worked reliably with Win-Test, even setting polling interval
> to 500 or 1000 ms. The COM port connections to the radio simply would not
> stay up. microHAM blamed Elecraft's serial port implementation, since it
> can fail to respond to serial commands for too long a period under certain
> conditions, such as changing bands. The microHAM buffers overflow, and
> that is when there are problems. The older firmware didn't support two
> serial ports per radio and didn't have these buffering problems.
>
> Using the same setup at home with two K3s, MK2R+, but with very old Router
> version 5 and corresponding older firmware (supporting only a single-com
> port per radio), the system is very reliable with two K3s, even when
> changing bands. In sum, the older Router works well, the newer Router does
> not, at least not with two K3s and Win-Test in Auto poll mode, though it's
> possible it may work better now than when I last tried it.
>
>
>> Disabling the 2nd COM port for the radio, in the Router Ports tab,
>>
>>> may help.
>>>
>>
>> microHAM have consistently warned against running two applications
>> that poll the transceiver heavily.
>
>
> Win-Test by default polls the K3 transceiver heavily. This is the Win-Test
> "Auto" poll setting.
>
> The only time the 2nd CAT port be used is with an application that
>> primarily listens - e.g. a
>> digital program looking for frequency reference or something like
>> SpectraVue for SDR-IQ used as a panadapter. Other software should
>> have their polling rate reduced as far as practical - if the primary
>> program polls at 200 msec, the secondary program should poll every
>> 1000 - 2000 msec to avoid collisions or overloading the serial
>> link to the transceiver.
>
>
> microHAM's documentation shows selecting a polling rate of 100 ms with the
> Elecraft K3 in the screen shots:
>
> http://www.microham.com/Downloads/MK2R_Win-Test_USB_Setup.pdf (Page 3)
> http://www.microham.com/Downloads/MK2R_Win-Test_LPT_Setup.pdf (Page 3)
>
> Is 100ms the recommended polling rate for K3s?
>
> Anyone here have reliable operation using two K3s, Win-Test, and an MK2R or
> MK2R+ with latest firmware? Maybe the problems have finally been fixed. I
> think I was trying to get Router version 7 to work but never could. W7RN
> is now using two micro2Rs, with the radio serial ports being handled by
> external USB-to-Serial adapters.
>
> A few weeks ago, I could I could not get SO2R operation to behave properly
> using the microHAM control protocol and Win-Test with a micro2R. I
> selected "Last One Wins," but WinKey messages would always complete on
> Radio 1 before the "interrupting" message was sent on Radio 2.
>
> I then switched the SO2R method from microHAM Control Protocol to Classic
> Auto Control, using virtual COM port pins, and everything started working
> perfectly, with Radio 1 / Radio 2 CW message switching occurring
> immediately, one message interrupting the other, as desired.
>
> Perhaps W7RN didn't have the latest micro2R firmware and Router software
> installed. But I reviewed all the bug fixes documented here:
>
> http://www.microham.com/Downloads/urouter_change_log.txt
>
> and I'm not seeing anything like this described there.
>
> 73,
> Bob, N6TV
>
>
>
> _______________________________________________
> Support mailing list
> support at win-test.com
> http://lists.f5mzn.org/cgi-bin/mailman/listinfo/support
>
More information about the Support
mailing list