[K3] MCU 4.47 serial recovery

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

[K3] MCU 4.47 serial recovery

Leigh L. Klotz Jr WA5ZNU
Administrator
I've noticed that the ability of the K3 to "recover" from disturbances in RS232 isn't what my memory of it used to be.

For example, if I start a second program sending query commands the the K3, I can never get it to answer any more serial commands.  My belief is that it used to eventually get back in sync.  I've verified this with my own serial programs and also with k3util.  For example, when it's stuck, it won't reply to "FA;" at all until power cycle.  (Other rig operation other than RS232 is unaffected.)

I don't know if it's related to the input parsing code in the K3 or if it's related to some DTR state that gets asserted briefly during the serial port open.  But I can't find any reliable way to clear this other than rig power cycle.  Resetting the K3 baud rate from the front panel, for example, doesn't cause it to come back.

Leigh/WA5ZNU
Reply | Threaded
Open this post in threaded view
|

Re: [K3] MCU 4.47 serial recovery

wayne burdick
Administrator
Leigh,

No firmware changes were made that should affect this. All I did was  
provide more robust rejection of ill-formatted commands that were  
coming from HRD and other programs when allowed to poll at too fast a  
rate. I heard from over a dozen users of various programs that this  
change completely cleared up the artifacts they had seen.

I suspect buffer overrun by multiple programs combined polling too  
quickly.

73,
Wayne
N6KR

On Jan 29, 2012, at 5:44 PM, Leigh L. Klotz Jr WA5ZNU wrote:

> I've noticed that the ability of the K3 to "recover" from  
> disturbances in
> RS232 isn't what my memory of it used to be.
>
> For example, if I start a second program sending query commands the  
> the K3,
> I can never get it to answer any more serial commands.  My belief is  
> that it
> used to eventually get back in sync.  I've verified this with my own  
> serial
> programs and also with k3util.  For example, when it's stuck, it  
> won't reply
> to "FA;" at all until power cycle.  (Other rig operation other than  
> RS232 is
> unaffected.)
>
> I don't know if it's related to the input parsing code in the K3 or  
> if it's
> related to some DTR state that gets asserted briefly during the  
> serial port
> open.  But I can't find any reliable way to clear this other than  
> rig power
> cycle.  Resetting the K3 baud rate from the front panel, for example,
> doesn't cause it to come back.
>
> Leigh/WA5ZNU
>
>
> --
> View this message in context: http://elecraft.365791.n2.nabble.com/K3-MCU-4-47-serial-recovery-tp7235321p7235321.html
> Sent from the Elecraft mailing list archive at Nabble.com.
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Reply | Threaded
Open this post in threaded view
|

Re: [K3] MCU 4.47 serial recovery

Leigh L. Klotz Jr WA5ZNU
Administrator
Thank you, Wayne.  

Changing the retry interval from 50ms to 250ms seems to fix the problem I encountered, which in more detail was this: When an unrelated program strobes the RS232 ports (looking for compatible devices) it gets the K3 out of sync. With the longer retry delay it gets back in sync with the rig control command stream.

Should I default the update the fldigi rigcat defaults for the K3 to 250ms retry, 10 retries, 100ms write delay?  We're presently shipping 50ms timeout, 5ms write delay, 2 retries with fldigi rigcat control.

Leigh/WA5ZNU

Reply | Threaded
Open this post in threaded view
|

Re: [K3] MCU 4.47 serial recovery

Dick Dievendorff
The K3 Utility waits 500 msec before retrying a command if an expected
response doesn't arrive. My rationale is that if the rig is too busy to
respond, me beating on it for a retry isn't going to help the situation.  I
retry most things up to 10 times. Some things aren't retried at all.

Dick, K6KR


-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of Leigh L. Klotz Jr
WA5ZNU
Sent: Sunday, January 29, 2012 8:13 PM
To: [hidden email]
Subject: Re: [Elecraft] [K3] MCU 4.47 serial recovery

Thank you, Wayne.  

Changing the retry interval from 50ms to 250ms seems to fix the problem I
encountered, which in more detail was this: When an unrelated program
strobes the RS232 ports (looking for compatible devices) it gets the K3 out
of sync. With the longer retry delay it gets back in sync with the rig
control command stream.

Should I default the update the fldigi rigcat defaults for the K3 to 250ms
retry, 10 retries, 100ms write delay?  We're presently shipping 50ms
timeout, 5ms write delay, 2 retries with fldigi rigcat control.

Leigh/WA5ZNU



--
View this message in context:
http://elecraft.365791.n2.nabble.com/K3-MCU-4-47-serial-recovery-tp7235321p7
235356.html
Sent from the Elecraft mailing list archive at Nabble.com.
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Reply | Threaded
Open this post in threaded view
|

Re: [K3] MCU 4.47 serial recovery

wayne burdick
Administrator
In reply to this post by Leigh L. Klotz Jr WA5ZNU


----
http://www.elecraft.com

On Jan 29, 2012, at 6:12 PM, "Leigh L. Klotz Jr WA5ZNU" <[hidden email]> wrote:

> Thank you, Wayne.  
>
> Changing the retry interval from 50ms to 250ms seems to fix the problem I
> encountered, which in more detail was this: When an unrelated program
> strobes the RS232 ports (looking for compatible devices) it gets the K3 out
> of sync. With the longer retry delay it gets back in sync with the rig
> control command stream.
>
> Should I default the update the fldigi rigcat defaults for the K3 to 250ms
> retry, 10 retries, 100ms write delay?  We're presently shipping 50ms
> timeout, 5ms write delay, 2 retries with fldigi rigcat control.

I would :)

Wayne

>
> Leigh/WA5ZNU
>
>
>
> --
> View this message in context: http://elecraft.365791.n2.nabble.com/K3-MCU-4-47-serial-recovery-tp7235321p7235356.html
> Sent from the Elecraft mailing list archive at Nabble.com.
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html