This very much could be a timing error. That¡¯s not completely consistent with what you¡¯re hearing and seeing, but it might be a start.
Could you do the following, please?
*) Open the LocoNet Monitor
*) Make sure ¡°Show Raw Data¡± and ¡°Show Timestamps¡± are checked.
*) Click clear screen, do a CV read from JMRI, and then once the timeout error happens, click Freeze Screen (to keep anything from changing)
*) Copy and paste the trace to a post here.
I can look at the timing to (possibly) see what¡¯s going on.
Also, when you read a CV from the Twin-Center keypad, does anything appear on the LocoNet monitor? If so, I¡¯d like a copy of that too.
Bob
On Jan 9, 2019, at 10:54 AM, cocco.bill.97@... wrote:
Hello Alan, reading CVs by directly using the keypad on the Twin-Center works as expected. I can hear the relay switching the program track on when I enter programming mode and all basic CVs read and write successfully on different locos.
I did some more testing with JMRI and I noticed that at times the relay clicked some (long) time after I pressed the command to read CVs on the PC but then I got the same timeout error. In this case, after the click, one time I saw the command station crash and reset itself, and another time on the console screen another error message came up regarding an unrecognized incoming loconet package. I'm sorry, in the hurry I forgot to copy-paste it this morning.
My guess would be that for some strange reason the command station takes longer to answer in programming mode when controlled via the serial port and because of this JMRI believes that the request has been lost and sends a new request, eventually hogging the buffer of the serial interface with requests, which would explain the errors broken incoming packages and the crashes on the Twin-Center. This is only my assumption based on what I saw though, I have no knowledge the internal workings of the thing, so if you know something more about it I'll be glad to be corrected.
Thank you,
--
Bob Jacobsen
rgj1927@...
Bob Jacobsen