Dave:

?

I put the "C" in the CMRI preferences.?This cured 90% of the problems. But I am still getting load problems. I sent a debugging info a second time with the newest problems. What I get is “Unexpected Error (exception while load (/C:user/dell/jmri/rvr%208-2018-3.xml) in adapter type jmri.jmrix.internal exception: invalid system name for turnout: IT needed IT”. I hit OK and things kept loading all the panels but then stopped and gave another error message “panel experienced errors during load. Check your system preferences. The console window contains error details”. I went to the preferences but could not find the console window. So I am lost again.

PJ

Bob:

? What Dave Sand told me to do solved most of the problems but I am sill getting Error Messages. I have send a new Debug Info.

PJ

开云体育

开云体育

Thanks Dave ? ? Now please someone else have a look. ? Can’t get to first base with this. ?I suspect it is something in the parameters or preferences or I have the wrong download?

Jerry:

The Arduini group (still on Y!) has the C/MRI interface described for a project and the MRCS site has product.

I've adapted the C/MRI I/O and their code to a number of Arduino boards though not the Mega. I forget just what I left on the Arduini photos and files sections...

Beware though the Mega and many other Arduino boards use 3v3 (3.3 volts to old timers) parts and there are strict rules on what can tolerate 5v... input to the analog pins is one biggie...

Jim Albanowski

P.S. You may contact me offline if needed.

<snip>

开云体育

I have built a physical panel with buttons and leds, using a CML Electronics DMT30 board. I want to use it in Slave mode, where it sends sensor messages in response to button presses, and I have written Logix to respond to those messages, to switch points and signals. These Logix's (plural of Logix?) represent the interlocking between points, signals and track circuits, and this works well.

However, there are a couple of applications where I wanted to use the 'use turnout feedback' feature of the DTM30 turnout cells to indicate when a function is complete, and for this I wanted to generate feedback messages.

The more I think about this, the less sure I am that I am going down the right path...

So, thanks for your time, but don't worry about it anymore!

Dave

See below:

On 1 Sep 2018, at 1:58 AM, Richard Knight <enginman@...> wrote:

Dave, Fine, Is it possible for you? or someone to populate the function labels the same way all the others populate the Function labels? Such as when I Identify a TSU, TSU2,TCS, Digitraxx, QSI or most any other decoder I read and have the function labels appear ?

Possible to load one generic set, as I once did but later removed.

Matt seems to be using a standard format for the Full Throttle files for Functions and function labels. I know about the free form of the sound slots but there is a pattern to what Matt is doing.

Matt convinced me to use a "standard" set some years ago and I did. As soon as Full Throttle came out they were not correct. Straight away we had a problem; there's no way JMRI can currently tell whether a project is FT or not. Then came the proliferation of third-party project developers, particularly in Europe and Australia, each with their own custom set. I was getting more and more user complaints on both this and the LokSound lists about decoders not behaving as per the Function and Sound labels in JMRI, so I took the decision to supply no defaults and instead make the user do their own homework. The "standard" set was too often misleading users, not helping.

Some time later a translator "helpfully" added F0 lights, F1 Sound & F2 Horn. He was from Europe and that's the European "standard"! I pointed out that the USA "standard" is F0 lights, F1 Bell, F2 Horn and F8 Sound, while the Australian "standard" doesn't have Bell, F1 is often repurposed for another function and some projects separate Sound Startup and Volume/Mute to different keys (say F6 and F8).

I compromised by having F0 lights & F2 Horn in all defaults and F1 Sound in German.

I'm not happy to mislead many users in order to make life easier to one particular group. JMRI should never mislead, if it doesn't know it should remain silent, not guess.

I'd like to implement user-editable Sound Labels, but that's not at the top of my JMRI ESU priority list. My top-priority one (reading the Sound Project name from the decoder) has been sitting there for four years already, despite being near completion then. My life has become much busier since then and attempts to get back are always being thwarted by even more urgent current priorities (one of which is the Essential Sound Unit, where there is a problem I am currently try to resolve with ESU).

I move functions, function labels, sound slots whatever, whatever I need to do. As to Decoder Pro Reading the Function Pane, I understand.

I'm glad you do.

The REAL point in this is that I went to the LOk website, downloaded the appropriate sound file, read the data of the decoder all was correct, loaded the sound file and ended up with a non Full Throttle sound file. How does that happen. This is the second or third time this has happened and this is before evr swtichuing to Decoder pro. Rick.

I have a few ideas how this happened:
- One possibility is that you didn't have "Overwrite Defaults with Current Values" set in LokProgrammer when you loaded the Project (you need to load sounds, CVs and Defaults). Otherwise a simple decoder reset will return function mapping (and everything else, including Project Name) to whatever was previously loaded in the decoder.

- However, I've just noticed what I now surmise is your particular problem; I noticed "downloaded the appropriate sound file, read the data of the decoder all was correct, loaded the sound file and ended up with...". That's almost certainly your problem, your "read decoder data" overwrote all the FT settings (including Function Mapping, Project Name, etc.) for the new FT project with whatever was appropriate for the old (probably non-FT) project previously in your decoder.

Dave in Australia

ok, fixed SSL logic issues and the 8 extraneous errors are gone. Original 2 on NT14 remain.
Panel icons loaded along with updated panel.

Allen

Thanks Ken, I get that people like the reminders, and I know you have enough to do, but as much as anything it's the crude format that makes the buttons too big to fit on the throttle that makes them a problem.

I guess that part of the software is pretty basic? With a decent format you could have numbers and labels. Just putting out my wishlist. ;-)

Peter


On 1/09/2018 10:37 a.m., Ken Cameron wrote:

It is all a function of the roster entry. Many people like the names or
icons instead of numbers because they don't have to remember what is what.
When you have a wide mix of decoders, particularly mixing US and EU based
vendors, the 'usual' function code isn't the same. And on some decoders you
don't have the option of universally mapping the sound to all functions. So
using names says you don't have to care that on one loco the dynamic brake
is F5 but some have it on F6 or F9.

---
This email has been checked for viruses by AVG.

Hi!
I am trying to program this decoder.
I keep getting? a 308 error, "No Ack From Locomotive"
The programmer is a DCC++ Command Station.
For every write, the light on the loco flashes.
JMRI tries 3 times to write it & moves on.
The flaked out PR4 did the same thing on this decoder.
The engine will not run either.
This is a new replacement from Digitrax.
Is there a timing issue in JMRI that does not wait long enough for a response?
I am stumped.
Jay

The number of backup files will not impact the performance of JMRI.? How large is the backup folder?? And how much disk space is available on your PC?

Dan

Dave, you nailed it, no system prefix in the configuration for the CMRI
connection. The LocoNet is the normal prefix of "L".

PJ,

So you will need to enter C in the system prefix of the CMRI connection.
Save and restart, you should do much better.

-Ken Cameron, Member JMRI Dev Team
www.jmri.org
www.fingerlakeslivesteamers.org
www.cnymod.com
www.syracusemodelrr.org

Peter,

It is all a function of the roster entry. Many people like the names or
icons instead of numbers because they don't have to remember what is what.
When you have a wide mix of decoders, particularly mixing US and EU based
vendors, the 'usual' function code isn't the same. And on some decoders you
don't have the option of universally mapping the sound to all functions. So
using names says you don't have to care that on one loco the dynamic brake
is F5 but some have it on F6 or F9.

Now if you are using phone throttles, I think one or both versions have a
settings option on the phone to use or ignore those labels.

-Ken Cameron, Member JMRI Dev Team
www.jmri.org
www.fingerlakeslivesteamers.org
www.cnymod.com
www.syracusemodelrr.org

Headed to an Op Session. I'll get back to you tomorrow. Thanks for your, Bob and Ken's help so far.

PJ

I have no intention to eliminate any feed-back, the system provides all the feed-back information as before under the conditions stated, achieving this with a reduced number of detectors sensing mobile decoders, not effecting others.

PJ,

Dashes are fine in user names, along with just about anything else.

What I noticed were the errors like the following:

2018-08-31 13:56:38,552 cmri.CMRISystemConnectionMemo ERROR - invalid system prefix in CMRI system name: T1 [AWT-EventQueue-0]

Normally the system prefix for CMRI turnouts, etc., is the letter C. The second letter is normally the type of item, such as T for turnout. As Ken mentioned, recent releases have been improving the validation and enforcement of system name standards.

I just did a quick test and I noticed a very interesting bug. It is possible to have NO prefix assigned to a connection. Check your CMRI and LocoNet connections and make sure that the Connection Prefix is set to C and L respectively. If the C is missing, that would explain T1 which probably should be CT1.


Dave Sand