Russ,
I've found it needed to keep the following types of programming connections
available:
1. Right from the command station (important for Tsunami 2, no booster)
2. Via a booster like the PTB-100
3. With a 1K across the programming outputs
4. With a 10K across the programming outputs
I try each of these until something works. If I'm seeing the loco twitch,
but feedback fails, I figure I have to write blind to that decoder.
-Ken Cameron, Member JMRI Dev Team
www.jmri.org
www.fingerlakeslivesteamers.org
www.cnymod.org
www.syracusemodelrr.org
|
Marc,? I checked trainboard.com, but it looks to be more for the hobby in general.? No 308 issues similar to mine.? When I first saw the failure of the Read full sheet, I had the motor disconnected, but subsequently reconnected when you mentioned the possibility of a current related issue.? I reconnected the motor and still get the same behavior.? Do you think that resistor across the connection might would be useful?? I don't know what value to try, but it would increase the current load.
If you have any suggestions for any other sites or boards I can post on for assistance, that would be great.
Thanks,
Russ
|
Locked
Re: java Exception when try to add a second internal sensor
Steve, Andy,
I am still working on this issue. ?The scope has expanded. ?
Start JMRI with CBUS Simulator, open the Sensor table, select Add, select MERG, type the letter a. ?The same java exception that occurs when adding a second internal sensor.
Current issues: - CBUS system name validation is not handling illegal system names properly. ?This is the actual problem.
- A residual listener is causing the default connection (first listed) to be invoked before trying the the selected connection.
- When the connection is changed, such as from Internal to MERG, the previous value should be reset to empty.
Dave Sand
----- Original message -----
From: Dave Sand <ds@...>
Subject: Re: [jmriusers] java Exception when try to add a second internal sensor
Date: Tuesday, November 26, 2019 2:54 PM
Steve,
I have identified a possible cause for the unexpected CBUS validation. ?There is a listener left when the first Add window is closed which is triggered when Add is selected again. ?When CBUS is the FIRST connection, it gets called to validate the saved input from the first add. ?The previous value is reused to facilitate entering multiple similar names.
If LocoNet, NCE, etc. are the FIRST connection, they get called instead of CBUS, but their validation is not as comprehensive as CBUS so the error does not occur. ?Creating a virtual LocoNet connection as the first connection is a temporary work around. ?A review of Preferences >> Defaults would be required.
I still need to identify if this is the true problem and figure out a solution. ?Another issue is that there is a change to validation in progress which may or may not affect this particular issue.
Dave Sand
----- Original message -----
From: "steve young via Groups.Io" <icklesteve@...>
Subject: Re: [jmriusers] java Exception when try to add a second internal sensor
Date: Tuesday, November 26, 2019 10:39 AM
Hi Andy,
Thanks for reporting, have been able to reproduce the same exception.
Will need to do a bit of digging around why any sort of Internal Sensor validation is going via CBUS Sensor validation,
Steve.
|
I have expanded the content of the zip file. ?It now includes a panel xml file which uses just a Layout Editor panel formatted to look like a typical Panel Editor CTC panel.
To simplify panel development, the required icons were copied to the user files location resources/icons. ? ?
Dave Sand
----- Original message -----
From: Dave Sand <ds@...>
Subject: Re: [jmriusers] PanelPro CTC tools
Date: Friday, November 29, 2019 7:14 PM
This is a complete profile with panels, demo content, the tutorial PDF file. ?If the file is unzipped and the resulting profile is placed next to existing profiles, the JMRI profile selector should see it. ?If you have the selector disabled, use Preferences >> Config Profiles to find it.
You can also just copy the panel xml files, the PDF document and the ctc directory to an existing profile.
Dave Sand
----- Original message -----
From: John Pearson <pearsonjohn99@...>
Subject: Re: [jmriusers] PanelPro CTC tools
Date: Thursday, November 28, 2019 1:38 AM
David
Me to?
Looking forward to your CTC ?tutorial
Thank you?
John Pearson
On 27 Nov 2019, at 22:02, Dave Sand <ds@...> wrote:
?
Nick, Greg,
The CTC tool is completely functional. ?It will generate all of the required sensors, a basic panel and the logic to put it all together. ?The basic prerequisite is a functional JMRI ABS signaling implementation.
I can put together a tutorial over the next several days. ?I will probably release it in phases.
Dave Sand
----- Original message -----
From: "Greg McCartney via Groups.Io" <gregmcty@...>
Subject: Re: [jmriusers] PanelPro CTC tools
Date: Wednesday, November 27, 2019 2:00 PM
Hey Nick,?
? ?I have tried to use the tools myself but I am not certain it is complete.? I haven't asked if it was complete, just got the impression after multiple times of trying to use it over different versions of JMRI.? I did get it to draw a panel once but it seems to be very hap hazard at this point.? I hope that work continues on it because it looks like an awesome product.
Greg McCartney
|
Hello,
This email message is a notification to let you know that the following files have been uploaded to the Files area of the [email protected]
group.
Uploaded By:
Dave Sand <ds@...>
Description:
A tutorial for getting started with the new CTC tool. Unzip the file and move the resulting profile and its contents to the JMRI profiles location. Avoid putting this profile within an existing profile. This is an updated version which adds a Layout Editor formatted CTC panel.
Cheers,
The 开云体育 Team
|
Conrad, Am 30.11.2019 um 15:50 schrieb conrad: For Klaus: My apologies for not responding to your PM of 11/13. Your RDC mod worked very well. I just got frustrated as I could not replicate your fix. I tried and tried with no success so I took a break from VSD and just ran the trains. My apologies again. Sorry for the confusion. Yes, I modified the RDC sounds. I did that as usual by cutting the sound files at a zero crossing. Then I sent you the files. But later I noticed the real reason for the clicks. For short: the RDC files had a wrong length value, which cause the click. Audacity can correct this very easily: open the file, save the file. This doesn't change the file length, but it does correct the data length. Then I sent you this simpler recipe. This background might explain why you can't replicate it. On the wrong-length-error (not an JMRI error): Matt Harris, author of SoundPro and much more, fixed the error in 2016. Now we have found an example produced by the old error. Klaus
|
Phil
For lots of VSD info search for VSD on this site.? Klaus is really the goto VSD guy and is very, very helpful.? I think he is in Germany so there is a time delay.
For notching I use speed tables to try and match the speed to the notch sounds.? Thus for switchers I'm going real slow at a speed step of say 50.? The 0,33,67,100 was just a first guess.? One thing to understand is that the accel/decel notch transitions in VSD have a fixed momentum value built in.? So a quick drop of the throttle will "slowly" step the notches down from say notch 5 to idle.? Therefore I try to match this with the loco's decoder momentum settings.?
For your S-1 I have initially set the engine volume to 0.2 and it seems to be a reasonable setting vs. the bell and horn volumes.
I'm an NTraker and don't have an S-1 but I love the sounds of the ALCO 539 in your S-1, and the 539T (turbo) in my S-2 and the 12-251C (also turbo) in my RS-11.? The RS-11 prototype uses a V12 251B turbocharged diesel while mine uses the 12-251C from a C420 loco.? I'm not sure of the difference.
Conrad
|
Hey, Conrad. Thanks for chiming in.? If, on my layout, I even got to speed step 30, I would be violating the speed limit.? I set this up for switching, and it's unlikely to get above 15. That's why I had the settings that I had.? When starting off from 0, the loco would notch up when it hit speed step 2 (or 3, I don't remember now), similar to what you suggested.? But I want to notch the prime mover up sooner than 30, otherwise I would always be in Notch 1.?
I am working on an Alco RS-3 VSD now, and am having some issues.? Next week I'll be able to spend more time and see what's wrong with my work.
Phil in gorgeous Young Harris, Georgia, USA
|
For Phil:
The diesel sounds good to me. One small comment on the idle file. As you said, I do hear the “pop-off” and right after a non-voiced tick. Is the tick an artifact of the file? It’s not really a problem but it is there.
On notching, I think on previous VSD post someone pointed out that a loco will move (depending on train load) with the diesel at idle. That is the diesel remains at idle (notch 0) and the field windings of the generator are energized thus powering the traction motors. This being said, I modified your notch settings to 0-33, 33-67 and 67-100.
For engine gain, a check of several config files shows :
<sound name="ENGINE" type="diesel3">
<fade-in-time>0</fade-in-time>
<fade-out-time>0</fade-out-time>
<gain>0.8</gain>
<start-sound>
So this is why I could never change the level of the engine sounds, interesting. Need to mod config for <engine-gain>0.8</engine-gain>.
Finally, you have found the VSD guru – Klaus Killinger!
For Klaus: My apologies for not responding to your PM of 11/13. Your RDC mod worked very well. I just got frustrated as I could not replicate your fix. I tried and tried with no success so I took a break from VSD and just ran the trains. My apologies again.
Conrad
|
Am 30.11.2019 um 01:07 schrieb Phil in gorgeous Young Harris via Groups.Io: Thanks Klaus.? I worked with my audio files (as you thought, the original was 44100, not 47k like I said, and they were stereo which may have been the real problem), added some more notches (I guessed at what it would sound like), made Accel and Deccel sounds, added those new notches to my config and tried again.? A new file is uploaded to ProblemsBeingWorkedOn/Phil Abernathy with a small write-up of the loco.
You're welcome, Phil. Sorry, I forgot to mention that the files have to be mono. You can use stereo files, but not for the engine sound. All of these sounds (start and stop, notches, accel, decel sounds) must be mono and must have the same frequence. That's all correct in your uploaded VSD files. The file size should not prevent you from using 16 bit files (it is an advantage of a computer to handle large files). Try e.g. EMD_645_Turbo.vsd to hear the difference. But it's harder to get good 16 bit sounds on the web! And you're right, in the end the typical sound character is important. I am still working on how to make the prime mover stop/start when I send an F8 (mute) function (I have it auto start now).? In fact, when I review my config file, I don't see what Starts or Stops the engine (I use the button on the VSD Manager Pop-up, but do not see that button mentioned in the file).
If I understand this right, you want to be able to start/stop the engine by a throttle function key, e.g. F8. You can do this in config.xml: <trigger name="ENGINE_STARTSTOP" type="THROTTLE"> <event-name>F8</event-name> <target-name>ENGINE</target-name> <action>NOTHING</action> </trigger> This must be declared in the tag sound-event name="ENGINE". I would love for someone familiar with the prototype to weigh in on the sounds though: I just guessed. Unfortunately, I don't know the prototype. Maybe somebody else will respond. Maybe you can find some examples on YouTube. I am hoping to try this at the Club within the next 2 weeks, along with my new SQ23 camera. That sounds good. There may still be time for fine tuning. Klaus
|
Locked
Re: Sluggish native panel response time compared to web
Randall Thanks for the background - I should have know that this problem was above my pay grade! Based on your suggestion I ran Windows within VirtualBox (the nearest I have to a Windows machine) and installed v4.16 JMRI to do a comparison. As you say the response is different - the windows native panel offers a more immediate response. We're only talking fractions of a second here, but it is noticeable. This is even though it is operating as a VM within the same desktop machine. The VM is running:
Apart from the OS, the version of Java is different between the 2, so I'm not sure if that has an influence also - more experiments required! Thanks for your help Andy
|
Thanks Marc, I'll check to see if there's a group out there with familiar with what settings I might be able to modify on the DCC++ side.
Russ
|
Dave, he is using DCC++? and it looks more and more like his DCC++ system has issues with reading the current draw and programming on Program Track is functional.
Marc? .
|
Russ, On 30 Nov 2019, at 2:43 PM, Russ Tom <russ.a3@...> wrote:
Sorry for the confusion. Through DecoderPro, I run an instance of a throttle launched from within DecoderPro on my Windows 10 PC.
When using the programming portion of DecoderPro through the programming track, that’s when the odd read behavior occurs. You still haven't grasped the problem. If you have a computer running DecoderPro, it can't magically talk to a loco on your layout to drive and/or program it. You must have some brand/model of command station and booster to enable you to drive trains or to use DecoderPro. You can't just plug two wires into some place on your computer and connect them to your track! You have to have some sort of DCC system for powering your trains/driving them/connecting to the computer. Please go to DecoderPro's Edit->Preferences->Connections screen and it will have some information about the brand/model of command station you are connected to. Maybe someone else can explain this better to Russ. Dave in Australia
|
Dave,
Sorry for the confusion. Through DecoderPro, I run an instance of a throttle launched from within DecoderPro on my Windows 10 PC.
When using the programming portion of DecoderPro through the programming track, that’s when the odd read behavior occurs.
thenks,
Russ
|
Locked
Re: Webserver layout editor panel showing incorrect block occupation for a crossover
|
Locked
Re: Arduino, JMRI, and relay-controlled turnouts
Ah ok.?
j - Jared F. Nichols bmwftw@...
Pro
toggle quoted message
Show quoted text
On Nov 29, 2019, at 22:13, Steve Todd <mstevetodd@...> wrote:
?Jared, be aware that this forum strips attachments. You'll need to post it in the files section (or elsewhere on the web) and include a link.
|
Locked
Re: Arduino, JMRI, and relay-controlled turnouts
Jared, be aware that this forum strips attachments. You'll need to post it in the files section (or elsewhere on the web) and include a link.
|
Locked
Arduino, JMRI, and relay-controlled turnouts
Hi all-
I’ve been at this for a bit, so I thought I’d just throw it out there and say that if you’ve got twin-coil type turnout motors (like the Peco PL-10 or PL-11) and you’re trying to drive them with relays and use JRMI/Arduino to control them, I’ve got the solution for you. Basically, you’ll need a spare Arduino and utilize CMRI. I’m attaching the sketch I made for it.
If you’ve been struggling with this or are curious, let me know and I’ll be happy to share what I’ve learnt.
j
-
Jared F. Nichols
bmwftw@...
Sent from my iPad Pro
|
Locked
Re: MIld panic: all ID tags appear to be gone
Paul,
Do you think this change will make it into the Dec production release?
Regards,
Allen
|
Ken Cameron
conrad
George Warner
ahouse3761