开云体育

Date

Locked Re: PowerCab issue - locomostive continues to crawl after speed 0

 

开云体育

Don mentioned knobs on his EasyDCC system. Reminds me of similar problems the Silk City Model Railroad Club has with their Digitrax UT4D throttles. Several of the throttles have the tendency to allow locos to crawl when the knob is turned all the way down. The solution is to turn the throttle up ever so slightly, then the loco stops. Clearly a physical problem with the potentiometer connected to the knob. The DT400D throttle doesn’t suffer this, presumably because it has encoders instead of potentiometers.

Mark Granville


Locked Re: PowerCab issue - locomostive continues to crawl after speed 0

 

开云体育

He also has a spare PowerCab that is v1.65 and has decided to take it home to try it with that.? I wanted him to leave it with me for another day to debug the issue but unfortunately he decided otherwise.? I would have liked to have tracked down the issue.? I may be back on this subject later if/when he has problems down the road.

Sorry about that.

Scott


Scott Jay

Summerside, Prince Edward Island

On 09/05/20 3:10 a.m., Dave Heap wrote:
Marc,

On 9 May 2020, at 3:13 PM, forfoum@... wrote:

How about this is speed Step related  ?   14, versus 28/128  . JMRI virtual throttle and WiFithrottle are known to have issues with speed step option.
I was actually thinking along the same lines...

- I discovered that both V1.28 and V1.65 Power Cab firmware have a known occasional issue with 28 speed step commands via the USB interface. The NCE documentation suggested to always use 128 step commands.

JMRI 4.19.5   now has an option to select between 14, 28, 128  for the JMRI virtual throttle.
- I thought that was available for JMRI Virtual Throttles but the option seemed to be gone for me in 4.19.5. We don't know what version Scott is using.

- Likewise I couldn't see how to do this with WiThrottle or Throttle preferences.

Everything was being sent as 128 step commands. Maybe this is to avoid the known Power Cab bug, but it seemed to be the same with other hardware profiles I checked.

The fact the E-stop does not work sort of points to a speed step setting.
Agreed.

But I did suggest to Scott that he posted a copy of the NCE Command Monitor output so we could see what is being sent. 

As an aside, let's not forget to debunk they mythology surrounding 14/28/128 step modes:


Dave in Australia






Virus-free.


Locked Re: Displaying the "Alternative Track Colour" when using Logix (or Lroute or anything else that works) #layouteditor #logix

 

I get three track colours in Logix by using a multiway sensor. You can define three different icons for each position of the sensor. I have grey, yellow and red track sections defined ?and by turning on the required sensor and the others off can change the colour of the track. My panels are somewhat basically programmed, with almost all done in Logix because I wanted full control over all the addresses to allow me to take modules in and out of the layout, and keep addresses for each module in set number ranges.

Mike Ruby


Locked Re: Can JMRI Issue a BASH command to set STTY?

 

Just for the record, regarding my script for a delayed setting ttyS0, I had added this to the /usr/local/bin folder with permissions changed for me to run as user, then had added it to the startup for the session.? I had tried to run it inline with PanelPro, but the "sleep" makes everything wait in the script rather than concurrent.? I have other ideas to try now though.
'
clear
sleep 10s
echo "Setting baud rate"
stty -F /dev/ttyS0 38400
echo " "
echo "Here is the speed"
stty speed
echo "."
'
Phil


Locked Re: Can JMRI Issue a BASH command to set STTY?

 

On Sat, May 9, 2020 at 07:40 AM, Ken Cameron wrote:
My question is why is the command needed? Why can't you set the right baud
rate in the connection config?
Ken, I don't know why but I have been unsuccessful.? From the JMRI help with MS100:
"... https://www.jmri.org/help/en/html/hardware/loconet/MS100.shtml
Linux hack to set the baud rate
Some users have reported success using Linux commands to directly set the baud rate while the program is running. To do this:
??? Start DecoderPro, and wait until you see the error message that the baud rate could not be set.
??? Run the following commands from the Linux command line:
??? setserial /dev/ttyS0 divisor 7
??? setserial /dev/ttyS0 spd_cust
??? stty speed 38400 < /dev/ttyS0
??? Then click on "OK" and continue on.
This has been reported to work with both RH9 and FC3..."

So I tried the process of opening a terminal window and issuing the command again.? On one computer (an Intel SR2400 Server with dual Xeon 3.2 GHz CPUs with 1 gig of RAM running Ubuntu 18.04.4 LTS and JMRI 4.19.3/Java 11.0.6 ), I do not have to send another command for the MS100 to work.? Just the Dell laptop.? With all my various searches, I have not found any conf file or setting that would hint at fixing this.

The laptop is not connected to the network (that has Internet), so I will retype the parts of the BASH PanelPro script ("PP"), leaving out the echo statements I added for screen messages.? I used 'sudo nano PP' and afterwards performed a 'sudo chmod 777 PP' so I can run it as a user.
'... cd ~
PATH="$HOME/Downloads/Java/jre/bin:$PATH"
cd ~/Downloads/Railroad/JMRI
./PanelPro '

I had downloaded and extracted Java and have crudely added to the PATH to get it to work.? If I have extracted it to the right folder (maybe /usr/local/bin where this "PP" file is) it would have worked better.

I have several ideas provided above, and I will try them later.? Thanks!

Phil in gorgeous Young Harris, Georgia, USA

ps, not even one admonishment for resurrecting a dinosaur!? And not one saying Buy a this or that instead!? I'm impressed.


 

Nice one!

You JMRI folk are way too efficient haha
--
Thanks
Mitch


Locked Re: Classification yards #operationspro

 

Phil, you are correct.

A copy of an existing route will be considered as a different route with regards to the C/I track default.? Note that you can always override the default by telling the program which routes or trains are allowed to pull or set out cars to a C/I track.? Better to use "Yard" track type in this case since there isn't any kind of sorting needed.

Dan


 

Mitch
Yep.
Im testing it now.
Steve G.

---------- Original Message ----------
From: "Mitchell via groups.io" <mitchell.scott93@...>
Date: May 9, 2020 at 1:05 AM


Hi Steve,

Did you end up finding out about the stopping speed bug?
--
Thanks
Mitch



Locked Re: PowerCab issue - locomostive continues to crawl after speed 0

 

Don't be too quick to rule out other problems. You might use your NCE controls more, watching for the same creeping problem. If it's an NCE problem, JMRI changes won't be able to fix it.

Of late, locos on my EasyDCC system railroad are sometimes creeping instead of stopping as before. Reversing direction then has no effect. This is happening while using the EasyDCC's plug in walk around throttles. I haven't yet solved why. Several locos do it, with more than one throttles, perhaps all. And, how rapidly or far the throttle knob is rotated to zero seems to matter: fast or farther is better. I do have a Raspberry Pi 3B for JMRI connected to the EasyDCC Command Station's RS-232 port, but this problem happens even with the Pi unpowered.

Don Weigt
Connecticut


Locked Re: NEW USER

 

Hi Peter

To define how roster entries and conflicts are managed between JMRI and your ECoS, you have 2 ways to do it:
  • Either from DecoderPro Main menu: Edit>>Preferences>>ECoS Preferences;
  • Or from DecoderPro Main menu: EcoS >> ECoS Preferences
To look at your ECoS Database, from DecoderPro main menu: ECoS >>Ecos Loco database. This helpful view tells you entry by entry which ECoS engine matches which JMRI roster entry
--
Alain355


Locked Is there any way to make JMRI throttle to use 32 functions instead of 28 #jmrithrottle #functions #esu-decoders

 

My Loksound 5 use 32 functions.


thanks,
James


Locked Re: PowerCab issue - locomostive continues to crawl after speed 0

 

I will get back to you a little later in the day.

Thank you for all of your help.

Scott Jay

On May 9, 2020, at 3:10 AM, Dave Heap <dgheap@...> wrote:

?Marc,

On 9 May 2020, at 3:13 PM, forfoum@... wrote:

How about this is speed Step related ? 14, versus 28/128 . JMRI virtual throttle and WiFithrottle are known to have issues with speed step option.
I was actually thinking along the same lines...

- I discovered that both V1.28 and V1.65 Power Cab firmware have a known occasional issue with 28 speed step commands via the USB interface. The NCE documentation suggested to always use 128 step commands.

JMRI 4.19.5 now has an option to select between 14, 28, 128 for the JMRI virtual throttle.
- I thought that was available for JMRI Virtual Throttles but the option seemed to be gone for me in 4.19.5. We don't know what version Scott is using.

- Likewise I couldn't see how to do this with WiThrottle or Throttle preferences.

Everything was being sent as 128 step commands. Maybe this is to avoid the known Power Cab bug, but it seemed to be the same with other hardware profiles I checked.

The fact the E-stop does not work sort of points to a speed step setting.
Agreed.

But I did suggest to Scott that he posted a copy of the NCE Command Monitor output so we could see what is being sent.

As an aside, let's not forget to debunk they mythology surrounding 14/28/128 step modes:
<>

Dave in Australia




Locked Re: Can JMRI Issue a BASH command to set STTY?

 

My question is why is the command needed? Why can't you set the right baud
rate in the connection config?

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


Locked Re: How to get decoder pro to work on an imac running OS catalina, a Locobuffer-USB and a digitrax DCS51 command station #mac-os #digitrax #locobuffer-usb

 

I did use the VCP driver,although it was the same as the original one I used from RR-Circuits - FTDIUSBSerialDriver_v2_4_2.dmg.
There was no prompt during the installation about setting security settings - this had to be set in Apple/
System Preferences/Security and Privacy/ under the "General" column. There was no indication anywhere that this was necessary, hence my original problem.
I guess Apple Mac Catalina experts would know this ?

So, yes I think it would be a good idea to update the instructions, may save others a lot of problems.

regards

Robert


Locked Re: Displaying the "Alternative Track Colour" when using Logix (or Lroute or anything else that works) #layouteditor #logix

 

Dave,

Thanks for the advice. I had been using Sensors for everything - since up until now they were really only on/off states.? Clearly I need to store the state of the Block - but that can be evaluated either as a Sensor or a Memory and since setting the state is not going to trigger some other action (as a Sensor could) it is better to use a passive Memory rather than an active Sensor.

More thoughts for the design process. I am still considering the implications of process of "reserving" Blocks and tossing around ideas of the need to actually have a further state of "locked" after the signal is cleared and until the Block occupancy sensor has reset.? I certainly will need "normal" and "reserved" (reflected in thrie display colours).

Why do I need locked?? Is it to prevent interference after the route is fully evaluated and set, or is there another purpose?? If only the first I am not sure that my small layout really needs it.

If you have time to comment or point me at some JMRI or other documents please do.? Your help is always to the point and appreciated.

Thanks again,

Iain


Locked Re: Can JMRI Issue a BASH command to set STTY?

Randall Wood
 

If your bash script is launching JMRI using one of the JMRI provided launchers, be aware that your bash script will wait until JMRI has exited before running the next command unless you explicitly put JMRI in the background.

You may need something like:

/path/to/JMRI/panelpro &
sleep 30
stty -F /dev/ttyS0 38400

Note the ampersand (&) after the call to JMRI.


Locked Re: AbstractAutomaton and stop() #scripting

 

Hi Bob,

While I'm sure nothing I add here will be news to you, I'm adding it in case someone else wants the information.

I have now carried out some testing, and I can confirm that your code snippets work.

So far, I have implemented a few main sections, which perform the following:
  1. Stop all locos, by grabbing the throttles one at a time and setting the speed to either 0 or -1 (a mix, which is setup in a list in the init() section of this class)
  2. Find all the threads that should be stopped
  3. Stop all those threads
  4. Perform some tidy up work on internal sensors
  5. Turn off the DCC power
When it came to the section on stopping the threads, I ended up with code along the lines of this:

? ? # This is section 2 from above
? ? for i in range(jmri.jmrit.automat.AutomatSummary.instance().length()):
? ? ? ??ThisName = jmri.jmrit.automat.AutomatSummary.instance().get(i).getName()
? ? ? ??if ThisName is not None:
? ? ? ? ? ??if "ClassNameExample1" in ThisName:
? ? ? ? ? ? ? ? KillList.append(ThisName)
? ? ? ? ? ? elif "ClassNameExample2" in ThisName:
? ? ? ? ? ? ? ? KillList.append(ThisName)

? ? # and this is section 3
? ? for i in KillList:
? ? ? ??jmri.jmrit.automat.AutomatSummary.instance().get(i).stop()

The code might not be the best (and if you have any suggestions I'd take them on board), but it seems to work.? When it is run, the threads are removed from the Thread Monitor, so I assume they are completely removed from the system.

I found that I needed to construct a list of thread names in the first pass (KillList), then kill them with stop() after I that, because if I tried killing the thread in the loop that was looking for the correct ones to kill, I was receiving errors about indexes out of bounds, which I *think* must have been because the list of threads can't be altered once you are in the "for i in range(jmri.jmrit.automat.AutomatSummary.instance().length()):" loop.

While I still need to test a bit more, and implement some more functions, so far it seems to work satisfactorily.? I have the emergency stop code executed as a result of a Logix, which is tied to an Internal Sensor and a Sensor Icon on my Panel.? Eventually I will investigate mounting physical buttons around the layout, tied to a real input sensor, so I can activate this emergency stop function from anywhere, not just when I'm at the PC screen.

Thanks once again for your help.

Cheers,
Trevor


Locked Re: Can JMRI Issue a BASH command to set STTY?

 

Hi Phil,

If you paste your bash script in here, the one you tried to use at start-up but didn't work, then I'll be happy to take a look and see if there's anything amiss. Also needed: where it is called from (the context of the call, if you will). I expect that to be a second script with one line calling yours - that script I'd like to see as well. If you've combined the two into one, then - fine, just the one.

Wouter

On Sat, 9 May 2020 at 04:33, Phil in gorgeous Young Harris via <philabernathy=[email protected]> wrote:

My question: Can JMRI Issue a BASH command to set STTY?

I am using a 9-pin MS100 connected to my 9 pin serial port on my Dell D610 laptop (with 2048 MB memory and a Pentium M #06D8 running at 1.86 GHz). I am running JMRI 4.17.7 with Java 1.8.0_211 on Ubuntu 18.10 (CodeName Cosmic). My Desktop is LXQt version 0.13.0. The LocoNet cable goes to my Command Station, a Zephyr DCS50.

To make this work (and it works well), as discussed in the JMRI help page for MS100, I have to first start JMRI, then open another Terminal and issue a command 'stty -F /dev/ttyS0 38400'.? Until I do that extra step, I can successfully send Track Power On or Off to the Command Station, but JMRI always says that the Power is Unknown and nothing shows in the LocoNet Monitor.

After I send the extra stty command, the system works great, including Programming: for example, I can read from and write to my Tsunami TSU-1000 equipped loco on the DCS50 Programming Track.

I had thought that I could use my Bash script that starts JMRI and merely add that command after JMRI in the script, but that does not work.?

I had written a script that executed at startup that would issue that command after 10 seconds, but that did not work.? (Hmmm, maybe I need to make that 30 seconds?)

So I thought, maybe JMRI can send out a command to the Linux operating system using a script or Logix!? So I am asking for any thoughts that might help.? If JMRI cannot do this, my extra step can always be done, or some other Linux trick might be the solution..

I also thought (apparently incorrectly) that somewhere a conf file could be tweaked to fix the tty, but no Google or duckduckgo searches bore fruit.? I am a Linux newbie, so I rely on Google more than I would like.

I am just trying to make the system boot up, run JMRI and connect to the DCS50 by itself, and although I am making progress with the auto login and scripts at startup, with this stty thing I have completely failed.

Thanks for any advice or helpful suggestions.? I have three PR3's and an LNWI, but was given this MS100 and since my old laptop has the 9 pin serial port, I got it working.? I just would prefer it to automatically run when booted.

Phil in gorgeous Young Harris, Georgia, USA


Locked Re: PowerCab issue - locomostive continues to crawl after speed 0

 

Marc,

On 9 May 2020, at 3:13 PM, forfoum@... wrote:

How about this is speed Step related ? 14, versus 28/128 . JMRI virtual throttle and WiFithrottle are known to have issues with speed step option.
I was actually thinking along the same lines...

- I discovered that both V1.28 and V1.65 Power Cab firmware have a known occasional issue with 28 speed step commands via the USB interface. The NCE documentation suggested to always use 128 step commands.

JMRI 4.19.5 now has an option to select between 14, 28, 128 for the JMRI virtual throttle.
- I thought that was available for JMRI Virtual Throttles but the option seemed to be gone for me in 4.19.5. We don't know what version Scott is using.

- Likewise I couldn't see how to do this with WiThrottle or Throttle preferences.

Everything was being sent as 128 step commands. Maybe this is to avoid the known Power Cab bug, but it seemed to be the same with other hardware profiles I checked.

The fact the E-stop does not work sort of points to a speed step setting.
Agreed.

But I did suggest to Scott that he posted a copy of the NCE Command Monitor output so we could see what is being sent.

As an aside, let's not forget to debunk they mythology surrounding 14/28/128 step modes:
<>

Dave in Australia


Locked Re: JMRI & Digikeijs DR5000

Jay
 

Hi,
I have been in contact with Ryan at Iron Planet Hobbies.
It looks like the DR5000 GUI has an issue with the latest release of Win 10.
Latest release was yesterday.
I got it to work with JMRI with his video.

Jay