Welcome to the forum   
Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error

Share
Options
View
Go to last post in this topic Go to first unread post in this topic
Offline MikeR  
#1 Posted : 11 March 2014 20:49:01(UTC)
MikeR

United States   
Joined: 26/08/2012(UTC)
Posts: 263
Location: Denver
I have the following hardware and software on my layout

CS2 60215 running v 3.0.1(1) and hardware version 4.33
60830 K83 decoders (16 K83s installed) for turnout control.

Railroad&Co TrainController Gold version 8.0B5

My layout is under construction and the layout completed so far is defined in both RR&Co and in the CS2 "Layout" component. All of the microswitches in the turnout motors have been shorted out so that they are not operational.

Manual running of trains and throwing turnouts can be done from either the CS2 "Keyboard", "Layout" definition and "Memory" and operates correctly with updates to the other systems being correctly processed. Requesting individual turnouts to be thrown from the RR&Co TC Switchboard also functions correctly with updates also being correctly processed on all displays and the physical turnout being thrown.

The issue I have is that when I run a schedule from RR&Co TC I have random turnouts not being thrown during the reservation of a route. I have stopped the trains and identified that the CS2 "Keyboard" reflects the RR&Co TC requested state of the turnout but the physical turnout is not being thrown. When running a "Refresh" of the layout from RR&Co TC after running a schedule there are always 1 or more turnouts that change their state confirming that turnouts are not changing when requested. It appears that the CS2 is reflecting the change on its "Keyboard" but the instruction to the K83 is being lost or is not being correctly processed. The problem with the turnouts is random with different turnouts giving problems on an apparent random basis. It takes approximately 3 to 5 minutes after starting a schedule for the first misdirected train to become apparent.

Since I identified the problem I have installed a separate bus for all K83 devices and am using a separate booster for this bus. This has not overcome the problem. I have also increased the interval between turnout instructions from RR&Co TC to the CS2 from the default 300 to 700 milliseconds without overcoming the problem.

Has anyone had a similar experience? and if so how has the problem been resolved.

Thanks
Mike
Digital - C track with CS2 and Railroad&Co TrainController; feedback using LocoIO via a Locobuffer
Analog - M track with solid centre rail (after C track layout is complete)
Collect all Eras - especially Crocodiles
Member of ETE
Previously a member of the Marklin Modellers' Group Johannesburg
Offline clapcott  
#2 Posted : 11 March 2014 22:10:05(UTC)
clapcott

New Zealand   
Joined: 12/12/2005(UTC)
Posts: 2,433
Location: Wellington, New_Zealand
I have not exercised RR&Co, however as far as commands to the CS2 is concerned, the software is responsible for sending both a activate (start pulse) and a deactivate (stop the pulse).

This is a bit counter intuitive but when porting software ( from .. say a 6021 or CS1) the first instinct is to just send the two commands in a row thinking you are sending them to the CS2 "Manager", and that it would manage the pulse duration using its settings for that turnout. In fact you are sending instruction direct to the hardware.

Therefore you have to manage the delay (i.e. length) of the pulse within the software.

The "PC" Display part of the CS2 has no concerns about sluggish motors and will happily display the state of the last command. But the actual grunt needed to drive the turnout needs to be a minimum duration (suggestions range from 100ms for a free C-Track mechanism to 1000ms for M-Track solenoids)

I would suggest researching for a setting within the RR&Co options for one that allows adjustment of the turnout pulse duration
There may also be a second setting for the time between switch commands (of a route) - Too short a gap will load the power needs (although you mention a seperate bus which is good)

Note: as a corrolollary..
Because you have overridden the end cutouts , then a reduced "Time delay between turnouts" along with a "longer plse" might actually be reintroducing a load problem.

For example, if you have set the pulse duration to 1000ms but have a delay between commands of 500ms (or less) you may be draining power for 3 devices at once.

While having the affected turnouts on different k83s might help (because each has its own CD unit), attempting to do this with points from the same k83 will result in overusing the resources available. Even a CD unit with plenty of grunt, needs time to recharge when its empty.
Peter
thanks 1 user liked this useful post by clapcott
Offline MikeR  
#3 Posted : 12 March 2014 08:05:26(UTC)
MikeR

United States   
Joined: 26/08/2012(UTC)
Posts: 263
Location: Denver
Thanks Peter for your response. The following are my pulse settings on the 2 devices:

RR&Co TC 100ms pulse length
300ms timing delay between turnout pulses
CS2 200ms pulse length (the shortest possible on the CS2)

I am using C track and the standard motors - a mix of the old and new. I assumed that I should keep the TC pulse length short and let the CS2 override with a longer pulse. This may however be confusing the CS2. Do you think I should change all TC pulse lengths to 250 or 300ms, leaving the CS2 at 200ms?

I notice that when I switch a CS2 route from its memory it has a very short time delay between throwing the turnouts - probably about 100ms. I left the default setting of 300ms in TC but will try longer settings to see if there is a problem with the CD recharging.

I will also do a more detailed study to try and identify any pattern of the turnouts not throwing including previous turnouts thrown in establishing the route.
Mike
Digital - C track with CS2 and Railroad&Co TrainController; feedback using LocoIO via a Locobuffer
Analog - M track with solid centre rail (after C track layout is complete)
Collect all Eras - especially Crocodiles
Member of ETE
Previously a member of the Marklin Modellers' Group Johannesburg
Offline Danlake  
#4 Posted : 12 March 2014 08:24:52(UTC)
Danlake

New Zealand   
Joined: 03/08/2011(UTC)
Posts: 1,571
Hi Mike,

Sorry to hear about your issue with turnouts.

I am using 250ms for pulse duration and 250ms for turnout interval in RR&Co

I have not modified the turnouts and using Viessmann K83 (with seperate power).

So far no issues.

Maybe try and ask on the Traincontroller forum as well?

It was my understanding that a command sent from PC will not be modified by whatever settings you have on CS2 (e.g. a higher or lower pulse duration for a solenoids)?

Brgds - Lasse
Digital 11m2 layout / C (M&K) tracks / Era IV / CS3 60226 / Train Controller Gold 9 with 4D sound. Mainly Danish and German Locomotives.
thanks 2 users liked this useful post by Danlake
Offline MikeR  
#5 Posted : 12 March 2014 10:42:22(UTC)
MikeR

United States   
Joined: 26/08/2012(UTC)
Posts: 263
Location: Denver
Thanks Lasse. I have followed your advice and done a similar posting on the RR&Co forum.
Mike
Digital - C track with CS2 and Railroad&Co TrainController; feedback using LocoIO via a Locobuffer
Analog - M track with solid centre rail (after C track layout is complete)
Collect all Eras - especially Crocodiles
Member of ETE
Previously a member of the Marklin Modellers' Group Johannesburg
Offline witzlerh  
#6 Posted : 12 March 2014 20:28:39(UTC)
witzlerh

Canada   
Joined: 25/09/2010(UTC)
Posts: 417
Location: Sherwood Park, AB, Canada
My first thought was that you may have been approaching the load limits of your transformer and need another transformer/booster. Switches not being properly thrown is the first sign that you layout is getting underpowered. You do have a separate bus but does it have a separate power supply? You probably do but you can check on your CS2 if the current draw is on the high side during normal operations.

I also think that you need to stagger the switching. I am not sure if the software or the CS2 give the command to switch 2 or more switches at once. This will also cause a rapid voltage drop during the impule. I would have 150ms switching impluse but have a 300ms+ stagger time just incase the software/CS2 begins the 300ms inter-switch delay at the start of the 150ms impluse. This way any impluse is done before the next one begins.

I too have shorted out most of my motors for greater reliability.

Also a suggestion. Run several times more than you need to verify if the switching problem pattern is indeed a pattern or just random....Pattern problems are easy to fix. Random ones are tough.
Harald
CS2 DB & Canadian Era 3-6
Offline kbvrod  
#7 Posted : 12 March 2014 20:53:44(UTC)
kbvrod

United States   
Joined: 23/08/2006(UTC)
Posts: 2,597
Location: Beverly, MA
Hi all,
More junk turnout 'motors'?

D
Offline MikeR  
#8 Posted : 12 March 2014 22:05:43(UTC)
MikeR

United States   
Joined: 26/08/2012(UTC)
Posts: 263
Location: Denver
Originally Posted by: witzlerh Go to Quoted Post
My first thought was that you may have been approaching the load limits of your transformer and need another transformer/booster. Switches not being properly thrown is the first sign that you layout is getting underpowered. You do have a separate bus but does it have a separate power supply? You probably do but you can check on your CS2 if the current draw is on the high side during normal operations.

I also think that you need to stagger the switching. I am not sure if the software or the CS2 give the command to switch 2 or more switches at once. This will also cause a rapid voltage drop during the impule. I would have 150ms switching impluse but have a 300ms+ stagger time just incase the software/CS2 begins the 300ms inter-switch delay at the start of the 150ms impluse. This way any impluse is done before the next one begins.

I too have shorted out most of my motors for greater reliability.

Also a suggestion. Run several times more than you need to verify if the switching problem pattern is indeed a pattern or just random....Pattern problems are easy to fix. Random ones are tough.


Thanks Harald for your comments. I have a 3 amp toroidal transformer for my CS2 and I have been monitoring the current draw which is about 0.8 amp when no locos are being driven. This is after the new bus and booster powered from a separate 3 amp toroidal transformer had been installed. I do not believe that this is approaching the limit on the transformer. I have also tested on a 60101 switched power supply and the result is the same.

I have shorted all of the micro-switches on the turnout motors as this has been highlighted in the forum many times as a potential problem.

This afternoon I increased the time delay between turnout instructions from TC to 1000ms and this seemed to give much better results. Unfortunately I was not able to complete this test as my grandchildren arrived to play trains! I will do more testing tomorrow with the higher stagger/delay times between turnout instructions to the CS2.

Regards
Mike
Digital - C track with CS2 and Railroad&Co TrainController; feedback using LocoIO via a Locobuffer
Analog - M track with solid centre rail (after C track layout is complete)
Collect all Eras - especially Crocodiles
Member of ETE
Previously a member of the Marklin Modellers' Group Johannesburg
Offline MikeR  
#9 Posted : 12 March 2014 22:08:43(UTC)
MikeR

United States   
Joined: 26/08/2012(UTC)
Posts: 263
Location: Denver
Originally Posted by: kbvrod Go to Quoted Post
Hi all,
More junk turnout 'motors'?

D


Thanks D. I assume you are referring to the microswitch problem mentioned in a number of threads. These have been shorted so that this problem should be eliminated.

Mike
Digital - C track with CS2 and Railroad&Co TrainController; feedback using LocoIO via a Locobuffer
Analog - M track with solid centre rail (after C track layout is complete)
Collect all Eras - especially Crocodiles
Member of ETE
Previously a member of the Marklin Modellers' Group Johannesburg
Offline craigericks  
#10 Posted : 16 March 2014 15:34:54(UTC)
craigericks

United States   
Joined: 04/11/2003(UTC)
Posts: 262
Location: ,
Mike,
I had similar problem after I updated 60215 CS2 yesterday. After the update, my once reliable turnouts became erratic on routes or schedules that required a long sequence of turnout commands I saw your post on RR&Co forum and here. I took Lasse's advice, " I am using 250ms for pulse duration and 250ms for turnout interval in RR&Co." The change from 0 to 250ms in the turnout interval seems to have solved the problem. I notice a definite delay in the sequence of the commands which ensures all turnouts get the proper command.
Regards,
Craig



Marklin:CS2 and TrainController Gold v10
thanks 1 user liked this useful post by craigericks
Offline MikeR  
#11 Posted : 16 March 2014 21:26:28(UTC)
MikeR

United States   
Joined: 26/08/2012(UTC)
Posts: 263
Location: Denver
Hi Craig

Thanks for mentioning your experience. I have also tried Lasse's advice and set turnout pulse duration to 250ms on both CS2 and TC. While this has reduced the incidence of turnouts not being thrown, the problem persists.

It seems to be fairly erratic but is mostly the last turnout in a route. The turnout works fine from the TC switchboard and when I activate the turnout as part of the sequence of the defined route. I have also loaded this route into Memory of the CS2 and it works flawlessly.

I have checked my transformer supplying my K83 bus and the voltage drops by about 0,15 volt during the route activation from the CS2 memory. This is less than the 0.6v measured when using a Marklin transformer.

Will you please advise what setting you have in the CS2 set up on TC for the interval between turnout instructions. Mine is currently 1000ms.

Thanks

Mike
Mike
Digital - C track with CS2 and Railroad&Co TrainController; feedback using LocoIO via a Locobuffer
Analog - M track with solid centre rail (after C track layout is complete)
Collect all Eras - especially Crocodiles
Member of ETE
Previously a member of the Marklin Modellers' Group Johannesburg
Offline MikeR  
#12 Posted : 16 March 2014 21:31:50(UTC)
MikeR

United States   
Joined: 26/08/2012(UTC)
Posts: 263
Location: Denver
Hi Craig

I forgot to mention that my problem started with the old versions of both TC and CS2 software. Upgrading to the new versions this week has not overcome the problem.

Mike
Mike
Digital - C track with CS2 and Railroad&Co TrainController; feedback using LocoIO via a Locobuffer
Analog - M track with solid centre rail (after C track layout is complete)
Collect all Eras - especially Crocodiles
Member of ETE
Previously a member of the Marklin Modellers' Group Johannesburg
Offline craigericks  
#13 Posted : 16 March 2014 21:44:42(UTC)
craigericks

United States   
Joined: 04/11/2003(UTC)
Posts: 262
Location: ,
Mike,
In digital system setup for TC for the CS, I have set a turnout interval of 250 ms. For the turnouts that gave me persistent problems, I put, in TC, a switch time of 250ms. I did not change from the default times in the CS itself.

Craig
Marklin:CS2 and TrainController Gold v10
thanks 1 user liked this useful post by craigericks
Offline Andreas K  
#14 Posted : 19 March 2014 20:33:12(UTC)
Andreas K


Joined: 04/11/2010(UTC)
Posts: 191
Location: South Africa.
Hi Mike
See post dd 24/10/2010 - Cs2 "forgets" to do part of routes? - under digital. This might help a bit to solve your problem.
Regards
Andreas
Pure Marklin - Steam is my Passion
thanks 1 user liked this useful post by Andreas K
Offline MikeR  
#15 Posted : 19 March 2014 22:34:03(UTC)
MikeR

United States   
Joined: 26/08/2012(UTC)
Posts: 263
Location: Denver
Thanks Andreas. The thread you referred to in 2010 seems to be the same problem I am encountering. I presume that this problem has since been corrected by Marklin on the CS2 as there is no further mention of this issue on more recent CS2 software versions.

After discussion with an electronics expert in our club I have decided to upgrade my computer as there is a slight delay between my actions on the computer and the physical events on my layout. My computer is currently an old Pentium P4 running Vista and has been giving me some blue screens. System software updates also cause problems and automatic update has been switched off. The new computer arrives tomorrow so I will spend the weekend upgrading and converting to Windows 7. I will keep the thread advised of progress.
Mike
Digital - C track with CS2 and Railroad&Co TrainController; feedback using LocoIO via a Locobuffer
Analog - M track with solid centre rail (after C track layout is complete)
Collect all Eras - especially Crocodiles
Member of ETE
Previously a member of the Marklin Modellers' Group Johannesburg
Offline MikeR  
#16 Posted : 23 March 2014 21:59:23(UTC)
MikeR

United States   
Joined: 26/08/2012(UTC)
Posts: 263
Location: Denver
Installing the new computer has resolved my problem as I have been able to run my trains continuously for abut an hour without any problems of turnouts not being set correctly. BigGrin BigGrin

I replaced my Pentium PC running Vista with a refurbished DuoCore PC running Windows 7. All turnouts are correctly set on all routes when running schedules set up in TC. I have also been able to reduce the time interval between turnout instructions to be sent to my CS2 by TC from 1000ms to 400ms. Turnout pulse length in both TC and CS2 is set at 250ms.

I have been underestimating the processing power required to run TC. Blushing

Thank you to everyone for your suggestions and assistance.

Mike
Mike
Digital - C track with CS2 and Railroad&Co TrainController; feedback using LocoIO via a Locobuffer
Analog - M track with solid centre rail (after C track layout is complete)
Collect all Eras - especially Crocodiles
Member of ETE
Previously a member of the Marklin Modellers' Group Johannesburg
Users browsing this topic
Guest
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.

| Powered by YAF.NET | YAF.NET © 2003-2024, Yet Another Forum.NET
This page was generated in 0.795 seconds.