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 siroljuk  
#1 Posted : 09 August 2018 16:01:06(UTC)
siroljuk

Finland   
Joined: 29/10/2010(UTC)
Posts: 374
Hello everyone,I need some help pleaseBlushing Blushing

Look at this video at 8 sec (one sec), at 18 sec ( one sec ) from 30 sec quite long time

https://www.marklin-user..._CS3Plus_Track_sniff.mp4

Symptom: Already in version 13.2 but especially after the last update my CS3Plus have had very strange behavior.

You can see from the video how CS3Plus is working at start quite normal way. at 8 sec a very short failure and also at 18 sec.
from 30 sec this failure starts a log period,my oscilloscope shows that in normal state there is some kind of protocol and voltage is average 19,7 volts. seems to me normal.

At above mentioned time spot something strange happens and + voltage almost vanish and protocol continues at - voltage.
During this kind of failure my locos start to behave strangely. Sometimes Loco will start to run at high speed and you cannot steer it at all, just cut the power with panic button.

Sometimes Viessmann 4558 powered point-motor start to change turnout forth and back.

This kind of failure can last from one sec to 15 sec randomly.

My CS3Plus is updated to 1.3.3, but odd behavior tarted already in previous version of sofware.

I have quite big layout app. 30 turnouts mostly Märklin but few Viessmann. I have seven S88 devices connected to L88 BUS1 and BUS2, I have Booster 60174 and I use 60145 terminal to connect L88 and booster to system.

I use Märklin 60061 power for CS3Plus and 60101 for Booster. I have checked all connections and I did not find any short circuits.

Still one thing: CS3Plus Track voltage is 19,7 ( average) and it varies in the protocol from Highest 25,7 to lowest -25 volts.
at the same time Track voltage of Booster is about 19 volt to -19 volt and it is smoother than CS3Plus voltage.

SO . . . Blushing Blushing Could anyone explain to me what is going on in my environment. PLEASEConfused Confused

I have not send this message to Märklin Service yet, because You Might Know better why this kind of problem exists.

Regards

Jukka

thanks 2 users liked this useful post by siroljuk
Offline MaerklinLife  
#2 Posted : 09 August 2018 19:07:57(UTC)
MaerklinLife


Joined: 03/02/2016(UTC)
Posts: 490
Originally Posted by: siroljuk Go to Quoted Post
I have not send this message to Märklin Service yet, because You Might Know better why this kind of problem exists.

Why would you think that? It cannot harm to send the question to service. I have heard from others about the points turning back and forth.
Offline clapcott  
#3 Posted : 10 August 2018 00:37:38(UTC)
clapcott

New Zealand   
Joined: 12/12/2005(UTC)
Posts: 2,401
Location: Wellington, New_Zealand
I am guessing that the timing may be related to the mfx poll/heartbeat (approx 30 seconds) and a decoders modulated response.

It appears like the decoder modulation back to the track is actually shorting out half the waveform.

While the video shows a single loco, were other decoders/lococ powered during the exercise?
What decoders ? manufacturers?

Just asking ... but do you have any railcom capable decoders? If these are incorrectly driving a signal back into the track you might have a similar symptom.

in summary, I suspect a rouge decoder.
If you have a booster then that would allow you to isolate the CS itself..




Peter
thanks 1 user liked this useful post by clapcott
Offline siroljuk  
#4 Posted : 10 August 2018 07:52:17(UTC)
siroljuk

Finland   
Joined: 29/10/2010(UTC)
Posts: 374
Originally Posted by: clapcott Go to Quoted Post
It appears like the decoder modulation back to the track is actually shorting out half the waveform.

While the video shows a single loco, were other decoders/lococ powered during the exercise?
What decoders ? manufacturers?




Thank you Peter, good thinking.

Only that loco was on the track at the time:

THW.JPG
THW2.JPG

As one can see the decoder is Märklin decoder and I have updated decoder's firmware yesterday but it didn't help.
I have isolated Booster and no effect to problem.
I took L88 and all S88 devices away from the layout . . . no effect to problem.
When I just am writing this reply CS3 started acting oddly again, make noise and Loco don't respond at all during that noise. Must cut the voltage by using STOP-bar and all is ok again.

This odd behavior is completely random and it starts when nothing is going on on the layout strange period can last only a second or ten seconds and during the problematic period I cannot do nothing but cut the voltage, wait few second and put power back.

Next what I am going to test is 60061 power device. I think that power from that device is not right . . .

I will write more

Jukka
Offline siroljuk  
#5 Posted : 10 August 2018 09:06:46(UTC)
siroljuk

Finland   
Joined: 29/10/2010(UTC)
Posts: 374
Hello.

I think I have found the reason for odd behavior.

Terminal 60125 is either broken or it is not compatible somehow with new version of CS3 programs.

I have to continue testing but if someone have similar thoughts please tell us of your findings.

I'll get back after better testing with all devices in my layout

regards

Jukka
thanks 2 users liked this useful post by siroljuk
Offline siroljuk  
#6 Posted : 10 August 2018 11:21:36(UTC)
siroljuk

Finland   
Joined: 29/10/2010(UTC)
Posts: 374
HelloagainBigGrin
I have found the reason for my problem.
My old terminal 60125 is some how broken, it interferes TFP3 protocol so that main controller CS3Plus start to behave very strange. Track protocol start to break after I connect 60125 to CS3Plus.

I have tested this now several times and always result is the same.

I have sent my question to Märklin service asking: What difference there might be between old 60125 and new 60145.
I have opened my old terminal and find out that there is only two components which may cause the problem:
LETEX LT210, Photo DMOS-FET Relay and some kind of transistor or voltage regulator, very small component.

I think that DMOS-FET relays is some how broken.

Well I will wait for Märklin service answer, but at the same time I will order new terminal 60145 from LIPPE.

To my layout terminal is necessary because of L88 and Booster.

It might be so that timing is the keyword if terminal would be unbroken, but if I start to change timing there might occur more problems.

Please comment my thoughts, Ill be Back

regards

Jukka


thanks 3 users liked this useful post by siroljuk
Offline clapcott  
#7 Posted : 12 August 2018 10:33:03(UTC)
clapcott

New Zealand   
Joined: 12/12/2005(UTC)
Posts: 2,401
Location: Wellington, New_Zealand
Ummm.

About 18m/2yr ago I had an issue with mFX registration. (was focused around a club get together with multiple CS2s that did have 60125s and an early L88)
After some (comparatively small amount) to and fro with Marklin Service they did "quickly" suggest that I had a faulty 60125.

Since you bring up the topic , and with a year of two hindsight, it is possible that the "faulty" 60125 is now either sitting ...
- no longer at the end of the string (so its self termination is disabled)
- is now at the end of the string (the supposed disabling of termination is not needed.

I will have to keep this in mind , and maybe do some testing with 60125s in different positions on the chain

Or. may be a different issue all together


Peter
thanks 2 users liked this useful post by clapcott
Offline siroljuk  
#8 Posted : 12 August 2018 12:50:51(UTC)
siroljuk

Finland   
Joined: 29/10/2010(UTC)
Posts: 374
HelloBigGrin

I did some changes to my layout. I did remove booster all together and put L88 direct to CS3Plus.
No problem all goes well.
I did order new terminal 60145 and it will arrive late next week or after that. The we see whether my 60125 is broken or not.

After my tests I am sure that my 60125 is at least partly broken. I seems to me that changing polling timer for L88 don't change the failure situation.

Now that I did remove 60125 my layout behaves as it should, no odd sounds from turnouts or locos running very slow, no blinking lights from laterns etc.

After week or so I can tell if I am rightBlink

Regards

Jukka
thanks 1 user liked this useful post by siroljuk
Offline siroljuk  
#9 Posted : 12 August 2018 12:57:19(UTC)
siroljuk

Finland   
Joined: 29/10/2010(UTC)
Posts: 374
Originally Posted by: clapcott Go to Quoted Post
Since you bring up the topic , and with a year of two hindsight, it is possible that the "faulty" 60125 is now either sitting ...
- no longer at the end of the string (so its self termination is disabled)
- is now at the end of the string (the supposed disabling of termination is not needed


Peter, what is that termination, I have only one 60125 and there is no any kind of plug in the socket to witch next 60125 should be connected.

Jukka
Offline clapcott  
#10 Posted : 13 August 2018 00:39:19(UTC)
clapcott

New Zealand   
Joined: 12/12/2005(UTC)
Posts: 2,401
Location: Wellington, New_Zealand
Originally Posted by: siroljuk Go to Quoted Post
Peter, what is that termination, I have only one 60125 and there is no any kind of plug in the socket to witch next 60125 should be connected.


Each 60125 (60145) provides the termination electronics.
If you pop the hood of the "Terminal" the small amount of electronics that you see relate to this.

However, while some of the components do the actual termination , the other half is listening "upstream" to see if a cable is plugged in.
If a cable is detected, it effectively switches off that "Terminals" termination - as the termination is only wanted at the ends of each signal buss

This detection is simply a loopback from the power sent to the terminal, back down the cable to the transistor/swtich.
I did draw up a schematic some time ago, will see if it I can find it however I suspect it was on paper only and I suspect this has been done more professionally by someone on this forum

Points to note.
- the CS itself needs this same logic so it may be that an external 60125 is fault and not terminating OR that the CS termination electronics is NOT disabled.
- Device (MS2,Booster, L88, 6021Adapter etc ) termination is a related topic, but a separate aspect to the main System Bus termination.
- Consider that the CS has a 60125/45 inside.
- - The SystemBus port extends externally to any 60125/60145(Terminal) or slave CS
- - and the 4 internal device ports service the front MS2 ports and 1/2 booster/L88 ports on the back , AND (probably) the GFP module as it is just a booster

It may even be that it is the cable (wires/pins that relate to the detection ) that are damaged.

Edited by user 13 August 2018 04:37:22(UTC)  | Reason: Not specified

Peter
thanks 1 user liked this useful post by clapcott
Offline siroljuk  
#11 Posted : 13 August 2018 07:15:11(UTC)
siroljuk

Finland   
Joined: 29/10/2010(UTC)
Posts: 374
Originally Posted by: clapcott Go to Quoted Post
It may even be that it is the cable (wires/pins that relate to the detection ) that are damaged.


Hello.
Thanks for the clarifying answer, I am going to buy those two components LT-210 Photo DMOS-FET and transistor BC846 and repair 60125-device.

I am sure that I have to order them because here in Jyväskylä the shop is so small.
I will tell afterwards what happens after repairing.

Terminal cable is not shielded so it might cause some induction and therefore some interference.

The video which I put n my post is not very clear but I did see som odd interference in the graph so i try to solve the cause of failure.

Jukka

Offline siroljuk  
#12 Posted : 14 August 2018 16:22:19(UTC)
siroljuk

Finland   
Joined: 29/10/2010(UTC)
Posts: 374
Hello BigGrin BigGrin

I did repair my old 60125 terminal and now it works as it should.

I searched that LT210 component from whole Europe and I didn't find not a single shop from which possible buy that component, so after Peter's mention of termination I had an idea:

"Let's open old MS 60652 and see if there might be that component too" YESSSThumpUp ThumpUp There was that and I just soldered it from the base board and put it onto Terminal's (60125) board.

It works fine now. Very easy and cheapSmile Smile .

That old 60652 Mobile Station was so old and had so old version of software that my CS3Plus didn't even reconsidered it any more.

Thaks all of you and specially Peter whose comment leaded me to right track.

Have Fun all of you

Jukka
Offline TEEWolf  
#13 Posted : 14 August 2018 19:37:42(UTC)
TEEWolf


Joined: 01/06/2016(UTC)
Posts: 2,461
Originally Posted by: siroljuk Go to Quoted Post
Hello BigGrin BigGrin

I did repair my old 60125 terminal and now it works as it should.

I searched that LT210 component from whole Europe and I didn't find not a single shop from which possible buy that component, so after Peter's mention of termination I had an idea:

"Let's open old MS 60652 and see if there might be that component too" YESSSThumpUp ThumpUp There was that and I just soldered it from the base board and put it onto Terminal's (60125) board.

It works fine now. Very easy and cheapSmile Smile .

That old 60652 Mobile Station was so old and had so old version of software that my CS3Plus didn't even reconsidered it any more.

Thaks all of you and specially Peter whose comment leaded me to right track.

Have Fun all of you

Jukka


You know that you destroyed your MS 1? Even with a CS 3+ you still can use a MS 1 60652.

https://www.maerklin.de/...emarchitektur_gesamt.pdf

I never heard before of your electronic component LT-210 Photo DMOS-FET, but perhaps at Reichelt you may find it.

https://www.reichelt.de/...f68022a7&LANGUAGE=EN

Or is it replaced by the LT211 component? May these web pages help you?

https://www.alldatasheet...hword=LT211&sField=4

http://www.datasheetcata..._pdf/L/T/2/1/LT215.shtml

But I think it is good to know, as a 60125/60145 is fixable, if you now the necessary spare parts.

Offline Minok  
#14 Posted : 14 August 2018 23:27:52(UTC)
Minok

United States   
Joined: 15/10/2006(UTC)
Posts: 2,204
Location: Washington, Pacific Northwest
The LT210 is indicated to be a

"The is a miniature 1-Form A solid state relay a 4 pin SOP package that employs optically coupled MOSFET technology to provide 1500V of input to output isolation. The Input of optical coupler been controlled by a highly efficient GaAlAs infrared LED and MOS FETs on the output side."

in a quick search. So if its not available anymore I'm sure there is a replacement part number with the same form factor (4 pin SOP) and same or better specifications.

You could see if you can track one down and replace the one you took from the MS1 and get it operational again.
Toys of tin and wood rule!
---
My Layout Thread on marklin-users.net: InterCity 1-3-4
My YouTube Channel: https://www.youtube.com/user/Minok1217/
Offline siroljuk  
#15 Posted : 15 August 2018 07:13:20(UTC)
siroljuk

Finland   
Joined: 29/10/2010(UTC)
Posts: 374
Originally Posted by: Minok Go to Quoted Post
You could see if you can track one down and replace the one you took from the MS1 and get it operational again.


Hello and Thanks.
Yes I could repair my old MS1 if I need it, but I few years ago I took connection cable from the device for another usage ( I made some experimental tests wit Raspberry mini computer) All ready at that time my recent CS2 Central station could not recognize that MS1. And I didn't have way to update old software.

So it is recycleBigGrin BigGrin

I know where LT210-componet is manufactured it is in Taiwan, I know all specs too and yes I have studied all possible information about Phot DMOS-FET Relays. There are only few different brands mostly manufactured in far east Taiwan China etc. But it is not forth to try to buy one this kind of component to Finland it is much more complicated than new 60145 Terminal from Germany. And I really don't need old MS1 any more.

Regards

Jukka




Offline siroljuk  
#16 Posted : 14 August 2019 17:14:12(UTC)
siroljuk

Finland   
Joined: 29/10/2010(UTC)
Posts: 374
Hello everyoneBigGrin BigGrin
This story continueBlushing

Even though I did buy a new 60145 about a year ago I have have used the old terminal which I did repair.

Now today this same old terminal 60125 started to disturb CAN-protocol again.
Symptoms, as you may remember, are:
Steady lights ( mostly led) which are connected to same power source as tracks are suddenly start to blink,'
If loco is running it start to behave oddly, sometimes full speed without touching speed knob and so on.
If there is several locos on the active track all of them can start to act oddly( but not at the same time and same kind of behavior)
AND after a while all goes fine again. And then again odd behaviors randomly.

I took first all possible devices away from CAN-bus but terminal 60125 was connected and odd behavior goes randomly on.
So then I took 60125 away and put my new unused 60145 on the system and no problem at all.Cool

This message is for you all who use terminals: Terminal can break even if it has only two active active components LT210 and some kind of transistor.
Other components are only resistors.

If your Digital based layout starts to odd strangely this might be the reason.

I tried today search replace component to LT210 but I am not such pro. who could find exactly same component but different name.
So If someone of you could help me with this I would be very grateful.

BTW I did updated several loco-decoders on the program-track earlier this morning and during this update work all this problematic behavior started. There was no locos on the main track when I updated decoders, BUT . . . main track was active and power on and there was about ten (10) lighted wagon on the track. Led light.

Regards

Jukka




thanks 2 users liked this useful post by siroljuk
Users browsing this topic
OceanSpiders 2.0
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 © 2003-2021, Yet Another Forum.NET
This page was generated in 0.528 seconds.