This is strange indeed... My old CS1 keeps the registrations, and the CS2 is supposed to be much better... ???

Do you have other mfx locos to test with the same train?

Is it only one mfx loco that does this?

Could it be a faulty decoder in the loco that always sends "Hi, I'm new here" or such?

If other mfx loks work ok, it seems like the fault is not in the CS2... Or is it?

Just asking these questions to get some brainstorming going, don't have a CS2 myself - But maybe others have had similar situations...

How are the trains stopping? Are you using a braking module or simply cutting the power using a signal?

Hello,

I had same problem and it was very hard to find out the cause.

I'm using 25 self made braking modules and I saw that on one the trains were not stopping correctly, by stopping directly in the braking section.

Solution: it was a non correctly soldered capacitor that caused this problem. Once soldered well this capacitor, the continuous mfx registering problem after stop was solved.

rgds

Stephan

I have hesitated a bit about responding to this. I do not have a CS2 and do not have access to one, so my contribution will be somewhat speculative, and I don't like speculation. However, if the loco is really re-registering (not just being slow to respond) it is likely because of an incorrect behavior by the CS2. There are two things that can force a decoder to re-register:

1. The PING command, if there is a mismatch between the actual decoder serial number and the serial number used in the PING command. This is not the likely cause of the problem.

2. The BAKE command, if the control station serial number or the BAKE command sequence number is changed. The BAKE command is sent out with regular intervals to inform the decoders of the serial number of the control station. I don't think the CS2 changes its serial number (that would be a very severe bug if it did) but it could change its BAKE command sequence number from time to time. If such a change happens while the loco is powered off, it re-registers, but if the decoder sees both the old and new sequence number without any power drop in between, the loco stays registered. Conclusions from previous threads indicate that the CS2 probably updates the BAKE command sequence number now and then. Don't know if there is a configuration option to turn this off. If the sequence number is updated for no special reason and there is no way to turn that behavior off I would consider it a bug. The MS1 at least does not do such a sequence number update unless you delete an mfx loco from its loco list. Probably the CS1 does not do it either.

Just a bit more speculation, since you have 2 CS2's, have you connected the track power output from both of your CS2's to your layout, and if so, have you made sure that you isolated the centre rail between the powered sections?

Hello All,
Thanks again for all you advice.
First of all , both CS2’s are connected with that special cord between the two CS’s. I am using a 60215 as a master and the 60214 as a slave unit. Each CS has its own transformer. The 60215 has a 60174 booster and the programming test track 78100 connected to it. I had to use a thumb drive to copy the info from the 60215 to the 60214. This is how it was connected ,when I tested all 12 engines for about 8 hours to make sure all bugs were worked out. By the way I used the same engines for the 8 hour test. I would add wagons along the way to see if I had any track issues.
Before calling Marklin yesterday I decided to remove all Lok’s from the layout and start over using just one main line and two MFX Loks and 2 standard loks.. Each main line has 3 stops in the staging area and one stop at a station. The 3 stops at the staging area are controlled by three 7039 home signals I use the voltage from a separate transformer for the solenoids and at the station I used one 76xxx signal connected to the main line. Reeds are used to trigger the 7039 home signals as each train arrives. The 76xxx signal is trigged by two 24994 circuit track. When I turned on the CS2’s, both MFX loks registered and I pressed the MFX button when it lighted on the CS2, to lock each one in.
I did notice one thing that might of caused the problem. It did re register at the staging area and the station area, but it did seem to work better in the staging area and a couple of time in the staging area it worked properly, but it did reregister at the station stop. One time it took almost 3 minutes to reregister. The two MFX loks I used were 39010 B1 and a 39590.
I talked to Jeff Simpson from Marklin after trying this. I told him I was going to default both CS units back to the factory spec. (By the way does anyone know how to do this? I could not fine anything in the manual on how to do it) after defaulting I am going to just power up the older 60214 and see if that solves the problem. Jeff also thought this was a good way to start. I bought the 60215 while in Germany and he said they are not here in the US so he was not familiar with the new unit.
I put a call into Tom Catherwood but have not heard back from him.
Thanks again for all your help. I will let you know the results of my changes later today.
Nelson77
Westminster, Md

In the configuration options of the slave CS2 you can specify what to synchronize. IIRC there are four or five check boxes, one of them for the loco list.
I saw it at the last club meeting but don't recall where it is or how it is called.

"Teppichbahning" is used to name huge floor layouts that spread across several rooms - typically with one digital controller per room.
Here's a video of a Teppichbahning event:

I didn't know that there is a 'special' cable for this purpose. I was under the impression that any ordinary networking / crossover cable can be used to connect 2 CS2's together!

Thanks for that Roger. Looking at the Marklin Systems Architecture diagram, I see that now. Can you connect 2 CS2's via the ethernet ports on each controller? If not, that seems rather daft on Marklin's part, as the CS1 is connected to a CS2 via the ethernet ports! Why not keep things consistent?

Hello All,
Did not have too much time yesterday to work on the layout. What I did do was remove the older 60214 including the special cable connecting the two CS2’s. It was easier for me to do this one instead of removing the newer 60215. I only left 60215 to power one main line and the 60174 booster attached to the 60215 to power another main line. All engines were removed from the layout. I put two mfx enigines and two regular digital engines on the main line. I did not remove and data from the 60215. The only way I know how to default the cs2 is to remove the data one engine at a time. After powering up, the two mfx engines did register as usual. After starting the trains, I held my breath after the first engine stopped at station ( Marklin 39590). When the second engine triggered the home signal to start the 39590 it started right away. The second MFX engine arriving at the station (Marklin 39010 B1) also worked as they should. Even arriving in the shadow yard all engines worked properly. They sat in the shadow yard for about 45 seconds before they are trigged again. I ran these 4 trains for about an hour without any problems. Today I will put four engines on the booster line and run all 8 engines.
After looking at my problem, it seems like the data on the second CS (60214) did not accept the data that I copied from the master unit (60215). If the newer 60215 is able to run several of my engines by itself while it is disconnected form the other one maybe I should delete all the data form the 60214 and than just connect the 60214 to a programming track and register all the engines into the 60214. I would of though that if an engine was registered into the main unit the slave would automatically recognize the engine.
Will give another update later. Any more suggestions would be appreciated. I still have some hair left on my head but falling quickly.
Thanks for all,
Nelson77
Westminster, Md

Another Update,
Turned on power to the layout. Kept the same configuration as I left last night. Cs2 60215 and two MFX lok’s and two standard Digital loks’on one main line. The two MFX were already registered with the CS2 and they started and stopped correctly right from the get go. I ran the four engines non stop for about an hour. I thought I would than go to the next step, add 2 more MFX loks and 2 standard digital loks to the second main line that was controlled by the booster that is connected to the CS2 60215. The two new loks on the booster circuit registered with the CS2 just like they always do. After one of the MFX loks arriving at the first stop the problem came back on both main lines. All the MFX lok’s on both lines started registering again after they came to the stop area and than when the track was triggered to go. Now I am back to the original problem.
My next attempt will be deleting all the loks form my older cs2 (60214) one at a time and start registering the loks on a programming track, using this unit as a stand alone. This should let me know if the 60215 is the problem or not. If it still happens to the second cs2 I won’t know what to try next.
Again open to any suggestions.
Thanks
Nelson77
Westminster, Md

I found a screenshot on the web:
http://nagoya-maerklin.fan.coocan.jp/SDIM0170.jpg

For the loco lists, check "contr." besides "Sync with Master CS2".

Since the problem shows up with only one CS2 involved, this won't solve the problem.

Ok, Another Update.

Put on line my old 60214 CS2. First I deleted all the lok’s from the data base, at lease they did not show the engine screen. But I found out later my test engines that were suppose to have been deleted from the CS2 started running as soon as power was supplied to the track, of course I did not have any control of the lok because it was not programmed in to the CS2. I programmed the running engine again into the CS2 but I still could not control it. There has to be someway to format or erase the complete data base from the CS2.
So I continued the test with two MFX loks and two standard digital loks on the main line. Again the MFX registered and all 4 engines ran like they are supposed to do.
As soon as I added a couple of MFX engine on the second main line, I had the same problem. Every time one of the MFX engines came to the stop they started registering again.
Now what should I try.
Thanks
Nelson77

...this second main line was connected to the booster as well ? ...same booster as before ? ...this booster is now connected to 60214 ?...
What is the reference of the booster ?

Hallo Jeehring,
Yes the same booster was used on both tests. All I did was swap the two CS2’s and leave the one not being used unplugged from the electric. Do you think maybe the BOOSTER is bad?
Now that you brought this up I remember reading somewhere that you should register the lok’s with the CS2 not on the booster line. I can’t read German that well it might have been concerning the old boosters I can’t remember for sure. I could try that tomorrow.
The big problem is how do I remove all the engine data that is registered in the CS2 and default to factory specs. When I remove each one at a time, it removes the MFX’s but not the older digital engines, even though their Icon does not show on the screen, they still start the loks on the track with out having any control.
Thanks for everyone’s help.
Nelson77
Westminster, Md

My English is very basic...
However I can say about each experience with 60215 vs 60214 that the common point is the booster 60174 . This booster is powering the main line on which you got the same problem with signal...
Despite it should be probably not easy to disconnect the electric feeder of the second main line from the booster to connect it to the central , I'd try my best to make it as a test...(so we can be sure the booster is not involved...)

What kind of track isolation have you between rail section powered by booster & rail section powered by station ?Did you replace an "old" booster 60173 by a new booster 60174 ?...
Despite I'm not using this material yet, it is mentionned everywhere that between "60174 booster section & central section " they have a common ground, only the center rail MUST be isolated, not the rails , if you use a 60215 CS2 as master.( This must be also valid when using a 60214 as master I think...)
Whereas "booster section " had to be completely isolated from "CS2 section" (no common ground) when using the former type of boosters (60173, etc...)

PS before calling Germany prepare the reference numbers of the LOKS that are involved in the issue, as well...

Edited by user 29 September 2011 14:43:09(UTC)  | Reason: Not specified

Hallo Jeehring,
Today I will disconnect the booster from the layout and just use one of the CS2’s. I made all my power connects in one area so it is easy to remove and replace equipment. I have a separate wire for each feed line run completely around the layout and one bare cable for a common ground. Each main line is isolated from each others (CENTER RAIL ONLY) with those little red plugs. I did have a 60173 before having it replaced by the 60214.
I was reading somewhere last night that you can RESET the CS to factory specs. I will try that also if I find out to do it.
Thanks for the suggestion,
Nelson77
Westminster, Md

Another Update.
Connected the 60214 and 60215 together with the special cable. Using the 60214 as the master, I copied the data and downloaded the info into the 60215 slave. Checked to see if the data was loaded in the slave and it looked OK. It automatically checked the AUX button. As soon as I turned the power, all the MFX started registering again, even they were all loaded into the cs2 master and ran great before connecting the two units.
Not sure what to do next. Any ideas will be apreciated.
Thanks
Nelson77

I would suggest an experiment:

You need 3 mfx locos and one controller.

1. Make sure none of the locos are registered in the controller.
2. Put locos 1 and 2 on the track.
3. Turn on the controller.
=> The two locos should register.
4. Lift off loco 2 from the track.
5. Put loco 3 on the track.
=> loco 3 should register.
?1 Does loco 1 re-register or stay registered?
6. Put loco 2 back on the track
?2 Does loco 2 re-register or does it keep its old registration?

To simplify, all locos should stand still during the experiment.
Repeat the experiment for both controllers, and for each controller repeat it twice, one for the main track and one for the programming track. What are the results?



Forget about the boosters. It is not likely at all that this problem has anything to do with the boosters. Boosters are simple repeaters of the commands. Locos only re-register as a result of a command sequence, and that will be the same regardless of whether there is a booster in between or not.

are you sure you plugged the 60123 cable in the correct exit in the master and the correct entry in the slave ?
Roger
I am known for stuffing this bit up

Hello Dodger0325,

Yes, I plugged the 60213 in the proper connection. After you do that and turn on power it tells you what cs2 is master and slave. If you reverse the cable it just does says the opposite.
Thanks
Nelson

Another update.
I talked to Tom Catherall today. He said that the two CS'2 should sync up.
I tried to sync the two CS2 together a few times. It did not work and it didn’t make a difference which CS2 I used as a slave or master.
When that did not work I decided to start over with just my 60214. I ran 4 trains on one line, they ran well after they all registered. I than added more trains on the next line. They also ran well. Than I swapped the cs2 and put the new 60215 on line. It registered all the loks but just about every time one stopped for awhile they had to reregister again. I will do more testing tomorrow and try to narrow it down but it looks like it is the 60215 that bad or they can’t sync with the older versions.
I will keep you informed.
Thanks Again for the forums help
Nelson SLinkman

...If there are 2 options : 1/synchronized or 2/ not synchronized

IMO It means that there is a choice between 2 ways of using 2 CS2 together.

( As a non owner of CS2 - not yet - I thought that it means option 1 = sharing the same memory, same track diagram, same keyboard etc.... Option 2 = on the second CS2 you may have a different track diagram for another part of the layout, different keyboard, different memory, etc...) MAY BE ?

Hello Perz,
I had all loks on the track with power when they were register the first time. After all registed in the CS2 is when I started running them.They only re-register when they come to a stop area (Stagging yard or Station area) and when the power is turned on from the Home signal that when it registeds again. I will do more testing today but I have a feeling it is going to be a problem withe the new CS2 (60215) I noticed when I used the data base to register a standard digital lok every time it came up with a BR03. I did upgrade both units 1.6.4(3) yesterday.
Thanks again for your help.
Nelson77
Westminster, Md

AFAIK this is a new feature, so it's not in the manual. Somewhere in the setup screens you can select the active protocols (look for a drop-down list box).

You can disable mfx - that should solve the problem with reoccurring registrations (just kidding).

If you activate MM + mfx only, DCC will be deactivated.