Despite locomotive callibration (15 points) I noticed my locomotives do not stop at the right place defined within the profile of the iTND. Often it goes too far than the stopping point and each locomotive has different behaviour and stopping point. Could you please advise me how to solve this issue.

Many thanks for your support.

Hello,

did you change the option "stop at contact" in the routes to another option like "stop at destination point"?

How long is your iTND and how fast do your locos enter the iTND. IF they are too fast it's impossible to stop them correctly.

Hi Olivier,

did you set the brake delay (CV4 for DCC decoders) to a minimum value (in most cases 0 or 1)?

If you did this already, you need to adjust the brake correction per locomotive experimentally in the locomotive measurement window (slider in the lower right corner).

Regards
Markus

Dear Stefan,

To answer your first question I think I choose the option "stop at contact" but I need to verify. My iTND are around 2 meters long, I guess that must be sufficient to stop them.

Dear Markus,

My Märkin locomotives have either Mfx or Motorola decoders, in the riding properties of the locomotives for the dynamic behaviour I setup the minimum figures in order to have fast acceleration and deceleration. I have done adjustment with break correction and it seems to improve the situation, I understand each locomotive needs to be tested individually.

According to the manual iTND is supposed to give you a very precise stopping point, based on your experience how precise can be this stopping point. I would be interested to hear your experience on this subject in order to setup my figures with the right error margin.

Thank you both for your answer and for sharing your experience.

Best regards.

Olivier

Hi Olivier,
in my experience, applying the solution Markus told above (no brake or acceleration delay on decoder, in order to let WDP act freely), I obtained very fine results.
Loco stops at the expected point with a tolerance (N scale) of about +/- 2 cm.
Obviously if the stop point is "stop at the contact", the precision will depend on the loco inertia

In hidden parts, to prevent locos going "long" or "short" I use intelligent train number display with stop at the center of contact, so you get a tolerance = (contact lengt - train lenght) / 2 , generally higher that loco brake tolerance.
In visible parts, if I want to stop at signal I set the stop point about loco brake tolerance + x before (for instance with the loco brake tolerance of +/- 2 cm I set stop point 5 cm before segnal, so the loco will stop from 7 to 3 cm before signal).

Kind regards
vittorio

Hello Olivier,

"stop at contact" disables the iTND! You must select one of the other options!

Dear Olivier,

the main part of my locomotives stop with +/- 1,5 cm.

The remaining ones (especially motors with bigger oscillating weight or some badder decoder types) can not be stopped with such a precision.

Regards
Markus

With the Marklin locos that have a software ABV or Accel-Decel delay (usually F4) I make sure that it is switched off - that is there is no acceleration or deceleration delay.  I then calibrate the loco using the defined test track and everything works perfectly, if I keep the F4 functional (i.e. no acceleration or deceleration delay) when the layout is running. If you don't do this then you have the decoder's deceleration plus the deceleration that the computer is trying to manage.

The bigger problem is where you have decoders with potentiometers - like the 37216 in your other thread. They do not have a F4 option. So you need to open the loco and turn the correct (you need the loco's manual to work out which one to change) potentiometer fully anti-clockwise to switch off the ABV. Then you do the calibration routine.

BUT - there is another factor.  I calibrated most of my locos with my eCOS (mono display), but when I upgraded to the eCOS2 (colour display) with the variable input voltage, I noticed that the locos were overrunning the stop positions. I had set the eCOS2 about 2 volts higher than that used by the original eCOS (technically the eCOS voltage was too low for Marklin).  It does not make much difference, but it was enough for me to have to relive the "joy" of recalibrating over 90 locomotives and railcars.    I won't be changing the controller on my layout for a very very long time - no matter how pretty they look. 

Zitat von: Adrian L in 26. Oktober 2016, 14:23:21
The bigger problem is where you have decoders with potentiometers - like the 37216 in your other thread. They do not have a F4 option. So you need to open the loco and turn the correct (you need the loco's manual to work out which one to change) potentiometer fully anti-clockwise to switch off the ABV. Then you do the calibration routine.

[

Hello Adrian,

thats not korrekt, also Märklin Decoder with Potentiometers have also the Option F4 to switch off the Acceleration and Decerlaration Delay!

Hi Adrian,
I executed a test with Ecos 2 to see how the voltage supply affects loco speed.
I found that it depends on the decoder type:
ESU LokPilot 4 micro and original Fleischmann decoder sound - no influence
Uhlenbrook and original Minitrix - speed varies with voltage
Kato - speed varies in opposite way to the voltage (very strange, but seen on two locos)

I agree, better not to change the controller, but what about the addition of a booster?
I hope , when I will be forced to add a booster to my installation, to be able to set exactly the same voltage as provided by controller.
In that case not even a new calibration will help between different sections.

It will be much better if all the decoders would behave like ESU and Fleischmann ...

Kind regards
Vittorio

Dear All,

Thank you very much for your additional explanation and experience, very much appreciated.

Best regards.

Olivier