Marklin/LGB decoders and JMRI

[ntpntpntp]

Out of purely academic interest, has anyone actually used JMRI (and SPROG or other hardware) to program a loco with one of the Marklin decoders now being fitted to recent LGB?

I don't have such a beast myself, but I've been reading through and trying to help on a couple of recent threads and it makes me wonder - I can't find any decoder definitions at all in JMRI for Marklin decoders?

I've read in a couple of places that Marklin's MLD and MSD (non-sound and sound decoders respectively) were developed by ESU. Opening up the ESU LokPilot V4 or LokSound V4 decoder definitions in Comprehensive Programmer mode, the Function Map tab looks closely related to the function mapping table for the Marklin decoders found here:

http://mediencms.maerklin.de/media.php/de/produkte/downloads/Umruestdecoder_mLD-mSD_CV-Liste_DCC.pdf

Could it be that JMRI will identify the Marklin decoders as ESU?

 

[ntpntpntp]

The above Marklin CV list, it seems to indicate the manufacturer code number = 131 which NMRA lists as Trix. Fair enough.

Has anyone tried reading out CVs 7 & 8 for one of these new Marklin decoders? Still no obvious sign of a suitable decoder definition in the JMRI download area.

 

[idlemarvel]

Sorry I can't answer your question but related to this I just got a new LGB V100 diesel and it comes with one of these Trix decoders fitted. At least I think it is Trix because although I can write CV to the decoder I cannot read, so I am relying on the documentation that says CV8 is 131 which as you say is listed as Trix Modelleisenbahn in the NRMA lists. Anyone know why i cannot read CV from these decoders? I know CV writing works as I have successfully changed the address. I have a Hornby Elite DCC command station with 4A transformer which can read and write CV to Massoth decoders but no joy with this Trix decoder. The only time I have seen this in the world of OO is when you are trying to program a decoder in a non-powered loco like a dummy DMU driving car, which has a decoder fitted just to handle directional lighting. Then you can write but not read as there is no load across the decoder. But I don't see why that applies here. I have tried on programming track and main track, and I have tried writing 0 to CV50 which is supposed to disable mts and DC protocols, but no joy. Any ideas anyone? Thanks.

 

[Brixham]

Some 10 years ago I read that some command stations can't read back certain decoders, as the current they draw is insufficient for this operation...as has been alluded to above.

The work round was to place a 1000 ohm resistor directly across the rails...this fools the command unit into reading back....

Worth a try

Malcolm

 

[idlemarvel]

Certainly worth a try Malcolm, thanks. I'll give it a go tomorrow and report back.

 

[Cliff George]

Id suggest trying turning off sound (which is on by default in these decoders), lights and smoke before trying to read CVs. Too much current draw can sometimes be a problem.

I have no problem reading CVs in one of these new decoders in a recently purchased LGB 24741 (Spreewald) with my Lenz system.

It is not usually possible to read CVs on the main track but it should certainly work on the programming track.

Sometimes I find programming works better by using crocodile clips to directly connect the loco skates to the programming track.

Hope this helps.

 

[PhilP]

Sorry I can't answer your question but related to this I just got a new LGB V100 diesel and it comes with one of these Trix decoders fitted. At least I think it is Trix because although I can write CV to the decoder I cannot read, so I am relying on the documentation that says CV8 is 131 which as you say is listed as Trix Modelleisenbahn in the NRMA lists. Anyone know why i cannot read CV from these decoders? I know CV writing works as I have successfully changed the address. I have a Hornby Elite DCC command station with 4A transformer which can read and write CV to Massoth decoders but no joy with this Trix decoder. The only time I have seen this in the world of OO is when you are trying to program a decoder in a non-powered loco like a dummy DMU driving car, which has a decoder fitted just to handle directional lighting. Then you can write but not read as there is no load across the decoder. But I don't see why that applies here. I have tried on programming track and main track, and I have tried writing 0 to CV50 which is supposed to disable mts and DC protocols, but no joy. Any ideas anyone? Thanks.

What equipment are you using to read the CV's?

I have had no problem reading values with either Massoth, or Gaugemaster equipment..
Setting them, and controlling a loco, yes. But not reading values.

 

[idlemarvel]

Thank you all for your input.

@Malcolm: I tried the resistance across the track but sadly didn't work.

@Cliff: I am using programming track so sound, lights and smoke (not fitted) are off anyway. I tried crocodile clips to connect the shoes directly to the programming output but no joy there either.

@PhilP: I am using Hornby Elite with 4A transformer. Unusual at this scale I grant you but DCC should work. it's okay with Massoth decoder.

 

[Cliff George]

@Cliff: I am using programming track so sound, lights and smoke (not fitted) are off anyway. I tried crocodile clips to connect the shoes directly to the programming output but no joy there either.

The decoder doesn't know it is on the programming track or on the main. Turn off anything that is going to draw current during normal operation, then try it on the programming track.

 

[GNB]

Out of purely academic interest, has anyone actually used JMRI (and SPROG or other hardware) to program a loco with one of the Marklin decoders now being fitted to recent LGB?
...
I don't have such a beast myself, but I've been reading through and trying to help on a couple of recent threads and it makes me wonder - I can't find any decoder definitions at all in JMRI for Marklin decoders?
...
Has anyone tried reading out CVs 7 & 8 for one of these new Marklin decoders? Still no obvious sign of a suitable decoder definition in the JMRI download area.

I have read the CV values for a LGB 25522 using a Massoth 8175101 reader and it's associated DCC program. It gives CV 7 as 0 and CV 8 as 123.

However the Marklin software fails to identify the decoder type and incorrectly reads CV 1 and CV 29.

This is despite these CV values being previously set by the DCC program. At that time the address was changed from the default of 3 to 5 and CV 29 from 4 to 6, to increase 14 to 28 speed steps. The new values were shown in the log and confirmed (the logged sequence was CV 29 value 6, CV 1 value 5, CV 17 value 128, CV 18 value 0, all OK).

The loco works fine at address 5 and 28 speed steps (Massoth remote, MTS III). However the current reported value for CV 1 is 15 and that for CV 29 is 30. All the other reported values agree with those given as the default in the table shown in the handbook (which fails to give the values for CVs 7 & 8 ).

The additional LGB 65006 sound card appears to have been updated at the same time and continues to function normally.

The light and sound were defaulted to off, the values were read on a programming track and the power supply was a LGB 50110 (5 amps). The DCC program is vers 1.4.16.1 - 14.04.2015, the PC module firmware has been updated to vers 3.2 and the XLS template was for a Massoth emotion M-XXL.

It's a bit puzzling and a little reminiscent of the problem reported for the LGB Allegra recently albeit with an apparent different type of Marklin decoder.

Don't know if that helps but any suggestions would be welcome.

 

[ntpntpntp]

I can understand if the software you used maybe isn't aware of the manufacturer and decoder model, but very worrying if your gear didn't manage those read those important CVs correctly. This is basic core DCC stuff. I can't think of any reason why those should give any more trouble than other CVs.

 

[PhilP]

My results from a 'green 100' Ge 4/4 (sorry, do not have model number to hand)..
*Both the owner, and shop staff, had fiddled with this before I had it to sort out.*

This has a new MFX Marklin decoder. The owner had tried to set the address using a Massoth CS and Navigator. - This failed..
This appears to be due to the order the Navigator et al do things in. - It also does more (if you use the 'set address' option) than just change CV1 (for a short address).
For a short address change, if you use the program CV option, then it will work.
For a long address, read CV29, and 32 to the value, and write it back. Then do the normal sums for CV's 17 and 18, add the value you calculate for CV17 to 192 and write this. Write CV18 as normal.

CV50 has two values shown in the manual (sorry, do not have this to hand, so a little vague but)..
One value is to enable/disable MFX, the other to enable/disable DCC.
The loco I had would NOT respond properly to DCC commands from Either Massoth, or Digtrax equipment. - Whichever function key you pressed, you got the 'whistle' effect for key 1.
The two values shown for CV50 are '2' and '8' (decimal). - When I read this back, I got '11'. I wrote '2' (this from memory) to this CV, to turn OFF MFX support, and all worked as it should. - NOTE: Reading back this CV gave '3' back!

Long Address, worked example;
To set a long address to 630, bit 5 of CV 29 must be ‘set’. – bit 5 equals 32 (decimal).
We want a long address, 28/128 speed steps, analogue operation permitted: 32+2+4=38.
So set CV29=38

CV17 value. The base address for this CV is 192.
Divide the long address by 256; 630/256=2.46 take the integer of this (2), and add it to 192 to get the value for CV17: 192+2=194.
So set CV17=194
CV18 value. The answer to the following sum; address – integer used in CV17 multiplied by 256.
630-(256x2) = 630-512 = 118.
So set CV18=118

 

[GNB]

The new LGB loco 28802, with onboard sound, similarly misreports the CV 1 & CV 29 values as 15 and 30 respectively.

However for this loco the reported value, using the Massoth PC Modul & DCC software, for CV 7 was 111 and that for CV 8 was 255.

This loco operates fine with its new address and 28 speedsteps. Here's an extract of the change log text for this operation:

12:02:57.578 MODUL COMAND Open Modul
12:02:57.843 SEND 004 004 10 08 00 18
12:02:57.859 RECEIVE 000 004 11 00 00 11
12:02:57.859 SEND 004 004 12 00 00 12
12:02:57.874 RECEIVE 000 004 13 00 31 22
12:03:26.703 MODUL COMAND Write CV 029
12:03:26.703 SEND 004 004 88 1D 06 93
12:03:27.202 RECEIVE 000 004 98 1D 01 84
12:03:27.202 CV WRITE CV 0029 Value 006
12:03:27.717 MODUL COMAND Write CV 001
12:03:27.717 SEND 004 004 88 01 07 8E
12:03:28.216 RECEIVE 000 004 98 01 01 98
12:03:28.216 CV WRITE CV 0001 Value 007
12:03:28.731 MODUL COMAND Write CV 017
12:03:28.731 SEND 004 004 88 11 80 19
12:03:29.230 RECEIVE 000 004 98 11 01 88
12:03:29.230 CV WRITE CV 0017 Value 128
12:03:29.745 MODUL COMAND Write CV 018
12:03:29.745 SEND 004 004 88 12 00 9A
12:03:30.244 RECEIVE 000 004 98 12 01 8B
12:03:30.244 CV WRITE CV 0018 Value 000

LGB seem to refer to the onboard chips for these locos as "MFS" and others, like the V100 diesel reported by idlemarvel above, as "MFX".

Any additional insight would be welcome.

PS Sorry PhilP I crossed your post.

 

[GNB]

Sorry I resurrected this thread. Seems that this topic has been well covered over a number of different threads....

Digital Allegra :
https://www.gscalecentral.net/index.php?topic=301747.0

New Marklin Decoder (LGB 24741) and Digitrax Super Chief (EU Wireless)
https://www.gscalecentral.net/index.php?topic=301158.0

LGB/Marklin new stock with Marklin decoder - ?problems?:
https://www.gscalecentral.net/index.php?topic=301011.0

Compatability MTS MFX DCC:
https://www.gscalecentral.net/index.php?topic=300574.0

...but surprisingly remains unresolved. An interesting read and retrospective thanks for your input.

 

[PhilP]

TO MODS:

Do we need a new area on the Forum?
Either under loco's or decoders, for new Marklin/LGB loco's and perhaps for the new MFX decoders??

There are a lot of CV values we could do with additional information on.. It would be useful if there were the equivalent to the tables at the rear of the Massoth manuals for this new decoder!

 

[Cliff George]

There are a lot of CV values we could do with additional information on.. It would be useful if there were the equivalent to the tables at the rear of the Massoth manuals for this new decoder!

The list of CVs in the manual for the LGB 24741 (Spreewald) seem comprehensive to me. I messed around with the function mapping, speed table, switching range, ABV etc and it all seems to work just fine. What CVs do you need further information on? This is the only MFX decoder I've got, do others perhaps have a different list of CVs? The only thing I could not get to work (yet) was the multiple station announcements but that doesn't bother me much.

 

[idlemarvel]

@PhilP: I am using Hornby Elite with 4A transformer. Unusual at this scale I grant you but DCC should work. it's okay with Massoth decoder.

FYI I sold my Hornby Elite on ebay and bought a Lenz LZV100 (60110) and PC interface (23151) and with the Lenz provided software CV-Editor I was able to read and AFAIK write all the CVs. Like Cliff I have not been able to get the multiple station announcements working, which unlike Cliff I would like to solve!

 

[PhilP]

The list of CVs in the manual for the LGB 24741 (Spreewald) seem comprehensive to me. I messed around with the function mapping, speed table, switching range, ABV etc and it all seems to work just fine. What CVs do you need further information on? This is the only MFX decoder I've got, do others perhaps have a different list of CVs? The only thing I could not get to work (yet) was the multiple station announcements but that doesn't bother me much.

Well, CV's 57 and 58 for a start..
CV's 112, 113, 114.
A plain English walk-through of the multiple values for many of the CV's would be good.

 

[MTheStrong]

Well, CV's 57 and 58 for a start..
CV's 112, 113, 114.
A plain English walk-through of the multiple values for many of the CV's would be good.

I cannot find any reference to CVs 57, 58,112, 113 and 114 on my guide supplied with the latest Spreewald LGB24741. Any idea what these CVs are used for?

Here with the link to the Marklin Instructions : http://static.maerklin.de/media.php/lgb/pdfs/Decoder-Ergaenzung.pdf

 

[idlemarvel]

I think 112, 113 and 114 are the mode, brightness and blink frequency of the forward lights.
(115, 116 and 117 for rear lights). And I think 57 and 58 are something to do with syncronising steam sound to wheel movement, here is an auto-translation for CV57 from German manual:
Steam locomotive : Soundabstand higher burst in the 1st ...
... With wheel sensor (pulse there is distance between the higher burst)
... without wheel sensor (distance of the higher burst at FS 1)
diesel locomotive / electric locomotive : Geschwindigkeits-Modulation
... gradually linear modulation
And for CV 58:
Steam locomotive : Soundabstand higher burst in the range >1 ...
... With wheel sensor (pulse count the higher burst)
... without wheel sensor (distance of the higher burst at FS >1)
diesel locomotive / electric locomotive : Geschwindigkeits-Modulation
... gradual linear, fixed distance between the gears
This is from a manual that ntp pointed me at.