Marklin/LGB decoders and JMRI

[PhilP]

Thanks for reply MTheStrong. I noticed your comments in an earlier thread (https://www.gscalecentral.net/index.php?topic=299019.msg318125#msg318125) that had a link to the new CS2 manual (http://www.rocousa.com/MARKLIN/60215_cs_gb.pdf).

It looks like an interesting system that provides a proprietary "plug and play" solution but with enough features to allow to delve into the inner workings.

Can I ask some supplementary questions?

Is it easy to read the CV values from either the MFX or MZS chipped locomotives?

How far does it allows you reset the setting on an MFX chipped locomotive to restore the default values for that locomotive? I might assume it might store the defaults from a dataset for locomotives that it recognises. I wonder how it handles the MZS chipped locos?

Quite easy to read conventional CV values from both derivatives.

From the course/presentation, if you 'default' a MFX decoder it goes back to how that loco was shipped, and NOT a full reset to as manufactured. - A nice touch..

Factory defaulting an LGB (or Massoth) chip will take it back to as manufactured. - Allowing for the different 'levels' of reset that you can do on Massoth decoders.

 

[MTheStrong]

Quite easy to read conventional CV values from both derivatives.

From the course/presentation, if you 'default' a MFX decoder it goes back to how that loco was shipped, and NOT a full reset to as manufactured. - A nice touch..

Factory defaulting an LGB (or Massoth) chip will take it back to as manufactured. - Allowing for the different 'levels' of reset that you can do on Massoth decoders.

Thanks Philip that is exactly what I have found. I have not tried to do a factory reset on the Dietz chips installed on my 2 railbuses. I have 2 MFX chipped locos the little KOF shunter and the Spreewald. With one other exception, a Lenz chipped 2015, the remainder of my locos are LGB/Massoth chipped.

 

[ntpntpntp]

just to add: "MZS" is simply the german abbreviation for "MehrZugSteurung", what we know in English as MTS = "Multi Train System" - ie. the LGB brand of digital train system. So references to MZS decoders should indicate equipment initially made by Lenz but mostly by Massoth.

 

[NCS from Qbyn]

The list of CVs in the manual for the LGB 24741 (Spreewald) seem comprehensive to me.

Cliff I also have this LGB 24741 model and have spent quite a bit of time looking at the CV doco. I also found it very helpful. I was also pointed at the standard Marklin DCC CV list [ie all of it, not just the subset shown in this loco's manual] Unfortunately, the only full version I could find was in German, but I was able to muddle through this with the help of BING Translate. One thing I did observe was that the Marklin DCC interpretation for LGB 24741, although different from LGB/Massoth and again different from ZIMO, is the same as Marklin uses for its other model scales.
It would be useful if DecoderPro3 could auto-recognise the Marklin decoder type in LGB 24741 - I intend to try using one of the HO Marklin profiles to read its CVs into, as I suspect that may do the job.

However, as you have this loco, you may be able to answer a question that I have about its sound functions.
I find that the sounds seem to be quite low in volume, compared to older LGB locos. All the sound control CVs seem to be set to 255 by default, which suggests they are already at maximum capable volume. The whistle is OK, but the bell is almost inaudible. The chuff sound is rather quiet as well, although it is not too bad.
There does not seem to be a volume control knob, so the only way to change the volume would appear to be via the CVs.
It has been suggested to me that the physical size of the loco and its speaker may just be such that it cannot produce any more sound than it now does.
Have you also found some of the sounds to be extra quiet, or is it possible there is something wrong with my particular unit that I need to follow up with the vendor? [It was brand new, when I bought it.]

I don't want to waste much more effort trying to get more volume, if the model is simply not designed to produce any more!
I just hope this sound profile is not going to be repeated in the Larger LGB locos with the MFX chip - such as the new Harz 2-10-2, which I have on order.

 

[NCS from Qbyn]

Just a note here. I have also noted that several independent threads in the past year have raised issues related to programming these Marklin/Trix MFX decoders. I have commented in at least 3 of them recently. I noticed a suggestion in at least 2 places that a special forum for dealing with issues related to locos with the new MFX decoders could be established.
As Marklin is now releasing more new locos this year with MFX, this is going to become more significant.
Rather than having to search all over the forums for information, it would be more convenient if we could go to one area for all discussion on this subject - at least for the next year or so, until the problems are understood better.
I am getting the feeling that the issues are not actual faults in the MFX, but simply that it doesn't respond, in the same way as other purely DCC decoders do, to the programming tools we have become used to.
Hopefully, we can pool our learning here to master these new decoders.
Could this forum be set up?

 

[ntpntpntp]

Maybe a good idea, and especially leading in with your discoveries about the MFX decoder seemingly wanting to be "initialised" on a DCC track before it will converse nicely with DCC programmers etc.

For me personally it's only ever been an academic interest as I have no locos with these decoders and no intention of buying any at the moment (all my G scale is packed away out of use). As folk has posted throughout the threads, it seems commercial suicide for Maerklin NOT to make these decoders compatible with NMRA DCC

 

[NCS from Qbyn]

am gradually digging up more info on these MFX decoders.

There is a bit of inference here, but bear with me.

The CV pattern in the LGB 24741 seems to fit the pattern documented for Marklin's mLD, mSD loco and sound decoders, as documented [in German] at http://www.maerklin.de/en/products/new-items/decoder/decoder-programmer-for-mld3-and-msd3/

This table of CVs has been posted elsewhere in this forum.

This table does actually fit in with the standard DCC protocol. However, the 'standard' allows for quite a number of manufacturer specific CV definitions, and Marklin has taken full advantage of this. So the term standard, when applied to DCC CVs, seems to be rather a movable feast. Each manufacturer [not just Marklin] has quite a number of manufacturer specific CVs in their list. These are not deviations from the NRMA standard, they are simply not nailed down by the standard! [This may be common knowledge to many readers, but maybe not to all of you.]

Unfortunately, although the manufacturer code for this decoder is TRIX, the CV pattern does not seem to be the one used by TRIX for their branded decoders.

Decoderpro does not appear to have a specific CV program for any Marklin decoder, let alone this mLD/mSD pattern.

The actual decoders in the Marklin/LGB MFX locos may, or may not be this mLD/mSD, but the pattern in the LGB 24741 matches this pattern, as far as it is documented.

Further digging reveals that the mLD/mSD decoders can be programmed using a Marlin USB dongle on a PC with some free Marklin software. I found this piece of text in a very long Marklin manual: [translated from German by Bing]

Programming:
-MDP 60971 decoder programmer, stick with USB and decoder interface.
Programming tool mDT3, software for fast programming of new
LokDecoder3 and SoundDecoder3 on the PC
Retrofit decoder with 21-pin interface:
-60972 decoder mLD3 (without sound module)
-60975 sound decoder mSD3 with preset Steam loco sound
-60976 sound decoder mSD3 with pre-set sound of diesel locomotive
-60977 sound decoder mSD3 with preset E-Lok-sound

I also found this link, thankfully in English, that describes that very device: http://www.maerklin.de/en/products/new-items/decoder/decoder-programmer-for-mld3-and-msd3/

It costs about 80 Euros from Marklin direct - probably cheaper elsewhere.

Now the questions are:

1) Is the decoder in the LGB locos able to be programmed with this device too?
2) Is it really necessary, as, armed with the list of Marklin DCC CVs from the earlier link, the decoder can be programmed using standard DCC tools [once it has been initialised]. I programmed mine with the single CV programmer on the Massoth Dimax via the Massoth 1200Z central station on my programming track.
3) Again, armed with the Marklin List, could the decoder be read, via Decoderpro, into one of the generic NRMA templates at the top of its list of templates, and then programmed using the menus. [I am going to try this later]

I would like to see Marklin actually publish the full name and version of the decoders it is using in its new MFX LGB locos, and indeed say if they are the same as these mLD/mSD items. I suspect they are probably not, as these are probably for HO models - but who knows?
At least it might make it easier to find published info about the LGB decoders, if we knew their names! Anyone know someone at Marklin?

 

[ntpntpntp]

Yeah, the sensible thing I'd expect to see would be that the decoder firmware is the same, just the large scale versions would have upgraded output stages for higher current and voltage.

 

[Cliff George]

Cliff I also have this LGB 24741 model and have spent quite a bit of time looking at the CV doco. I also found it very helpful. I was also pointed at the standard Marklin DCC CV list [ie all of it, not just the subset shown in this loco's manual] Unfortunately, the only full version I could find was in German, but I was able to muddle through this with the help of BING Translate. One thing I did observe was that the Marklin DCC interpretation for LGB 24741, although different from LGB/Massoth and again different from ZIMO, is the same as Marklin uses for its other model scales.
It would be useful if DecoderPro3 could auto-recognise the Marklin decoder type in LGB 24741 - I intend to try using one of the HO Marklin profiles to read its CVs into, as I suspect that may do the job.

However, as you have this loco, you may be able to answer a question that I have about its sound functions.

I will see if I can help. I got out my Spreedwald today and this is what I found. I haven't done anything with it for a year or so. Not sure this is going to help you much.

I find that the sounds seem to be quite low in volume, compared to older LGB locos. All the sound control CVs seem to be set to 255 by default, which suggests they are already at maximum capable volume. The whistle is OK, but the bell is almost inaudible. The chuff sound is rather quiet as well, although it is not too bad. I set the master volume CV back to 255, still much too loud for me!

Yes I agree that all the sound volume CVs are set to the max 255 by default. For me I obviously found the sound too loud as I had set the master volume in CV 63 down to 220 from 225. However I often reduce other older LGB sound locos volume by as much as a half so that would seem to provide some anecdotal evidence to indicate that indeed the max volume of this loco is lower than for older LGB sound equipped engines.

There does not seem to be a volume control knob, so the only way to change the volume would appear to be via the CVs.

I agree.

It has been suggested to me that the physical size of the loco and its speaker may just be such that it cannot produce any more sound than it now does.
Have you also found some of the sounds to be extra quiet, or is it possible there is something wrong with my particular unit that I need to follow up with the vendor? [It was brand new, when I bought it.]

Sorry but I cant really advise you, just to say that for me the sound is loud enough.

I don't want to waste much more effort trying to get more volume, if the model is simply not designed to produce any more!
I just hope this sound profile is not going to be repeated in the Larger LGB locos with the MFX chip - such as the new Harz 2-10-2, which I have on order.

The Spreedwald is the only MFX equipped engine I have, sorry I can't say if there is any difference in sound volume between different models.

As far as what CV programming environment to use, I have used decoderpro in the past but I am equally happy programming CVs one at a time which is what I do most of the time now. I'm a bits and bytes kind of guy I guess. Personally I wouldn't spend money on yet another CV programmer if I already had a DCC system that could do the job one CV at a time. Just my view.

 

[NCS from Qbyn]

Cliff
Thanks for this reply. I am starting to think that the loco is just a bit quieter than others. One thing more, did you notice if the bell - fn 3 - was noticeably quieter than the other sound functions on this loco? It is by far the quietest on mine.

 

[Cliff George]

Cliff
Thanks for this reply. I am starting to think that the loco is just a bit quieter than others. One thing more, did you notice if the bell - fn 3 - was noticeably quieter than the other sound functions on this loco? It is by far the quietest on mine.

Yes it is.

 

[NCS from Qbyn]

Yes it is.

Aha, well that's it then. At least it's not just me. I don't use the bell much anyway. At least I can stop searching for ways to make the sound louder. Thanks
Nigel

 

[Cliff George]

Hi Nigel,

I can see no obvious speaker grill in this loco, and I've not seen a schematic or had this engine apart but to my ear the sound is perhaps coming from the rear bunker? Perhaps a few hidden holes or fake coal load with holes would let more of the sound out?