Hi Lutz,

Could you please let us know the dimensions? And are these manufactured inhouse?
And when did Marklin start fitting them to the loco's.

Thanks,

Mike.

Very interesting,Lutz...!
Will that means Marklin can or and does producing better quality of decoder from now and on...?
Next step will been in the sound too...?
For Trix company too...?

Goofy

A big positive that I see from this is that more and more Marklin loks will come with 21 pin sockets. This makes it easy to switch to any decoder or replace in them in the event of failure.

Another waste of money. Why develop a new decoder when you can get it relativly cheap and good from ESU?
37940 should have one installed.

Indeed...!

Goofy

It were about time. But I wouldn't bet my money on this

OK, I have a T44 on the way from ETS. Shipped a few days ago so it will arrive next week. I hope I will have time to test it thoroughly when I get it.

Now can we hope for an mfx function decoder to be produced? This is needed for the 41928 Rheingold coaches set, for example - the instructions for this show Märklin colour conventions for the wiring of a decoder, but I think there's no decoder currently in production (by Märklin or anyone else) that conforms to these.

Any inside information about this, Lutz?

Best wishes,

Nigel

Thanks Lutz,

I appreciate the information.

Do you have any idea if they will produce a more basic version similar to the decoder in the 60760 kit?

Cheers....

Mike.

I have got my T44. I have only started the tests, but I can already say:

No. This is not an improvement. Regulation is not as good as on previous models. And it has the old work-around for the Alzheimer problem, with the side effect that the lok starts very abruptly if the power is turned off and then on again (e.g. at an old style signal). It also does not respond to attempts to program CV 73, so I can't set it to "Alzheimer mode" to see if it really suffers from that or not.
I have also had the speed bar blinking (indicating it does not get mfx feedback), even on my small test track. Haven't ever seen that on any of my other mfx loks except when I deliberatly worsen the conditions a lot for testing purposes.

It looks as it has a more linear default speed curve, 27 speed steps the old Märklin way in MM mode and more a "slope" than a "delay" when changing speed. More back to the 6090x way of working. It requires much more tests before I have really verified these things. So far it is just based on the "feeling" when driving it. Those things I regard as improvements, but it can not help up the bad regulation and the not fully fixed handling of power off/on situations.

This is completely the opposite development compared to my 37900, which was at Märklin for an upgrade, and came back with a perfectly smooth regulation, a perfectly correct handling of both short and long power breaks, a heavily non-linear speed curve, only 14 speed steps in MM mode, and more "delay" than "slope" when changing speed.

Why can't they make everything right in one and the same lok ?

Note: I have used a Mobile Station and a 6021 to test with so far.

Another note: The lok is still not bad, in absolute terms.

Nothing, of course. Would you expect anything else ? No instruction booklet I have ever seen from Märklin has ever mentioned CV73. Märklin have mentioned it in their FAQ though, but of course not with reference to this specific model. You know this as well as I do.

Anyway. Test results are rather discouraging. Bad regulation (relatively seen), and non-fixing one of the most annoying problems they have, even removing the option to choose between the two previously available flavours of that problem.

If it is going to stay this way, the ESU mfx decoders is clearly a better alternative, even if they aren't perfect in all respects either.

You can test for yourself if you have one of these models.

I am a bit confused about the behaviour of the T44.

It has a Soft Drive Sinus motor, and according to what other people have said in this forum, the motor regulation for the motor is not in the controller but on a separate driver board. Yet, this lok does not have at all as good regulation as the other SDS (or upgraded New C-sine) models I have seen. In some respects it is even worse (or to be more correct "less good") than the 6090x.

What differs between this one and previous SDS models should be the decoder I think, not the SDS driver circuit. But I may be wrong there, I am not an expert in distinguishing between different decoder or driver circuit types.

---

And to Lutz:

I am not "complaining" about the missing CV73. I am just telling the facts, it is not there. However, I am complaining about that they haven't fixed the problem that made CV73 a meaningful feature.

Perz,
I am a bit confused here. Has a Post been deleted where you are accused of being Biased????

Nev

[Relatively offtopic] I have just ordered a couple of Zimo MX64 decoders to try out, now that CS2 is also DCC. They even provide special decoders for Märklin's Softdrives, even if there are special precautions to take when using those decoders with Softdrive driver boards.[end offtopic]

I actually bought these decoders for my glockenanker conversions. As to driving SDS with any other decoder other factoy spec ones (not only Zimo but ALL decoders), you just need to locate the protection resistors and check if they are actually shunts or resistors and replace them with resistors in case of them being shunts.

If one doesn't have SMD desoldering/soldering skills, just add common resistors in parallel to the shunts, via thin wires.

No, I don't have an ecos, and I haven't seen this problem. I haven't done any testing to verify non-existance of the problem either.