Hello!

With an mfx loco, the CS shows you all the current settings: maximum speed, acceleration and braking delay, etc.
And acceleration and braking delay are set as seconds.

The meaning of writing 20 into CVs 3 and 4 is different - depending on the decoder.

With an Intellibox you can read the contents of CVs 3 and 4 (and others, of course) and check the decoder manual to find what it means.
OK - a combination of PC and IB with a decoder database could do that work for you and show seconds in the GUI.

However, the CS is far from being perfect.
When selecting a loco from the database, it always sets "Motorola 14" - it's up to the user to know if the loco supports 27 or 28 speed steps.
You can always set the volume for MM locos - even if the volume can't be adjusted or the loco has no sound at all ...

Well, regarding the CS - I almost fall asleep while it boots... Even our company's mobile PC:s with the Win XP Embedded custom OS I have made are much faster to boot to a ready state... But with Linux, they are just as slow as the CS, unfortunately... So it's obviously a software issue...

I mean, an IB + PC with appropriate train controller software does not take much longer to boot than the CS... I find this a bit annoying regarding a dedicated hardware box...

A dedicated controller should start much faster, in my humble opinion...

Another thing that annoys me is that the multitasking seems a bit under par, if I have a dialog box open on top of the background display - like selection of something - the control knobs do not respond... Strange software priorities...

But in general it's a good controller, even if I am a bit IB-biased...

Doing it with one wire is indeed very difficult.
Doing it with two wires, as is the case here, is not trivial, but at least practically doable.



Again, it is not "single wire" even if it is sometimes called that. There are always two wires to the track.



The reverse channel of mfx uses the principle of modulating the current draw of the decoder. The current draw is modulated with 57 kHz which is the standardized RDS modulation frequency. The coding of the reverse data is superimposed on the 57 kHz carrier frequency according to the RDS standard. This makes it possible to use a standard RDS decoder chip to decode it in the MS/CS. To extract the signal, the track suppply is fed through a signal transformer, and the signal is picked up on the secondary winding. This principle allows you to detect very minor current variations, even if the total current is huge.

The same principle of current draw modulation is used in the Digitrax feedback system. The principle was invented and patented by a man named Ireland (don't remember his first name). The patent is licensed to Digitrax, but is written in a way that it only protects the situation where the replies from several sections are multiplexed. I.e., it stops non-licensees from using the principle for train position identification.

/lw

Damon, is it like "multiplexing"?

OK Damon.
I lost it at "Err"

1. http://www.alice-dsl.net/mue473/mfxmenue.htm

2. By putting together the information from numerous posts in this forum by several people over a long time.

3. Personal investigations.

4 General knowledge about electronics.

I must say that I have not yet made any successful attempts to extract the reverse mfx signal myself. But there was someone in this forum who described it rather detailed long ago. Don't remember who it was.




It means that there is some device that takes in and samples the signals from all the different sections in parallel, and then sends back the sampled results one after the other on one line to the Control unit.

It does not have to be implemented exactly like this physically, but as I understand it, the patent protects the use where you can distinguish that a lok is present on one segment as opposed to another segment. It does not seem to protect the usage where the signals are not "multiplexed" i.e. just used to identify if the lok is present anywhere on the layout or not.

Very true Juhan. Compared to myself Perz is a genius. I just like everything to work as the instruction book says as to how it works is way beyond me. I can change a light bulb but anything else requires an electrican

David

Ditto for me. I can't even begin to count the number of times I've driven home from the night shift (10:30pm) with good intentions to get down into the train room and enjoy, only to find myself waking up on the couch at 2:30am after reading the newspaper and somehow nodding off.....

You'd think with the wife and kids (3 kiddies aged 8 and under) all fast asleep it would be prime time for me!!!!

But alas.........no [V]

I am familiar with that site, but being (unfortunately) mono-lingual, I had forgotten about it.
I assume "Fenster für Bestätigungs" refers to the idle time that the mfx decoder communicates back to the controller.



OK, here I want to tread very carefully, in case I just misunderstood your "tone of voice" (Damn, I hate internet forums -- you can never hear the other persons voice...[:(][:(])
I have been an Electronic Engineer for 20 years, designing all sorts of microcontroller & portable devices. I have difficulty finding information about mfx, I don't have difficulty understanding it.
My apologies if I took that the wrong way...[:I][:I]



Reverse Engineering a protocol like mfx is not trivial.



OK, thanks for the explanation.

Bummer, I had previously thought of breaking my future layout into small sections, and using mfx to locate decoders. Of course, if I'm not going to sell the technique to anyone...

Per, I assume you are in the electronic or software design field. I myself get a bit frustrated with Märklin's "explanations" of digital train control. I suspect it is relatively simple, but trying to get past the "lowest common denominator" factor means that I get lost...[:I][:I]
I wish there was some "techo/geek-speak" documentation.

I am not trying to be sarcastic. It is, as you say, difficult to find information about mfx. A person with good knowledge in electronics will understand it, if he/she gets all the pieces of information. But in this case the pieces can be that somebody mentioned one little observation three years ago and somebody else mentioned another little observation two years ago. Not so easy to catch and remember everything. It is kind of a detective work. However Rainer Müller has done most of the hard work for us, don't know how he managed to find out everything so quickly.



Unfortunately this is not such a good idea. Or rather, the idea is good, but it won't work so well with mfx. The CS or MS send "PING" messages regularly to the loks, but the "PING" in mfx is addressed to a specific unit and the answer is just a simple "yes" or not. This means two things:

- The sample interval for a specific lok will be quite long if you have many loks. The CS/MS can not just send "PING" messages, they mainly send "useful" messages. I think there would be a problem with slow detection because of this.

- You can detect that you get a "PING response" on a specific section, but to correlate it to a specific lok, you will need to decode the "PING" message that the CS/MS has sent out as well. Doable but rather complicated.

The Digitrax system uses a broadcast "PING" and a coded response, I think. This makes the detection much faster, but on the other hand makes it less useful if two or more loks are on the same segment.



Yes, I am a chip designer. Nominally that is electronic design, but with modern design methods it is often more like programming.



The traditional 6015 / 6017 booster is just an amplifier/repeater. It does not decode and re-encode the signal. It is however built for amplifying digital signals. A "1" will give a "1" (+20 V) out and a "0" in will give "0" (-20 V) out. If you feed it with an analog signal (e.g. a sine wave) you will probably not get a nice analog signal out. I can't say for sure that you don't but it seems unlikely given the design of it. The Delta (6604 or 66045) is a more complicated device, which "emulates" the simple repeater function. Therefore, the delay through the Delta unit when used as a booster is significantly longer than trough a 6015 / 6017.

When it comes to the new mfx capable booster I don't know the details, but the output portion of it is most likely just as simple as in the old booster. For the return channel, I think the booster will extract the signal as I described it, decode it, and communicate the result over the CAN bus back to the CS. Note that there is some speculation in this since I haven't seen those boosters nor have I seen any descriptions of their internals.

I have seen the suggestion that you could use a 6017 booster and extract the reverse signal with a signal transformer on the track feed from the 6017 booster. Then amplify it and let it cause a modulated current draw on the direct CS segment. It should work, at least in theory.

Hi John I wonder just how many forums he causes trouble. That site looks good but problem is the translation which may distort what is being said.

David

Thanks for information, dear Steven . You're a never ending source ...

David, Sire - will send you an translated excerpt, soon.

Funny isn't it John, how the same issues keep raising their ugly heads, no matter where you go? What does that say to you?

A lot was lost in the Google translations, but I kept seeing the words "Lutz" and "Forum Ban" in the same sentence....

Thanks [:I], friends. Just our daily job ...

And re the promotion (Chief of MBI sounds great, Macca ) -
but I'd rather be Director ( =$$$ [:p][:p][:p]) of <u>our</u> company.

Adrian - alas, having never met our dear Nev I know his picture, at least ...
So I can imagine quite well (it jumps at you !!!) what guy you're comparing with. Not fair

Yes, the likeness is astounding. Plus, the writer is having a go at Lutz as well, something about a banning from a forum!

Sounds like our Nev!

I think you're right, David.

I wonder when Lutz will call it "discontinued".
It is not available, it is not actively marketed - but it's not discontinued.

It's no longer listed at Modellbahnecke, it's listed as a discontinued item at ETS.
I think the traffic light in M*'s product database will turn grey from yellow without showing green again.

M* SYSTEMS is discontinued (except for the MS maybe), now we have M* DIGITAL!

No regrets, Tom ...
Their chaotic adventure brought us grey hair and them huge losses, presumably [}:)]
That's the only cause they scrapped it - after having released it just before [xx(]