Welcome to the forum   
Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error

Share
Options
View
Go to last post in this topic Go to first unread post in this topic
Offline hmsfix  
#1 Posted : 05 April 2005 19:58:53(UTC)
hmsfix


Joined: 06/02/2005(UTC)
Posts: 1,383
Location: Darmstadt,
I have a question to all the electronics freaks out there, concerning mfx: does anyone know how the loco communicates with the MS when it registers ? Well, there is a controler ID transmitted from the lok to the MS with a complicated protocol. But speaking purely about volts and amps, how could I make that visible on my oscilloscope ?

There must be some kind of electric impulses emitted from the lok decoder through the tracks, but on the other hand, it's always the control station that determines the voltage across the rails, otherwise there would be confusion with the whole digital system (as always when two are speaking at the same time)

Presently I have no mfx at my disposal, otherwise I would have test it myself. So, If any of you has already looked at this, I would appreciate to know how it works.

Hans Martin
Offline Sam  
#2 Posted : 06 April 2005 21:47:05(UTC)
Sam


Joined: 04/02/2002(UTC)
Posts: 799
Location: Phoenix, AZ
It would appear that nobody here is technical enough and/or owns an oscilloscope to do this experiment.. Try the Marklin Bar and Grill types as well, some of those guys are very technical. Apart from Gregor, I don't know anyone here that this in-depth.

Sam
Era I-V / HO & 1 Gauge / C-Track & Mobil Station, with Central Station.
Offline Mikael  
#3 Posted : 06 April 2005 23:07:17(UTC)
Mikael

Denmark   
Joined: 10/09/2004(UTC)
Posts: 959
Location: Copenhagen, Denmark
I would love to help you out with this. I do have a digital storage scope (and the knowledge to operate it wink). I also have a mfx lok or two. Unfortunately, I do not have a MS or any other controller that talks mfx, so there's not much I can do [:(]
Offline Webmaster  
#4 Posted : 06 April 2005 23:51:18(UTC)
Webmaster


Joined: 25/07/2001(UTC)
Posts: 11,165
Sam, I would not say like that.... Rather the opposite.... wink

Haven't touched a real scope myself in 20 years, but several in the group do it for a living... biggrin

The free software scope I occasionaly use is limited to 20KHz bandwith so it is of no use here...
Juhan - "Webmaster", at your service...
He who asks a question is a fool for five minutes. He who does not ask a question remains a fool forever. [Old Chinese Proverb]
Offline Lars Westerlind  
#5 Posted : 07 April 2005 00:55:20(UTC)
Lars Westerlind


Joined: 19/10/2001(UTC)
Posts: 2,379
Location: Lindome, Sweden
Sigh.
The attitude of some americans...
/lars
Offline Sam  
#6 Posted : 07 April 2005 01:00:48(UTC)
Sam


Joined: 04/02/2002(UTC)
Posts: 799
Location: Phoenix, AZ
Certainly maybe true Juhan... The problem with the B&G guys is they talk about 10% marklin and 90% everything else, including DCC...Which is why I don't go there anymore myself.

Era I-V / HO & 1 Gauge / C-Track & Mobil Station, with Central Station.
Offline jorge_vilarrubi  
#7 Posted : 07 April 2005 01:01:08(UTC)
jorge_vilarrubi


Joined: 15/12/2003(UTC)
Posts: 655
Location: Buenos Aires,
Quote:
[size=1" face="Verdana" id="quote]quote:Originally posted by Webmaster
<br />The free software scope I occasionaly use is limited to 20KHz bandwith so it is of no use here...

Hello Juhan, many years ago I built my own scope but it didn't work for more than ten years. However, the 20KHz free soft could be good for my audio frequency needs, could you please give me some info about where to download it from?
Thans a lot.
Best regards,
Jorge Vilarrubí
Buenos Aires
ARGENTINA
Offline hmsfix  
#8 Posted : 07 April 2005 01:09:24(UTC)
hmsfix


Joined: 06/02/2005(UTC)
Posts: 1,383
Location: Darmstadt,
Hi all,

I should explain that beyond model trains I have another, rather strange hobby: any new thing I get usually comes under oscilloscope observation until I understand what is going on: handy, toaster, rasor, computer network, and of course model trains. It's fascinating to see how it works and I feel less helpless in our technical world.

My point is, when I add some self-made electric circuit to my layout, could it eventually spoil the mfx signals from the lok ? There is no info on the M* webpages. Only chance is the oscilloscope.

Best regards

Hans Martin
Offline Webmaster  
#9 Posted : 07 April 2005 01:13:09(UTC)
Webmaster


Joined: 25/07/2001(UTC)
Posts: 11,165
Juhan - "Webmaster", at your service...
He who asks a question is a fool for five minutes. He who does not ask a question remains a fool forever. [Old Chinese Proverb]
Offline perz  
#10 Posted : 07 April 2005 01:20:25(UTC)
perz

Sweden   
Joined: 12/01/2002(UTC)
Posts: 2,578
Location: Sweden
Quote:
[size=1" face="Verdana" id="quote]quote:My point is, when I add some self-made electric circuit to my layout, could it eventually spoil the mfx signals from the lok ?

It depends on what you add of course, but as long as you add reasonable things it shouldn't be a problem. There are lots of things added to the electric circuit anyway on a normal layout. The mfx protocol must be robust enough to operate with a lot of different loads, like different decoders, lamps, smoke generators, brake circuits etc. Adding some minor extra load for measurement purposes could not be a problem.

If I had an mfx loco, I would do measurements. But I don't have any mfx loco yet.
Offline Webmaster  
#11 Posted : 07 April 2005 01:32:06(UTC)
Webmaster


Joined: 25/07/2001(UTC)
Posts: 11,165
When it comes to mfx-locos reporting to the MS/CS I think (I do not know...) that the communication goes something like this using a MS/CS broadcast at intervals:

MS/CS - "Hi there on the track, send me your personal info"

Lok 1 - "Ok, here is my data"

Lok 2 - "Ok, I'm here"

and so on...


I do not believe that the lok data is sent without a request from the MS/CS like this:

Lok 1 placed on track - Starts sending "Hello, here I am!!!"

Lok 2 placed on the track - "Howdy MS/CS, here I am"

MS/CS - "Hi guys, welcome to the track. You can stop shouting now, I see you!"


Why do I assume this? Well, if loks were to send a lot of extra traffic in a 6021/IB environment without any acknowledgement, the tracks would be filled with traffic from unsatisfied loks and that would slow down the performance...
Juhan - "Webmaster", at your service...
He who asks a question is a fool for five minutes. He who does not ask a question remains a fool forever. [Old Chinese Proverb]
Offline Lars Westerlind  
#12 Posted : 07 April 2005 10:02:52(UTC)
Lars Westerlind


Joined: 19/10/2001(UTC)
Posts: 2,379
Location: Lindome, Sweden
I agree Juhan,
in addition, I would expect the central occasionally to send out info ". I know about xxx,yyy and zzz. Are there any others", and sometimes, "would xxx be so kind to send it's attributes?". When a new loco appears, it's serial number will be sent to the central, and an internal (shorter) ID will be assigned and sent to the loco.

/lars
Offline digilox1  
#13 Posted : 07 April 2005 12:02:58(UTC)
digilox1


Joined: 28/05/2003(UTC)
Posts: 719
Location: ,
Juhan,

Quote:
[size=1" face="Verdana" id="quote]quote:Why do I assume this? Well, if loks were to send a lot of extra traffic in a 6021/IB environment without any acknowledgement, the tracks would be filled with traffic from unsatisfied loks and that would slow down the performance...


Why? I guess the mfx decoder will recognize a non-mfx environment and therefore suppress any log on messages.

I also think that the decoder will start the log on procedure, as the MS/CS never know what`s happening next and therefore would have to transmit a continuous stream of log on requests, thus clogging
data traffic on the tracks.

Speculation only...

Regards,
Manfred
Offline Lars Westerlind  
#14 Posted : 07 April 2005 13:31:05(UTC)
Lars Westerlind


Joined: 19/10/2001(UTC)
Posts: 2,379
Location: Lindome, Sweden
To me it makes sense if not every decoder is allowed to transmit at any time. When there are lots of mfx's present, and most of them are known by the central, it would make sense that the central ask's specific decoders, meaning all others should be silent. So I have the same thinking as Juhan, but for a slightly other reason.

The speed bar on a Mobile Station blinks when 'contact has been lost' it's said. This could be achieved in different ways of course, but the simplest is that it has asked a question and didn't get an answer.

Thinking based on how I think I would have implemented the features we know (or think we know) only (speculations).

/lars

Offline digilox1  
#15 Posted : 07 April 2005 15:40:23(UTC)
digilox1


Joined: 28/05/2003(UTC)
Posts: 719
Location: ,
Sorry,
missed the point.
I only thought of the log on procedure, not normal operation acknowledgement messages or requests.

Regards,
Manfred
Offline john black  
#16 Posted : 07 April 2005 16:37:57(UTC)
john black

United States   
Joined: 22/04/2004(UTC)
Posts: 12,139
Location: New York, NY
Quote:
[size=1" face="Verdana" id="quote]quote:Originally posted by Webmaster
<br />MS/CS - "Hi there on the track, send me your personal info"

Lok 1 - "Ok, here is my data"

Lok 2 - "Ok, I'm here"

and so on ...


Juhan, many thanks - this is great information ! Do they really talk in such a polite manner, or is cyberworld communication rather like real life: "C'mon guys, need your data - fork it over!" [}:)] ???
Long time ago we had an Deputy Director who sounded exactly that way ... [xx(]

John

I hope no one visits a poor Southener's layout in Brooklyn. Intruders beware of Gators.
AT&SF, D&RGW, T&P, SP, WP, UP, BN, NYC, ARR, epI-III - analog & digital Marklin Classics only.
CU#6021 FX-MOTOROLA DIGITAL SYSTEM. Fast as lightning and no trouble. What else ...
Outlaw Member of BIG JUHAN's OUTSIDER CLUB. With the most members, worldwide

Offline digilox1  
#17 Posted : 07 April 2005 17:40:49(UTC)
digilox1


Joined: 28/05/2003(UTC)
Posts: 719
Location: ,
John,
I think it sounds rather like Buzz in our beloved cartoon.

Regards,
Manfred
Offline john black  
#18 Posted : 07 April 2005 20:49:37(UTC)
john black

United States   
Joined: 22/04/2004(UTC)
Posts: 12,139
Location: New York, NY
biggrin
I hope no one visits a poor Southener's layout in Brooklyn. Intruders beware of Gators.
AT&SF, D&RGW, T&P, SP, WP, UP, BN, NYC, ARR, epI-III - analog & digital Marklin Classics only.
CU#6021 FX-MOTOROLA DIGITAL SYSTEM. Fast as lightning and no trouble. What else ...
Outlaw Member of BIG JUHAN's OUTSIDER CLUB. With the most members, worldwide

Offline Rudi Bekkers  
#19 Posted : 07 April 2005 20:58:55(UTC)
Rudi Bekkers


Joined: 15/02/2005(UTC)
Posts: 132
Location: Vught,
Hi,

I have not been able to observe the protocol myself, have a oscilloscope, a MS but no MFX loco's. Anyway, it would need a digital storage scope to see something useful, and even then it might be tricky to determine which messages came from where (the MS, the various loco's and other devices allowed to transmit data).

Nevertheless, it seems very likely the protocol used has many things in common with protocols found on two-way telecommunications networks, and with mobile phone networks in particular. The challenges the systems face is very similar. Very simplified, the protocol works like this (assuming a shared media for up- and downlink):

Messages are exchanged within so-called logical channels (which may share the same physical channels). These channels have their own features, and might be downlink or uplink. Advanced systems may have quite a lot of different type of logical channels

One may distinguish between logical channels of a random access type, and those that do not. The latter category are planned communications, Sender and receiver know which channel (or time slot) has been reserved for which particular message, and who will be originating it. In mobile networks, the network (base station) typically determines the planning. It sais: ok, in this logical channel, and this given time slot, I will send data to terminal 5, or terminal 7 will send some data to the base station.

Then there are these random access channels. For these it is, by definition, impossible to know in advance who wants to communicate something. It might be a new terminal, one of the more than 1 billion phones on earth. In this random access channel (often called RACH), any terminal might start sending, either at a random moment or according to some predetermined moments in time. The network will confirm successful reception, and will take the steps necessary to plan regular future logical channel assignments if necessary. Of course, it is impossible to prevent that two terminals happen to start sending at the same time during the RACH. In that case, none of them receives the correct acknowledgement. They each have to wait a random time and might start again. There are many, many variances on this model but the essence is usually the same.

Given the way that a Märklin MS has to cope with several known loco's and unpredictable new loco's on the track, it very likely that the protocols used are rather similar to the one sketched above.

/rudi
Offline David Dewar  
#20 Posted : 07 April 2005 22:21:22(UTC)
David Dewar

Scotland   
Joined: 01/02/2004(UTC)
Posts: 7,449
Location: Scotland

Maybe I am missing the point here but does it really matter??
I am just happy that it works.
David
Take care I like Marklin and will defend the worlds greatest model rail manufacturer.
Offline jorge_vilarrubi  
#21 Posted : 07 April 2005 23:51:01(UTC)
jorge_vilarrubi


Joined: 15/12/2003(UTC)
Posts: 655
Location: Buenos Aires,
Quote:
[size=1" face="Verdana" id="quote]quote:Originally posted by Webmaster
<br />Jorge
http://polly.phys.msu.su/~zeld/oscill.html


Thanks Juhan, nice tool!
Best regards.
Jorge Vilarrubí
Buenos Aires
ARGENTINA
Offline hmsfix  
#22 Posted : 08 April 2005 00:28:16(UTC)
hmsfix


Joined: 06/02/2005(UTC)
Posts: 1,383
Location: Darmstadt,
Hi all,

As Rudi I also believe that there is such a highly developed protocol running between the MS and the mfx decoder. I shall not try to resolve all its details with a simple oscilloscope.

However, what I want to understand is much more primitive: how does the mfx decoder send digital signals at all ? The MS, CU, IB etc. send signals via a serial protocol, switching the output voltage between +20 and -20 volts. But how can the mfx decoder do the same, i.e. how can it affect the voltage across the rails ? It has no own power source. In contrast, the MS output works through a brute force power amplifier. Finally you can run locos at this output! Even currents of several amperes have no significant effect on the signal voltage. If the tiny decoder chip would try to send signals this way, it would be fried within seconds.

So, as you see, my problem is still unsolved. Perhaps the decoder sends high frequency signals that cannot pass to the MS power amplifier. Don't know but would like to do.

Regards

Hans Martin
Offline Gregor  
#23 Posted : 08 April 2005 01:17:48(UTC)
Gregor

Netherlands   
Joined: 17/04/2003(UTC)
Posts: 997
Location: Netherlands
Hi All,

Although flattered by Sam's remark, I have to admit that I am not realy into electronics at all, yet I do own an oscilloscope. But it's an ancient model, one channel only, and with radio tubes in it. It has helped me in some cases to identify whether there is a digital signal present or not, but that's about it. So if anyone with real experience and real equipment finds out.... please share !!

Gregor
Offline nico van zon  
#24 Posted : 08 April 2005 02:08:26(UTC)
nico van zon


Joined: 25/07/2004(UTC)
Posts: 202
Location: ,
Quote:
[size=1" face="Verdana" id="quote]quote:Originally posted by hmsfix
<br />..... The MS, CU, IB etc. send signals via a serial protocol, switching the output voltage between +20 and -20 volts. But how can the mfx decoder do the same, i.e. how can it affect the voltage across the rails ? It has no own power source.....

I'm not quite sure about the workings, but I do have an idea about it. I once connected an oscilloscope to an MFX system. What I saw was alternating "old" Motorola protocols and MFX protocols. In between the voltage was held for a while at one of the levels, as far as I remember it was a negative voltage on the studs with respect to the rails. So it's power supply without a digital message. What I think (not sure) is that this power supply between MS protocols enables the decoders to send their messages, superimposed on the power level but with a much smaller voltage. At that time I was not able to go further into it. It was not my own railway and there was not enough time to do further investigation.

Apart from the fun of all this reverse engineering I'm of the opinion that Märklin and/or ESU should release all the details about the protocols. I am always very reluctant to buy a system when the manufacturer wants to keep the principles secret.
Offline Lars Westerlind  
#25 Posted : 08 April 2005 09:59:28(UTC)
Lars Westerlind


Joined: 19/10/2001(UTC)
Posts: 2,379
Location: Lindome, Sweden
Makes sense to me. In this way the "power" always comes from the central, and the decoder varies the current consumtion, which may be measured at the central (or booster). I've never really adopted the idea proposed that the central should now and then stop supplying power, and in thes gaps the decoder should send stored energy to the track.

As you clearly can se I'm not an electrichal engineer; rather more interested in physics.

/lars
Offline nico van zon  
#26 Posted : 08 April 2005 11:13:45(UTC)
nico van zon


Joined: 25/07/2004(UTC)
Posts: 202
Location: ,
Lars, DCC also has the possibility that decoders talk back. (Not yet implemented by any manufacturer)When you read the DCC specs (to be found on www.nmra.org) you can see that it works different. With DCC the power from the central and the boosters is switched off completely in certain intervals and the decoders talk back on stored energy. They don't talk back to the central however, but to the section occupancy detectors. And the message is much simpler. The decoder just says: "Hi, I'm number 47". And so the occupancy detector can report back to the central: "Hi, I'm block 14 and number 47 is here".
So the technique is somewhat similar, but the functionality is completely different.
Offline Lars Westerlind  
#27 Posted : 08 April 2005 11:40:19(UTC)
Lars Westerlind


Joined: 19/10/2001(UTC)
Posts: 2,379
Location: Lindome, Sweden
Thanks. Still, I don't *LIKE* the DCC feedback idea. As I say, I'm not knowledge enough, but it feels wrong, to have componets with unclear responsibilities in this way. Isn't there a clear risk that the timing goes wrong, and several will emit energy at the same time, and thus, having a great risk of causing damage to eachother? With just "power consumtion", this risk is clearly lower I think.

Besides,
I like my Lissy. To have "feedback" clearly separated, doint exactly what I want it todo (not only showing presence of a loco, but also causing event like sound or speed changes etc to the correct place, in a way which is far easier to understand than the DCC approach. And Märklin uses "feedback" to perform a task that I'm very willing to do manually (registry) [V].

/lars
Offline Rudi Bekkers  
#28 Posted : 08 April 2005 11:57:23(UTC)
Rudi Bekkers


Joined: 15/02/2005(UTC)
Posts: 132
Location: Vught,
I can think of two or three ways in which the loco's can communicate back with the system.

In the first two ways, a specific period in time (time slot) must be used in order to allow this uplink (either using a predetermined pattern, or following a system message). Then the two ways are:
- The central unit stops transmitting and listens for a while. Loco's send messages using power stored in a capacitor or so. It's a relatively short message, probably a few sets of bytes will do. They will be faced, of course, with the resistance that other power consumers pose to the rails. If the message is shore enough, and the received at the central is sensitive enough to recognize these (alternating) signals, it will work. If the bytes are send out rather fast (e.g. at a high frequency), than the resistance might be lower.
- The central unit acts as a continuous power source. It then looks carefully at the load (current drawn). Loco's that want to transmit something, do so by a transistor that short-circuits the line shortly in order to send bits (think of an open-collector design). Also here, there is also the load of other power consumers (motors, lights in cars, etc.) but this other loads are relatively stable. And, as we know from the digital braking practices, these other trains have no problems being fed with -15 VDC, especially if this is only for a short time.

A third, rather different way is that no random access time slot is reserved at all, but a separate, independent way is sought for uplink communications. This could be done by having the loco's send their messages on a much higher frequency (that's also what Hans suggests). In fact they would employ RF transmitters, modulating their bytes upon a carrier wave. However, as someone noticed seeing two different protocols alternating on the line, this seems not to be the case. If you would use a simple frequency analyzer (for example FTT - Fast Fourier Transformation = real-time analyses of frequency components) you would be able to see such protocols.

Once we understand the protocol, we could have a follow-up discussion what the new MFX-compatible boosters exactly do (there are several choices/issues there as well!).

Agaim, all of the above is speculative only, not having analysed the protocol myself.

I agree that it is not necessary to understand this (as long as all runs, many of us will be happy), however, it is also nice and challenging to understand.... It could help you solving problems, determining what is allowed when adding things (esp. circuits) to your layout, and may even allow you to build MFX-compatible things yourself.


Quote:
[size=1" face="Verdana" id="quote]quote: Apart from the fun of all this reverse engineering I'm of the opinion that Märklin and/or ESU should release all the details about the protocols.
I cannot agree more. It gives trust to consumers, prevemnts lock-in, allow more freedom and would facilitate allow the entry of third party producers (which might, eventually, be in the interest of M* too). And, for such relatively simple protocols, keeping secrets is not so usefull anyway because with reverse engineering other competitors or third parties will find out anyway. (Assuming to patents on Märklin's side, i did a quick scan on www.espacenet.com but that's no patent search in any way). Some of you might know somthing of the famous IBM System/360 legal fight for access to specifications, very interesting in this context.

Ok, time to stop, I'm drifting off.

best, rudi


Offline Rudi Bekkers  
#29 Posted : 08 April 2005 12:09:58(UTC)
Rudi Bekkers


Joined: 15/02/2005(UTC)
Posts: 132
Location: Vught,
Me again, found something after all that might interest you. M* does have some patents, though not filed at EPO but in Germany only (afgter all, the biggest market for their products?)

I saw some eight patents:

DE10251086 Verfahren und Fahrtregler zum Steuern mehrerer Fahrzeuge einer Modellbahnanlage
DE10302046 Stromabnehmer für ein Modellfahrzeug
DE10204540 Vorrichtung zur Verbindung zweier Modelleisenbahnfahrzeuge
DE10236099 Modellbahnfahrzeug
DE10204539 Spielzeugmodell
DE10204538 Modellbahndampflokomotive
DE10222019 Verfahren zum Steuern eines Modelleisenbahnzuges sowie Modelleisenbahnzug zur Durchführung eines Verfahrens
DE10204542 Lichtsignalvorrichtung für eine Modellbahn

Several might involve their digital and/or MFX protocols or more general architactural matters related to that. You can look at full-text (facsimile) patents at http://de.espacenet.com/. Though in German only.... And for those of you that are not used looking at patents: dont's expect to fund a full MFX description here!

I'll read some of these over the weekend, did not look for corresponding patents or other patents yet.

If some of you know of more patents on this matter (by M* or others, I think that Lenz had an important but now expired patent).

best rudi

Offline McLae  
#30 Posted : 08 April 2005 16:48:37(UTC)
McLae


Joined: 16/07/2002(UTC)
Posts: 1,575
Location: DeSoto (Dallas area), TX
I expect one of those patents will be the hinge in the middle of the frame of large steam loks. (BR50, Br85, etc) This is why so many non-Marklin Steam Loks do not handle R1.[:(]
The McLae
IB digital, DB, OBB, SBB epII-V
Providing a home for little lost 'Gators
Offline john black  
#31 Posted : 08 April 2005 17:36:22(UTC)
john black

United States   
Joined: 22/04/2004(UTC)
Posts: 12,139
Location: New York, NY
Quote:
[size=1" face="Verdana" id="quote]quote:a system the manufacturer wants to keep the principles secret


Quite sure there'll be no secrets anymore - as soon our Forum's Electronic Gurus get their hands on the CS [}:)]

John

I hope no one visits a poor Southener's layout in Brooklyn. Intruders beware of Gators.
AT&SF, D&RGW, T&P, SP, WP, UP, BN, NYC, ARR, epI-III - analog & digital Marklin Classics only.
CU#6021 FX-MOTOROLA DIGITAL SYSTEM. Fast as lightning and no trouble. What else ...
Outlaw Member of BIG JUHAN's OUTSIDER CLUB. With the most members, worldwide

Offline john black  
#32 Posted : 08 April 2005 17:41:36(UTC)
john black

United States   
Joined: 22/04/2004(UTC)
Posts: 12,139
Location: New York, NY
Quote:
[size=1" face="Verdana" id="quote]quote:Originally posted by McLae
<br />This is why so many non-Marklin Steam Loks do not handle R1 [:(]


Bad luck - the early bird catches the worm ... Cool

John

I hope no one visits a poor Southener's layout in Brooklyn. Intruders beware of Gators.
AT&SF, D&RGW, T&P, SP, WP, UP, BN, NYC, ARR, epI-III - analog & digital Marklin Classics only.
CU#6021 FX-MOTOROLA DIGITAL SYSTEM. Fast as lightning and no trouble. What else ...
Outlaw Member of BIG JUHAN's OUTSIDER CLUB. With the most members, worldwide

Offline nico van zon  
#33 Posted : 08 April 2005 18:02:40(UTC)
nico van zon


Joined: 25/07/2004(UTC)
Posts: 202
Location: ,
Quote:
[size=1" face="Verdana" id="quote]quote:Originally posted by Rudi Bekkers
<br />.... And for those of you that are not used looking at patents: dont's expect to fund a full MFX description here! ....

Of course not. Because those patents don't describe mfx. And that's because mfx is not patented at all.
Apart from that, if it was patented then it would be a patent from ESU, who is the real inventor of mfx.
Offline hmsfix  
#34 Posted : 08 April 2005 18:17:30(UTC)
hmsfix


Joined: 06/02/2005(UTC)
Posts: 1,383
Location: Darmstadt,
Hi nico,

The patents cited by Rudi are all from the years 2002 and 2003. We can't know what M* has invented in 2004 and 2005 because in Germany and most other countries new patents are kept secret for the first 18 months after their application. Therefore we eventually shall not see the full mfx specifications before 2006 or 2007.

So there *** are *** secrets

Hans Martin
Offline nico van zon  
#35 Posted : 08 April 2005 18:37:35(UTC)
nico van zon


Joined: 25/07/2004(UTC)
Posts: 202
Location: ,
Hans Martin,
I'm sorry to say that you are totally wrong.
As soon as an inventor applies for a patent, the description is made public. And that's because everyone else must have the possibility to check if the patent violates his rights.

Your statement is a common mistake. Patents are not secret, they are public knowledge right from the start by legislation.

So the simple fact that no patent or patentapplication for mfx can be found is for me the ultimate proof that no patent exists or is applied for.
Offline john black  
#36 Posted : 08 April 2005 18:58:26(UTC)
john black

United States   
Joined: 22/04/2004(UTC)
Posts: 12,139
Location: New York, NY
Quote:
[size=1" face="Verdana" id="quote]quote:Originally posted by nico van zon
<br />if it was patented then it would be a patent from ESU, who is the real inventor of mfx


Well. Excuse me, Gentlemen - but here comes my dumb question: If MFX is really "The Big One" M is producing so much secrecy about - why didn't it's inventor ESU bother for a patent, then ??????
My only conclusion: perhaps MFX isn't that important at all - except for M ... biggrin

John

I hope no one visits a poor Southener's layout in Brooklyn. Intruders beware of Gators.
AT&SF, D&RGW, T&P, SP, WP, UP, BN, NYC, ARR, epI-III - analog & digital Marklin Classics only.
CU#6021 FX-MOTOROLA DIGITAL SYSTEM. Fast as lightning and no trouble. What else ...
Outlaw Member of BIG JUHAN's OUTSIDER CLUB. With the most members, worldwide

Offline nico van zon  
#37 Posted : 08 April 2005 19:10:01(UTC)
nico van zon


Joined: 25/07/2004(UTC)
Posts: 202
Location: ,
John,
I think that ESU didn't bother for a patent just to keep it secret.
Offline hmsfix  
#38 Posted : 08 April 2005 19:29:34(UTC)
hmsfix


Joined: 06/02/2005(UTC)
Posts: 1,383
Location: Darmstadt,
Hi John,

I think you are right with your statement: for me mfx is of minor importance, I am more interested in overall quality of the models, and up to now I didn't have any problem to remember the ID numebers of my 4 digital loks. Nevertheless it's a challenge to solve this mfx enigma.

Nico,

I should indeed have ben a little bit more precise about patents. You are right: when a patent is granted, it must be published such that anybody else can see whether it affects him. But between application of a patent by the inventor and the time when it is granted there can be up to 7 years (!), at least according to german patent rights. In the first 1 1/2 years of this period the application is secret, then it is published. The owner has the priority on the invention from the date of application, though there is not yet a protection. Protection starts with being granted. Thats a big trick in industrial research: you can never know what your competitor has in mind.

Best regards

Hans Martin
Offline perz  
#39 Posted : 08 April 2005 21:20:21(UTC)
perz

Sweden   
Joined: 12/01/2002(UTC)
Posts: 2,578
Location: Sweden
Regarding communication speed from the mfx decoder, we have already figured out that it uses the "RDS" system, which means 1.1 kbit/s data rate sent with a 57 kHz carrier. See older threads. We haven't found out if it uses stored energy or modulates the power consumption, but given the low bit rate and the probably rather long message, the stored energy method is less likely.
Offline Rudi Bekkers  
#40 Posted : 13 April 2005 13:58:18(UTC)
Rudi Bekkers


Joined: 15/02/2005(UTC)
Posts: 132
Location: Vught,
Hi Perz,

Funny, it indeed seems to be the RDS protocol as has been long used to send data with radio programs. I also saw this at the bogobit site (maybe you are involved in this site, I don't know).

This means that M* could use the rather affordable semiconductors already available. For the real freaks, however, it would allow them to decode the signal easily, and maybe even design their own senderswink

/rudi
Offline hmsfix  
#41 Posted : 13 April 2005 22:47:35(UTC)
hmsfix


Joined: 06/02/2005(UTC)
Posts: 1,383
Location: Darmstadt,
Hi Rudi

Thanks for that very interesting bogobit link. Now I know what I have to look for.

Hans Martin
Offline perz  
#42 Posted : 13 April 2005 23:47:50(UTC)
perz

Sweden   
Joined: 12/01/2002(UTC)
Posts: 2,578
Location: Sweden
I am not involved in the "bogobit" site, but I look there.
Users browsing this topic
Guest
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.

| Powered by YAF.NET | YAF.NET © 2003-2025, Yet Another Forum.NET
This page was generated in 0.870 seconds.