TMCC serial sending unexpected data? - Deep down in the technical nitty-gritty stuff

Isn't there a reverse function available on the remote when spinning fast CCW? On the Legacy controller it's called Velocity Throttle Direction Control or V.T.D.C. and can be set ON or OFF. And there is also Velocity Throttle Automatic Brake, V.T.A.B. ON or OFF.

It's not VTAB.

Some CAB1's implemented an internal relative speed counter. When that counter decremented to zero after "enough" counter-clockwise spins, these CAB1's  will cease outputting relative speed commands and instead send absolute speed zero commands. Not all CAB1's have this behavior, but it seems yours does.

WOW, another interesting tidbit about the TMCC internals.

So in a short answer, it IS normal behavior for the system to send absolute speed zero commands after some period of turning the wheel counter-clockwise?  

Am I correct that TMCC engines do not read absolute speed instructions and should take this CCW spin absolute speed zero as an instruction to slow down another speed step?  Or should it be read as a stop moving command?  

JGL

I was under the impression that one of the "improvements" in performance with Legacy was the fact that it uses absolute commands for TMCC.  Since running in TMCC mode, the steps are very consistent, the operation seems to bear that out.  I believe that the R2LC knows how to handle absolute step commands, it's just that the CAB1 was originally designed to just run the PowerMaster at first and the TMCC locomotives came after the CAB1.  It was also stated as the reason that the original Odyssey worked better with Legacy than the CAB1.

Let's see if my guess is correct.

Rudy?

Also, there are differences in the throttle in ACC mode for certain cab-1's.  When the cab-1 is the newer version and sends Absolute zeros after 32 throttle downs, it also is able to send throttle commands in ACC mode.  The older cab-1 would send throttle down commands and not switch to sending Absolute zeros for ENG or TR mode; but also did not send any throttle commands for ACC mode.

A bit of trivia, as mentioned by the OP, the Absolute zero would cause the locos to come to an immediate stop.  Frustratingly, all of the newer locos have to ignore the Absolute zero and convert the Absolute zero to relative -2 down to prevent the abrupt stop.

Jon, thanks for the updates, it's interesting how many little details are buried in the operation.  I didn't realize there were multiple versions of the CAB1.  Is there any way to recognize if you have a new or old version?

Wouldn't it be great to have a history and as well technical book written about Trainmaster and Legacy Command Control. And even more interesting and useful for those of us that have committed our layouts to Lionel TMCC and Legacy.

Yes and all the hidden codes and in and outs of the R2lC-C XXs.  My understanding is that technical documentation was not fully kept during those years.

A little like some of the software in QSI Proto Sound chips.  The engineer wrote it, but never really documented what he did.  G

Thanks everyone so far. 

As I understand it now, a TMCC locomotive WILL respond to absolute speed commands, and this is how Legacy sends speed instructions to TMCC engines, so to be compatible with Legacy a device would need to be able to read absolute speeds as well as relative commands.  

In addition, some, but not all, Cab1 remotes will send 32 relative speed down commands, then send absolute speed zero on any further turning of the wheel.  From here the locomotive should read those zero commands as relative speed minus 2, NOT as a command to stop?

JGL

Hi John - A lot of excellent in formation. I do have a question.

What are the procedures to read, serial data coming off my TMCC command base?

I run Cab-1 with my base. I am lost at the station. 

trainroomgary posted:

Hi John - A lot of excellent in formation. I do have a question.

What are the procedures to read, serial data coming off my TMCC command base?

I run Cab-1 with my base. I am lost at the station. 

As far as how to read the data, you can connect the serial directly to a serial port on a computer and use a serial monitor program to look at the data in real time.  

For my purposes I have an Arduino micro-controller connected to the TMCC base's serial port.  The command base output's data in a sort of half-a##ed RS-232 standard used by most modern electronics.  True RS-232 uses a negative 15 volt signal as a high, and positive 15 volt as a low, with the large voltage difference the best way to send reliable data when the standard was developed in the 1960's.  Modern computer serial ports are still built to accept this standard, but most modern devices have replaced the high and negative voltages with 0 volts for High, and 5 volts for a Low.  This is how the TMCC base sends data.  This causes a little problem when reading the data into non-RS-232 serial devices, such as those used on pretty much all micro-controllers... the highs and lows are backwards.  I solved this simply enough with a 7404 inverter to swap highs for lows and vice versa.  From there I can process the incoming  16 bits of data in whatever way I like to read or write any tmcc command.  

To the last part, "I run Cab-1 with my base. I am lost at the station. ", I'm not exactly sure what you mean.  Are you asking why someone would want to read the data, or simply saying that for you, like 99.99% of other users,  there is no need or want to delve deep into the technical workings?  

For the latter, that is mostly why I titled this post as I did, to let people know it wasn't a typical sort of 'how do I add an engine to my remote' sort of topic.  The former, I will be writing up a rather detailed post in the near future with all the fun details and information, once I have a worked out all the kinks.  In general, folks the the serial connection to the command base to use a computer to control their railroad,  or to build their own switch/accessory controllers and the like.  You can also gain access to extra locomotives, switches, acc's, routes, and train/tracks.  Tmcc supports the next higher binary step for each address than what the remote is capable of addressing, so for example with serial communication you can have 16 routes and 128 engines... Need to double check that though because I think address 0 on each type is not a valid setting, so maybe 15 and 127... 

In the end, if you're happy with the name brand products and such,  this stuff really doesn't matter.  If you like tinkering with things and building custom devices, it can be a lot of fun to dig into.  

JGL

P.s.  Where abouts Detroit are you hailing from, Gary.  I'm about 2 miles away from a LHS on Groesbeck that I've seen you mention in posts a while back.  

You got all that Gary!    G

JohnGaltLine posted

 In the end, if you're happy with the name brand products and such,  this stuff really doesn't matter.  If you like tinkering with things and building custom devices, it can be a lot of fun to dig into.  

JGL

P.s.  Where abouts Detroit are you hailing from, Gary.  I'm about 2 miles away from a LHS on Groesbeck that I've seen you mention in posts a while back.  

Hi John - I took several electronic classes in High School & College, know just enough to be dangerous. Live in a historic home and have done a lot of electrical projects from major to minor jobs. Even have a Michigan Historic Marker, as you drive into the neighborhood.

My layout has 4 main lines, and 3 blocks to each main line so I can run the old stuff. This project you are talking about I find interesting, but it is over my pay grade.

I live about 40 miles West of our LHS, I was in there on Dec. 26th., to pick up two built to orders. One from Lionel and the other from M.T.H.  I did not pre-order the MTH / WiFi / DCS, but he had them in stock and I took a pass, with the two pre-orders I did not want to go over budget.

I am still interested in the MTH / WiFi, but I want to see what happens with the MTH design out of St. Joesph, MI. If he runs out, I will catch the next, "Slow Boat From China".

Gary and

The MTH wifi also has my interest going as it seems a 99 cent wifi module would make for a good way to get a micro-controler talking to DCS, as opposed to the fairly complex methods that I've seen documented thus far.  MTH sure doesn't like folks digging into the programing the way Lionel seems to welcome it with easy access to the basic information needed.  

JGL

Be aware that the Legacy base uses more or less true RS-232 voltages rather than 0 - 5v.  A MAX232 or equivalent chip would be recommended for interfacing a microcontroller to the Legacy base.

If you want to read the TMCC serial data with a computer instead of a microcontroller, you can use a USB-RS232 converter cable and read the serial data with a terminal program.

Professor Chaos posted:

Be aware that the Legacy base uses more or less true RS-232 voltages rather than 0 - 5v.  A MAX232 or equivalent chip would be recommended for interfacing a microcontroller to the Legacy base.

If you want to read the TMCC serial data with a computer instead of a microcontroller, you can use a USB-RS232 converter cable and read the serial data with a terminal program.

For my purposes I do want to talk to a microcontroller, but it is good information to have.  I have not done a ton of research on legacy yet, but will be getting into it further soon enough.  Does the Legacy base still use 9600 baud, or has it's speed been raised as well?  I seem to recall that you need something between  the base and other devices for them to work correctly, but not sure.  

JGL

JohnGaltLine posted:

The MTH wifi also has my interest going as it seems a 99 cent wifi module would make for a good way to get a micro-controler talking to DCS, as opposed to the fairly complex methods that I've seen documented thus far.  MTH sure doesn't like folks digging into the programing the way Lionel seems to welcome it with easy access to the basic information needed.  

JGL

Please build it.  Why is everything so simple to make.  While TMCC access was granted, I don't believe Legacy was.  Instead of producing this stuff maybe the manufacturer can just sell every one a plan/diagram.   G

As GRJ likes to say Nothing is so easy as the job you imagine someone else doing.  That said, I've found working with TMCC protocol, arduino electronics, and wifi/bluetooth modules to be fairly straight forward, and to have a pretty easy learning curve.  As to what would or would not be allowed to come to market, that is a whole other thing, but the information on Legacy's command structure is freely available.   From what I've seen here from comments in other threads talking about it, the DCS gurus don't seem have nearly the same open minded philosophy to letting people know how to work with the system.  

JohnGaltLine posted:

As GRJ likes to say Nothing is so easy as the job you imagine someone else doing.  That said, I've found working with TMCC protocol, arduino electronics, and wifi/bluetooth modules to be fairly straight forward, and to have a pretty easy learning curve.  As to what would or would not be allowed to come to market, that is a whole other thing, but the information on Legacy's command structure is freely available.   From what I've seen here from comments in other threads talking about it, the DCS gurus don't seem have nearly the same open minded philosophy to letting people know how to work with the system.  

 

Actually Stan said, and John adopted it.  I guess that is the reason for patents, and how an inventor can make a profit for his efforts.  Should John give out his super chuff code so other can make his board?  I do think he put out the diagram for the board.     G

GGG posted:

JohnGaltLine posted:

As GRJ likes to say Nothing is so easy as the job you imagine someone else doing.  That said, I've found working with TMCC protocol, arduino electronics, and wifi/bluetooth modules to be fairly straight forward, and to have a pretty easy learning curve.  As to what would or would not be allowed to come to market, that is a whole other thing, but the information on Legacy's command structure is freely available.   From what I've seen here from comments in other threads talking about it, the DCS gurus don't seem have nearly the same open minded philosophy to letting people know how to work with the system.  

 

Actually Stan said, and John adopted it.  I guess that is the reason for patents, and how an inventor can make a profit for his efforts.  Should John give out his super chuff code so other can make his board?  I do think he put out the diagram for the board.     G

Actually George, that was my line originally, well before I even joined OGR.  I just missed the boat patenting it I guess, you're already trying to attribute it to someone else!

JohnGaltLine posted:

Professor Chaos posted:

Be aware that the Legacy base uses more or less true RS-232 voltages rather than 0 - 5v.  A MAX232 or equivalent chip would be recommended for interfacing a microcontroller to the Legacy base.

If you want to read the TMCC serial data with a computer instead of a microcontroller, you can use a USB-RS232 converter cable and read the serial data with a terminal program.

For my purposes I do want to talk to a microcontroller, but it is good information to have.  I have not done a ton of research on legacy yet, but will be getting into it further soon enough.  Does the Legacy base still use 9600 baud, or has it's speed been raised as well?  I seem to recall that you need something between  the base and other devices for them to work correctly, but not sure.  

JGL

The TMCC base with the 0 to 5 volt output worked well for multiple TPCs, etc. When the Legacy base came out, the new RS232 which I think goes -8 to +8 v not could not drive more than one TPC and maybe one other device like the DZ2500 switch data driver. Incidentally, the true RS232 spec was a no change data state for the receiver for levels -3 to +3 volts.  So the original TMCC output was clearly not RS232. When the true RS232 in the Legacy base came out and was a unable to drive previous loads, DZ reduced the current draw for the photo-transistor input of their data driver. This helped but not enough for many users so Dale M on this forum built a booster amp and offered it in limited quantity to the forum members. When the SER2 came out from Lionel which is a booster amp capable of driving multiple loads, Dale referred the members to Lionel's fix.

I believe the SER2 is also not truly RS232 compliant either, it sports a uni polar output signal.

Cjack, as I said earlier in the thread, the original base used half-a## rs-232.  any computer made in the last 25 years will still read the 0/+5 signal correctly and that is what is used by almost all serial devices made in the last quarter of a century for home computers.  The problem in interfacing with home projects is that RS232 standard sends data backwards from any common sense, TTL serial data.  The high voltage of a true RS232 device is useful only when you are dealing with signal loss from long wires, but won't actually do any good if current drain is your problem.  That can only be solved by placing a transistor or other current amplifier on the output.  Having not opened up either base this is only a guess, but I think the problem with the signal on the legacy base comes from the fact that something like the MAX232 was probably used to provide true RS232 signals where as the TMCC base probably uses a TTL type chip allowing it to drive much more current.  I really respect Dale M, and wouldn't want to cut into his sales in any way, but I would be surprised if his booster has much more in it than a power supply and a pair of transistors.  

My comment on 'something between the base and devices' is not on the degrading signal, but that I seem to have heard or read somewhere that the serial on the LCS system will not talk to some other TMCC devices, and you have to have a wifi or ser2 in order for everything to work.  I'm unsure of the details on this at the moment.  

JGL

The Legacy base is 9600 baud.  You don't need any additional hardware to talk to it (other than something to deal with the negative voltage) though as mentioned above the current available to drive multiple receivers is lower than the TMCC base.

JohnGaltLine posted:

 I really respect Dale M, and wouldn't want to cut into his sales in any way, but I would be surprised if his booster has much more in it than a power supply and a pair of transistors. 

Actually, I'm pretty sure that Dale actually stated as such when he build these, and I believe there was even a diagram available of what was in it.  He didn't make any outlandish claims, just that he was solving a problem.

Indeed, John, I am slowly learning that the value of a device is not in the components inside, but rather in the solution it provides.

Actually, when I totaled up the time spent taking orders, answering questions, preparing the supporting documents, packing, shipping and billing the Boosters, all the time and parts I spent building and testing the devices was for far below minimum wage!  The end result is that I don't build them anymore, leaving folks with a need for a Booster forced to use another probably more complicated solution.

Well said Dale.  

By the way, I'd like to thank you again for the information you provided me back a little over a year ago when I was just starting to dig into all of this.  A year later it all seems quite simple to me, but when you first dip your toes in the water just knowing where to start is sometimes hard to figure out.  You are a fantastic resource for all of us here!

JGL

JohnGaltLine posted:

My comment on 'something between the base and devices' is not on the degrading signal, but that I seem to have heard or read somewhere that the serial on the LCS system will not talk to some other TMCC devices, and you have to have a wifi or ser2 in order for everything to work.  I'm unsure of the details on this at the moment.  

JGL

If you are using any LCS components, the command base's serial port is used for that connection and therefore, if you want to connect something like a serial cable/computer/terminal emulator program SIMULTANEOUSLY with the LCS device(s), an LCS SER2 is required.

If there are no LCS devices connected to your command base, your computer serial cable can go directly to the DB9 connector on the command base.

Railsounds posted:

JohnGaltLine posted:

My comment on 'something between the base and devices' is not on the degrading signal, but that I seem to have heard or read somewhere that the serial on the LCS system will not talk to some other TMCC devices, and you have to have a wifi or ser2 in order for everything to work.  I'm unsure of the details on this at the moment.  

JGL

If you are using any LCS components, the command base's serial port is used for that connection and therefore, if you want to connect something like a serial cable/computer/terminal emulator program SIMULTANEOUSLY with the LCS device(s), an LCS SER2 is required.

If there are no LCS devices connected to your command base, your computer serial cable can go directly to the DB9 connector on the command base.

 

Is this because of issues with two way communication between the LCS devices and legacy base?  As I think back more and more I'm remembering that the issue was that a SER2 was needed to send Legacy's 9 bit commands into the legacy base, and that it would not accept them directly in the base's serial port?  

Dale Manquen posted:

Indeed, John, I am slowly learning that the value of a device is not in the components inside, but rather in the solution it provides.

Actually, when I totaled up the time spent taking orders, answering questions, preparing the supporting documents, packing, shipping and billing the Boosters, all the time and parts I spent building and testing the devices was for far below minimum wage!  The end result is that I don't build them anymore, leaving folks with a need for a Booster forced to use another probably more complicated solution.

I've learned that lesson with my little product base, it's surprising how much effort it requires!  I do have a better appreciation of why some products are priced like they are.