Cruise commander in Polar Express

1. Yes, you can keep the standard bulb for the headlight, and the rear light output puts out a nominal 12 volts, so any bulb that handles that will be fine.  Note that the lighting outputs are DC negative in respect to frame ground.

 

2. The only thing you get with the Cruise Commander is a 5V logic signal when you trigger the whistle from the remote.  There are ways to fool the conventional whistles into thinking they have track power, that will require additional circuitry.

 

2. Yes, the CC documentation recommends you not route the smoke unit common through the board, keep it grounded to the frame of the locomotive.

Thanks for the info John. Do you have any more info on adding the extra circuitry for the whistle steam? 

Thanks, Liam 

Well, my plan would probably be to use one of the 10ma 5V coil relays I have on the logic output of the Cruise Commander to power the whistle steam smoke generator.  That presumes it's a separate smoke unit like my T-1 Duplex and K-4 are.

The whistle steam is the same smoke unit. It's a a typical postwar puffer unit with a little hole in the metal. Then the fan sucks the smoke out when the whistle button is hit. Do I attach the RS boards to the logic output to for the whistle? Do you have a diagram with the relay?  

Thanks, Liam 

Here's a relay that handles 1/2A and has the 10ma 5V coil.  This is more than enough for the whistle fan.  Wire the coil direct to the logic output, and note that the relay coil is polarity sensitive.

 

http://www.digikey.com/scripts...&y=0&cur=USD

 

I don't have a diagram, as I haven't actually done this.  I can't see an issue, you simply wire the relay coil to the logic output, and control the fan using the relay contacts.

Hello all,

This is really interesting topic as I wondered if it would be possible to add cruise and keep the whistle steam function working.
I don't follow how this works though. Sorry
Can some help explain?
What is the logic output? And how will the smoke fan know to go when the whistle is sounding?
Thanks everyone!

Pin 3 of the 4 pin RailSounds connector on the Cruise Commander is normally unused.  However, ERR has put a 5V logic signal on this pin that is active when the whistle on the remote is activated.  Given this fact, if you run the whistle fan with this output, you'll get the whistle steam whenever the whistle is sounded.  The relay is just an electronic switch to turn the fan on when the whistle is sounded.

Folks the Cruise Commander does NOT put out a logic signal on pin 3 of the RS connector!  The AC/DC commander does.

 

The best solution for the air whistle is a circuit that can read the RailSounds Serial Data and trigger the Horn/Whistle.  There are two ways to trigger the sounds,

1) create an offset on the power going to the sound circuit already present in the tender

2) create a logic level to close a relay to operate the sounds -or- "tickle" the correct point on the circuit that is used for offset detection.

 

ElectricRR does not have such, but this is not hard to implement.

BUMMER!  I thought that was in the CC as well! 

 

I guess the question is, why did that get left out?

 

I think the "circuit to read the RailSounds Serial Data" is a mouthful for most folks here.  I happen to have one that's close to ready, but I'm still working on it.

I've been interested in doing this as well but have no electrical experience.  If you do get this to work, detailed instructions would be greatly appreciated!

What is the "this" you're interested in doing?

 

Adding TMCC capability to the Polar Express with Whistle Steam and maintaining all features (sounds, whistle steam, etc.).

+1     

Whistle steam is problematic as you have no way of triggering it with the Cruise Commander.  If you're willing to accept operation without cruise, it's much easier with the DC Commander, it has a logic output for when the whistle is activated.  So, you can have whistle smoke or cruise control, your choice.

 

Jon, What would you propose for the circuit to detect the whistle signal on the serial line?  G

Hello, I just found this forum a few weeks ago.  Lot's of good stuff here.

 

You're moving into the realm of microcontroller programming. A couple years back, I did this in my PE which I had upgraded with an ERR Cruise Commander.  I used a small Microchip PIC12F635 (programmed with PIC-Basic) to decode the serial data stream from the CC board.  That seems to basically be the protocol described in the TMCC manual with address headers stripped off. I decoded four signals - Horn, Bell, Aux1-2 and Aux1-7. I fed three of  those signals to some optocouplers and LM317 regulators (wired in pairs as AC regulators) to generate the appropriate DC offsets  for the TrainSounds boards in the tender and PE Announcement car.  This circuit also reduces the AC voltage on the trainsounds boards which get too hot if run continuously at high track voltages typical in TMCC operation. The fourth signal (Aux1-7) I used with another optocoupler to drive the Lantern in the PE Announcement car.  

 

Though it required some upfront familiarity with the microcontroller and serial communications, it was fairly straightforward and actually works quite well.  I have since used the technique in several other engines & operating cars.   

 

Cheers,

Jim

Care to share the code and your wiring diagram with the crew here?  It would be nice to be able to get some programmed parts to build some of these.

 

I have somewhere on my plate the intention to decode the stream with the Arduino, but it hasn't come to the top of the queue yet.

 

 

(11-15 Updated schematic to RevA)

 

Here is a PDF of my schematic for the design. Basically datasheet/application note  implementations of the PIC & regulators.  The idea of using the opto isolators with the LM317's to generate  the DC offset I found on another website some years back. It's basically a minor mod to National's early 1970's app note on using the LM317 as an AC regulator.   One caveat - I find this circuit works  better with old traditional transformers like the ZW than modern triac based one's like the CW-80 (it does function, just not as well).

 

I'm leary of posting the code for the microcontroller.  Unlike the some other model railroading & 3D printer communities (which have embraced open source design as a way to benefit the future of the hobby), Lionel & MTH seem old school in that they are patent and lawsuit happy and have not embraced open source design (much to the detriment of the hobby's future, IMHO). It is fairly easy to see what is going on with a scope. Also, the various patents for the hardware (easily found on the US Patent Office and many other websites) describe the structure, timing and decoding of the serial data quite well. It's unfortunate they chose not to document it in the manuals.

 

Though I originally designed this to work with the TMCC CAB-1 system, it does seem to work fine with the newer Legacy/CAB-2 system in TMCC mode.  I haven't made any attempts to make any of this work in Legacy mode yet.

 

I've been looking into doing things with the Arduino & RaspberryPi  boards as well.  Arduino & PIC are pretty comparable and good for most things.  RasberryPi looks really powerful and would be interesting for custom sound or visual effects, but it's just big enough to be inconvenient for O-gauge.

 

Too much fun stuff to play with, too little time to do it. :-)

The TMCC command codes are an openly published standard, not like Legacy.  I can't imagine there's any problem posting code to use it. All sorts of companies have designed stuff to interface with TMCC, and they're actually charging money for their products.

 

The usefulness of the schematic is drastically reduced without the code...

 

Legacy mode is a whole different kettle of fish (and commands), but there is a glimmer of hope that at some point they may be available as well.

 

Good points.  It might take me a day or so to dig up the file - I was using an old parallel port PIC programmer when I did this so I think that code is sitting on my old Windows 98 computer. It needs to get copied to the new computer anyway.

 

It would sure be nice if they publish the data for Legacy - that would be a whole lot easier than reverse engineering it.

You should find most of the TMCC codes in Lionel's publication 'The complete guide to command control'.  This is available directly from their website too.  

That's the source I was referring to Nicole.  I'm pretty sure there's no problem using the TMCC interface, however they have yet to release Legacy interface info.  I'm getting the feeling that Legacy will be a much more "controlled" release when it happens, probably to folks that sign some kind of license.

 

That's the document I've been using for the codes - it does a good job describing the data payload (we need the same thing for Legacy).  There was additional information in the patents describing the the more technical details of the data as it's received by the R2LC in the train and, by extension,  what goes out of the R2LC to the other boards. If I remember right, the R2LC seems to strip some of the addressing fields off and just forwards commands. It also described algorithmically how to actually detect and recover the data in the microcontroller.  That's the stuff I was more concerned about.  Now that I went back and looked, it's pretty basic microcontroller stuff and not particularly unique to trains.

 

Cheers,

Jim

I've programmed tons of serial stuff for other applications, nothing unique about this data except the odd baud rate.  Did anyone else find that 3000 baud was a bit odd?

 

Found it.  Here are the files for the code I used. It is written in PicBasic Pro from Microengineering Labs.   The .hex file is the programming file complied for a PIC12F635 

 

Yes, the baud rate was definitely odd.  The timing of the button presses  could be better in this.  I've been meaning to tweak on the order & delays in the routine at the end. Also, it occasionally misses button presses - I think some minor changes could be made to improve noise immunity. 

 

Disclaimer - I'm a hardware engineer, not a programmer.  There may well be better ways of implementing this. :-)

 

Cheers,

Jim

jj1138, Very elegant hardware design.

 

The Baud rate is 3030, not 3000.  The rate is related and timed to the incoming Track-Link Signal timing. (330usec / bit)

Thanks Jon, I was estimating from what I timed on the 'scope.  3000 would be plenty close to decode the bytes, but knowing the exact timing is obviously a good thing.

 

Your statement about the incoming signal makes perfect sense, I wondered why the odd rate.

 

If either you, gunrunnerjohn, jj1138, etc. would be interested in adding tmcc to my PE with whistle steam effect (for a fee), contact me via email as I'd like to add it while also being able to utilize my PE conductor announcement car, trainsounds/railsounds in tender, whistle steam effect, bell, etc.

I've been wanting to have this done since purchasing the engine.

(BTW, might adding either PS 1.0, 2.0, or 3.0 be a better way to go rather than attempting to add tmcc? I'm set up to run both tmcc and dcs.)

Thanks for your input and any assistance on this.

Kenn ogaugeguy  

I don't know that you'd be able to get steam whistle effects from PS2/3 any better than from TMCC.  The real issue is you need a digital signal when the whistle blows.  It seem to me the only hope for doing this is by decoding the TMCC data stream, as that's available and documented. 

 

I don't know of any digital signal in a PS2/3 locomotive that is available when the whistle blows, and there's no way I know of to tap the serial data to generate one.

 

 

Thanks Jon, that's good to hear from someone at Lionel. As an design engineer, I have been impressed with the simple, elegant, modular design of TMCC - in my experience it really works quite well. I have tried to follow suit with the electronics I've built and added to my layout.  I wish I could take credit for the clever AC regulator circuit but I actually found that years ago on a website where someone described how they had put a DCC board into a Lionel engine with TrainSounds.

 

I'm curious if you saw my earlier comment where I was somewhat critical of Lionel for not being more open with the technical details of TMCC & Legacy? It would seem to me that encouraging individuals and small companies to build into your command environment would result in more users buying into that environment, yet the O- gauge manufactures seem to be taking the opposite approach. Of course there are plusses and minuses to anything, so I would be interested in hearing someone from Lionel's thoughts on that.

 

Cheers,

Jim

 

 

Unfortunately my day job and some volunteer work I'm doing with a First Lego League Robotics competition group doesn't leave me much time for that now.  That's the kind of thin I'm looking forward to doing after retirement someday.  

 

You would be best to stay in the TMCC world for this. While DCS may have some more capabilities, this engineer's opinion is that it is way over complicated for what it does. I haven't near as good of experiences with MTH's electronics as I have with Lionel's.  

Jim, We plan to open up the protocols on our products on several items coming to market shortly.  Our WiFi interface to Legacy has been in the field for a year in beta form; and we feel that the WiFi hardware (soon to be generally available), and the protocol we plan to open for folks to use, will promote a unified way for the hobby to work towards a common control system for the good of the Hobby.

 

That's really great news Jon. I look forward to seeing it.

 

Cheersm

Jim

Liam,  let me give some thought on how to build this circuit for you based on Jim's hardware.  Email me off forum. 

 

Jon, I'm interested in this circuit as well, I'd like to be able to trigger a horn/bell if the board was small enough.

I built it on a piece of perf board using thru hole parts and it came out about 1x1.5".  I thought about a PCB but wasn't building enough to pursue it.  With SMT parts this thing could be on a really tiny PCB. The TO220 package regulators generate some heat but not enough that I needed to heatsink them.  With smaller packages, that may come into play. 

 

Also, the opto isolator driving the lantern is probably not necessary.  A transistor should work. 

 

I also just noticed that optoisolator (U103) is flipped on my schematic. It won't work as shown.  I'll fix it and repost.

I updated the schematic in my earlier post to RevA to fix the error with U103.