Skip to main content

Got an interesting repair in.  An A-B-A that has two powered A-units, each with a K-Line cruise board.  Of course, you have figured out the story, the trailing A has fried it's cruise board!  OOPS, there are no replacements for those.

The logical replacement is an ERR Cruise Commander M, but the speed curves will be totally different, so that's probably going to be a significant issue.  The K-Line cruise and ERR cruise do not play well together.  So, I figure to replace both the working and non-working K-Line cruise boards.  Each unit will get a Cruise Commander M board.

Now comes the fly in the ointment...

The K-Line cruise boards just use serial data, but the CC-M wants serial data, the two PWM direction outputs, and the +5V DC from the R2LC!  That's three extra pins! That could get ugly!  I examined the K-Line cruise 1--pin tether, and to my amazement there are four ground wires out of the ten pins!  They have separate grounds for the coupler, lights, smoke, and the standard frame ground!  I just found my three wires!

Next up is the issue of driving two motor control boards from the R2LC PWM outputs, it doesn't have the drive to get it done.  Each motor control board has a set of opto-couplers to receive the PWM data, paralleling them causes all sorts of issues.  The serial data is also marginal, especially after it travels through a few tether connections to the trailing unit.

I've actually run across this before, but this time I decided that I should consider a way to do multiple powered units without having totally separate electronics.  I suspect this isn't the last time I'll see this same configuration.

Time to fire up the Circuit Board Design tools.

This is a board that will allow me to boost the serial data for the long trip to the rear, and also provides two buffered outputs for the PWM drive from the R2LC.  The buffer has sufficient drive to power three or four opto-couplers, two won't be a challenge.  This board will feed both the lead A unit CC-M and the trailing A unit CC-M with buffered PSM inputs.  The independent 5V power supply will also prevent us from overloading the R2LC 5V power output since it has very limited capability.

This board (when it comes back from OSHPark) will provide the buffering to tie these units together for proper operation.

TMCC Motor Driver Buffer 1.1 Schematic

TMCC Motor Driver Buffer 1.1 3D N1

TMCC Motor Driver Buffer 1.1 3D N2

Attachments

Images (3)
  • TMCC Motor Driver Buffer 1.1 3D N1
  • TMCC Motor Driver Buffer 1.1 3D N2
  • TMCC Motor Driver Buffer 1.1 Schematic
Original Post

Replies sorted oldest to newest

I've found that I can't do that, I have the example sitting on my workbench!  The direction control was "iffy", and about half the time the trailing A unit would suddenly go full throttle.  When I put a 'scope on the serial data, it's amplitude was less than 3V, that's getting down where the CC-M starts missing commands.

Since I'm delivering this to a customer, I like to make sure it's 100% bulletproof before I ship it.

Its a good idea. I was surprised mine worked. My backup plan would have been to just add a second radio board but using a single one makes for seamless running every bit as smooth if not smoother than two Legacy engines lashed up.

Likely not a drop in for every engine though. Mine was not cruise and came with a 6 pin tether so that had to be replaced with a ten pin.

Pete

I'm relatively illiterate when it comes to electrical circuits, but what you've done here seems pretty amazing. Because of my illiteracy, if I had this problem I would consider just taking the motors out of the trailing A unit and converting it to a dummy. For my use, at least, a single dual can-motored engine would suffice for my  pulling needs. I think so, anyway, given that I have single dual-motored engines pulling loads like six aluminum passenger cars and decently long freight consists without problems. On the other hand, if I had this situation and really liked the A-A set,  I'd probably send to GRJ for a complete makeover.

Out of curiousity, I have a K-Line A-B-A set with the two A units powered, but the engines don't have cruise control. Just to confirm, it is just the presence of the K-Line cruise that creates the problems, and adding ERR CCMs to a pair of K-Line A-A engines without cruise would be a straight-forward installation of the CCMs, wouldn't it?

In the case of this engine, when I put two CC-M modules in parallel, the trailing unit was not happy.  After looking at the pictures on the 'scope, I could see the signals were marginal.  I have no idea if it's just this particular set or a general issue, but since I've experienced a host of issues with the serial data stream that buffering it solved, I'm just added the boost for the PWM signals.

FWIW, I also use the PWM boost on my Smart Motherboard project, I found that the single DCDR (or CC-M) loaded down the signals enough that they couldn't be recognized by the on-board logic.

The issue with the serial data stream and CC-M problems are the stuff of legends, that's been talked about for years.  Even with the most recent CC-M, more than one person has encountered serial data issues.

Whenever you add things in a non-standard configuration, you have to think about the possible issues you may be introducing.

Breez, in your case one option would be to install a Cruise Commander in each engine. Your engine has two DCDRs not Cruise motor drivers driven by a single Radio Board. The DCDRs don't require serial data only direction (PWM). It becomes problematic when you have to use serial data to drive multiple loads which is the case if you put in Cruise Ms.

Johns circuit allows use of less expensive Cruise Ms rather than the more expensive Cruise Commanders.

Pete

It would work, however the beauty of a single "command" unit is no MU configuration and no getting out of sync with each other.  One of the constant issues with TMCC/Legacy MU configurations is a power interruption makes them forget they're in an MU and the revert their direction to "forward".

Yes, you can wire them to run backwards for "forward", but that makes them an oddball.

That being said, it would be possible to just remove the tether and do that.  However, I always wanted to do one of these this way, and now I'll have my little board to make it happen.

Add Reply

Post
OGR Publishing, Inc., 1310 Eastside Centre Ct, Suite 6, Mountain Home, AR 72653
330-757-3020

www.ogaugerr.com
×
×
×
×
Link copied to your clipboard.
×
×