2-8-0 Only runs forward

[RobBob]

Great running Spectrum 2-8-0 I've had a few years suddenly only runs forward and won't go in reverse.  I'm running DC using a tech-II power pack, momentum is off.  Put the 'dumb plugs' in and it runs ok, put the decoder back and it won't reverse.  Weird.   Anyone seen this before?  Also noticed it seems to run smoother on DC with the decoder plugged in.

One more thing.. the decoder plug pin marked '1' is oriented such that in can only fit in the pc board pin hole marked 8, or 4 if turned around.  If I plug it so pin 1 matches hole 4 it wont move.  Pin 1 in hole 8 only runs forward, no reverse.

Happy Railroading!

[Doneldon]

RB-

I don't recall seeing this or hearing of it but it's certainly true that non-DCC operations seem to be better with the decoder out and the shunts in. You are the first person I've heard say that their loco runs better on DC with the decoder in than with it out. I'm not sure I understand your question, though. Are you saying that you get better operations with the decoder in but then your Connie won't run in reverse? That sounds real goofy. Does it mean you must choose between smooth opertion and no reverse?
                                                                                                                                                                    -- D

[RobBob]

  Yeppers, that's exactly what I'm saying.  I'll try it again tomorrow and try to re-create the results.
I wonder if I blew somtheing up  by accidently putting the DCC plug in backwards.
Thank you.

[Jim Banner]

A typical decoder uses an H-bridge output.  Looks something vaguely like this:

        +     +
   F₁  |     |  R₁
        - M -
   R₂  |     |  F₂
        -     -

In the diagram, M is the motor and vertical lines | are transistor switches, When switch F₁ and F₂ are on, then the locomotive moves forward.  The left side of the motor is connected via F₁ to the positive side of the decoder's power and the right side is connected via F₂ to the negative side of the decoder's power.

When R₁ and R₂ are on, then the locomotive moves backward.  The right side of the motor is connected to + via R₁ and the left side is connected to - via R₂.  This is exactly the reverse connection and the reverse motor polarity compared to forward.

If either F₁ or F₂ fails, then the locomotive cannot move forward.  But if R₁ or R₂ fails, then it cannot move backwards. 

Big fleas have little fleas,
Upon their backs to bite 'em,
And little fleas have lesser fleas,
and so, ad infinitum

The transistor switches controlling the motor are in turn controlled by smaller transistors at the outputs of the decoder's microprocessor.  And these smaller transistors often are controlled by still smaller transistors, although unlike the fleas, this does not continue to infinity.  If any of these little transistors fail, the transistor it controls cannot turn on and so the locomotive cannot run in the corresponding direction.

So what would cause a transistor to fail?  Their biggest enemy is heat.  Some decoders protect themselves from overheating; some do not.  This begs the question - what decoder is/was in the 2-8-0?  As far as I know, the problem with the plugs only occurred in the first generation of 2-8-0's and they were not sold by Bachmann with decoders on board.  So if you have a first generation one, I suppose it might be possible that the decoder is a little runty in the amps department for the locomotive.  Often an under sized decoder works okay to start with, only to fail a few years down the road when the locomotive loses its lubrication and/or picks up hair/lint/dirt which causes binding in the mechanism.  To keep up speed, the motor draws more current.  More current means more heat.  More heat can pop a transistor.  Which transistors feel the heat first?  Those four in the H-bridge.

Excess voltage can do in transistors as well.  This is not likely to occur unless you have a surge on your power line while running trains.

Yet another possibility is a faulty motor.  In addition to extra mechanical drag, motors occasionally draw more current  as they age, caused by a build up of carbon between the sectors of the commutator.  This is the only non-catastrophic cause of motors drawing more current when old than they drew on their stall current test when new.  The increase is only marginal, but if the decoder is marginal as well ...

And sometimes transistors just fail.

For more information on the plug problem in the first generation 2-8-0's, click on this link:
http://members.shaw.ca/sask.rail/dcc/2-8-0/index.html

Jim

[RobBob]

Thank you...you nailed it! That's exactly right.  Thanks for that great explanation.  In the interest of full disclosure, I'm running "cab control" with block selection by DPDT switches.  My other power pack is a CMI 20 M.  The CMI looks exactly like an MRC Tech II but it has an additional knob to adjust the track voltage.  Back in the day when I was running N scale I developed the bad habit of crancking up the voltage if a train stalled to kick it into running.  Obviously this will not work with the DCC decoder installed.  The part number for the decoder is Bachmann  44915 on the shrink wrap, I do not see this in there current on-line parts inventory.
You 'da Man!

Happy railroading!