Runaways on NCE – reasons and fixes.

by Mike Beckemeier on the 18Dec05. I have added some extra items..

 

The top reasons for the runaway conditions are:

1. Not disabling DC mode in the decoder.

2. Noise on the track bus & rails.

3. Noise on the Cab bus.

4. Cheapy decoders.

5. Multiple cabs controlling one loco or consist.

6. Using Radio encoder Procabs with cab addresses of 4, 7, 8 and 17 with the old 99 eeprom or using with the new 04 eeprom with radio fix OFF.

7. High DCC track voltage.

 

Fixes:

1. Make sure all decoders are set up to respond to only DCC commands and not DC.

 

2a. Keep wiring clean and simple, and preferably using twisted pairs for track bus feeds.

 

2b. Use a terminator on the ends of the power bus. A 0.1 uf capacitor & 130 ohm 3 watt resistor. (This has been posted many times on this list). It is amazing how much this "cleans up" the DCC signal. it keeps the voltage spikes down, and makes it easier for the decoder to interpret the signal it is seeing.

 

2c. Keep power buses less than 30 feet long. See what can happen with the DCC signal with longer power busses at: http://www.wiringfordcc.com/track_2.htm#c2  If you have long power busses and you don’t have any problems, leave everything as it is, no need to rewire. Not all layouts show the same results.

 

3a. Keep cab bus wires away from track bus.

 

3b. Use a different terminator on the ends of the cab bus if it is a very long bus. Not usually required. Mark Gurries has recommended 0.01uF 10V to 25V and 100 Ohms 1/4W

 

4. Cheap decoders that do not do a very good job of distinguishing between seeing a DCC signal vs a DC signal. (Hence the reason for fix #1).

 

5. Make sure at the end of running a train, that the operator selects "loco zero" to clear previous loco from his cab.

If this is not done, and another operator using a different cab selects the same loco that was in use, now has control of the train too! Not instead of the other cab, but in addition to the original cab! The warning that sometimes comes up saying "Loco in use.. Take??", will give you control of that train, but does NOT delete it's ability to be controlled by the original cab that was controlling it! If more than one controller has control of the loco, obviously bad things can happen. Two operators could be fighting over control, or appear as though there were two operators fighting over it, if there is noise on the cab bus sending contradictory commands.

 

6. Use other Proocab addresses. I have confirmed this problem with the purchase of a new radio Procab (Dec 05) and assigning it addresses 8 and 17 with the new 04 eeprom, with Radio Fix - OFF (selected OFF due I operate a potentiometer Procab and loose control of loco - Procab locks up - Bug in new eeprom). Selected the Radio Fix ON and this eliminated the runaways but gave me a "double time display". This second time display was instead of the "LOC:----" display in the top L/H corner. Using other cab addresses gives a normal display and no runaways.

 

7. DCC Operators have reported reducing their “problems” (loosing address, consists not running etc) and runaways by just reducing their track voltage to about 13 volts DCC.  

 

Each of these "fixes" contribute to the remedy. The question is which are the most productive. Based on experience, and the history of others on this list, I would implement fixes:  1, 5, 2b, 3a, and 4 as the primary items to do.

If it is still a problem after that, resort to implementing the other fixes.

 

Mike Beckemeier

http://www.NapmLtd.org  HO Club In Milwaukee WI

 

While most modelers don’t have any of these problems, a few are having some strange “goings on”. Mark Gurries, Allan Gartner and others have contributed in the past to these situations. So these are some of the things to look at if you are one of the unlucky ones having problems with these things. If wiring a new layout, some “better practices” have be learnt from other people’s experiences. These were not thought of at first and it need actual running trains and all different layout configurations for some of these problems to materialize.

 

For a new layout, recommendations would be:

  1. Power bus and cab bus separated as far as possible.
  2. If power busses are longer than 30 feet, twist at 3 twists per foot on as much of the bus length as possible.
  3. Fit terminators to both cab bus and power bus “ends”
  4. Fit booster at a central location to keep busses as short as possible.
  5. Never joint power bus ends to another part of the bus (parallel paths).
  6. Fan out with cab buses like a wheel and start with a splitter from the command station to go in both directions.

 

For existing layouts, first:

  1. Fit terminators to cab and power busses.
  2. Separate cab and power bus.
  3. Rewire a long power bus with twisted wire at the start of the bus. A long power bus is generally one that does not have any track feeders attached for the first 20 or so feet. Rewire this 20 or so feet with twisted wires and connect to existing power bus where all the track feeders join.
  4. Use some of the above if possible.

 

Always operate with DC (analog) OFF.

 

Hope this helps if you are one of the unlucky ones to experience runaways.

 

Marcus

 

 

 

--- In NCE-DCC@yahoogroups.com, "Andrew Keeney" <hunter48820@y...> wrote:

> 

> Hi Mike,

> Thanks for the reply.  Your suggestion is worth looking into.  I'll

> take a closer look at the possibility of doing that when I get home!!

>

> A follow-up question to my original concern.  Are the highspeed

> runaways more likely caused by problems with the track bus rather than

> the cab bus?  Or could it be caused by either one?

>

> I have very little, or no problem with throttle commands when plugged

> in.  Is that an indication that I should be looking at the power bus

> and leave the cab bus as it is?

>

> Thanks again.

>

> Merry Christmas and Happy Holidays!!

> 

> Best,

> Andy Keeney

> Dewitt, MI

>