Getting "Crickets" from the Bachmann forums so I'll try here...

Model Railroader Electronics and DCC
1

Hello All,

I have three (3) Bachmann HO GP30s: two (2) with factory-installed (OEM- -Original Equipment Manufacturer) DCC-ready Printed Circuit Boards (PCB) with NMRA 8-pin sockets, one (1) with the OEM DCC PCB decoder.

These three (3) units are MUed; originally with Universal consisting.

I am using the Bachmann Dynamis system with the ProBox and 5 Amp booster. I have a stand-alone programming track through my desktop computer running JMRI DecoderPro interfaced with a Digitrax PR3.

Originally I installed Bachmann 2-function (36-552) decoders in the two (2) DCC-ready units with NMRA 8-pin sockets to match the Bachmann OEM PCB decoder.

Because of the limitations of the Bachmann decoders I wasn’t able to use “Rule 17” dimming and Advanced Consisting.

I replaced the two (2) Bachmann decoders with Digitrax DH126P decoders in the units with 8-pin NMRA sockets.

Then I replaced the Bachmann OEM PCB DCC decoder with a “DCC Ready” PCB that I ordered from Bachmann, and added a Digitrax DH126P decoder to the last unit.

After programming all three decoders to their respective addresses- -3001, 3002, 3005, I also changed CVs 49 & 50 to activate “Rule 17” dimming.
Now all three (3) are “stuck” in “Rule 17” dimming.

Using JMRI DecoderPro I confirmed that all the CVs are set to “Normal” function of the lights; FoF & FoR; CV 49 & 50 > Value 0 (zero).

At first, I thought it was a fault with the Digitrax decoders.

I replaced all three (3) decoders with new-out-of-the-package ones- -still no change in the lighting functions!

After attempting to reset all three- -CV 8 > Value 008 through the Dynamis DCC system and DecoderPro- -, still nothing.

Unfortunately, Digitrax tech support was of no help. They suggested “resetting” the decoders; CV 8 > Value 008.

I finally replaced the Digitrax decoder with a 2-function Bachma

2

Update…

After the above post, I decided to “sleep on it”!

This morning, to confirm that it is the PCB circuitry and not the decoder, I unplugged the decoder and put in the DC 8-pin plug.

I got out my DC controller and attached it to my stand-alone test track after unhooking the DCC DecoderPro.

When I powered the unit up in DC THE EXACT SAME THING HAPPENED! - -“Rule 17” dimming- -in DC!!!

This is a real head-scratcher!

If “Rule 17” dimming is decoder dependent why is it happening in DC???

Thoughts?

3

Does “008” work for a reset for CV8 using the Dynamis? It’s just “8” with my NCE Power Cab.

You can also reset it using Decoder Pro. If you do just type in “8” rather than “008”. Sometimes leading zeros (or lack of) makes a difference whether something works or not.

Tom

4

Could it be the resistor specification in the PCB not suitable for the bulbs you have in your locomotives?

5

So you’re running a straight DC locomotive, and getting “Rule 17” dimming. I am guessing that this means bright in the direction of travel and dim at the other end.

Seems to me a way to do that is to have two LED’s in parallel in each position, but with the polarities switched. By adding an appropriate resistor in series with each LED, it seems you would get either bright or dim, depending on the polarity of the supplied voltage. Since you reverse a DC locomotive by changing polarity of the track, that could change the polarity of the supplied power.

That’s how “Rule 17” could happen on a DC locomotive. No decoder. No CV’s.

So it seems to me.

Ed

6

It would change the polarity of the supplied power but one LED would be out, not dimmed.

Maybe it’s how one defines Rule 17. Digitrax simply says the forward headlight is dimmed when the engine is stopped. I don’t think that’s what JJD is describing.

DCC Wiki says:

Rule 17 is actually a collection of rules that govern which lights are to be illuminated on a locomotive at given times and to when they should be be dimmed. These rules may vary from railroad to railroad but generally follow a similar pattern. The general purpose of the rule is to make sure that the locomotive is visible, without creating undue glare for others in the area.

  • Except when an engine is clear of the main and stopped, both the front (and rear if so equipped) headlights should be on.
  • The light in the direction of travel should be a full brightness except:
      1. At stations and yards where switching is being done.
      1. When the engine is stopped close behind another train.
      1. In non-signalled (dark) territory, when the engine is stopped on the main track waiting for an approaching train.
      1. When approaching and passing the head end and rear end of a train on the adjacent track.
      1. At other times to permit passing of hand signals or when the safety of employees requires.
  • The opposite light should be dimmed.

No idea how that happens in DC.

7

Exactly. That’s the plan.

Remember, there’s two LED’s in each light. We want only one at a time on. Either the bright one, or the dim one.

It could be done with bulbs, too. You’d need to add a diode for each bulb.

Yes. We do.

Ed

8

I prefer to follow Rule 17 manually, by having the lights be controlled by the engineer, not automatically by circuitry.

I’ve set up a couple of engines (in DCC) to have each headlight to be able to be off, dim, or bright. Separately. Like the prototype.

I like it.

I’m also thinking of going in the other direction with my switchers. Since they don’t go out on the road, I was thinking of simply having both headlights on dim, all the time.

Ed

9

You need to trace out the blue, white and yellow wires. They need to be connected as shown in the Digitrax decodermanual.pdf If they run thru LEDs or bulbs rated less than 12 volts there will be resistors. There needs to be one front bulb (LED) and one rear bulb (LED).

10

Hello All,

Thank you for all the great responses, suggestions, and questions.

I have reset the decoders with both JMRI through my desktop computer to a stand-alone programming track and through the throttle with the programming track.

Not only have I used CV 8 to value 8, but Digitrax also recommends “CV08 to 009 will also preserve the current sound project selected by CV60.” Even though I don’t use sound.

All the decoders reset to factory CVs which I checked through DecoderPro. I also confirmed the decoder resets by running them on my pike.

However, the lighting effects did not reset to “only in the direction of travel”.

That’s why I kept the OEM PCBs- -and installed a PCB I bought from Bachmann in the third unit which originally came with a PCB-type DCC OEM decoder.

All of them have LED bulbs installed from the factory along with the 8-pin NMRA plug.

To clarify…

All three (3) units are run in DCC.

Yes, Rule 17 Dimming is bright in the direction of travel and dim at the other end.

Rule 17 Dimming is a decoder function.

I pulled the decoder and reinstalled the 8-pin DC plug and attached the DC controller to my programming track to eliminate the variable of the decoder for troubleshooting purposes.

Even after reverting the unit to DC and

11

“Rule 17” style headlight usage in the real world was really intended for yard engines, which normally keep both headlights on for visibility and identification (a headlight on the end coupled to cars can be extinguished).

For road engines/consists, it doesn’t have meaning. You controlled the headlights “manually” for each end. Might have the back end light on when backing up to make the hitch on the train, but then you’d kill it and from that point only the leading light would be used.

I don’t use multiple-unit consists on my small layout, so I haven’t really experimented to see how the headlights can be set up “in consist” through different decoders. However, I’d expect it can get touchy if they’re not all the same. Might not be possible to get a “rule 17” setup at all.

I’d want my consist setup so that I could control the leading and trailing headlights independently of each other. Same as it works on the big engines…

12

I believe I explained how “Rule 17” lighting could work in DC. So the theory could be wrong.

Thinking on it a bit more, I can see other methods.

Ed

13

When I look at Rule 17 in a copy of The Consolidated Code of Operating Rules , it is plainly directed at trains on the main line.

It is Rule 18 that applies to yard engines.

I believe it does have meaning for road engines/consists. That’s why it was placed in the CCOR.

NCE, anyway, is set up so that the headlights on each end of a consist can be controlled exactly the same way as for a single engine. So if you’re happy working a single engine with their Rule 17 protocol, you should be happy working a consist.

I believe Digitrax is supposed to do the same thing, but I have yet to pull it off.

Ed

14

There is obviously a mix of diodes and resistors at play here on the factory board to create the effect you are getting. The decoder is just providing the voltage to the factory board which is doing the actual action.

The only way to completely correct it is to trace out the factory board and eliminate / bypass the offending diodes.

Mark.

15

The only OTHER way to completely correct it is to remove the factory board.

Ed

16

That’s what I normally do regardless. There are always un-needed components on a factory motherboard that are not required when using DCC and can sometimes even cause problems. Scrap everything and start from scratch so I know exactly what I’m dealing with.

Mark.

17

Hello All,

YES!!!

Thank you for understanding my situation!!!

Now the question becomes which ones…???

As I said earlier I’m getting nothing from the Bachmann forums.

I was hoping there was an easier solution.

Without…

That might be the only option in this situation, short of sending all the OEM PCBs to Bachmann for “Factory Service”.

Or…

I can just live with the lighting effects.

Again, thank you for all your responses, suggestions, and questions.

Hope this helps.