MM 201 control rear marker lights remotely under DCC

[murphaph]

Problem:

Rear red marker lights on the 201 are permanently illuminated with the front marker lights unless the manual switch under the loco is operated. This kind of defeats the purpose of DCC etc. It would be nice to be able to remotely disable the red marker lights, while leaving the front white ones on. It looks unprototypical in some circumstances to have the marker lights on.

(my) Solution:

desolder the wires leading to the current switch and slot in a relay instead of the switch and trigger the relay using one of the remaining free aux outputs on your decoder (in my case I used aux 5), if you have any. Hint: a Lokpilot v4 that many units will have fitted does not have any spare aux outputs to do this. I have fitted a Lokpilot/Loksound v5 for this, but any standards compliant decoder with an aux 5 output should do.

(kind of ) tricky bit:

The NMRA spec calls for the aux 5 and above outputs on the 21MTC connector fitted decoders to output at TTL (transistor-transistor logic level, a kind of low power used for microelectronics and not strong enough to switch anything) so we need to take this TTL voltage and amplify it using a small signal transistor such as a BC107 or (as I used) a BC549. The amplified signal is then used to drive the coil of a sub miniature relay (I chose the omron G5V 12V, but any 12V relay that fits should be ok) where the normally closed output of the relay is connected to the rear marker lights + terminal on the 201 PCB, while the common of the relay is connected to the decoder + (again, K1 on the 201 PCB). The Aux 5 terminal on the 201's PCB does not go anywhere so you have to solder your wire directly to the pad of the 21MTC connector (please don't start soldering stuff to the decoder even if it appears easier, we want to modify the loco itself, so decoders can still be changed in and out without de-soldering stuff)

Other bits:

You need to fit a resistor between the aux 5 output on the 201 PCB and the base of the transistor (marked b in circuit diagram attached). The transistor collector connects to the other side of the relay coil  and the emitter of the transistor connects to the decoder - (minus). I used the RHS of the resistor marked R5 on the 201 PCB rather than solder to the MTC21 connector, though you can do that too of course, just be sure to use decoder minus and not track minus or something. 

After desoldering the wires to the switch I black tacked the relay on it's head beside the switch. I then superglued the BC549 also upside down to the side of the relay. This is then easy to solder everything to. 

A so called flyback or clamp diode should be fitted across the coils of the relay even though ESU never mentions this in their own documentation about fitting a relay. This is purely there to suppress any induced voltage spikes during switching but should not be omitted as some decoders will be damaged over time if no suppression is included here. Note the flyback diod has the negative end connected to the decoder +. This is not an error, it's how it works. If this is reversed it will provide no protection.

This all sounds more complicated than it is probably. The circuit diagram should clear it up. I've included a couple of pics too. The colours of the wires in the diagram don't always match the actual colours I used I'm afraid. The white wire I used is coming from aux 5 on the 21MTC connector via a resistor that is difficult to see because it is enclosed in heat shrink where it's soldered to the base of the transistor. Don't forget to use heat shrink or otherwise adequately insulate your wiring, especially at the transistor if you choose to do it my way. YOu could use a small piece of veroboard but space is limited in the loco so Iike my way with almost everything stuck directly to the upside down relay.

You then need to map some free function key to aux 5.

Briefly the circuit works by using the transistor as a switch. When the base-emitter junction is saturated, the resistance between the collector and emitter falls to almost zero, allowing current to flow from the decoder + through the relay coil to decoder -, this switching the relay. When the relay is not energised, the red marker lights + has a path to the decoder + so its circuit is complete and the red lights will be on if the marker lights are on. When the relay is de-energised the path is interrupted (you can see nothing is connected to relay pin 10) and so the red lights cannot illuminate, just like with the manual switch, except it's controlled remotely now.

 

I've left the old switch in there as I can use that to isolate a stay alive capacitor (during programming) later. 

Any questions, feel free to ask. Any suggested improvements are also welcome. I know a FET would probably be a better job instead of the NPN transistor but I had them in my bits box.

No warranty, modify your loco at your own risk, incorrect wiring can fry your stuff etc.

 

[Noel]

Fascinating. The latest Murphy Model 121 loco can do this (ie separate the operation of the forward and rear marker lights) using DCC (ie LokSound V5). Sounds like you are having fun probing and exploring the world of DCC and microelectronics on locos. A pal of mine joined MERG https://www.merg.org.uk and built all manner of DIY DCC systems, controllers, cabs, accessory decoders, servo operated signals with bounce, etc. A whole new world of exploration. Have fun. Enjoy these posts. 

PS: Fair do's figuring out how to dismantle a 201

[murphaph]

Thx Noel, Yeah I have built a couple of kits from openDCC (kind of the German MERG), DCC central station, booster, an occupancy detector based on current and some homemade feeback boards and sensors (s88 bus) over the years. I've recently put together a test oval so I can get familiar with RocRail as the control software using current sensing to detect when the train enters the block and photosensors between the rails to detect when the loco is at the stop point. Even on my little test oval with passing loop it's fascinating to watch the computer randomly route a pair of locos around. I can say I've learned a bit, though none of the hardware will not be used in my "big layout". I have already settled on hardware implementing the BiDiB protocol, which has good support here in Germany. Kits are available which pushes the cost down when you need a lot of inputs.

[Strimmers]

Is this working off a 6 function decoder? 

 

[murphaph]

It should do yes. You need a decoder with an aux 5 output for the above circuit to work. Aux 5 is the first unused aux on the 201 so I used that. The 6 function decoders have aux 1 to aux 6, with aux 3,4,5,6 being TTL as long as they are nmra spec compliant.

[Strimmers]

Impressive stuff maybe I'll understand this by the time I'm finished college