Adding a second decoder to a loco to control lights

[Gwdriver16]

I'm thinking of adding a second decoder to a tts fitted loco, to have independant control of the lights. The wiring is straight forward enough, but what about programming the second decoder?

Would I have to unplug the tts, then program the second decoder, then plug the tts back in.

I would give them the same address for ease of operation,

Cheers, Rich

[Howbi]

How can you have independant control over the lights when you can only use one DCC controller........when you press F0 for front & rear lights they will work via the TTS decoder anyway........can't see the point in disconnecting the lights from the TTS and then fitting another decoder just for lights.......or have I totally misunderstood your post?............HB

[Chrissaf]

As HB says, the TTS decoder has inbuilt lighting functions. It has Directional front / rear lighting functions controlled by F0 and an Aux lighting function controlled by F25.

.

The directional forward lights are on the TTS white wire (switched negative) and the directional rear lights on the yellow wire (switched negative). The F25 Aux wire (switched negative) is I believe green, but may be purple. The positive return will be the blue wire. Remember with LEDs, each LED will need a current limiting resistor included.

.

Because the lighting is controlled by F0 and F25. They are already independently controlled from the motor function even though they are just other pins on the single TTS decoder.

.

In the scenario that you have outlined and in the absence of any further detail that explains why, I cannot see any advantage of fitting a secondary decoder.

.

 

[Gwdriver16]

Yeah, I would rewire the lights so that the reds each end are switched on/off independently of the white lights. This needs 2 aux functions, tts only has 1. I've done a 47 where I've disconnected the cab lights from aux 1 and connected the no2 end reds to this, so when pulling a train, the tail lights can be switched off, but the no1 end will still light up when travelling in opposite direction. This is ok for running light engine, but won't be correct if a train is connected to no1 end. If I had a 4 function decoder, I can wire reds and cab lights both ends independently, and whites are switched on using F0 dependent on direction selected. 6 function would allow headlights to be independent of markers for running on depot as opposed to mainline.

Cheers, Rich.

[Gwdriver16]

The necessary resistors are already on the light pcbs in the nose of the loco, you just disconnect from main pcb and wire to the aux tabs. On Bachmann 37 & 47 that I have, you have 1 wire for reds and 1 wire for whites (headlights and markers combined) going from light pcb to main pcb at each end (orange and red wires), these are the negative feeds. Then you have 1 blue and 1 green, these are common positive. Tested the circuits using battery and leads, it was the orange wire on mine that fed the reds. Disconnected this from the main pcb at no2 end and attached to aux1 tab on main pcb, disconnected the cab lights (Don't tend to use these). The existing common connections are left in place. Now F0 turns on whites and no1 end reds depending on direction, no2 end reds switch on/off using F25.

Cheers, Rich.

[Fishmanoz]

Same address for ease of operation - you do realise if they are on the same address and you issue a function command that both decoders will respond to the extent they can?  The only thing easy will be programming the address as you can have both connected for that. 

 

Has as anyone checked to see if there is a 4th function on the TTS chip but with no wire attached to it?  If there is and you can solder a wire to it, you’d have to figure out what function it is by trial and error.  I ask as people have said the motor/function drive is an 8249 variant and so probably 4-function. 

[Gwdriver16]

I assume that you use F0 and aux1(F25) on tts and aux2,3,4 etc on the second decoder. If all  the functions are used up with the tts sounds, then giving the second decoder a different address would be the easy option,

Cheers, Rich.

[ellocoloco]

Hi GWdriver,

 

I believe I fully follow the thread above......Interestingly, Hornby Int'l locomotives have recently been upgraded to have full independent control of the lights through DCC which took me some getting used to. These have all utilised 21pin sockets. I don't know if Hornby have upgraded their OO models in a similar way yet as this website has very little technical information quickly to hand. 

 

I reckon the 4 function decoder is fine for this however they must be set to different addresses or similar commands will activate the same functions on each decoder. 

 

Let us know how you get on.

[Rog RJ]

Maybe consider unplugging the TTS decoder from the socket and replacing it with the alternative decoder.  Then make connections from the pickups, or socket, to the red and black wires of the sound decoder.  The sound decoder should produce sounds as normal and the motor drive and functions are now controlled by the non sound decoder.

 

Edit.

Just thought,. This may not work very well as the TTS decoder uses back emf to adjust the sounds?

[96RAF]

If you need the added functionality of a second decoder for complex lighting then consider this...

 

Disconnect the 4 inner (function) wires from the TTS socket, then wire in another socket for the second decoder in parallel with the TTS socket connecting only the track-pickup and motor pins to both sockets. Wire the second socket inner (function) pins with the 4 inner wires removed above.

 

Operationally there will now be problems as the TTS functions will also operate the second decoder functions in that F1, 2 and 3 will play sounds on the TTS decoder but also operate the associated lighting from the second decoder. Not what is required, therefore a workaround as described by others may be the best option.

 

Or... consider adding a relay operated by the TTS aux to flip the lights from one state to another, but only if you can stay within the TTS function output current limit.

 

To answer Fishy’s question. I can definitely confirm there are only three physical outputs (directional lights fwd and aft and Aux1) on the TTS decoder as opposed to four outputs (directional lights fwd and aft and Aux1 and Aux2) on the R8249. There is no ‘spare pin’ that could be used as there is no feed output from the actual PIC chip.

 

Rob

[Gwdriver16]

Hi RAF96, I was thinking along the lines of fitting a relay to replace the tail light on/off switch fitted in the fuel tank, that would enable me to use Aux1 to switch the feed to both sets of tail lights to on or off. That would be ok, but if I wanted to keep the cab lights, then more functions are required. On my 37, it only has headcode markers and tail lights, so thats what I'll probably do and ditch the cab lights, on my 47 which has headlights aswell, I might go with a Zimo (21 pin 8 function!). I was just curious if anyone had tried 2 decoders, you can get function only decoders, but they are more expensive than a basic 4 function decoder, So you might as well use a cheap 4 func decoder and don't connect the motor feeds. The cost of a tts decoder and a basic 4 func decoder, would still be a cheaper way of getting sound. Using the second decoder, you would have to assign F26/27/28, as the rest have sounds assigned, but you would also have F0 and F25 on the tts, so you have 5 available. If you gave the second decoder a seperate address, you would get 7. F0 on both, aux1 on tts and aux1-4 on your second decoder. 

Thats assuming you can fit all this in,

[Chrissaf]

This previous post details some transistor circuits that were developed to allow the F25 Aux function to be used to switch some complex lighting features:

.

https://www.hornby.com/uk-en/forum/post/view/topic_id/19094/?p=1

.

You might be able to use some elements of these circuit schematics to obtain your requirement.

[96RAF]

If using a second decoder as functions only on a second address then you would likely need a 100 ohm resistor across the socket motor pins to allow programming (decoder and controller dependant) but then you would also have to figure a way of disabling motor output to prevent accidental throttle application to that address burning out the resistor. If made a suitable wattage to handle such motor current then you risk melting nearby plastic stuff.

 

I know you can disable output in the decoder firmware but wouldn’t know how to do it via CVs unless you could set all speed associated CVs to zero (Vstart, mid and max as well as accel/decell).

Rob

[Gwdriver16]

Thanks Chrissaf, very interesting. The first lot of diagrams show a tts decoder and seperate 4 func decoder as I'm suggesting, but more complex lighting options requiring resistors to double up on number of functions. LMSfan72 said it worked, but the circuit diagram shows that his 4 func decoder isn't connected to a motor. RAF96, you suggest that this would need a resistor across the motor outputs to be able to program it. What about the tts HST double decoder set? How is programming the decoder in the dummy power car going to work? That won't be connected to a motor. I'm only just getting into the electronics side of things, so apologies if I'm being a bit thick. This is basically what I would have, but both decoders in same vehicle (Only 1 of them would be a sound decoder tho).

Thanks again, Rich.

[Gwdriver16]

/media/tinymce_upload/a121287c5730efe68afa2e3b4cc14611.png

This is LMSfan72's circuit diagram. What I'm suggesting is near enough the same, but omiting the part to switch between day/night headlight (Most of my locos only have 1 headlight each end).

Thankyou LMSfan72 and Chrissaf for this. 

[96RAF]

GWD16

 

During programming the controller expects an acknowledgement from the decoder to confirm it has applied the command. This is normally by way of a short pulse,into the motor circuit which manifest as a small current surge. Some decoders will program without a motor load. Some decoders that will so program may not if the controller is not also capable.

 

As an example v1.4 of the Elite firmware intorduced support for dummy car programming and a version of the R8249 decoder was supplied with a 7-pin plug unique to the Pedolino which uses two decoders, one in the dummy car for lights, although the blurb recommends the dummy car and motor car are programmed together to the same address. Hence if trying to program a solo dummy car R8249 with a v1.3 or earlier Elite then it won’t work.

 

If a decoder will not program solo without a motor then a simple 100 ohm resistor across the ‘motor’ pins of the socket will allow the decoder to pulse into this false motor circuit to provide an acknowledgement to the controller that programming is going OK and to proceed to the next bit. This resistor only needs to be temporary for the progrmming phase and could/should be removed upon completion.

 

In general try it as it comes and if it won’t take the address then opt for either simultaneous car programming or bung a temp resistor across the dummy socket.

 

If you want to get deeper into the electronics of installing prototype lighting mods then this is a good site...

http://s374444733.websitehome.co.uk/railindex/trainsindex.htm

which initiates from here...

http://www.bromsgrovemodels.co.uk/decoderinstallation4mm.htm

 

Rob

[Gwdriver16]

Thanks RAF96, very well explained. I've seen that website, it's superb, that's where I got info on the circuits in the Bachmann locos, which has helped alot. So what you're saying is, if the decoder doesn't respond when trying to program it, put a 100 ohm resistor across the motor outputs on the decoder, program it, remove the resistor and it's ready to go, is that right? That'll be fine. Once it's address has been programmed, the rest of the CVs should be ok at default settings, because it's only the function outputs that will be used.

Thanks, Rich

[96RAF]

Correct GWD16. Fit the resistor for the initial programming, then remove it.

 

You can also try different methods of programming if supported other than Direct mode, e.g Paged, Reg or Ops. Reg can be a good alternative programming method if Direct proves unreliable. Ops mode (programming on the main) is useful for changing some CVs in real time but there is no readback in this mode. Handy for volume adjustment as you hear the change immediately,

 

Rob

[LMSFan72]

Hi GWD16, I have now fitted around 5 locos with auxiliary lighting decoders. To avoid the problem of not having a motor connected I tend to use a function only decoder which doesn't have motor connections. I typically have fitted additional sockets so I can disconnect the appropriate decoder for programming. I tend to run the second decoder as a second address. I also have started to have the decoders lighting opposite ends, i.e. 1 decoder runs end 1 and the other end 2. This way I don't always have to use the second decoder if I am pulling/pushing a train. If I want to light both ends you can either consist the decoders or run them on 2 throttles. Consisting works nicely for me.

[Gwdriver16]

Hi LMSFan72, thanks for the response, it's good to hear from the horses mouth so to speak, as you've done this with success it seems. I'll be trying this with my tts decoders, for locos that I can't get tts sounds for, I'll probably get zimos, as some of these have 8 functions. I'd heard that you could use a second decoder for extra functions like smoke etc, but was struggling to find info on the details like programming 2 together. Thanks again to you all for your help, when I eventually get round to it, I might post some pics or a video.

Cheers, Rich

[Gwdriver16]

Got this worked out now. On Hornby TTS and standard 4 function decoders, F0 energises white and yellow wires, white is energised when forward is selected on controller, yellow when reverse is selected. This means you have white wire to markers, headlights on No1 end and tail lights on No2 end. You have yellow wire connected to markers, headlights on No2 end and tail lights on No1 end. This uses up 2 functions. The green wire is the 3rd function, aux1 (F25 on TTS and F1 on standard) and purple wire is function 4, aux2 (F2, only on standard decoder).

I'm going to keep both sets of markers on F0 on TTS and put No1 end headlight on aux1 (F25).

The standard decoder will have both sets of tail lights on F0, N02 end headlight on aux1(F1) and both cab lights on aux2 (F2). I will give both decoders separate addresses. During normal running (Train attached No2 end), only TTS decoder will be on (Good idea LMSFan72). This will reduce heat.

I've used a bread board to set up a circuit to program the standard decoder before fitting (Only needs an address change). I've described this in a separate post (Awaiting passing, contains photos).

Cheers Rich

[Fishmanoz]

Rich, what you are describing here for F0 plus white and yellow wires is the standard directional head/taillight arrangement in all decoders. So remember in your standard decoder you will have to control running direction to ensure your lights on yellow and white are switched on/off as required.  Also ensure you have a current limiting resistor in every light circuit. 

 

Also note that all decoders connected to a DCC supply are on all of the time.  However, they draw less current if none of their outputs are operational (train at standstill or lights etc off).

[Gwdriver16]

Thanks Fishmamoz, on the Bachmann locos i have, all of the necessary resistors are on the lighting PCBs in the nose ends. I tap into the circuit where the wires connect these lighting boards to the loco main PCB. I've done a trial on a class47, by connecting the tail lights at one end, to aux1 (Green wire), all worked fine, the second decoder will give me the necessary extra functions.

The decoder will be set up so white is energized in forward direction, I've done a simple bread board circuit to programme the decoder before installation, this includes a check for all functions.

I'll be making a permenant one once I've scorced the parts. See my post on this forum for bread board decoder tester. RAF96 has already done exactly what I was thinking of and has supplied a link to his web page detailing it. My permenant one will be nearer to RAF96's design, using 8 pin socket and motor attached to orange and grey wires. My bread board one was just a rough mockup to test if it worked.

Check it out, it'll allow me to pre-programme the function decoders with new addresses and check all functions are working, then fit them after I've fitted and programmed the TTS decoder.

Thanks, Rich

[Gwdriver16]

Hi everyone,

Just had a thought. With the 2 decoders, can both blue wire + commons be connected to the same circuit, or will they have to be separate. I was wondering if I could connect them both to the main PCB, or would one have to be connected to a separate circuit?

Thanks, Rich.

[Fishmanoz]

Interested if anyone thinks you can connect them. The blue in each decoder is a tap off the positive side of the bridge rectifier in each and, as such, is isolated one decoder to the other. While the volts should be the same or similar, I would keep them quite separate.