Monday, 30 March 2009

Brafferton Update 20: Adventures in DCC

Having successfully forced my way into the Farish DMU yesterday and installed a Lenz Silver, I today set my mind to programming. A second chip was inserted in the trailer unit; this seems a rather expensive option just to control the lights. However it seems to be the most straightforward so I'll just gulp and accept it.

The chips were then given consecutive address numbers; 0405 for the power unit and 0406 for the trailer. I must be getting good as I then worked out how to join them in a multiple unit. It went a bit downhill after that; the two ends display the same colour lights at the same time! The power car is correct, but the trailer is wrong. I'll need to work out what to do next; presumably altering the direction CV will correct it. Unfortunately, at that point the kids arrived home from school and any pretence at concentrating was lost!

A job for an evening perhaps. For now, though, here is the Liverpool service coming toward you:


Edward Sissling said...


Both decoders should have the same address for simplicity (seems like hard work making up a consist for what in effect is a single unit).

I struggled to read or write an address to the chip in the trailer car (Lenz Silver Mini), I think owing to the lack of a motor, so I did the programming for both chips via the driving car and then moved one into the trailer car. Works fine and lights all behave as they should.

One downside of the Farish 108 is the 'orrible blue/white LED illumination of the destination blinds etc. I cured mine by giving the transparent plastic light guide mouldings a thin wash of yellow enamel paint. This gives a much more realistic 'tungsten filament' glow typical of these 1960's vehicles.

Nigel Cliffe said...

I agree with Edward, using a "consist" for a multiple unit is creating work for yourself.

You might be able to use decoder locking to stop programming the "wrong" one (depends on decoder maker, and I haven't looked up Lenz). Though using separate cars on the programming track is the simplest method.

The lack of "read" of the trailer car is because there is no motor load on the chip. The chip will have accepted the programming instruction, but cannot send an acknowledgement reply back to the command station. Most chips will acknowledge with just lighting load, but a LED might be insufficient. You can fix this by clipping a circa 100ohm resistor over the motor wires (orange/grey), though I recommend removing the resistor in normal running (or it gets hot!).

The simplest way to reverse direction (which should change the lights) is with CV29. There is a calculating tool for CV29 in the "articles" bit of the 2mm website.

The more advanced way to alter the lights is with function mapping, but to be honest I don't recommend this via a DCC handset, instead take it to someone with a computer interface and JMRI (eg. Mick, which means he can learn how JMRI works !).

You might be able to tone down the "horrible LED blue" with the output level CV's for the decoders. Or possibly replace the LEDs with different ones, or try with a larger series resistor before the LED (makes them dimmer).

If you have something else to chip, I suggest retrieving the relatively expensive Lenz from the trailer car and fitting a function only decoder. Or a cheaper one. Not sure of the smallest function only decoders, many are a bit big, but a Digitrax 125 loco decoder is small and a lot cheaper than a Lenz.

- Nigel

mick said...


I did as with my 108 as Edward did with his. During the experimental process I seemed to lose the active DCC address (didn't write it down and forgot what I'd given it...!) but when I acquired the Sprog (JRMI) I was easily able to read the decoder and do a full reset then change the number to what I wanted. I did both decoders separately then when put on the running track together everything was fine. The decoders are cheapy Digitax DZ1251N which would easily do for the dummy car. I may replace the driven one for a Zimo or Lenz.

I'll bring Sprog along on Saturday.