The following article is made available with the kind permission of malc-c, who undertook this project. Have you been working on anything interesting regarding Poole Graham Farish locomotives / diesels? Would you like to share it with thefarishshed community? If so then you will find details at the end of this article.
This is by no means a definitive method of conversion, and is still a work in progress. Things I would do to improve the conversion are mentioned as I work through the stages, but as I really wanted to do a conversion that others might want to try, and whom may not have fine soldering skills etc I kept things simple.
The first consideration was which decoder to use. Space inside the body is taken up by the chassis, so the decoder couldn’t fit on top of the chassis. The only option I could see was to place it in one of the cab end, so the decoder had to be small. I’m on a budget so opted for a Bachmann E-Z Command 36-556RA decoder, and promptly placed my order for one and a pack of sockets.
The first thing I did was perform some surgery on the chassis, by grinding off the rounded lip on one of the end plates of the chassis. This was done with a cutting disk (shown below) and then finished off with a file.
This allowed the wire harness to sit flush on the top of the chassis, and then to allow the body to sit correctly on the chassis the corresponding standoff inside the body was ground off as well.
Now as you will know, with DCC the supply from the track needs to be isolated from the feed to the motor. This is simple when you have a motor with two terminals on the end, but in these older models, the chassis forms one side of the motor collection, and the pickups make direct connection to the chassis. So how best to isolate the motor from the rails. Following a suggestion I received from an N gauge Forum member, I looked at using copper clad PCB. The problem is that most of the PCB available to the retail market is too thick to be placed on the chassis so that the pickup contacts still functioned. Now you can get PCB in all manner of thicknesses, even down to flexible films. I opted for something that still had some rigidity and sourced a sample of 0.2mm thick, with 1oz copper grade. This is thick enough to take the pressure of the pickups, but thin enough not to restrict the movement of the bogies. Four small pieces were cut and fixed to the chassis using superglue.
The decoder was then placed in the harness and offered up for fitting. I used a hot glue gun to secure it to the chassis, making sure that the end part could still be remove. Holes were drilled through the plastic battery box moulding that isolated one half of the motor connections, and the motor wires, and one pick up wire fed through them. On test fitting the body it was a snug fit, and the thickness of the socket still meant that the body sat a little high. If I was confident enough that this conversion was going to work first time I would solder wires directly on to the pins and then apply heatshrink to them which would allow the body to sit just that little bit lower. But as I may need to change the decoder I stuck with the harness. However the difference is only a fraction of a millimetre as can be seen in the pictures.
The next stage was to make the connections. The motor wires were terminated by soldering one to the copper strip, and the other was wrapped around one of the screws that held the plate to the chassis. The capacitor was removed. The red pickup wire was then soldered to the inner PCB, with a jumper wire soldered to the other inner PCB at the rear bogie. The black pickup wire was taken down the cab end and soldered to the outer PCB.
To test the connections, the decoder was removed and the loco placed on a test track with 12v applied. There was no tripping, and the loco failed to move, which was a good sign. Then using two thin bits of wire the connections between the pickups and motor wires were bridged at the harness, and when 12v was applied the loco trundled down the track. Job done !
The decoder was then fitted and the body replaced. To see if the height was a problem I coupled a coach and it looked OK. When this is running on a layout it would be hard pushed to notice the extra half a millimetre.
I have the excellent MiniDCC controller designed by Robert Côté breadboarded on the bench. The advantage of the MiniDCC controller is that it has four control knobs that can control four DCC trains directly, just like having 4 analogue DC controllers, so less need to hit the emergency stop button so much when things get complicated ! Anywaye I wired up the test track and tuned the power on. Using the oscilloscope I confirmed the presence of a DCC signal on the track and then placed the loco on the test track and advanced the control knob for the loco with ID3, and as I did the loco pulled away and trundled down the test track. – I was chuffed that my first attempt to convert the Class 25 worked.
Anyway, hope this helps anyone looking at converting an old Poole manufactured Graham Farrish loco.
Are you interested in posting for thefarishshed? Have you undertaken a project on Poole (UK) built Graham Farish stock that you would like to share with us? If so then use the contact us page to tell us about your project and get it published!