Flaws in Both Design and Manufacturing

With Canopus finally finished and a little modelling time opening up now that my son is back at nursery I had a look at what I could work on next. Now I do have a number of projects already on the go, but they all require some thought or scratch building. What I wanted was something a little simpler where I could just "follow the instructions". Looking through my stash of un-built kits and I've picked the Narrow Planet kit for the Baguley McEwan Pratt ‘677’ 10HP 0-4-0PM. This is actually the most recent kit I've bought, but as it uses a ready-to-run chassis should be a little easier than most of the other kits I have to hand. Famous last words!

The kit is designed to use one of the tiny Japanese N gauge chassis, specifically the TU-DB158 from TGW (Tsugawa). As you can see from this official diagram the chassis really is tiny.

While using a ready to run chassis like this does make things easier, I did know that I needed to do some work to it first, but it turned out to be a little more than I expected.

Firstly, as with many of these tiny chassis, the motor is not designed to be used with the standard 12V track supply. In fact the motor in this one is rated for only 4.5V. One option would be to fit a resistor to make it safe to use with 12V, but to be honest the tiny motor isn't brilliant and the kit instructions suggest changing the motor and recommends using a 0615 from Tramfabriek. This is actually really easy to do as you just need to unsolder the existing motor wires, slide the motor out, swap the worm gear from the old motor to the new, slide the new motor back in, and finally solder the wires on to the pickups. If it all goes smoothly then it would take probably all of about five minutes.

As you can see the new motor is a little longer, but given it will still fit inside the body then it's worth it to have a really good 12V motor in use. Now as I said that modification was nice and straightforward. Unfortunately there is still one design flaw and one manufacturing flaw that I had to deal with before I could move on to the body.

Let's start with the design flaw. The design work for the kit was almost completed when it was discovered that while the chassis would run fine on plain track it had a nasty habit of derailing when passing through point work. Both axles of the chassis are powered and so there is a series of gears linking them together. These gears are module 0.4 with 11 teeth and have a maximum diameter of 5.2mm. Unfortunately the wheels on the chassis are 5mm (excluding the flange) meaning that the teeth of the gears protrude 0.1mm below the top of the rail. On plain track that's not a problem but the gears catch on the rails when passing through points. I've no idea how this wasn't spotted during design, testing, or manufacture of the chassis, but clearly it's something that needs fixing.

It's difficult to see the problem, especially in photos, but with the chassis on the wrong tracks of my dual gauge (14mm and 9mm) test track you can just about see that the gear rests on the third rail and actually raises the chassis up slightly. The solution is fairly easy but you do have to be careful and take it slowly. Once I'd got the decent motor in, I secured the chassis upside down using some blutak and then with the wheels turning gently rested a small file against the gears to grind away the ends of each tooth. I think I've taken off enough to fix the problem, but the only OO9 points I have are currently a bit difficult to access in the garage. Rather than taking off too much I'll just revisit this later if it's still a problem.

While fixing the gears I spotted the manufacturing flaw, in that one of the wheels wasn't perpendicular to the axle. If it had just been slightly off I might have left it, but there was a serious wobble. Fortunately the chassis uses split axles, with each wheel only having a small axle stub which is fitted into the plastic axle. This means that I could remove just the single problematic wheel. Using a modelling knife behind the wheel it was easy to prise it off. I then removed the pin holding the coupling rod so that the wheel was completely separate to the chassis. It took me a while to find a way to hold the wheel that would also let me straighten the axle stub. In the end, I used one of the collets from my lathe, which highlighted just how bent the stub axle was.

I'm not sure what the wheel and stub axle are made from, but it's quite strong whatever it is, and took a fair amount of force to bend it. It's still not perfectly straight but it's the best I could do, as I didn't want to risk snapping the axle from the wheel. Once straightened putting the wheel back into the chassis was easy, but it took a long time to get the coupling rod pin back in. It's a tiny pin and it's a tight fit into the hole in the wheel. I tried lots of different ways of holding it and pushing it in and all failed. I had to be careful as the last thing I wanted was for the pin to ping off into oblivion. In the end I lightly superglued the pin to the end of a piece of acrylic rod and then pushed it home. As it seated itself properly in the hole the glue gave way leaving almost nothing to clean up.

After all that I was quite happy to see it trundling up and down my test track without any problems. It would benefit from a little weight (which the body will provide) but it runs a lot better than it did before I started to fix it, so I'm now happy enough with it to start work on the body.