Welcome on Henk's website

find all about fiNe-scale modelling
in a scale of 1 : 160


DIY loco building

a few construction photos

The best of both worlds: massive split frame construction

Let's suppose you are interested to add a certain prototype to your layout. But alas this turns out as not available as r-t-r model or only as unsatisfactory model not suitable for modification. For the latter decision all information is helpful, that includes photos taken under unfavourable conditions of similar loco's.

type 64
The only remaining original NMBS type 64, happily under roof in Treignes.

and drawings of course, leading to:

Why not construct my own chassis kit?

chassis parts
The chassis kit components: a few pieces of metal with some holes at the right place, some wheels bought in the 2mm shop, 2 gears + wormwheel, some screws and a few pieces of plastic plundered from an old loco or directly cut from sheet

Try to get this parallel groove into the rods! there must be alternatives for skrawking by hand in 0.5 mm steel?.

After spending some time a near complete chassis, assembled and ready for some tests.

Now if you are here wouldn't the rest be easy meat either?

top part component construction

other parts
Start with turning major and minor parts, some less conventional, but nevertheless found on the chosen prototype.

smokebox door
In the mean time solder up some other components, here the smokebox door pieced together, just a further 23 pieces of flat metal soldered into a single unit.

A short note on the current state and why it takes so long. The cab is etched in NS silver and was fitted to the boiler and footplate. Most turnings are ready to put into place when running is satisfying. The simplest of problems to solve was the short created over the Walschaerts rods and the link hanger. This was done by creating insulated eyelets on the connecting rods using some copperclad epoxy board. One of the original wheels was damaged during turning and showed considerable wobble. This required replacing the wheels with new ones. Next testings showed that the loco slipped when running forward. Part of the problem is that the weight with massive boiler was too far forward. Drilling out the smokebox with a 10 mm hole 10 mm deep helped getting the center of gravity closer to the second coupled wheelset. Laying the bogie a tat deeper may solve the slipping entirely as it doesn't slip without bogie. However the gearing chosen seems to give a rather low top speed, changing this for a module 0.3 wormwheel with same diameter will require a total dismantling into parts to get access to the wormwheel axle. The tender frame was replaced with fixed axles of the type with closed bogie sides. This enables to get some weight of the tender onto the back of the loco to enhance the pulling power when necessary.

For its 12th anniversary I made some progress on building a correct tender. The rivet counter will come to 858 and may conclude that some still lack.

There are some 30 pieces of metal shaped up now and it slowly starts to look like a tender. The rivets were impressed with a special tool a similar process as with this tender wrapper here only this time using the CNC mill. This allows higher accuracy, goes faster and can produce smaller rivets. The process took only 12 minutes for 858 imprints. The horizontal rivets are spaced about 7% further apart, this in order to see individual rivets instead of becoming a full line.

combination Testing the combination, the tender looks to sit a bit low here, the loco body dusted off, all loose items failing, most of them sit somewhere in a box (hopefully).

The major reason why this project was shelved was because of the noise the gearbox produced. Thorough inspection showed several sources, first of all the space between worm and wormwheel was too large and there was axial play in the worm axle. Together this caused the worm to vibrate. As speed was still too high the decision was taken to strip the loco again and replace the combination of worm and wormwheel with a different reduction. It now has an 18t skew cut M0.4 brass wormwheel with a nylon worm giving a ratio of 34 to 1 and no longer generates noise. The bits for the Walschaerts motion need installation again. This means that I can finish the loco body by applying handrails and appendages etc. but there are a lot of those and numerous tubes between them.

A new approach for a faster finish

The succesful print of a new cab and boiler for the type 25 (see here) did me wonder whether it was possible to take it a bit further and try to print all appendages and connecting tubes in one go. No better model than trying to finish than the type 64, which already had a running chassis, but a body that was still missing all those gubbins. Thus I delved up my prototype information and made a hand sketch with dimensions measured against an easy reference point. I choose the front of the cab as zero and set out the dimensions and angles of all appendages on boiler and cab. I Also studied the available photo material for the Belgian ACFI preheater system that was installed on most of the loco's in passenger service. With this information I made a new 3D drawing, this effectively costed a smallish 18 hr. This lead to a new 3D model that can be printed:

The type 64 with ACFI as seen in Cinema4D

Then came the test by loading the Anycubic with the sliced STL file and surprise ...

print right
print left

Although not a 100% print succes, but it is close to it, that is not due to the printing itself. On the left hand side the lower part of the pump section has disappeared because I overlooked that it floated above the footplate. But all those other vessels and tubes for the ACFI are there printed and well. Clearly the whistle sits a bit too close to a temporary post that holds the weight on the safety valve thus probably will get lost and I think I ripped off the lower part of the sand tubes on the right hand side during cleaning the print. Some extra posts under the cab front will help to keep the intended curve too. But this result looks a good shortcut to arrive at an acceptable loco body for a complex loco. Even this first print already looks to be quite usuable with only some minor smoothing with 600 wet and dry necessary.
Because the boiler is now a hollow tube I can fit a sound decoder and the remaining space can be filled with lead, it probably won't be as heavy as a massive brass body but the total weight will suffice as it already has a massive frame. The bonus is of course that the next loco with ACFI will take only a fraction of time as it all can be lifted and draped onto another boiler, the same holds for the Knorr double phase airpump and other small components as tube flanges, snifter valves etc.

print left

Something gained something lost, 2nd print with pumpsection but a piece of tube needs to be inserted.

Next step will be finishing the whole lot, with some 0.2 mm fine steel handrails on posts, produce a fitting weight piece and program a sound dekoder.

copyright: Henk Oversloot
updated: 10 June 2020