Tuesday, August 14, 2018

Insulated Driving Wheels: An Experiment

While the build of Ivor might be on hiatus, I now have four wheels which are effectively scrap (given that they are no no longer the same size) on which it is safe to experiment. The final two pieces of the wheel design I needed to finalise were the insulation from the axle and the crank pin.

Let's start with the insulation first. It's difficult to see in the photos but what I've done is to simply take the test axle from a few posts back and fit this to one of the wheels. This involved opening out the axle hole in the wheel to 2.8mm which is a nice tight fit for the insulating bush I took from a pair of hornby disc wheels. I'm not sure this is the final way I'll go (I might try turning my own insulating bushes) but it is nice and straightforward and seems to adequately insulate the wheels from the axle at little effort or expense. Obviously on the real wheel I'd trim the axle back to the face of the wheel.

When I was originally working on the 3D model for the wheels my plan was to use a 14BA screw as the crankpin with a hole through the wheel tapped for 14BA so the screw would actually screw into the wheel. I had wanted to cast a recess into the back of the wheel to seat the cheesehead bolt but the wall thickness requirements meant that wasn't possible. So my plan was to screw the bolt into the wheel and then cut the head flush with the back. A little drop of loctite helping to hold the bolt in place. Unfortunately while I managed to open out the hole big enough to tap, I then broke one of my 14BA taps and stripped the cutting threads from another. With no taps left I opened the hole up slightly further so that the bolt was a tight sliding fit. Rather than holding it in place just with loctite on the body of the bolt I decided to drill a recess for the bolt head anyway as there seemed to be enough material. With that done the bolt sits nicely in the back of the wheel and the bolt head helps to keep the bolt perpendicular to the wheel while also providing more surface area for loctite to hold it in place. Once the loctite had set I did try pushing the crankpin out and couldn't do it by hand. Even with a set of pliers it took a lot of force to break the seal, so this definitely looks like a workable solution.

Friday, August 10, 2018

Ivor is Going on Hiatus

So unfortunately work on Ivor is going on indefinite hiatus.

After doing the maths this morning I thought I'd go ahead and turn the wheels down to 20mm from their original 21mm diameter. If you remember when I originally turned the wheels I said it was hard work to get them down to 21mm, well getting down to 20mm seemed nigh on impossible. I continued to take it very very slowly but the work involved looked like being too much for my lathe without substantial extra work to the wheel (thinning the depth and reducing the width with a normal cutting tool first etc.) but with perseverance I got the first wheel to 20.5mm and so decided to split the difference and stop there. All (reasonably) good so far.

Having let the wheel cool (they get quite hot during profiling) I went to undo the wheel holding fixture only to find it was quite tight. I gently used a pair of pliers to start loosening it (I did this last time when it was too hot to touch) only for the threaded rod to snap in half! This means that the wheel holding fixture is now useless. More to the point I now have one 20.5mm diameter wheel and three 21mm diameter wheels so Ivor is going nowhere anytime soon.

I think the best thing to do is simply put the kit away for a while and focus on something else. This will give me time to rethink the wheels (I'll redo the 3D model for a smaller diameter and to make them easier to profile) and probably to save up to replace the wheel holding fixture. I could continue on with the body, but I think I'd just get frustrated knowing I couldn't fit it to a working chassis (plus it might need modifying to fit the gearbox etc.) so will probably turn to something else entirely. Not sure what yet, but I'll be sure to post when I've done something new.

Determining the Wheel Diameter

I've been having a bit more of a think about the wheels for Ivor after yesterdays discovery that they are probably too big to both fit inside the splashers and next to each other. It's also been pointed out to me that, normally the centre line of the buffer on standard gauge stock sits 3' 6" above rail height, which would be 14mm in 4mm to the foot scale. While Ivor isn't exactly based on a real loco, it would still be useful to have him sit at roughly the right height if I don't want him to look daft against standard 4mm wagon kits etc. The upshot of this is that I've been doing some maths...

The upshot of all that is summarised in the bottom right hand corner. My original 21mm diameter wheels would put the centre line of the buffer at 14.5875mm above rail head, so about 0.6mm (or almost 2" in real life) too high. To get it to sit at exactly 14mm I need the wheels to be 19.825mm in diameter (excluding the flanges).

I think given this bit of maths (and if anyone fancies checking my working it would be most appreciated) that I would be better off reducing the wheels to 20mm (or 5' in real life rather than the suggested 5'3") which would give a total diameter including the flange of 21.27mm which given the wheelbase of 22.5mm should leave a gap of 1.23mm between the flanges of the two wheels.

Does that seem sensible to everyone?

Thursday, August 9, 2018

Is It Just Me?

I'm beginning to wonder if either I'm drawn to kits with design issues, or if I'm just not very good at building kits accurately.

Having finally battled all that was weird and wonderful with the Canopus kit I thought the kit for Ivor the Engine looked quite a bit easier. Okay I made life more complex for myself by deciding to make my own wheels, but in itself that shouldn't affect the design of the kit. Having built up the chassis (fairly easily I may add) I've now made a start on the body, which in turn has led me to discover a couple of, potentially, catastrophic issues.

So far I've formed the splashers and attached them to the footplate, which also includes the front buffer beam. What you can see, in the photo above, is the chassis slotted into the footplate containing one of my wheels on the test axle I turned.

The first problem was that the chassis wouldn't fit into the footplate as it was slightly two long. After a lot of careful filing I've made it fit and the fixing holes appear to line up pretty well. The second problem though is that the chassis wouldn't fit with the wheels attached. In the photos you can see that I've bent the front of the splashers that are inside the model to get the wheel to fit, and it still feels like it's catching when I turn the wheel. I think the solution to this will be to cut away the extra metal, as it can't be seen once the model is assembled, and isn't needed to keep the chassis central once the screws are in place. Before doing that though I wanted to figure out why the wheels didn't fit.

Looking again at the instructions, which are just a sequence of photos with no textual description, I think all the photos show the 7mm version of the kit and not the 4mm version. It's difficult to be sure as most of the photos don't include anything to give a sense of scale, but they all seem to have previously appeared in this RMWeb thread discussing the build of the 7mm kit. The one page that is clearly for 4mm gives the size of the finished model and is the place where it suggests to use Ultrscale OR Markits 5'3" 16 spoke wheels with 8 spokes removed. As I was making my own wheels I took the 5'3" as a starting point and produced 21mm wheels (5.25*4). As I made the wheels to RP25-110 the flange depth is 0.025" or 0.635 giving a total wheel diameter of 22.27mm. The instruction page also states that the wheelbase is 22.50mm which as far as I'm concerned is just a bit too close to 22.27mm for comfort. In fact placing two wheels against the chassis I'm not sure they will fit without the flanges touching!

Reading through the thread on RMWeb I did notice that the suggested wheels in 7mm scale are actually 5'2" diameter not 5'3". I know that would only change their diameter by a small amount (they would be 21.94mm over the flanges) but that extra clearance would probably be great for both the splashers and between the wheels.

I can't be sure that the flanges will touch until I turn up a second axle (I've just the test axle so far) but if they do then I guess I'm going to have to try and reduce them in size slightly which is a real pain, although it might mean I can remove the slight flats on the flange from removing the wheels from the sprue. Mind you that isn't really the point; if I'm following the instructions for a kit am I just picking kits with issues, or am I just not assembling them accurately enough?

Monday, August 6, 2018


Next up in the sage of making wheels for Ivor is the bit I've been dreading the most; profiling the wheels to add the flange. In theory this should be easy as in essence all it involves is reducing the diameter of the casting using a profiling tool, but that glosses over a whole bunch of issues.

Let's start by defining exactly what we mean by "to profile" a wheel. Put simply it's to shape the tread and flange of the wheel to a specific shape. There are lots of different profiles all defined by different guidelines and associations. For Ivor I've settled on using wheels profiled to match RP25-110. This means I'm using the RP25 profile, as defined by the NMRA, with code 110 wheels, i.e. wheels that are 0.110" in depth (or for those who prefer metric, 2.794mm). Now a lot of the work of making the wheels fit this profile was done during the design of the 3D model as it set the depth etc. but it is impossible to 3D print thin enough and accurately enough to produce the flange.

To add the flange (and to ensure a nice smooth turned tread) I'm going to finish the wheel profile on the lathe, using a specific profiling tool. You can get profiling tools from numerous places but I picked up this one from Carbide Solutions as it had a square shank which made mounting it in the lathe easy.

Even with a sharp profiling tool (I'd not used mine before) this is actually quite hard work for the lathe and involved taking very very tiny cuts and quite a few breaks even during each wheel to ensure I didn't over heat the lathe or the wheel (it got exceptionally hot).

As you can see I've held the wheel in the lathe using a tool designed specifically for the job (another purchase from fohrmann-WERKSEUGE). The tool ensures the wheel is perpendicular to the lathe bed and clamped tight, although I did manage to over tighten the clamp on two out of the four wheels which unfortunately opened out the axle hole in the wheels slightly -- not a problem as I'll fit insulating bushes to these two wheels so will need to open the hole further anyway.

You'll also notice that I've left as little of the profiling tool hanging out of the toolpost as possible to try and reduce chatter, and I clamped the crossslide tight so as to remove as many sources of movement as possible. On the first wheel I started with the lathe on a slow speed but that was a disaster as the tool tended to dig in and bounce around, even with the lathe locked as solid as possible. I reconfigured the belts to run the lathe on it's fastest setting (4000 rpm) and with very very tiny cuts this worked a lot better, and fortunately I figured this out before destroying the first wheel. Once I'd reduced the first wheel to the correct diameter (21mm for 5' 3" wheels at 4mm to the foot) I took a note of the position on the cross slide, and then repeated the process for the other three wheels.

It took me all morning (in blocks of a couple of minutes here and there) but I've managed to profile all four wheels. While I'm not going to claim any of the four as perfect, I'm happy enough with them all to try and assemble the loco. All four have a difference of no more than 0.02mm when measured and all are slightly above 21mm rather than below, so they are close enough that I think they should sit level. The main issue with them is that when I removed them from the sprue, and filed the excess away, I didn't take into account that the edge of the flange was the edge of the casting (something I'll change if I produce any more) and so in a couple of places there is a slight straight edge on the flange, but hopefully not enough to cause a problem, if it is I can always order another set and replace them but fingers crossed...

Friday, August 3, 2018

Axle Hole

Having separated all four wheels for Ivor from the sprue they were attached to during printing the first step is to drill out the central hole to take the axle. Normally if I was trying to make a hole to fit an axle I'd drill it slightly undersize and then open it out slowly with a reamer to get a tight push fit. In this case though I'm going to turn the axle as well, so I can make the axle fit the hole. This means I can use a cheap drill to make the hole and not have to worry about buying an expensive reamer. That isn't to say this step doesn't include expensive tools as not only did I do the drilling using my lathe but I bought a set of step chucks specifically for the task.

Specifically I bought the 20.00mm to 24.50mm step chucks from fohrmann-WERKZEUGE. As you can see these allow me to hold the wheel (or any other appropriately sized cylindrical object) in the lathe for accurate centre drilling. In this case I've used them to widen out the axle hole to 2mm in each of the four wheels.

Visually the wheels all look good and the hole appears to be nice and central so hopefully that's the first step successfully achieved, and now they all have an identical sized axle hole I should be able to mount each in turn in the same position in the lathe which should make turning the flanges on each in turn a little easier.

Thursday, August 2, 2018

Flangeless Wheels

Having checked that I could machine the silver steel rod to make axles I've turned my attention to the actual wheels. I've actually been working on these for a while now. The main problem I had is... what do Ivor's wheels look like? While they always seem to have the same number of spokes they seem to differ slightly from scene to scene, mostly around the thickness of spokes. However, in some shots the wheels have balance weights and in some they don't. In the end I decided to work from a specific drawing in my childhood book.

The plan was to make a 3D model of the wheel, bulked out in diameter slightly. These would then be printed in brass, and then finished on the lathe using a profiling tool. So far I've designed the model (see above), had them printed as a sprue of four wheels (cheaper than printing each separately) and separated the wheels from the sprue.

Next up will be drilling out the central hole for the axle and using a profiling tool to reduce the diameter slightly and add the flange.

Stepped Axles

Having assembled the basic chassis the next step is to figure out the wheels. Unfortunately as the prototype is a cartoon character there aren't any wheel sets I can just buy off the shelf. The only suggestion in the instructions is to use "Ultrascale or Markits 5'3" 16 spoke wheels with 8 spokes removed". Having looked at both of these options I wasn't happy with either, as even with the extra spokes removed I didn't think they really matched up with those you see in the cartoons. In the end the only thing I could think of was to make my own wheels. Clearly this isn't going to be a quick and simple process so it's likely to go on for quit a few posts as I detail the different steps and experiments. In this post I'm going to concentrate on the axles.

As I mentioned previously the kit is designed for use with axles with a diameter of 1/8". The wheels on Ivor don't have a huge central boss, and certainly not something big enough to accommodate a 1/8" (or bigger for an insulating bush) hole so my plan is to use stepped axles. Most wheels seems to use steel for axles so I've bought some 1/8" diameter silver steel rod (from Eileens Emporium) but having never tried to turn anything other than soft brass rod (CZ121 brass to be precise) I had no idea how well this would work.

For a quick experiment I've turned down a short length of the silver steel rod to take one of the OO gauge wheels I used to build the Clayton. These wheels fitted on a 2mm axle so I had to remove roughly 1.25mm from the diameter of the rod. I actually found machining the silver steel possibly easier than brass and the turnings come off as long thing ribbons rather than lots of dust. All told (including setting up the lathe) this took about 10 minutes. Now obviously machining steps on both ends of an axle and getting the back-to-back correct is going to be more time consuming but at least now I know I can work with the materials I've got, and I've 26" of rod (minus the short test piece) to play with and I only need two short (approximately 19mm) axles.

Wednesday, August 1, 2018

In the Top-Left Hand Corner.....

You may remember that back at Christmas 2016 one of my presents was the OO gauge kit from PH Designs for Ivor the Engine. It's taken me a while but I've now made a start on assembling the kit.

The first step was to separate the main chassis component from the etch and then solder bearings into the frames. The kit is designed to use wheels with 1/8" axles but the 1/8" bearings I already had wouldn't fit in the holes in the frames as the walls of the bearings were way too thick. As you can seem I'm using a set of bearings from Nairnshire Modelling Supplies which fit perfectly within the etch.

There were some odd issues with little tags on the etches (not where the tabs hold the part to the fret but elsewhere, almost as if it hadn't etched properly) but in general the cleanup was easy and the bearings soldered into place without incident. I did have a bit of an issue when folding the chassis though. While folding down the frames was easy enough, and two lengths of 1/8" silver steel rod kept everything square, it was impossible to get the step down at the back of the chassis to fit properly. In the end I snapped off the final part of the fold, reduced the length ever so slightly and then soldered it back into place -- hence the slightly messy solder joint. It might just be that I didn't quite get the fold right in the middle of the line (quite wide fold lines) rather than an issue with the part itself. Anyway so far so good.

Next up will be sorting out the wheels...