Ian_C's workbench - P4 and S7 allsorts

[Ian_C]

I've been gradually making up the motion for the 8F and thinking about how to fix the return crank to the driving crankpin.

There's no advice in the instructions and the Geoff Holt books (Wild Swan) don't say anything much about the subject apart from a reference to being quick with the soldering iron. Not sure I fancy my chances of soldering the return crank arm to the end of the crank pin; crank parallel with wheel, phased correctly relative to crankpin and not melting the crankpin insert or clogging up the motion. Besides, how do you get the motion apart after the return crank is soldered on, apart from unsoldering of course? Here's one way of doing it.

The prototype driving crankpin is 6" diameter, or 3.5mm in S7. Much larger than a typical model crankpin, but provides a better opportunity for fixing the return crank to it. The steel crankpin insert in the Slaters S7 wheels is threaded 10BA and provides a good, solid fixing for the pin. There's no reason to change that, so the end of our new crankpin should have a 10BA male thread to match. A 10BA thread has an outside diameter of 1.7mm so it's quite possible to tap a 10BA thread into the end of the 3.5mm crankpin to secure the crank. It should be possible to solder a 10BA stud to the back of the crank. Then we'd have a crank we can screw into the end of the crankpin and secure with a medium strength thread lock. We have to set the angle of the crank correctly relative the crankpin but the chances of getting the threads phased accurately to go tight at the right angle is about zero. If we assumed a worst case of getting the crank tighten 180 degrees from where we wanted it then we'd need to adjust the thread engagement of the crank stud by half a thread pitch, or about 0.17mm. We could put the crankpin back in the lathe and skim a little off the length or we could put a shim on the back of the crank to phase it the other way. So if we machine the crankpin to be a little longer than necessary we can adjust the phase of the crank relatively easily by carefully shortening the pin. Seems plausible, so the next step is to make one and try it out.

The crankpin is machined from steel rod, I'm guessing it's EN1A or similar. You really need to do it in a collet, it's far easier and more accurate than a chuck for these tiny things. Turn down to 3.5mm , then the threaded section to 1.7mm (for a 10BA die) and beyond that a short section at just greater than thread core diameter, 1.4mm will do. The 1.4mm diameter section acts as a lead in for the die so that it starts to cut easily and is lined up nicely before it starts to cut the real thread. Square up the shoulder, under cut the thread, reduce the thread to the correct length. Gets you one of these...

Cut it off and put it back in a collet the other way round. I confess to cheating, I cut it off with a junior hacksaw rather than shred my nerves trying to use a parting off tool. Face it off and take it out of the collet to measure the length of the pin. Put it back into the collet and machine off to the right length, 4mm in this case (being the combined thickness of coupling and connecting rod bosses, plus some clearance, plus some extra length to allow for crank phasing). Dimple with a tiny centre drill then drill 1.4mm (tapping drill for 10BA) to the required depth. Tap 10BA with a taper then a plug tap. There's your crankpin...

The crank is a laminate of an etched and a half etched overlay and they come to about the right scale thickness. It's a Stanier 4 stud fixing and the kit has two different overlays, one correct (ish) and one with a hole in the middle and the 4 studs too far apart. Goodness knows what that's for, so I used the correct (ish) one. There's a hole etched in the main thickness of the crank on the crankpin centre so we can use this to locate a stud. The stud is threaded 10BA to fit the crankpin, has a thin flange to solder to the crank arm and a small spigot to locate in the etched hole. Made this from brass. A picture's worth a thousand more words...

Now we can assemble the parts...

The starting angle of the threads in the wheel crankpin insert, on the crankpin and the return crank stud are uncontrolled and quite random so the angle that the crank arm ends up at relative to the wheel crankpin when everything is tight is also random. This doesn't prevent the model valve gear from working in a fashion ,but we should take the trouble to get the crank arm phasing and throw correct (trailing the crankpin by 90 degrees and with the end of the crank arm at a radius of 8" = 4.7mm on an 8F). We can do this by noting the angle between where the arm ends up and where is should end up. The pitch of a 10BA thread is 0.35mm, which takes a 360 degree rotation of the thread. 90 degrees would be about 0.08mm, and so on. We can use the angle error to calculate how much length to take off the end of the crankpin to get crank arm to sit at the right place when tightened up. Pop the crankpin back in the collet and reduce the length by the amount you calculated. Go cautiously, if you take off too much you either have to make a very thin shim or reduce the length nearly one pitch and have another go, and then you may end up with the pin too short. From this point on the crank arm and crankpin become specific to an individual wheel, so mark them and keep them together. We end up with this...

I plan to use a higher strength thread lock to keep the crankpin in the wheel and a lower strength thread lock to retain the crank arm in the crankpin. That way it should be possible to unscrew the crank arm to take the motion apart. I'm hoping that the cranks arms don't unscrew in normal operation. If the motion and valve gear move freely then the crank arm won't need to transmit much torque. Fingers crossed, we'll see.

Sketches (sorry, not quite BS 8888) from the notebook if anybody wants to adapt it to their own circumstances...


I should add an observation about small taps and dies. I recently bought some budget 10BA and 12 BA taps and dies with this project in mind. They were 'carbon steel' and a real bargain. except they weren't a bargain when it came to using them. The threads and cutting edges were so badly formed that the resulting threads were very poor. I coughed up for some proper HSS taps and dies from Tracy Tools Ltd.

 

[Ian_C]

While I'm waiting for a couple of odd sized drills to turn up to complete the 8F motion I thought I'd make a start on the chassis. Getting chassis to fabricate properly with axles at correct centres and square across the chassis hasn't been something I've found easy with etched kits in 4mm. It's my first attempt at a 7mm locomotive so I want to get it right (and it's too expensive to get wrong!). I've checked out some of the chassis jigs you can buy these days. They look handy, but any that look like 'proper' engineering are expensive. Here's my simple version of proper engineering.

Fundamentally all we need is four rods pretending to be axles exactly the right distance apart, parallel to each other and square to a flat surface. A metal plate, 4 holes and 4 lengths of 3/16" ground silver steel seems like the simplest way of doing it. I managed to get an odd end of 3"x 1" extruded aluminium bar locally, probably HE30 or similar. It's reasonably flat but you can see a little daylight under a ruler in some places. I had a friend at work with a milling machine skim it both sides and drill and ream the holes at the correct centres for a 7mm 8F (thanks Paul !). I want the axle rods to be a good push fit in the plate so there's no free play, but I also want to be able to remove them. Turns out that intermediate sized reamers are available 0.05mm increments and 4.75mm is about right for a light push fit with ground 3/16" rod. Simple, and as accurate as a modern machine tool will allow, which is plenty good enough for this job. Sure, it's not adjustable but I can just have some more holes put in at different centres for another loco (if I ever finish this one). Yes, the slab is long enough to take five holes for a 9F.

I made two sets of rods in the lathe. One short set to use for making up the coupling rods, and one longer set for chassis assembly. Here's the short set...

The ends of the rods are turned down to the diameter of the holes in the coupling rods, 2.3mm for axles 1,2, 4 and 3.5mm for axle 3 (see earlier in the thread for details of the driving crankpin). And here they are pressed into the slab...

There's one downside to having a aluminium brick under your chassis; it'll suck all the heat out of any soldering work. I cut a spacer from a bit of tufnol sheet to provide some insulation. On the MOK chassis the horn blocks are located to the chassis by tabs that project through the chassis plate and are removed after soldering. The spacer allows the tabs to project through the chassis and keeps them clear of the plate.

I was expecting some conflict between the exact centres of the jig and the assembly of the etched parts. Amazingly they match exactly. Horn blocks and bearings just slide down the rods and push exactly into the slots in the chassis. Gobsmacked. You could have just pushed the tabs into the slots, soldered it up and got the same result. I imagine the etch design was done on CAD so it ought to be correct, but the masking, registration and etching are spot on. That gives me some hope that the coupling rod centres are good too (they're etched on the same sheet, which helps). Congratulations to Dave Sharp at MOK and whoever he uses for photo etching. Here are the main chassis plates, the horn blocks, the compensating beams and the bearings all set up for soldering...

After some messing around with a 40W soldering iron that wasn't quite up to the job I went and found an ancient iron in a box in the shed that's been around for years, and of great and mysterious calorific output. Mysterious because there's no marking to indicate the power. You just plug it in and it gets very hot very fast. 60W? 100W? Dunno, but you could heat the room with it. It's enough to do this job neatly. After a bit of cleaning up and a dip in the ultrasonic tank (and it looks like a 7mm 8F chassis will fit in the tank - bonus) here's the result...


A hint on removing excess solder - a triangular scraper is far quicker and tidier than files and/or emery. With a sharp scraper and a bit of practice you can shave off any amount of solder really quickly and leaving hardly a mark. Not that I ever end up with excess solder of course. Just saying.

 

[Ian_C]

Another day on leave. The problem of how to spend the day was solved by a packet dropping through the letterbox containing a 2.3mm drill. That was all that was standing in the way of coupling rod completion (nearly). Here's what I ended the day with...

Being a first loco in 7mm,and S7 to boot, I'm anticipating some mither around coupling rod and crankpin clearances behind the motion and slidebars. On the prototype the clearance between the leading crankpin and the rear of the slidebars is nominally 11/16". That's about 0.4mm in 7mm. The prototype had recessed crankpin nuts on the leading 2 axles, the return crank on the third axle and a normal crankpin nut on the fourth, trailing, axle. Seems to me that my odds of success would be greatly improved by having some kind of recessed crankpin nuts on axles 1 and 2. the S7 Group sells a set of crankpin nuts to complement their Slaters S7 8F wheel set. I bought a set. Thing is the set provides normal crankpin nuts for all the axles apart from the driving axle. Not gonna work in S7 is it? And I have read on a forum somewhere that other folk have had clearance problems to overcome as a result.

So, recessed crankpin nuts then. Some challenges on the way; making new nuts with a 12BA thread in them, shortening the S7 Group crankpins or making some new ones, making a counterbore in the coupling rods to house the nut. After a day's work it ends up like this...

The nut is turned from a bit of nickel silver bar. I used NS because I wanted the nuts to match the rest of the motion. Don't know why I bothered, because the bar turned out to be more yellow than the etched rods. Looks like brass in the end, and it wasn't fun to work with either. Drilling a 1mm hole down the bar to tap 12BA was particularly patient and nerve shredding work and mercifully I didn't break a drill or a tap. I eventually solved the mystery of getting the parting off tool to work properly so I can now slice them off at exactly the right thickness. The counterbore means that the brass bush needs to be shorter, easier to make new ones than try to shorten the ones supplied, and the plain diameter on the crankpin needs to be shorter and the thread extended down to the new shoulder. I had sketched up a replacement crankpin to machine from scratch but hadn't quite worked out the practicalities of making the thing so chickened out and modified the existing ones which turned out to easier then I'd feared, albeit requiring some delicate work with the 12BA die. In passing, the way I held the crankpin to extend the 12BA thread was to clamp the 10BA (wheel) end in a 1-2mm collet in a collet holder held in the bench vice. It is possible, with gentle tightening of the collet nut, to get a good enough grip on the pin to hold it against the die and not to damage the 10BA threads.

The hole in the rod is 2.3mm through, simply because that's the diameter of the brass crankpin bushes that come with the S7 crankpin set. The laminated rods were drilled through 2.3mm with the new, sharp drill, trusting that the drill would find centre, which it seemed to do. The prototype recessed nut is quite a large diameter and extends out nearly to the diameter of the boss, and although I'm not exactly replicating the original arrangement I want the model to look similar. Therefore the counterbore diameter is 4mm. So here's the next challenge. Where can you buy a 4mm counterbore tool with a 2.3mm pilot? Answer = you can't. I'd considered using a 4mm slot drill but experience suggests that you need the work rigidly clamped to make the drill go where you want, and an off centre counterbore would look pants and cause clearance problems around the nut. Besides I couldn't see how to clamp the rods effectively. While searching for small counterbore tools I came across a couple of model engineering forums (fora?) where folk had made their own custom counterbore tools. One method seemed easy enough with the tools at my disposal so I had a go. Here's the result...

The tool is made from a length of 3/16" silver steel, axle material actually. Turned to c/bore diameter, drilled out pilot diameter, flats milled across to form two 'pegs' or cutting teeth, teeth filed to provide some rake, hardened (cherry red & water quench), teeth stoned to a sharp edge, pilot rod loctited in the hole. Sometimes you get lucky, and the pilot rod is the shank of an old mini drill bit that ended up in the scrap box. Exactly 2.3mm and Ti-N coated as well. Tested on a bit of aluminium first before attempting the rods. Worked very well, much better than expected. Very carefully worked out which ends and faces of which rods needed to be counterbored (no easy remedy if you get a c/bore in the wrong place), rods held on a piece of aluminium plate with a 2.3mm pilot hole drilled in it so the pilot is guided first by the hole in the rod and then by the hole in the plate before the cutting edges start to work, rod (and plate) held down by fingers (you get some sense of what's going on when you're providing the restraining torque with your finger tips), drop of oil on the rod, slowest speed on the drill, hold your breath and out comes a ribbon of swarf. Taken easily a bit at a time until the counterbore was 0.8mm deep and the nut tightened down on the brass bush and left the rod free to rotate with a little side play. Scrapping a rod was something to fret about but it worked OK in the end.

The end result...

Not exactly true to prototype, but looking similar. And anyway, it's a mid sixties 8F so you'll not see it under a coating of brake dust, ash, road grime and oil. This approach does reduce the rod/crank pin bearing area on the leading two axles, there's 1.1mm left, I'm guessing it won't be a problem. It'll be some time before I discover if I have 0.4mm clearance to the slide bars, or another challenge.

Oh yes, 'nearly' complete. There's still the pinning together of the forked joints to work out.

 

[Ian_C]

Like the chassis, the etched rods are spot on for centre distance. Having drilled the crankpin holes through to the final diameter they dropped straight onto the rod setting pins in the baseplate. I'm still amazed that etching, soldering up and drilling through produces parts so dimensionally accurate. The holes for the rod articulation pins lined up perfectly too, so no messing about required.

(full size, just for you daifly)
The articulation pin holes are 1.6mm so I could have used a length of 1.6mm wire, but since I have the Wild Swan LMS Locomotive Profiles No.8 - The Class 8F 2-8-0s I can see how it's done on the prototype and I'm afraid that wire just won't do. Measuring, sketching and off to the lathe again to make some little rivetty pin things.

The pins are steel. I left a 1mm diameter pip on the head that I filed to an approximate hexagon, just about visible in the photo. The opposite end of the pin is drilled out 1mm for a about 1mm deep to make kind of tubular rivet end that can be spread out gently with a centre punch and a hammer.

The final job is to drill 0.6mm into the top of the oil reservoirs to take some 0.6mm wire to represent the corks. Couple of hints here. The rods are laminated from 3 thicknesses, so if you cut a notch in the middle laminate where the cork will go it will fill up with solder during the lamination process. Then it's much easier to centre spot and drill down this deposit of solder than virgin nickel silver. For work this small I usually make the centre spot with the end of a scriber as it's easier to position than the end of a centre punch. A vigorous poke with the scriber is often enough to start a small drill but the scriber divot can be used to locate a centre punch if you want a bigger crater. The wire is just a push fit into the hole, no messing about with solder or glue at this stage. To make 0.6mm brass wire a force fit in a 0.6mm hole gently squeeze the end of the wire in pliers or a vice to flatten and widen it slightly. Force it in with pliers, snip off and file down to whatever height you think an oiling cork is in 7mm.

Bit of an epic, lots learned on the way and I know they exactly match the axle centres. Very happy.

There's a bit of me wondering if I couldn't make a steel set from scratch, hewn from a billet of Baron von Krupp's personal stock. They'd be easy enough to model on CAD. Anybody out there with a tiny CNC mill?

 

[Ian_C]

After much distraction and diversion the chassis parts got dropped onto the jig and soldered up. After all the planning and preparation it all went together without fuss and took about 15 mins to complete. I was pleasantly surprised to find that the chassis axle centres matched the jig exactly. Design, artwork and etching spot on again- congratulations MOK!

One other advantage of using a hefty slab of aluminium for the jig...you can turn it on its side to complete the soldering from above and below.

I must say, the tab-slot-twist construction on the chassis works very well, although filing off the projecting tabs after soldering is somewhat of a chore.

Finally for this posting, a mystery part...

It gets a mention as a part to remove from the etch when building the ashpan / frame x-member sub-assy. It doesn't feature in any diagrams and never gets mentioned again elsewhere in the 'structions. It looks to me like the front lower firebox plate riveted to the foundation ring but I'm damned if I can figure out how it fits in that location. It's not off the specific S7 etch so it may be a bit narrow, but still, I'm stumped. Suggestions?

Anyway I'm not sure you'd actually see the firebox rivets/stays on the prototype because they were covered with a layer of 'plastic magnesia' below the running plate for insulation.

 

[Ian_C]

Finally some cylinders.


It was a bit of a hard work getting the slide bars aligned properly for soldering. In the end they were soldered on close to aligned and the slide bars slightly bent afterwards to get them perfectly parallel . Likewise the valve spindle guide. It probably wasn't the best plan to attempt this late on a hot and airless Saturday night. Sweaty and stressy with patience carefully rationed. Having learned the lessons, the opposite side was done in about 30 mins this morning without stress or sweat.

Some things to note...

As cast the slidebars looked a bit rough and they really need to be straight, clean edged and shiny in 7mm. I briefly considered making some replacements from a scrap of 3mm N/S plate. In the end they cleaned and polished up OK. The separate steel rod for the piston rod is a nice touch and that also will get polished before final assembly. All this heavy metal in the kit looks prototypically chunky.

Noting some advice elsewhere on this kit I took the time to get the relative positions of slide bars and valve spindle guide correct so that the combination lever passes the front of the crosshead with (hopefully) some clearance. We'll see shortly.

The holes in the rear castings for valve spindle and piston rod are undersized as cast (which is OK) and not well aligned with the axes of the cylinder and valve (which is less OK but understandable in a casting). They are also quite long so they naturally align the spindle and rod in a direction slightly different from the slide bars. It's difficult then (OK it's impossible) to get the piston rod and crosshead sliding freely over the full stroke. The simple remedy is to drill out the hole inside the casting oversize (I chose 3mm) to within a couple of millimetres of the gland. That way only the short section of hole at the gland provides any guidance for the rod and a slight clearance here provides enough accommodation to allow the crosshead and rod to be guided by the slidebars without the rod binding up. All of these little snags need clearing up in order to get a the valve gear working smoothly.

All but the early locomotives had two rectangular access covers on the top shoulders of the cylinder cladding. There's no provision for this in the kit so far as I can see so they were made from tiny rectangles of brass shim with rivets popped in the corners. I found it very difficult to make four identical covers this size so I made about twice as many as needed and chose the best matching pairs (and fed a couple to the carpet, naturally). Looking at photos (Wild Swan LMS Locomotive Profile mostly) there seems to have been some variation in size and shape, with some of them in later years being downright scrappy. So maybe you can get away with a slight lack of uniformity. The circular cover came in two different sizes; smaller for early locos and larger for later. I'm not sure which is represented by the etched detail on the kit but I elected to leave it as is.

The valve guide castings benefit from having the backs skimmed in the lathe to help them sit flat and square on the cylinder ends.

There will be some detailing of lubrication pipes using fine copper wire and there are still the the cylinder drain castings to add. They'll go on when all the mechanical assembly and fettling is done.

Motion brackets and associated clutter are the next job.

 

[Ian_C]

Motion brackets were a bit of faff on account of the instructions not being specific about the location of all the parts. In particular the lateral location of the reverser lift arms and crank had to be worked out from the drawings in the Wild Swan book and measurements taken from the model. The model does not correspond exactly to the prototype in this area. The differences are small but enough to make scaling directly off prototype drawings invalid. Still, a pleasant morning at the kitchen table in the sun measuring etched parts and sketching out the arrangement helped to clarify things.

I'd also add that the improved instructions for this kit by DavidinAus referred to in his thread Stanier 8F in S7 were a considerable help in working out which other parts to take off the etch for this assembly.

I think the intent of the kit is to thread the reversing shaft supports, arms and crank onto 0.8mm wire and solder them in position. Sure, you could do that and it would be relatively straightforward, but I don't think it would capture the look of the prototype well. You could argue that the area under the motion bracket is in the shadow of the footplate and difficult to see on the model but I like to photograph models from around scale eye height so it'll show up there a bit. There are a few areas where the kit can be improved if you want to spend the time on it:

1. The reverser gear shaft was quite a hefty size in real life and it scales out to about 2mm diameter, noticeably bigger than the default 0.8mm wire. So new shafts were made for the LH and RH assemblies.
2. The reverser shafts were supported off the back of the motion bracket casting by substantial castings of their own. The etched plates in the kit don't represent this very well; looks too plain and empty up there. Some 'looks like' castings were fabricated from odds and ends and added to the outside of the support plates.
3. The prototype lifting arms were machined forgings or castings and are not well represented by the pair of etched arms on the wire shaft. The etched arms were soldered to a spacer to make a more substantial sub-assembly.

I found it useful to model up the main parts of the motion bracket assemblies in CAD to understand positions and clearances and design the new reverser shaft parts. If you're puzzling your way through this job then these pics should help.

The screen shot above shows the LH motion bracket assembly upside down along with a section of the adjacent chassis side plate. The etched back plate is not shown for clarity. The reverser arm should sit between the two motion bracket etches (with all the oval holes) and be equally spaced from the inside of each. Note that the four arm etches have different numbers on the fret, but they are all the same. The crank arm (dark green) is made by laminating the two etched arms together (it is the reach rod that has the forked end on the prototype), and needs to be positioned close to the fixing angle bracket (nasty violet colour) and between the two support plates (orange). It will be impossible to see the connection between the reversing reach rod and the crank so it doesn't matter if they don't eventually match up exactly. It does pay to get the crank at roughly the right angle depending how your gear is to be set. I pondered how to set the gear and decided that I'd set it in neutral since most of the photos I'll take will be of the loco at a standstill, and the driver should leave it in mid gear when parked. The downside is that when in motion the valve gear won't move the valve spindle much. Well, you can't have it all! Actually you can have it all because somebody, somewhere on WT has worked out how to operate the gear with a tiny servo and DCC. A bit beyond me though .

The screen shot above shows the RH gear. Upside down again and showing the rear plate (brown) this time. No crank arm to complicate matters on this side.

The screen shot above shows the RH gear the right way up this time. The cosmetic support casting is shown here between the arm and the outer support plate. I modelled it up to look like the casting you can see in some photos with a view to having some 3D printed, but in the end I wanted to press on and made a pair from a turned boss and bits of scrap etch. You can see them in a later phot0, They fill a space and I think they'll pass muster in the shadowy, filthy gloom beneath the footplate.

The lifting arms were drilled through 2mm and soldered to a small turned spacer. Plenty of solder and some work with the thin end of a round file gets them looking more like a casting.


Rather than try and juggle the position of all the parts along the reverser shafts when soldering up the assembly I elected to make stepped shafts and different sized holes in some of the parts. The shafts take a little longer to make than a plain 2mm diameter pin, but the assembly then is much easier and everything ends up in the right place and square to the shaft. Having modelled up the various bits & bobs in CAD it was easy to design the shafts. If anybody's daft enough to follow suit the drawings are below.

Finally all is assembled to the chassis, alignment checked, and soldered up. A couple of minutes in the ultrasonic tank and you end up with this...



Note that the arms are not yet fixed to the shafts. That'll be done when I assemble the motion and align them with the radius rods. The reddish tint that is sometimes apparent on the cleaned nickel silver parts is due to a bit of electrochemistry going on in the hot ultrasonic bath. The hot solution and the agitation leaches some nickel or zinc from the surface leaving a coppery sheen.

Not sure what comes next. I'm waiting for Mr MOK to send me a replacement drop link casting that I'm missing, so I can't complete the motion just yet. Maybe I'll browse Stanier 8F in S7 for inspiration...

 

[Ian_C]

A couple of firsts this week.

The first first: in order to find some missing late-summer modelling mojo I popped the chassis on wheels, dropped the motor /gearbox in and ran it under power for the first time.

There you go - a Stanier 2-2-2-2, or a may be better described as a 1-1-1-A. No pick ups of course, it's wired directly to the controller. No rods or motion fitted so it really isn't much of a step forward. But it's alive for the first time, I get a sense of the thing and it's the first 7mm loco I've (not yet) built. Followed by 20 minutes of playing trains, just driving it up and down a few inches of test track. Actually took a few seconds of video of it, but not compelling viewing. So far so good.

Interestingly the ineradicable wobble of the wheels (earlier in the thread) isn't particularly noticeable when installed, and I guess it'll be even less noticeable when the rods and motion are on. Still some engineering issues to address with the wheels though, so still not decided what I'm going to do with them. There's a telescopic axle experiment brewing...

The second first: I actually wore out a piercing saw blade this week. That's the first time I've managed to do that. It's usually 'ping' long before blunt.

 

[Ian_C]

The union links have been a pain and a subject much fretting. There are etched links in the kit that are to be soldered together and the ends opened up and bent to make the forked ends. I struggled with this. Couldn't make them consistent and couldn't get all the holes aligned and at the correct centres (8.8mm on most of the 8Fs). Some ineptitude on my part no doubt, and I know that others have managed it. The one thing I couldn't get my head around was the length of the link. I think the etched parts make a longer link than 8.8mm however you go about it.

Thought about other ways of making the things. What we want to end up with is this...

...and it's a very difficult item to make from scratch. The real thing scales down to quite a delicate little thing in 7mm. Because it has to match up with other kit parts that are slightly over size (limitations of sheet thickness etc) it ends up a little chunkier than true scale. I considered machining them from a lump of N/S, but couldn't work out a good way of doing that. I had a look at the Wild Swan Geoff Holt book. No real guidance there, but a close look at a photo of one of his Stanier locos suggested that he fabricated them from ickle bits. Lightbulb moment (fuzzy low wattage light bulb) and a way forward. What we're actually doing is this...

A bit of a fiddle and the important thing is to get all the holes aligned and at correct centres. The ends were cut from the parts supplied in the kit (after I'd straightened out my abortive attempts to bend them to shape). The sticks in the middle were cut from scrap etch.

Small soldering jig made by drilling 2 holes in a scrap of tufnol and using two 0.9mm drills as location pins. A good way of preventing the link parts from ending up soldered to the drill bits is to powder some pencil lead on wet & dry and rub that on the drill bits. Graphite makes a very good solder mask, better than a film of oil I find.

Thread the first layer over the pins and line them up with the tip of a scalpel. Solder the two centre parts together before placing them on the link ends...


Starts to get tricky now, and a couple of blobs of blu-tack hold the ends in the correct orientation while scalpel and soldering iron are applied...

Build up plenty of solder because it will be used to sculpt the right shape later. Then add the other ends and solder them up likewise.

And you end up with this nasty little nugget of N/S and solder...

Which is then cleaned up and sculpted to the right shape by needle file, cocktail stick and wet & dry and finally a rub with a Garryflex block to polish it up.

And the end result is a pair of passable (but not perfect) links that just need the fork ends cleaning out a little to fit the mating motion parts.


A note on soldering. I prefer to use a paste flux for this sort of thing for a number of reasons:

• It 'sticks' the tiny parts together and holds them steady while you nudge them into place with a scalpel.
• It doesn't boil and spit like fluid flux when the heat is applied, so it doesn't 'rearrange' your carefully assembled parts
• If you have to linger with the iron it persists and and continues working after a fluid flux has boiled dry.

I use Fluxite from a small tin I inherited from my dad (and it'll probably last longer then me). It has the same active ingredient as the dreaded Baker's Fluid, zinc chloride. It's not a problem if you clean it up properly afterwards.

 

[Ian_C]

Like learning to swim. Having read all the books and watched the instructional videos there comes a point where you just have to jump in and wave your arms about. So it is with going off piste with wheels and axles.

Some notes from my first attempt at telescopic axles for Slaters wheels in 7mm. There are WT folk better qualified then me to write a ‘how to’ on this subject , but since I’ve never seen it written up, and since it has proved to be a viable approach, I thought I’d share it.

As previously related I wasn’t delighted with the Slaters wheels for this project on account of the run out of the driving wheels and the appearance of the standard Slaters axle end. Following encouragement from eastsidepilot that it was possible to re-axle Slaters wheels I thought I’d have a go. I chose to experiment with the pony truck wheels because they don’t need quartering and they’d be the least costly to replace if it all went horribly wrong.

Some design considerations first…

1. the diameter of the axle should reflect appearance of the prototype as it projects from the wheel boss
2. the inner section of the telescopic axle has to be large enough to accept a taper pin
3. the wall thickness of the outer telescopic section cannot be too thin
4. the diameters of the various holes have to correspond with available drills, reamers etc
5. we must be able to produce the bearing diameter to match the axle

Starting with point 1, the prototype axle end scales to very nearly 3.8mm in diameter. A bit of rough design on CAD suggested that would be viable with a 2.8mm diameter inner axle. The overall length of the axle assembly was made slightly shorter than theoretically required to provide a bit of length adjustment when setting the back to back measurement. The assembly method is to set the wheels on the axle ends first, getting the axle end slightly proud of the wheel boss per prototype, then set the back to back dimension and then drill and pin the axle sections. There’s also a small flat on the outer axle where the taper pin will be. That reduces the tendency of the small drill bit for the pin to wander off when starting the hole. I’ll also use it as the means of identifying the thick end of the taper pin for assembly and removal. There are axle part drawings attached at the bottom for those interested.



The axles were turned from bright mild steel rod (because silver steel is a pain to turn in small diameters and we don't need the properties of sliver steel). The flat on the outer axle section was done on the milling machine in a tiny precision machine vice held in a much larger machine vice.



The next job was to work out how to bore out the Slaters axle insert to 3.8mm diameter and get the bore concentric with the tyre and perpendicular to it. A tool was made on the lathe with two recesses that are a gentle push fit for the wheel flanges (the other, larger, recess is for the driving wheels - ever the optimist!). Holes were tapped for clamping screws and a hole was made in the centre to provide clearance for drills, boring bars and to give the swarf another way out. The material was a slice of 50mm aluminium extruded bar left over from another project.

The 3 jaw chuck won’t centre the work perfectly but the turned recesses will be centred perfectly and the tool will stay in the chuck until all the wheels are bored through. If you can’t use the chuck for anything else for a while then, depending on how you are equipped, that might dictate a certain order of work. For example, if you had intended to turn the axles in the chuck then you’d need to do that first. In fact it is a good idea to turn the axles before you bore out the wheels so that you can use the axles to gauge the bore accurately to a gentle press fit as you open them out. As it happens I have a collet chuck that fits in the spindle so I used that to turn the axles. The position of the 3 jaw chuck is marked on the spindle nose so that it always goes back on the same way. The concentricity errors arising from exchanging the chuck and collet holder this way will be acceptably tiny.




With a wheel mounted in the tool and the tyre running true the first job is to reduce the projection of the wheel boss and insert to the prototype dimension. I’m guessing that the Slaters wheel boss projects from the face of the wheel as far as it does simply so that they can get all the axle end features into it (countersink, thread, square hole with sufficient axle engagement etc). As it stands the boss projects too far and as well as not looking quite right this also causes clearance problems in S7 on the driving wheels behind the slidebars (thanks to Davidinaus for the heads up on that one!). A tiny, pointy tool with a modest depth of cut (low cutting forces) was used to face the boss down to within 0.25mm of the face of the tyre. Easy enough.

The boring out of the insert requires some care. On the pony truck wheels the insert is brass so it won’t put up much of a fight but the features in the insert may not be concentric with the tyre or perpendicular to it (which is kind of the point of the exercise !), and some of the hole through the insert is square. I’ve no idea what the insert looks like or how it is engaged with the moulded plastic (but I’ll find out when I ruin one!) and all the machining forces have to be transmitted by the plastic spokes. Simply poking a drill through the middle might not be the best plan as it would have a tendency to centre up and follow the existing countersink, and a 2 flute drill opening out a square hole could be a bit rough. I started by gently running a sharp 3mm slot drill through the insert. Since it will cut on the end and the flutes it will find its own way through and not be particularly influenced by the existing geometry. That left a round hole through and was followed up with a 3.5mm stub drill.




A note on drills here. Most of the holes we drill are not very deep and a typical jobber type drill bit is much longer than necessary, and as a result quite flexible, which doesn’t help when trying to drill holes accurately. Often it doesn’t matter, but sometimes it does and that’s when stub drills are useful. They are essentially shorter versions of the standard drill bit and consequently less flexible and more inclined to drill in the direction you point them. There’s one other difference that’s not so obvious - quality. A lot of the drill bits available through the modelling or hobby engineering trade are very ‘reasonably priced’. They can also be of variable quality in terms of actual diameter and cutting geometry. You’ll probably have to buy stub drills from a more industrially oriented supplier. They cost more but are of much higher quality, they’ll have a more sophisticated cutting geometry and as a result they’ll cut more cleanly and accurately. I’ve accumulated a few HSS or solid carbide stub drills and I use them when I need to drill holes accurately. Recommended.

So, there’s now a 3.5mm diameter hole through the insert and there’s no turning back. I don’t have a 3.8mm drill or reamer, but I do have a tiny carbide boring bar for 3mm+ bores. It’s an amazing little thing and I came across it on eBay where Kyocera USA dispose of their surplus machine tool stock. There are even smaller ones for micro machining work. Cost a few dollars and a few dollars for shipment. Now, at last, I’ve found a use for it. Carbide tools are very hard and resistant to wear, but they’re not as tough as steel tools. I imagine that a sudden load on this tool would ‘plink’! It took some care to set the tool correctly in the 3.5mm bore and with held breath it was given a small cut and advanced through the hole. Very good result and no ‘plink’, so breathe out again. I’ve no way of accurately measuring bores this size so work proceeds by gradually opening out the bore until the axle (you’ve already made one, right?) just slides in with a bit of a push. If you make the hole too big you’re scuppered, so tiny cuts and repeat passes with no cut to clear out the tool deflection. It helps to put a tiny chamfer on the end of the axle. You can tell when the bore is close to size when the chamfer starts to nearly enter the hole. That worked out OK in the end and the second wheel was done much quicker.



The axles and wheels were thoroughly degreased and the wheels were fitted to the axle ends using Loctite 601. It is easy enough to adjust the axle projection on the face of the boss by eyeball before the Loctite gets a grip. Putting the wheel and axle halves back in a collet in the lathe showed that they were running true and concentric. Well, as true and concentric as the collet which is the pretty good and good enough to be undetectable on the model I think. Here’s the difference between the standard Slaters appearance and the modified wheel and axle. Much closer to prototype appearance I think. Very happy with that so far.


(Looking at the photo as I post this I think the prototype axle centre drilling is bigger than I have it . Easy enough to pop them in the lathe and make it a bit deeper. )

The next challenge is to set the back to back correctly and pin the two axle halves together. I don’t fancy my chances of juggling a back to back gauge (assuming I ever get to purchase one - the S7 Group stores seem to be mighty uncommunicative these days) and a small drill. Since I’ll have more of this to do (the 8F driving wheels, and you never know, I might even finish the 8F and start another S7 loco, if I live long enough) a bit of tool making seems to be required. A picture’s worth a thousand words (or possibly more in this thread) so here’s the tool I came up with.



It is essentially two back to back gauge faces with a V groove between them to hold the axle parts, some tapped holes for axle clamping strips, a hole through the middle where the taper pin drilling and reaming will take place and a couple of rebates on the bottom face to locate it in a machine vice. (The two tapped holes in the bottom are for the next part of the plan. The quartering jig). Easy enough to make if you have a milling machine. Impossible to make if you don’t. You need to take care to have the gauge faces parallel and within S7 B2B tolerance, and the groove has to be perpendicular to the gauge faces. Any geometrical errors in those respects and any little inaccuracies in assembly of the axle will tend to increase the actual B2B dimension so I chose to make to gauge distance toward the lower end of the tolerance at 31.22mm (well, that’s just where it ended up after the final cut so I left it at that). The material is aluminium, a chip off the same block that was used to make the chassis alignment jig earlier in the thread. The drawing for the tool is shown in the following post.



With the wheels and axle parts clamped in the tool (yes , I did line up the spokes by eye even though you’ll never notice on the model - OCD) and the tool clamped in the machine vice the hole for the taper pin can be drilled and reamed. Taper pins and taper reamers is new territory for me so thanks to those folk on WT who answered my questions on the subject. I ended up using imperial 1/16” taper pins with a matching 1:48 taper reamer from Tracy Tools. A point of note is that a 1/16” pin doesn’t need a 1/16” hole. The thin end of the pin measured about 1.2mm so the hole was drilled 1.2mm diameter (very, very carefully using a light touch on the fine feed) and then opened out with the reamer until the end just projected through the axle by about 1mm. The pin was gently pressed home before removing the assembled wheel set from the tool and cutting off the surplus pin. There’s about 0.5mm projecting from either side of the axle.





It took a couple of weeks of odd hours but in the end it wasn’t particularly difficult. I’d always felt there was an unwritten commandment ‘thou shalt not mess with wheels and axles because it’s a bit like engineering and you’ll probably make an expensive mistake’.

If anybody wishes to have a go the drawings for axle parts and the setting tool are attached to the entry after this (can't upload any more pics on this post).

With that out of the way I can press on and complete the pony truck.

Driving wheels next, and the challenge of quartering.

 

[Ian_C]

Here are the drawings for the various bits and bobs in the previous post

 

[Ian_C]

With the wheels and axle sorted out I've been able to finish the pony truck. The etched parts went together very nicely. I changed the way the axle bearings fit. The bearings supplied are meant to fit inside the frame and the spring castings are to be located on the projecting part of the bearing. A few issues with this -

• It'll be too narrow for S7 anyway, and this is one part not covered in the S7 conversion etch.
• The holes in the spring castings are much bigger than the turned bearing, so not a good locating feature
• The backs of the wheels will rub on the spring castings, with bags of lateral free play in S7
• You have to file off some of the bearing flanges to fit them inside the frame
• I chose to make my own axle and it's not the same diameter as the supplied bearings

I decided to make some replacement bearings that are spigoted fit to the outside of the frame, are the correct diameter to locate the spring castings, match the axle diameter and project far enough to bear on the back of the wheel bosses and control side play. The springs end up a bit further from the inside of the wheels than on the prototype but I can live with that in the murk below the footplate.



With the spring castings fitted...

...wheels fitted and complete...

 

[Ian_C]

Here's the quartering tool. It follows the well known principle of having two surfaces set relative to the axle centre that contact the crankpins such that they end up 90 degrees apart. The wheel gauging and axle pinning block shown in a previous posting is used to hold the axle. The block can be used independently, as previously described, or installed in this quartering tool.


The quartering tool is designed around a 3/16" diameter axle and 3.0mm crankpins. Of course, once they're set 90 degrees apart you can use any diameter crankpin you like; they'll still be 90 degrees apart. I've made a pair of dummy crankpins specifically for this tool and they're stored in a couple of 10BA holes in the base plate when not in use. The horizontal reference block extends far enough each side of axle centre line to allow the crankpin to be set forwards or backwards for RH or LH lead.


If you use an axle diameter other than 3/16" then the axle centre height in the V groove changes, and that in turn alters the angular relationship of the crankpins dictated by the horizontal and vertical reference faces. Not by much, and theoretically you can get away with a couple of degrees either side as long as all axles are the same. The vertical reference block is fixed through slotted holes in the base plate so it can be adjusted by 1.5mm in either direction to restore the 90 degree relationship for different axle diameters or different crankpin diameters if you wish. The degree of movement and the resulting position on the base plate can quickly be plotted in CAD for any combination.



The accumulated errors from making several parts on a milling machine that isn't to toolroom standard mean that I won't get exactly 90 degrees. But it'll be very close to 90 degrees and, more importantly, all axles will be the same. Once I've done all four axles and checked running with the coupling rods on then I'll find out whether all the theory is borne out in practice!

If anybody wants to see the drawings then let me know and I'll post them.

I'm getting bit ahead of things here because I haven't posted the approach to modifying the Slaters driving wheels and making new axles. I modified a pair, seen here unfinished, as a trial. It worked out mostly as planned. I'll do the rest of them when I get the time and post the learnings.

 

[Ian_C]

The confident assertion I made in an earlier post about being able to remove and replace the lathe chuck and wheel holder without introducing any out of true error turned out to be slightly wrong. For a reason I don’t understand the wheel holding tool did not run quite true when the chuck was replaced on the spindle. The wheel holding tool had not been removed from the chuck meantime. It was only a little bit out of true but it should have been near perfect. Not a big problem because it was easy to reface the part of the tool where the wheel sits.

The issue where, in S7, the Slaters wheel and crankpin boss projects far enough from the face of the wheel to cause clearance problems behind the slide bars was flagged up in a DavidinAus post some time ago. David opted to take 0.4mm off the face and it seemed to work for him. Taking a close look at the prototype drawings in the Wild Swan book revealed that…

• the driving wheels measured 6-13/16” overall from front of axle and crankpin boss to inside face of rear boss
• the rear boss on the prototype wheels measured 15/16” from rear face of tyres to inside face of boss (there's no rear boss on the Slaters wheels, we'd use a washer or two on the model)
• the tyre was 5-1/2” in width

…and converting these dimensions to 7mm and doing the arithmetic I conclude that the model wheels ought to measure 3.42mm from inside face of tyre to outside face of axle and crankpin boss. The Slaters wheels measure 4.23mm. So there’s the missing clearance of 0.81mm each side. The prototype ‘as drawn’ clearance between inside of slide bar and leading coupling rod pin was a mere 11/16” , or 0.4mm in 7mm world. If the MOK kit has the slide bars the correct position (and it should have because there are parts on the S7 conversion etch specifically for this) then half a revolution and CLANG is the guaranteed outcome.

As it turns out you can’t sensibly face the axle and crankpin boss down by 0.81mm on the Slaters wheels. The inside ends of the spokes flare outwards to meet the moulded boss on the Slaters wheels to a greater extent than on the prototype. I suppose it makes the over thick Slaters wheel centres look in proportion. I’d never have noticed if I hadn’t checked, and in 0F it may not matter. It is possible to take 0.55mm off the thickness before you start to machine into the end of the spokes and the face of the boss starts to look like a starfish. If DavidinAus got by with 0.4mm and I can manage 0.55mm then I figure I’m in with a chance provided that I have appropriate coupling rod and pin projection on the leading axle.

Mount a wheel in the holder and check that the tyre runs true. On some of the wheels you could see by eye that the axle insert was a little off true when the lathe was spinning. There's the root cause of the 'Slaters shimmy', as I've heard it called.



The first part of the job is to face off the crankpin and axle boss by 0.55mm using a sharp, tiny, pointy tool. This is the same tool described in the pony truck wheel thread somewhere above. The small point combined with a small depth of cut (0.15 - 0.15 - 0.15 - 0.1 as it happens) gives rise to small cutting forces which can be handled by the plastic spokes and which won’t upset the steel crankpin and axle inserts. A collateral benefit of this is that the moulding sink marks in the original surface of the moulding are removed and we have a nice flat boss like the prototype.



Next operation is to clean out the countersink and square axle hole with a sharp 4mm slot drill.


The 4mm bore is opened out to about 5.0mm with the tiny boring bar. The rationale here is that a boring bar will cut true to centre where a 5.0mm drill might wander a bit depending on how accurately it’s been sharpened.


The last operation differs slightly from the conversion of the pony truck wheels previously described. Scaling from the drawing, the prototype driving axles are about 9” diameter where they project from the driving wheel boss. That’s 5.25mm in 7mm scale, or 5.2mm for convenience. So the outer end of the axles must be 5.2mm in diameter and the bore in the wheel insert must be a gentle push fit onto the axle. For the pony truck wheels I’d made the axles to the required diameter and bored out the wheels to fit. That wasn’t straightforward as I couldn’t measure the bore accurately and had to take tiny cuts with the boring bar and keep trying the axle fit. I did it the other way round on the driving wheels. The wheel inserts are all reamed out to 5.2mm (I have a 5.2mm machine reamer) and the axle will be machined to fit the wheel. It is easier to measure the axle diameter accurately than the wheel bores and we'll get to the finished size faster and with more confidence, and since there are 8 axle ends to do so that’s a worthwhile time saving. The 5.2mm reamer works nicely through a 5.0mm bore.


With a bit of organisation it took me about 15 mins to do a wheel, so in about 2 hours one evening all the wheels were done.

Before and after...


..and all done...


Suppose I'd better make the axles next. This also makes me think about balance weights, which in turn means I need to decide which locomotive I'm actually modelling, which in turn might have consquences further downstream. I fancy a late inhabitant of the mighty 18A Toton, mid sixties, last knockings of steam on the Midland main line and I was a kid in short trousers growing up nearby. My brother reckons he remembers seeing 8Fs on the low level approach through the old Long Eaton station behind the Co-op car park. Must have been' 64 or '65 or at a push '66.

 

[Ian_C]

After a bit of research I found that the choice of late Toton 8Fs wasn't that big. I also discovered that Toton stopped being 18A and changed to 16A in September '63. Good job I haven't acquired an 18A shed plate already! By the end of 1965 it was almost all over for steam at Toton, and those that were resident in '63 and '64 were mostly transfers passing through after a few months to make a last stand elsewhere, or withdrawn to the scrap line. Check out brdatabase.info for this kind of detective work, very interesting. I eventually settled on 48142 as the most promising subject. Having decided that, the next challenge was to find some decent photos of 48142 in the 60's and work out what state it was in. I figured the chances of finding a good photo from the right period in a class this large would be a bit needle and haystack. As it turns out there's an excellent photo of 48142 in the Wild Swan book (and it can be found online too), and, serendipity being what it is, the photo was taken in 1963 at Uttoxeter station, which is now my local station and about 3 miles from where I sit typing this. The loco is in very clean condition and looks like it was recently ex-works. That being the case, I'm assuming that this was its final condition and was unchanged between 1963 (when it was based at 16D) and it's stay at Toton from August to September 1965. The Appendix B in the Wild Swan book 'Variations as built' was a very good starting point for the detail picking. There were a lot of variations that won't show up on a model, such as bronze coating on the regulator rod, but for a class you tend to think of as uniform there are a lot of visible and modellable differences, and since I'm doing a portrait model I'll have to swallow them whole. This is what I've learned so far...

• 8142 - Built at Crewe - Lot 154 - Order E424 - February 1942. Withdrawn in November 1966.
• Was at Toton (16A by then) from August 1965 to December 1965.
• Original tender 9869, should be Mk2 welded type, and still was a Mk2 welded in 1963.
• AWS was fitted in June 1961.
• Short connecting rods, longer slide bars, union link and piston rod (8126 onwards) - so I got the union link length correct then!
• Was one of the few fitted with ball bearing and cover on the eccentric rod big end (and still had it in 1963). Goddam - that needs changing.
• It was built with flush rivets on the front buffer beam, but by 1963 it had snap head rivets.
• Cab commode handrail fixed to the wing plate top and bottom with pillars.
• It was built with hollow coupled axles, but the 1963 photo appears (can't be 100% as most of the axle ends are covered by motion in the photo, apart from a glimpse of the rear axle) to show solid axles on coupled and pony truck. They did mix them up somewhat, but most of those built with hollow coupled axles seemed to keep them into BR days. 48142 appears to be an exception.
• Atomiser steam cock in the higher position on the smoke box.
• Double row of rivets on side of smoke box indicating hinged smoke box cross bar.
• Rivets on lower arc of smoke box front ring indicating liner plates fitted.
• Single cone ejector, with single pipe from cab front. And the single cone ejector body differs from the more common double cone type. Must check the casting in the kit - bet it's a double cone!
• Footsteps attached with flush rivets.
• Rectangular access panels in upper shoulder of cylinder clothing. Got that right pre-emptively as well!
• Larger circular access covers in side of cylinder clothing - I already guessed wrong on that one but I'm not inclined to change it and I'll bet that nobody notices.
• Curved reversing reach rod and tall steady bracket (just missed the straight ones that cut in with 8146 and all the NBL built ones as well).
• Built with built up balance weights, but appears to have all wheels with cast balance weights by 1963 - which may be related to the solid axles noted above. The cast weights covered a different spoke pattern and they have a noticeably different edge profile at the tyre, much sharper. That'll take some working out. Again, they did mix them up in BR days and some locos had a mixture of built up and cast balance weights (and solid and hollow axles too).
• 40% balance presumably, and no star on cab side in 1963.
• Sloping throat plate, domed boiler.
• Top feed pipes semi recessed into boiler clothing.
• Was built with higher mounted tube cleaner cock but by 1963 had the lower mounted tube cleaner cock with external feed pipe from higher up smoke box.
• Access plate in smoke box saddle front.
• By 1963 it had the later pattern top feed cover with ‘top hat’.
• Steam sanding.
• Straight fluted combination lever (and union link forked both ends, mercifully)
• Cross head drop links with 3 stud fixing.
• Pony truck oil boxes on outside of front frames.
• Extra stops and guides on cab gangway doors (not sure what that entails - more photo picking required).
• Stiffening brackets on cab wing plates.
• Separate blower and sanding valves in cab.
• Access doors in platform for steam chest covers (no idea what this means - more research).
• Split cast iron valve spindle crosshead guides.
• Altered brake hanger top bracket pins (but I haven't spotted the difference yet)
• One of the batch fitted with hopper ash pan and drop grate. Operating spindle behind LH trailing wheel. There are some useful photos of other locos with this detail.
• Cab window beading half round section with curved lower window corners.
• Smokebox door locking handle probably plain when built but had a collared end by 1963. Who knew?

That's what I think I've established so far. A surprising (to me at least) number of visible differences to model. And I haven't really taken a close look at the tender yet - I know there were some differences.

Sorry - no photos this time. Normal service will be resumed...when I get my act together and make something. Balance weights probably.

 

[Ian_C]

Following the discovery that 48142 was fitted with ball bearing eccentric rod big ends with the distinctive brass cover I set about modifying the previously made eccentric rods. The Wild Swan 8F book does not have a drawing of the ball bearing rod end but there is a drawing in the Wild Swan LMS Locomotive Profiles N0.6 - The Mixed Traffic Class 5's (page 38). Some Black Fives were so fitted and I'm assuming that the LMS had the wits to use the same details on the 8F. Scaling from the drawing showed that the ball bearing ends were quite large compared with the the more common pin joint, about 9 inches in diameter. Surprising. I had to check my numbers again and stare at a few photos before deciding that was correct. Here's the 'story board', commentary below.

1. The original rod and a couple of lengths of 1/4" brass rod.
2. Some scraps of N/S from the etch fret the same thickness as the rod laminates, flattened and soldered to the ends of the brass rods.
3. Brass rods with N/S ends turned down to the diameter of the rod ends and bearing covers, 5.3mm.
4. Hole drilled through N/S disc and no deeper.
5. N/S discs unsoldered from end of brass rods.
6. The existing big end has the oil box filed off and half the thickness filed away where the first disc will sit.
7. N/S disc soldered to rod located by a drill shank into a piece of Tufnol.
8. Second disc sawn and filed to fit around remaining big end on other face of rod. Lined up and soldered. Plenty of solder added to file back and form the fillets.
9. Cleaned up. Rod complete and looking like a lollipop. Looks a bit goofy, but they really were that size!
10. Brass rods back in the lathe and faced off to remove solder and drill dimple. Set up vertically in a vice in the mill, centred, and three 0.5mm holes drilled down the rod a little deeper than the thickness of the bearing end cap. Back in the lathe to turn the end cap profile and parted off.
11. Wire soldered into the three holes and filed to length to represent the nuts holding the brass cover over the bearing. Sweated into place on the end of the rod.

What I neglected to photograph was the short length of 12 B.A. thread projecting from the back of the big end. This was cut off a brass screw and soldered into the big end hole before the cap was sweated on. It will pass through the hole in the crank and be secured with a 12 B.A. nut behind the crank where it is not visible. Well that's the theory. Assembling the motion will require the patience of a saint...

One feature I didn't reproduce (before anybody points it out!) is the little bump on the cover where the oiling/greasing point was. Couldn't see an easy way to do this and I felt that I'd be better off with clean detail on a simplified part. It could be cast in brass I suppose, but I think the smooth machined surface captures the essential character of the prototype better than the surface texture you'll get on even the best casting. I am my own harshest critic and I'll have nightmares about the missing oiling point. I'll live through it, I'm sure there will be worse compromises to come...

 

[Ian_C]

Here's the detail of the assembly that I failed to cover in the previous post.

There's the 12BA stud in the back of the eccentric big end, a small spacer collar to allow the nut to be tightened without gripping the crank, and the crankpin. I think there's enough clearance for the nut but it could be filed down if required.

And here it is assembled on a wheel (crank not phased properly yet). Hmmm....that crank could be improved, those 4 studs really ought to be sitting in counterbored holes. Job for another day.

 

[Ian_C]

48142 was originally built with built up balance weights and hollow axles but by 1963 it appears to have had a wheel set with cast balance weights and solid axles.

The MOK kit has etched parts for both types of balance weight. If you're looking at the MOK instructions page 15 and wondering what's what, the upper diagram shows the cast and the lower the built up types. The parts on the etch for the cast weights have the 'B' suffix. The plan is to laminate two etched weights together and fix them to the wheels.

As I've come to expect there's a bit more to it than that. Firstly the etched weights don't actually match the Slaters wheels. They're a bit small on diameter and don't extend as close to the rim as on the prototype. Also the weights on the prototype are not all of the same thickness; the weights on axle 3 (the driving axle) are noticeably chunkier than the others and project further beyond the rim. On the wheels with integral cast balance weights you can see that the weights extend the full depth of the spokes, which isn't the case with the built up type where you usually just see the plate edges. New balance weights were made from styrene sheet.

1. Blanks were cut from plasticard. 30 thou (0.75mm) seemed about right for axles 1,2,4 and 40 thou (1.0mm) for the fatter weights on axle 3.
2. Damned if I was going to carefully file a load of round discs. Blanks were clamped on a length of M6 studding and mounted in the lathe.
3. A couple of minutes and a load of plastic fluff later they're all perfectly round and 28mm diameter which matches the Slaters wheels nicely .
4. All the blanks I need and load of spares besides.
5. The shape of the weights was scribed on the blanks using the etched weights as a template. You can see the diameter difference here between the etched weights and the plastic blanks.
6. The spokes covered by the weight had to have the flare of the spokes near the centre boss carefully removed with a sharp scalpel to allow the weights to sit flat on the wheel. The weights were attached using rapid epoxy generously applied to the back and between the spokes.
7. They start to get a bit of character when the weights are attached. Each axle is different.
8. The thickness of the cast weight between the spokes is added using epoxy putty (Milliput Silver Grey). The putty is squashed into the gaps between the spokes behind the weight until it bulges out beyond the edge of the weight, followed by lots of sculpting and fiddling with a wet cocktail stick to blend it with the spokes. The Milliput adheres very well to the wheels and sets hard enough to file and sand to a final shape. The better you can finish it with the cocktail stick the less cleaning up you have to do. The first wheel takes an eternity, but it gets quicker with practice.
9. Once the front side is tidy it's easier just to leave the back to set as it is. A bit messy but once the front is tidy it's better to leave alone.
10. When the putty had set rock hard the back was cleaned up down to the spokes on the mill with the rotary table (a splendid bargain some time ago and I've finally found a use for it). Overkill of course, it could be just as well done with a file and a scraper and some sanding.
11. A bit of tidying up with scalpel and wet & dry. Cleaned with IPA and a coat of grey primer applied to see how it all looks.

There's still the permanent fixing of wheels to axle ends, quartering, gauging and pinning to complete but the end is in sight. The wheels have been a bit of an epic but they run nice and true and they actually look like 8F wheels now, so I think it's been worth the effort.

And please don't think I'm knocking Slaters. I had to start from somewhere unless I was going to make them all from scratch (yes, it did cross my mind but that's a project for another day).

I have a load of circular blanks left over, enough for another set. If anybody wants them just message me.

 

[Ian_C]

Couldn't resist posting this...

The driving wheels have had the tyres chemically blackened (Birchwood Casey Gun Blue) and a coat of epoxy etch primer all over. Set them aside to harden off and got on with some actual kit building. The front chassis and footplate sub-assembly and the rear chassis and steps sub-assembly went together with no drama, and the MOK tab/slot/twist assembly aid really makes it easy.

Tackled the smokebox core and the boiler today too. Parts starting to come off the bodywork etches, feels like progress is being made. I don't have any rollers for the smokebox and boiler so other methods were tried. Rolling with a bar on a soft surface made little impression. The material is quite thick compared with what I'm used to in 4mm, and quite springy. Bars from 1" to 1/2" on a cutting mat made no difference at all. Bars on a thick and squishy 2011 plumbing supplies catalogue (now in recycling!) made hardly a curl. Bars on a stack of Scalefour News (heresy, probably) no good. 1/2" bar plus bodyweight on one copy of Scalefour News on top of a a kitchen towel made some impression (cut to thumb on corner of etch and plenty of blood on the kitchen towel - I'll hear more about this I'm sure). In the end I went out to the workshop, put a length of 1" bar in the big bench vice and wrapped the etched parts around that. Worked out OK in the end. The smokebox is a bit tricky on account of the big holes in the etch for the chimney and the steam pipes. The bend radius tends to go a bit wonky near the big holes. Those areas were evened out with a nylon mallet over the bar. Utter blacksmithery, but effective if done with a bit of care. Having soldered them up, I couldn't resist just arranging some of the bits and bobs into a loco shape. Yes, it's progress.

The basic shape of the firebox and boiler does have a whiff of Churchward about it don't you think? Stanier out of Churchward. 28xx and 8F, like mother and daughter. A S7 28xx might be a nice project, and I wonder if Finney 7 will re-introduce the 28xx kit? A late model 28xx (or 38-something I suppose) with a proper cab with side windows as opposed to Mr Churchward's engineman's shelter. Shame they only ever had that dinky Edwardian tender behind them though .

 

[Ian_C]

48142 had a liner plate fitted to the inside of the lower part of the front ring pressing and the rivet heads are visible on the outside of the smokebox. The front ring and the smokebox door are rendered in the MOK kit as a single brass casting. Quite a nice casting too, but it has no liner rivets. The positions of the rivets can be worked out/estimatedfrom the drawings on page 46 of the Wild Swan book. The chances of me marking the rivet pattern on the casting accurately by hand are about nuffink, so I sketched it all out on CAD and plotted the rivet positions relative to the centre of the door.


The rivets appear to be 1/2" diameter on the drawing. Googling rivet suppliers suggests that 1/2" rivets usually have a head diameter of 7/8", which looked about right on CAD. They're quite subtle on the prototype. You'd miss them if they weren't there but they don't stand out and I wanted them to be similar to the rivets already cast into the part. 0.5mm wire scales out about right for a 7/8" rivet head. The casting was set up on the milling machine with the door straps aligned with the x-axis and the centre found and set to x=0, y=0. The hole positions were cranked up on the DRO and drilled with the 0.5mm tip of a small centre drill. Which was OK, except that in my eagerness to get on with it I had the casting 180 degrees around and started drilling then pattern in the top half of the ring. Homer Simpson moment. Set it up the right way round and put all the holes in the right place. Short lengths of 0.5mm brass wire were soldered in the holes and then filed down to represent the rivet heads. The wrong holes were filled with solder and cleaned up (except one escaped - you can see it in the photo - I'll get it next time the soldering iron is hot) Residual solder was removed with a glass fibre brush.

Worked out OK. The outer rivets ended up a bit close the the radiused edge. I'd move them in a touch if I did it again. They'll hardly be visible amongst the ash, corrosion and flaky paint, but they're there and I can cross that job off the list.