East Barnet Joint Stock - a workbench

[Chas Levin]

9 hours ago, Daddyman said:

That's absolutely superlative, Chas. Your following of the line is very impressive, and there's nowt wrong with adjusting the line with thinners - the big boys do it too: https://www.westernthunder.co.uk/threads/painting-and-lining.2310/ (are you aware of this thread?) 

 

On the next loco you'll have a glassy-smooth gloss airbrush finish () and it will be even better - and easier to line!

Thanks David for your kind words and no, I wasn't aware of that thread or even of that forum: some evening reading there, I think!

And yes - The Future is Airbrush (sounds like a car ad slogan). The replacement tank arrived a while ago, but I haven't been able to tackle the swap-out because this stupid shoulder problem is still there; it is much better, but progress is annoying slow.

I've just been working this afternoon on the rear cab roof beading run, slow work but it's coming along well. Tidying up the outer edge of the white - where it meets the black - by nudging the black up to it is proving quite effective too: I'll post photos when there's a decent amount to show...

[Chas Levin]

The cab roof is now lined in black and white. I was able to edge the black up to the white so as to thin the white lines in places where needed, which was pleasing. The black surfaces do show the unevenness of the underlying green, but it's accentuated in these brightly lit blow-ups and looks much better in reality and at normal viewing distance. It's also brought out by the very reflective gloss, which will recede under some flattening satin varnish, nearer the later stages. I left the spectacle openings until last, thinking I'd find them much more difficult than the beading, but in the event they weren't too awkward to do - I'm not sure whether that was because I over-estimated the difficulty, or because I was well practised after doing the other parts!

 

 

 

 

 

 

 

 

 

 

Edit, April 8th 2022: while replacing photos after the RM Web outage, I should note that I subsequently realised I'd missed a line, one that runs along the top of the front curve that forms the lower edge, between the two spectacle openings - I added it in later on!

 

Edited April 8, 2022 by Chas Levin

[Chas Levin]

I decided the next area for attention should be the boiler bands, partly to get as much practice as possible with the bow pen before attempting the side panels, partly because the white band each side of the boiler bands will need time to dry and harden thoroughly before the other side is done, as the bow pen compass's guide wire will be moving in partial contact with that first band to do the second one, on the other side.

 

So far then, I've done approximately a thirtd of the boiler band lining - I say 'approximately', because the two end bands only have one white line, the two between the cab and the done are less than half the full boiler circumference and the one that runs 'underneath' the dome is a fraction of that, so calculating how much more there is to do becomes complicated... but the white lines are in place on the two end bands and one of them is done on each of the other bands:

 

 

 

 

 

I did them by using the opposite side of each band as a guiding shape for the bow pen compass guide wire, as recommended by Ian R in his book. I'd forgotten that he'd written about that method in the book and asked him how he'd done the boiler bands on one of his extrordinarily well painted locos featured on tony Wirght's thread a while back; he commented that you need a well-defined and smooth band edge to be able to use that technique and I thought at the time that my decision to stick with the brass bands on this loco - perhaps a somewhat questionable one in terms of prototypical dimensions and appearance - would have the distinct advantage that I could at least use the bands to guide the pen: every cloud...!

 

It does call for some careful manoeuvering: both pen tips and wire need to be in constand contact and at the right angles, such that the pen doesn't scratch the base paint and does deposit a good line, while the guide wire needs sufficient downwards and sideways force applied through the pen body for it to follow the line of the band edge, but not so much force that it slips out of the very shallow groove formed between the side of the band and the boiler surface, and you also need to allow for the odd obstruction:

 

 

 

Not to mention those places where multiple obstructions and intersecting structures prevent easy access - though in places like this, I found it easiest to persist with the bow pen for even very short sections of the line, so that I was then filling in between those sections with a brush and had the maximum amount done with the pen as a guide: you'd be surprised at the small spaces you can get the pen into with care:

 

 

 

I was using a new arrival, a compass bow pen by Kern:

 

 

 

This isn't quite as it arrived though (New Old Stock, less than £20): the larger knurled wheel  that tightens the blades had no markings other than the maker's name, so I swapped it for one of the ones from the Minerva set shown in some pics further back in this thread, as that wheel has numbers which make replicating a setting easier. The other side's wheel and threaded shaft for holding the guide wire are also swapped in from the Minerva set, as the one that comes with the Kern doesn't actually tighten against the wire (or compass point, if you're using that), it just pulls against it and I wanted the wire to be absolutely immobile. Also, the wire you see there is nickel-silver rather than brass: the angle against which pressure is applied to that wire when doing the boiler bands is partly inwards, so the wire isn't fully supported along it's length by the compass leg in the way it is if you're pressing outwards against it and it occasionally bent a little: nickel-silver is a little harder and seemed to work better, though that may have been more due to my increasing confidence with the banding. I picked this one up because I was still having diffculty persuading paint in the Minerva compass pen to flow easily and smoothly, in spite of honing and practising: this Kern one worked absolutely perfectly, first time, straight out of the box, which made me feel I;d done the right thing!

 

Leaving those band linings to harden fully for a while (as explained above, so the guide wire hopefully won't damage them when I do the white on the other side) I returned to the drag beam and filled in the black edge, around the white line - not easy to see here, as the black is very thin and blends with the darkness under the beam and the brown pipe on the right-hand side:

 

 

Edited April 8, 2022 by Chas Levin

[Coal Tank]

Thats beautiful 

[Chas Levin]

7 hours ago, Coal Tank said:

Thats beautiful 

Thank you .

[RANGERS]

My impression of the C12 was as an antique, workaday loco seeing out its years on branch services in the East Midlands in black, usually filthy but this sheds a whole new light on them.

 

Exquisite.

[Chas Levin]

12 hours ago, RANGERS said:

My impression of the C12 was as an antique, workaday loco seeing out its years on branch services in the East Midlands in black, usually filthy but this sheds a whole new light on them.

 

Exquisite.

Thank you - they must once have looked this way, fresh from the paint shop... for at least a few days, anyway!

Edited to add: Also, as far as I know we have no colour photos of original GNR livery C2s and the GNR livery in black and white doesn't look quite as striking...

Edited November 7, 2021 by Chas Levin

[Chas Levin]

Having said I was going to let the single white boiler band lines fully harden before doing those on the other side (which length of time to me means about a fortnight, having read that changes in the paint structure continue to occur for that long after application) I decided to push on with finishing the remaining white lines after only a few days, as I've found that once I'm 'in practice' with a particular technique, I get better results than if I let the facility go off a little through lack of use. The white components of the bands are now all done and to take stock, here are some pictures of where I've reached with the whole loco:

 

 

 

 

 

 

 

I've also done the black edging around the white on the buffer beam, where I was able to thin the white lines just about enough (they'd come out a little wider than those at the rear) by taking the black up over the edge of the white:

 

 

 

Next, the side tank and bunker lining...

Edited April 8, 2022 by Chas Levin

[Bucoops]

Looks absolutely stunning.

[Chas Levin]

Thanks Rich, I'm pleased with how it's going so far .

[MikeTrice]

So you should be.

[Chas Levin]

Quick update on lining pens: having had success with a new Kern compass one and found it so much better than the older Minerva one I'd been using, I thought I'd see what Haff pens are like (highly recommended by various knowledgeable folks) as they're available new too. We're not talking major investment here by the way: I ordered two different models to try and the total outlay including shipping and whatever charges were levied, Germany to the UK, was £33.

 

I wanted to try one of the types with sideways pivoting blades as cleaning these things is never fun, but I was interested to see how well they hold their blade separation before and after pivoting the blades. I mean no disrespect here to Haff who have of course been making these and many other excellent and highly professional tools for far longer than I've been using them, but I couldn't help wondering...

 

Here then is the Haff 228, alongside my first efforts with it - pretty variable but it's surprising how different the different brands and models are to get used to using:

 

 

 

And as viewed from the other side, where you can see how the pivoting blades are held in place during use - the dimpled point in the shorter moving blade locates with a positive click into the groove on the fixed blade:

 

 

 

Like the Kern, it worked extremely well and very consistently straight out of the box, though you need to be careful about the preceise angle at which the blades meet the surface:

 

 

 

I guess they're different in the way that different fountain pens feel different and you have to get used to them... for those old enough to remember fountain pens!

 

One of the reasons I decided to try Haff was because I finally discovered why the older pens I'd been using were giving inconsistent results and failing to produce a line quite often and it's to do with blade flexibility. Ian R in his book mentions one of the advantages of pens like the Haff ones is that the blades are very rigid (they're hardened steel, as I now know), preventing the flexing that can sometimes occur when a thinner or less rigid blade is placed against a straight edge, the flexing acting to close the gap between the blades. I'd read this and filed it in my mind... and then forgotten about it (age!) until I was using the vintage nickel pen that had been my dad's and which works very well most of the time... until it doesn't! I was sighting down the blades as they moved along against the stright edge and I realised I could see the paint reservoir between the blades bulging slightly as I pressed the blades a little harder against the ruler. Probably sounds very obvious but it's one of those things I thought "couldn't apply to me" until I saw it there - I realised that with a gap of a fraction of a mil, if the blades were moving together enough to cause a visible bulge in the paint reservoir, the gap must be closing and stopping the paint flowing: eureka!

 

And as for the consistency of the gap with pivoting blades, I'm sure no-one will be surprised to learn that my doubts - borne or inexperience and mistrust - were completely unfounded: I can confirm that the Haff 228 can be opened for cleaning and closed again multiple times and it will keep a very consistent width.

 

I think this is actually pretty impressive and I guess it must be partly due to the fact that the blades are hardened steel and therefore don't deform. There must clearly be some slight but temporary deformation while the dimple is moving across the land either side of the locating gloove, but the fact that the blades return to such accurate position afterwards must indicate how well made and well designed they are: jolly good job to Haff, very pleased as it does make cleaning so much easier...

 

Once I'm comfortable with the new pen I'll steel myself (sorry!) to start on the side panels. They are a bit daunting to be honest, because you're facing flat, empty areas with no guides. The boiler bands are their own guides and the buffer and drag beam edges and the chassis edges are just that: edges. With the side-tank and bunker sides, you've just got large flat areas to place lines on.

 

I'd intended starting with the rear of the bunker, as it's the least conspicuous, but I realised that the upper and lower outer lines along the sides are the ones that need to be absolutely right and everything else needs to align with them, so they need to be done first, unfortunately .

Edited April 8, 2022 by Chas Levin

[Jon4470]

Hello Chas

 

I have no doubt in my mind that you will make an excellent job of lining the side panels. Your work so far has been fantastic- fully repaying your patience and persistence.

 

Keep up the good work!

 

Jon

[Chas Levin]

9 hours ago, Jon4470 said:

Hello Chas

 

I have no doubt in my mind that you will make an excellent job of lining the side panels. Your work so far has been fantastic- fully repaying your patience and persistence.

 

Keep up the good work!

 

Jon

Hello Jon, thank you for your kind words and encouragement, much appreciated!

 

I think I'm going to spray up a practice panel and practise the whole pattern - parallel white lines, black infill, incurved corners, dark green surround, the lot! I need to find a way of reducing my nervousness about hitting those Great Big Open Spaces!

[Daddyman]

There are some really useful lessons for using/buying these pens here, Chas - thank you. 

 

Some nice English too!  

41 minutes ago, Chas Levin said:

a practice panel and practise the whole pattern

 

 

[Chas Levin]

1 hour ago, Daddyman said:

There are some really useful lessons for using/buying these pens here, Chas - thank you.

Thanks David; Haff pens are very highly regarded but I'm not sure how many people realise they're still available and very easily obtainable, too. They seem to get mentioned in the same context as other vintage pens - they've been around a long time of course - but in ways that somehow suggest they're no longer manufactured. I didn't think they'd still be available - Ian Rathbone's book is from 2008 after all - and I was surprised to find they are: https://www.haff.de/drawing-pens/ . Very helpful and friendly company to deal with too (usual disclaimer, no connection, just a happy customer!).

The Haff 135 - the one Ian R mentions as being a favourite in his book - is no longer a current model, but when I bought the 228 they still had some old stock so I picked up one of those too - as I mentioned in my last post, the cost of them was so low I thought it might be interesting to try, especially if the swivelling-blade 228 didn't quite work as I'd hoped... but so far I haven't even got the 135 out of its box! Enormous fun, experimenting with them .

[Chas Levin]

2 hours ago, Daddyman said:

Some nice English too!

Ha - thank you Sir . Keeping the old traditions alive...

 

Damned if I'm going to begin every sentence with "So..." (my current bugbear!).

[Jol Wilkinson]

1 hour ago, Chas Levin said:

Thanks David; Haff pens are very highly regarded but I'm not sure how many people realise they're still available and very easily obtainable, too. They seem to get mentioned in the same context as other vintage pens - they've been around a long time of course - but in ways that somehow suggest they're no longer manufactured. I didn't think they'd still be available - Ian Rathbone's book is from 2008 after all - and I was surprised to find they are: https://www.haff.de/drawing-pens/ . Very helpful and friendly company to deal with too (usual disclaimer, no connection, just a happy customer!).

The Haff 135 - the one Ian R mentions as being a favourite in his book - is no longer a current model, but when I bought the 228 they still had some old stock so I picked up one of those too - as I mentioned in my last post, the cost of them was so low I thought it might be interesting to try, especially if the swivelling-blade 228 didn't quite work as I'd hoped... but so far I haven't even got the 135 out of its box! Enormous fun, experimenting with them .

Chas,

 

I've got a selection of pens acquired over the years. My first was an Ecobra, followed by a Kern, which is still my favorite (like the Haff 133). I have a Haff 135 but haven't fully got to grips with it yet, the blades are much fattier than the Kern. It draws super lines, but you need to change your approach slightly.

 

I felt prompted by your post to get a to get a Haff 228 but then I remembered I've also got a vintage Ken set (from Golden Arrow, a present from a friend for doing some CAD design) which I haven't even tried yet. The test sheet supplied with it has fine, crips lines.

 

If I needed a first pen, I would get both the the Haff 133 and 228. At those prices and given the availability of decent bow pens, it's worth it. The one thing I have found is that the cheap ebay ones, even if you follow Ian Rathbone's guidance on honing them, never match a decent one like a Kern or Haff.

 

Jol

[Chas Levin]

17 minutes ago, Jol Wilkinson said:

Chas,

 

I've got a selection of pens acquired over the years. My first was an Ecobra, followed by a Kern, which is still my favorite (like the Haff 133). I have a Haff 135 but haven't fully got to grips with it yet, the blades are much fattier than the Kern. It draws super lines, but you need to change your approach slightly.

 

I felt prompted by your post to get a to get a Haff 228 but then I remembered I've also got a vintage Ken set (from Golden Arrow, a present from a friend for doing some CAD design) which I haven't even tried yet. The test sheet supplied with it has fine, crips lines.

 

If I needed a first pen, I would get both the the Haff 133 and 228. At those prices and given the availability of decent bow pens, it's worth it. The one thing I have found is that the cheap ebay ones, even if you follow Ian Rathbone's guidance on honing them, never match a decent one like a Kern or Haff.

 

Jol

Hello Jol, I'd not heard of Ecobra before, I'll look them up. Do any of the other models you mention have swivelling blades for cleaning - I think that's my new Favourite Feature!

 

Where you said "The one thing I have found is that the cheap ebay ones, even if you follow Ian Rathbone's guidance on honing them, never match a decent one like a Kern or Haff" - exactly, and that's what prompted me to get these. The boiler bands were done with a Kern compass model I also recently got, again very cheaply. I'd previously spent quite a lot of time and effort on honing cheaper models (as detailed in previous posts) and probably learned quite a bit about how these instruments work in the process, but still not been happy with the results, so even if the likes of Kern and Haff were quite expensive, I'd still have moved on to them I think.

 

The 228 has quite fat blades too and that's another part of how all these different models feel different, work differently and have to be used differently, as you say, but I suspect the bulkier blades are also a factor in their lack of flexibility during use, which - as I recently learnt - is very important in keeping a constant flow.

[Chas Levin]

Modelling time's been a little limited over the past couple of weeks for various reasons, so I thought it would be a good time to put up some pictures of other projects, earlier builds that I'd only shown as finished models, but which include things that might be interesting or useful shown in more detail...

 

First up then, let's start with the vehicle that appeared at the start of this thread, the D&S kit of an ECJS 46’6” Luggage Brake Van. This was a by-the-instructions build except for two areas, the side lanterns and the roof.

 

I decided I wanted to use one of Train-Tech's motion-sensor triggered lighting units; these coaches had a side-lamp each side with (according to some sources) white lights facing forwards, red facing rear. I didn't like the idea of trying to use two units - hiding even one unit in such a way that it didn't obtrude into view too much but still allowed a quick and easy battery change (I'm not keen on dismantling models to change batteries if it can be avoided) looked to be enough of a challenge. I spotted a note in the instructions for the AL2 Flickering Tail Light that both the Hi and Lo outputs can be used simultaneously. Tests showed that while there is of course a difference in brightness it's not too dramatic; unless you look down from the top you don't see the two sides together and in real life I'm sure there was a fair amount of variation between oil lamps: problem solved!

 

The next thing was where and how to mount the unit. Clearly the battery holder side would need to protrude through the floor and a little examination and measurement showed that putting it parallel to the direction of travel would work best, as clearances mounted sideways were tight and the edges of the coin cell would be more visible:

 

 

 

I cut a suitable sized slot in a business card by trial and error (it took several business cards actually!) to be the right size to grip the unit as steadily as possible, whilst still allowing its removal in the future, should it need replacing:

 

 

 

This was then used as a template to cut the slot out of the carriage floor, seen in the photo at a slightly later stage but it gives a clear view of the slot:

 

 

 

Immediate problems occurred: cutting the slot length so as grip the lower ends of the battery holder prevents the unit dropping down through the slot... but does nothing to stop it moving upwards or hold it in place when a new battery is pushed up into it. I also realised there's need to be some sort of 'sides', above the slot, to hold the Train-Tech unit in place vertically. I therefore added two 'walls', small L-shaped pieces, soldered by the short legs of the 'L' to the carriage floor either side of the slot:

 

 

 

Also visible in the photo are four small holes, two in each side-piece. These are my solution to holding the unit securely in place, including during the removal and replacement of batteries. The battery is helped in place by the sprung grip of the ends of the prongs either side of the opening, through which a new battery is pushed. As supplied, it's a pretty tight fit and you have to grip and pull the spent battery very firmly to extract it; likewise the new battery, which must be pushed in with considerable force. No system for holding the unit that I could devise would have coped with these forces, so the first thing was to carefully file down the inside surfaces of those two prongs, a very little at a time, testing every few strokes, until the degree of grip had been reduced to a minimum: enough to stop the battery falling out, but still allow easy removal.

 

At this point I'd only fixed one of the two side-walls in place and I then drilled two holes through it and, holding the Train-Tech unit in place, marked where those holes lined up on the plastic unit's sides. The holes were then drilled (very carefully!) through the plastic (I had set the holes' positions in the brass so that they would align with the plastic 'land' areas on the T-T units to each side of the metal innards, where there is enough plastic in the 'shoulder' on each side to accomodate a 1/2mm hole) and by threading pieces of NS wire through side-wall and the unit, it can be held strongly enough but also removed when needed:

 

 

 

With the other side-wall in place, the wires threaded right the way through both and a battery in the holder, you can begin to see how things will work:

 

 

 

Mounting on bogies showed that there was plenty of clearance underneath for the battery, which hang down considerably less than the gas tanks, dynamos etc:

 

 

 

 

 

While sitting the body on top also looked fine at this stage:

 

 

 

I also positioned the plastic roof that's supplied with the kit so as to check that there would be clearance for the wires that will protrude from the top-mounted sockets. I didn't use this plastic roof and I'll show how I made a brass replacement shortly, but the clearance was the same and allowed me to see that it wasn't a problem, though I could see that providing the ends of the wires with right-angled sections to go into the sockets would be a wide precaution:

 

 

 

Having sorted out the power and sensor unit, the next thing was the actual lamps...

Edited April 8, 2022 by Chas Levin

[Chas Levin]

The side lamps were done using what was then a new Modelu product, hollow bodied LNER-style side lamps. They are, I can't deny, fairly fiddly to work with, particularly as at that time I hadn't yet found an effective and comfortable magnifier headband. First, they were mounted – in a fairly snug pressure fit – on the ends of cocktail sticks that had been chamfered and squared off to the right size, so that they could be primed (with Halfords grey Plastic Primer) and then painted back (Halfords Matt):

 

 

 

Then the lens opening either side (which had been masked during painting) were cleaned up and with the lamps positioned with one opening facing upwards, a tiny line of epoxy was laid round the circular opening and either a clear or a red lens (they're supplied by Modelu too) dropped on. I didn't want to use cyano because of the probably fogging, especially bearing in mind that the amount of light the 1mm LEDs supply wasn't going to be huge. The amount of epoxy also had to be kept to an absolute minimum to prevent it spreading out over the inner face of the lens – there's only a millimetre to play with, after all! With care, this is the result, each lamp having one clear and one red lens, on opposite sides:

 

 

 

These hollow-bodied lamps were only on the drawing board when I first spoke to Alan them about at Modelu at the time, and it was his excellent suggestion that one 'wall' or side of each lamp body be left missing, so that an LED could be inserted. I had sourced some warm white 1mm LEDs and carefully glued two to a couple of pieces of thin plasticard the same size as the missing lamp wall – here's a shot of one the lamp bodies where you can see the inside, next to one of the LEDs which has been glued to its plasticard backing - sorry about the poor focus in some of these shots, I had difficulty explaining to my phone's camera that I really did want it to focus on the smallest thing in the shot:

 

 

 

I found the easiest way to do that gluing was to glue the LEDs to a much larger piece of plasticard and then trim the card down afterwards, round the shape of the LED. For this job I did use cyano, partly because I've had issues with epoxy not bonding well to plasticard, partly because of the need to have it bond quickly and strongly, especially given the difficulty of holding something as small as these 1mm LEDs still on the ends of their springy wires.

 

The idea is that the LEDs are glued in place of the 'missing' rear wall of each lamp, so although they fire forwards, sufficient and equal amounts of light will go left and right, through each lens. Before gluing the LEDs in place I also painted the insides of the lamps white, to help with light levels – here are the two lamps with their new false back walls, LEDs on the inside.

 

 

 

The plasticard backing pieces are then also painted black and the lamps powered on, which allows you to spot tiny areas where the black paint hasn't quite covered the joins between the lenses and their mounts, or any of the walls, including the back walls:

 

 

A tiny hole was then drilled in the top surface of each ducket, centrally front-to-back but quite near the roof side looked at sideways and the bare-ended wires from the lamps were threaded through from the top so that pieces of wire of the right size for the Train-Tech unit's sockets cold be soldered onto the wire ends. A tiny dab of UHU under each lamp base secures the lamp bodies to the tops of the duckets (I used UHU rather than cyano or anything stronger partly to avoid fogging, partly in case the LEDs fail and the lamps need to be removed and replaced). Here's a shot with the wires connected (the wires pieces at the ends have been insulated with Starbrite paint-on electrical insulation) – it's not all that clear but they go up inside the duckets on each side:

 

 

 

And by painting what does show black, it's rendered even less noticeable (a trick I learned when adding Train-Tech carriage lighting strips to Hornby LNER stock); here's a shot of the finished carriage viewed from only slightly above eye level and I hope you'll agree that it's pretty unobtrusive, especially with all the other underframe gubbins:

 

 

 

From above, the slight difference in level between the two sides is again something I think is a very acceptable compromise for the fun of the full effect:

 

 

 

 

 

So there we are - motion-activated ECJS / GNR / LNER style side-lamps! I plan to fit identical ones to a D&S Howlden 45' brake kit that's sitting on a shelf waiting to be built. A fair amount of work involved, to be sure, but they do look very effective.

Edited April 8, 2022 by Chas Levin

[Chas Levin]

The other part of this build that deviated from the instructions was the main roof span, that part that sits under - and supports - the clerestory. With absolutely no disrespect intended to Danny Pinnock of D&S, a man who - in the classic phrase - has surely forgotten more about building models than I'll ever know, I didn't quite like the idea of plastic for this structure. I know lots of people do this quite happily but I have also heard tales of brittleness with plastic and one of the things I like about working in metal is the removal of that possibility, so I found the idea of constructing this coach entirely in brass to be very appealing. Plus it was an opportunity to try rolling my own roof, which I'd not done before.

I started with a piece cut to the maximum length (measured between the high areas of the carriage ends) and quite a lot wider than the final width, in a set of bend bars with a steel 3/8" rod taped along the length:

 

 

 

The tape was necessary because without it, the steel rod bulged out in the middle and it was still a bit of a struggle to prevent it doing so, even with the tape. I then had to introduce a stout piece of wood and a pair of g-cramps, in order to get enough purchase / leverage to make the folds each side, over the steel bar; this photo also shows the positioning of the steel bar along the edge of the brass, probably about half an inch in, so that when the brass is folded over it (using another piece of wood along the top surface) the curve is formed with the diameter of the steel bar:

 

 

 

And, after filing broad notches each end to accommodate the upward-protruding top parts of the ends, this was the result:

 

 

 

You can see, along the upper parts of the side aprons, the first of my scored lines to mark where to wield the Stanley knife, which was done slowly, gently and carefully, to avoid deforming the brass. I'd left the piece much wider than it needed to be because of the impossibility of calculating the final width / depth after folding and in order to avoid the appalling disappointment of finishing up too short:

 

 

 

There then followed a lot of careful filing of the ends - very small amount off, trial fit, very small amount off, until I had this:

 

 

 

 

 

Next, the clerestory...

Edited April 8, 2022 by Chas Levin

[Daddyman]

A superbly accurate bit of bending, Chas! 

[Chas Levin]

24 minutes ago, Daddyman said:

A superbly accurate bit of bending, Chas! 

Thanks David! I had experimented beforehand with some pieces of scrap brass and the two or three different diameter pieces of steel or iron rod I had (including some very old Meccano iron rod!), but it still takes some care to do, as I found that you can get quite varied radii from bending around the same piece of rod, depending on how tightly you make the bend, or how closely you're able to follow the rod's diameter. That's why I ended up with the piece of wood and the g-cramps, in order to be able to exert sufficient leverage on the rod that most closely matched the desired curve.

 

The end result was well worth it though, as it made the whole vehicle so much more solid, rigid and stronger. I'll add the details about the clerestory later today...

[Chas Levin]

Having securely soldered the main roof span to the body (which also had the advantage of allowing the filling of small inconsistencies in fit between roof and sides or ends with solder), I wanted a nut and bolt fixing system for the clerestory, to allow for putting it in place after painting and glazing - and possible removal in the future if the glazing slips - but it also needed to be able to hold the clerestory flat and in intimate contact (in the engineering sense, not emotionally, of course!) with the main roof span for its entire length, maintaining the impression of the two sections being one. Three bolts raised the possibility in my mind of slight gaps appearing at the midpoints between bolts, so it had to be four. That made inserting the glazing more fiddly, but dividing each side into two with the join parallel to one of the upright dividers between glass panels in the clerestory worked well.

The clerestory as supplied is a simple fold-up; here it is - upside-down - with the first two folds made and you can see where the second two will create a flat strip along the inside of each side, strengthening the side and providing more area for glue, which is how its designed to be secured to the supplied pastic roof:

 

 

 

I cut four small rectangles of sheet brass, equal in width to that between the inside walls of the clerestory when carefully positioned with the sides truly perpendicular to the top. These then had 10BA bolts put through suitable sized holes and soldered in place, with washers soldered underneath. The washers were to match the thickness of that folded over inside bottom strip, so that when the nuts were tightened on the bolts, they wouldn't pull the brass rectangles down onto the roof surface, deforming the sides by even a small amount, which would be visible if you looked along the side of the clerestory and it wasn't perfectly straight.

 

 

 

The four rectangles were then soldered in place, up inside the two side strips - a slightly awkward job because you're trying to pull the small brass pieces downwards against the upper surfaces of the base strips but also trying to avoid getting solder on the undersides of those strips... Unfortunately, I didn't take a really clear photo of this stage, but here's a cropped section from another picture that hopefully shows what was done and how the soldered bolts and the soldered brass rectangles now become strong and rigid fixing points on the clerestory. The brass rectangles also act to strengthen the clerestory itself, which is otherwise still quite flimsy and as it's not going to be glued to the roof, it needs to be a strong assembly in itself:

 

 

 

The clerestory, complete with downwards protruding bolts, is no used to mark the positions for drilling the four holes in the main roof span through which they will be secured. This apparently simple operation proved more difficult to do accurately than I'd anticipated! In the end, I taped the main roof, upseide-down, to a piece of fine planed batten, slightly narrower than the roof, so that by feeling with my fingers all the way along each side I could position the roof centrally. I also tack-soldered two short brass pieces to each end of the upturned roof to act as alignment aids to hold the clerestory against, before marking the bolt ends with white paint and drilling on the marks:

 

 

 

 

 

I still ended up havng to turn a couple of the holes into ovals of course, but by taping the clerestory centrally in place from above and soldering the washers you see under the roof in place (having first bent them into very slight curves to conform to the inside of the main roof, so as not to deform it when the nuts were tightened) I was able to arrive at an acceptable result:

 

 

 

And the final result - the whole vehicle in brass - was well worth the extra effort:

 

 

 

In the above photo the clerestory was bolted in place for some test running, but not fully tightened so the gap between it and the roof is more obvious.

The main body and the clerestory were then masked and painted as separate units, testing the fit every so often to be sure nothing had altered or any clearances become obstructed by paint coats. In particular, the outermost end overhangs of the clerestory sit directly on top of the raised central sections of the carriage ends and I found that too many primer and paint layers where those two surfaces met would lift the clerestory just enough to prevent that intimate contact between it and the main roof that I mentioned earlier - you can see that in this photo, taken with the various different parts at different stages of painting and only the earliest stages of the teak being started:

 

 

 

The end result though passes muster I think, and the key reason is the ability to tighten those four nuts up inside the roof (with some threadlock applied too of course), drawing the clerestory down to the roof along its whole length. When viewed close up and at eye level the join is evident, but as the two following shots hopefully show - with apologies for the poor depth of focus on the daylight one - it looks fairly good at normal distances and angles:

 

 

 

 

 

Edited April 8, 2022 by Chas Levin
A photo was in the wrong place!