Atlantic's works: Portable layout - Scenic details next

[Atlantic 3279]

Oddly enough, firm as the transfer previously seemed, the R just pushed back into place with finger tip pressure this evening. I've also added those rain deflectors and door fittings.

[kimballthurlow]

Hi 3279,
That is a magnificent model on so many levels ....

I agree that the tissue printing has better definition.
The lazy R is no detriment to the model at all. It is one of those things that may attract attention and invite further scrutiny.
Which is probably a good thing. That is assuming "ordinary" persons would notice.

Being invited to scrutinize your model and wonder at its allocation and purpose, I was easily taken back 90 years.
So I could not help thinking what it was like to work in that kitchen. Was it a turn of the century fit-out with coal burners,
slow-combustion ovens, ice boxes, straw insulated pantries ....?
Now the layout could equip a provisioning siding, the platform lined up with victualling trollies, wicker baskets, coal, restaurant staff etc ready for the next shift.
Great job!!

Kimball

[Atlantic 3279]

Distractions and demands on my time having eased a little for the time being, I can report on some progress with the making of moulds to reproduce as nearly as possible in cast resin the exquisite master parts that Mike Trice so kindly printed earlier in the year.

I started with the side frames for the 11'9" wheel base Doncaster 6 wheeled carriage bogies. Mike had tried out several versions of these, and the evolution of design had resulted in both a flat-backed variety designed to stick onto a thin etched fold-up inner frame such as Dave Sutton's or possibly Bill Bedford's (the latter I've not tried) and a version with locating pips and tabs on the back to permit an all-resin assembly for which Mike had also printed an inner frame master. I was trying to make most efficient use of time especially and of materials, also trying to guess what was most likely to suit any likely users. Being keen to initially make only one set of the minimum necessary number of moulds, but to keep options open in case my expectations proved to be wrong, I decided to base my side frame moulds on the version with the pips and tabs on the back, reasoning that these are easy enough to trim off to suit fixing to a metal inner frame and are essential for use on a printed / resin inner frame. At first I cursed the thought that I would therefore have to make a two part mould in order to create the pips and tabs on the back, but then I realised that a two part mould would have been necessary even for flat-backed frames if I wanted to at least produce dimples in the correct places to accommodate axle bearings, whether those are soldered into a metal inner frame or glued into the resin side frame! I decided, after some agonising, to postpone production of a resin inner bogie structure pending convincing evidence of demand.

After examining the printed master frames I decided to add some little rectangles of plastic to the backs of the spring hangers where those project below the lower edge of each frame. This was to ensure that when casting there would be good access for resin to find its way in from the back, equally for air to find its way out. I also filled the fully formed bearing holes almost completely with plasticine so as to leave just a shallow conical dimple (sorry Mike, I wasn't confident that they would be reproduced reliably if left full depth). I then made a plastic base for a moulding box, with some slots and holes to accept the tabs and pips on the backs of the frames, then stuck the frames down as firmly as was possible on a couple of layers of double-sided tape. I also added some blobs of plasticine around the edges to later create alignment aids for the two parts of the mould.

The printed master frames had all cured with slight matching curvature which I did not want to reproduce in the cast resin versions. Once stuck to the base of the moulding box alone, that also tended to curve. I therefore made the sides and the joints in the completed box strong enough and stiff enough to hold the base flat. That was good as far as it went, but I hadn't thought about what would happen when I had poured the first part of the mould, stripped down the moulding box and kept (or reinstated) the frame masters in the flexible rubber. Of course, the frame masters assumed curved form again and so did the rubber mould! I had to think of a way around that problem, and it occurred to me that if I allowed the first part of the mould and the masters to remain in a curve while I produced a suitably thin and fully cured second part to the mould, everything would be okay again when the completed mould was straightened out by simply placing it on a flat surface. In order to allow the first part of the mould to sit temporarily in the same curvature as the master sides I supported it as shown below on two round paint brush handles and very carefully built sides around it once again.

Here's a look at the first part of the mould which has to capture all of the surface detail:

Here's the thinner second part with the recesses that create the tabs and pips, plus the shallow mounds that produce the bearing hole dimples. I've also bored some little breather holes to ensure that the recesses fill with resin rather than trapping air:

The mould certainly produces straight castings, although the first attempt to cast in it revealed some problems. The most obvious one was that it was a struggle to thoroughly coax the resin in and the air out of all of the recesses within the viable working time of the resin mixture. Four frames in a single mould were too many. At least one of the spring hangers failed to form properly as a result. Air bubbles had also been retained in the mould rubber in the narrow spaces around the leaf springs, resulting in annoying little beads of resin appearing in those areas in the castings. As a one-off defect in the first casting attempt I also had a badly defective axlebox cover, the result of a stray flake of rubber that had fallen into the mould....

The resin beads can, with some care and patience, be largely picked and scraped away with a small sharp tool, and the hidden backs of the castings are adequately formed.

After slicing the mould into two smaller moulds for separate production of two frames apiece it is easier to get the resin in fully. I may in the end decide to re-make this mould completely, taking care to dislodge the air bubbles before the rubber sets, but that is for later consideration.

More about the separate bolster end pieces for the bogies, the buffers, and those nice bucket seats when I can find more time to type and upload!

[Atlantic 3279]

The next items I addressed were the bolster end pieces with the earlier double leaf spring arrangement. In common with the later coil spring type, these are designed to have their upper parts stuck to the inner faces of the main bogie side frames and their lower parts stuck behind the tie rods, so that the spring detail appears in the right places in the gaps. In an earlier version of the master prints, Mike had arranged for the upper part of each bolster end to sit in a recess in the back of the bogie side frame, but when the idea of using metal inner frames arose there was no room for the original arrangement. Were the bolster ends to instead be stuck to the inner faces of a metal frame (if used) then as printed they would actually sit too far inboard. I think this would be particularly noticeable with the leaf spring version as the spring ends should project quite a way out of the bogie sides. I have therefore thinned down the tops of the bolster ends where they attach to the bogie frames, which was a rather fiddly job, but it ensures that they will fit correctly within a metal bogie frame, and can be made to fit in the right way when no metal inner frame is used if a little piece of 0.015" or 0.020" plasticard is stuck between the bolster end top and the inner face of the main frame. Confused? I hope not. If so, I apologize...

Anyway, apart from flattening any imperfections on the backs of the bolster ends before trying to stick them to the base of a moulding box, the challenge I faced with the leaf-spring type was that of reproducing the open space behind / between the two leaf springs. I had to let the mould rubber into this space yet still be able to extract the master parts and in turn the castings from the rubber of the mould. I actually managed to do that using only a single-piece mould, although I had firstly to stick some tiny triangular slivers of 0.005" plasticard into the V formed by each pair of leaf springs, thus dividing the space into two halves. You might just see the piece of white plastic in one example:

I produced a mould to create eight of these at a time, enough for two six-wheeled bogies.

When the rubber had cured and the masters were extracted I also put some part depth cuts into the rubber, in line with the divisions behind the springs that my slivers of plastic had created, making it possible to open up the mould somewhat both when pouring resin in and when attempting to extract undamaged castings.

The first castings out of the mould had poorly filled spring ends.

I've addressed that problem both by deepening the cuts in the mould to allow it to be opened up even more to admit resin, and by paying more attention to the spring end areas when coaxing the resin into the mould. Although the flash is still in place in the image below, the completely formed spring ends are evident.

[Atlantic 3279]

The pieces for the coil-sprung bolster ends seemed a far simpler proposition, requiring only a shallow flat-backed one-piece mould, produced in this box.

Typically, even the seemingly simple jobs produce a snag. One awkward air bubble in the rubber has caused a problem with the casting of one of these eight bolster ends, but even if I have to remake this mould it involves hardly any time and trouble, so that's a relief of some sort.


For the GNR pattern extendable / retractable buffers I'm trying a two piece mould design that will actually split around the edge of each buffer head. I do not think the bases of the buffers are so large as to be really difficult to cast in resin or to extract from the mould if they are formed a few millimetres within the rubber. The moulding box initially looked like this, with buffer heads stuck down firmly onto / into several layers of double-sided tape, and some pieces of square plastic rod also stuck to the base of the box to shape the bottom of the mould so that (when inverted) the other part of the mould would "key" firmly and accurately in place. Once the first part of this mould had cured, I made some part depth cuts in it in the same way as I had cut the mould for the leaf-sprung bolster ends. Although I can't yet show a picture, the cut mould opens up enough to get the resin in properly and the first castings out of the mould are quite reasonable.

The last items for the moment were the bucket seats. Mike had created a block type base to these to simplify the task of making a mould and casting in that mould. I took the added precaution of putting a little plasticine under the edges of the seat base to ensure that any undercut in the mould would have a bevelled edge.

The seats cast with little or no trouble in the resultant simple mould.

Enough for now. Almost beer o'clock......

[MikeTrice]

Someone has been busy. Good to see an explanation of how you are going about these.

[Atlantic 3279]

Thanks Mike. Given the length of time I've taken to actually make a start on these parts, I thought a good account of the overdue developments was owed.

[Atlantic 3279]

Here's one of the cast copies of a leaf-sprung bolster end, previously shown in the rough with flash obscuring detail, now cleaned up with the thin film of resin removed from the space between the two leaf springs.

The mould for the later coil-sprung bolster ends and a set of hastily poured slightly imperfect trial castings. There was an unwanted ball of resin on one of these, easily removed , but removal revealed a trapped flake of rubber torn from the mould so I think there will be a bigger problem when I cast again in this mould. As I said yesterday, this mould won't be hard to re-make if necessary.

The completed moulding box, the mould and the first set of castings for the bucket seats. Reassuringly straightforward.

The cast seats in close up.

The two self-aligning parts of the buffer mould, a rough view of the way I've cut the lower section to ease introduction of resin and release of air or cured castings, plus the first set of quick trial castings.

A couple of the better ones from the casting trial cleaned up a bit.

[bucoops]

Massively productive! I was thinking the other day I don't know of a source for GNR coach buffers. Back to gentle encouragement for being able to supply (and the seats) to the great unwashed

[Atlantic 3279]

That's the idea.....

[Woodcock29]

Terrific work Graeme, as usual.
Andrew

[Atlantic 3279]

Although I had proven that the main castings for the bogie sides could be relieved of the unwanted resin balls formed in the unintended trapped air bubbles in the rubber mould, I wasn't particularly happy with the thought that I, or somebody else, would have to tediously clean these defects from every casting, nor with the plain fact of the reduced fidelity of the casting to the superb master. Given the slenderness of some of the gaps between deep-relief details of the master sides, I have some concerns that even a perfect new rubber mould may have limited "perfect" life, but I felt that it was nonetheless desirable to have another go at making the mould, or at least the critical part of it, in the hope of more nearly perfect capture of details.

As I had not disturbed the original plasticine registration blobs on the base of the original moulding box, and had provided precise location for each master side by virtue of the slots and holes in the base of the box, I hoped that I might get away with remaking just the first part of the mould. The snag was, that some of the plasticine blobs that I did not now want to disturb lay on top of the double sided tape that had originally anchored the masters to the flat base, and having been released once from that tape the slightly bowed masters refused to stick back in place sufficiently firmly. I've tried to solve that problem using some my old-favourite 0.9 x 5mm screws, attempting to place these so that they do not damage the masters and so that any cast reproductions of the tiny screw heads will form in places in which it is easy to trim them from the proper casting.

In making the new mould, I initially introduced just small blobs from the liquid rubber mixture into the areas where defects had previously formed. Using a narrow probe I coaxed the rubber very thoroughly into the recesses that had retained air bubbles at the first attempt. I actually found that I had spent so long doing this that the main bulk of the rubber mix was beginning to thicken before I had poured it into the moulding box, but I still had just enough time to get it poured in before it refused to level out!

To my relief, the new part-mould that I removed from this box this morning seemed to have all of the details around the spring ends cleanly represented by hole-free rubber. The other parts of the original mould, with their rubber registration lumps, also fitted neatly in place. Learning from experience last weekend I have therefore divided the new part of the mould in two so as to produce just two sides at a time, and resin is currently curing in it. I hope to be back later with some images of clean castings......

I've also re-made the smaller, simpler mould for the coil-sprung bolster ends, again initially coaxing the liquid rubber into as many recesses as I could.

[Atlantic 3279]

Here are some results, firstly one of the new side frames complete, still on the part of the mould that forms the inner face of the side.

Close scrutiny of this side, and one of the others from the new moulds, reveals some very tiny beads of resin due to air bubbles in the rubber mould, but not nearly the number, the size, nor in the awkward locations that featured in the castings from the first mould.

One under the right side of the spring here, in the upper corner of the horn guide.

One above the right hand spring hanger rubber here.

Those tiny beads will be easy to flick away with the tip of a suitably narrow tool.

The coil-sprung bolster ends are okay too. You'll see that in several cases trapped air in the rubber has resulted in a squashed bead or "sausage" of rubber in the upper very narrow parts of the spaces between the springs and the vertical links, but these are not sticking out from the surface. There's a round bead of resin in one of the wider spaces too, which will need to be dislodged or trimmed away, but that should be minimal work. The thing perhaps to bear in mind is that each of these bolster ends is less than 6.5 mm in height and in width. The flaws are therefore very small.....

[Atlantic 3279]

Those bogie sides do of course have a full length wire cast into the tie rod to address the risk of snapping......

[iainkirk]

Very nice...