A new Gauge 3 garden railway

[Mike W]

The plan is to lay 3mm rubber sheet (sold in rolls) onto the steel deck and the track on top of that, then ballast and fix with PVA. I hope the rubber will reduce noise and also give a realistic height to the track above the cess at the sides. When bolted together and supported on tubular steel posts I hope the effect will be a smooth and neat edge all around the circuit.

The design does need some tweaking, for example the ends are recessed when they would be better flush, the holes in the ends will be changed to the Gauge 3 standard of two M10 holes at fixed centres, and I think the sides protrude above the deck just a little too much, even when the track is laid on 3mm hard rubber sheeting. Note the drain holes at the edges (thanks Osgood).

The main and unexpected problem however is distortion of the deck during welding which means it is stressed and dips 2mm in the centre. Pressing on the underside makes it twang and bow the other way – being higher in the centre. If they cannot resolve that I will have to abandon the whole concept. It also makes me wonder, even if they can prevent the distortion during welding, whether it will distort again in bright sunlight.

Sorry about the photo quality. One shows it upended, seen from the top.

 

[jamiepage]

Mike,
This is all very interesting; indeed, timely.
Would thickish metal mesh/ perforated sheet material be an option for the deck, a la Mike May? (As I recall, that is.)
Built in drainage of course, but perhaps less liable to distort during welding? Or perhaps it would distort more..
Jamie

 

[Mike W]

You may be right Jamie, I don't know. I wanted a flat deck on which to lay the track direct save for the rubber which is for noise reduction as much as anything else. I also want ballast, which sort of goes against too many holes. Ted Sadler's Gauge 3 line has very good ballast, so it can be done in our scale.

I'll seek advice from the fabricators as maybe a thicker top would help.

Mike

 

[mickoo]

Ask them to put a full depth diagonal plate brace from corner to corner on the underside and then spot weld the centre of the deck panel to the brace, that'll stop it flexing up and down. It's an extra piece of metal, cost and weight but could stop the floor buckling.

Even better, make the diagonal brace L shaped with the flat against the bottom of the floor, that way the brace will have it's own sustaining strength added to the floor.

Mick D

 

[Mike W]

Interesting idea Mick.

With baseboards about 6ft long and 15in wide do you think the "diagonals" would be better longitudinal beneath the tracks?

Mike

 

[mickoo]

Mike,

Definitely, but if your going for longitudinal stringers then they would certainly benefit from being L shaped, a better option might be a U channel but upside down so the base of the U is attached to the floor of the panel.

I don't know how deep the sides are below the floor but lets say they're 6" and the panel is 15" wide then I'd opt for a U channel of say 6" x 3", or maybe two L shaped runners, each 6" x 2" and spaced 5" apart, but with the feet of the L toward the centre.

Which ever is the easiest and cheapest, not sure who's building the parts and what equipment they have, but a simple bending brake and guillotine will make L and U shapes.



At a risk of increasing the fabrication you could make the runners fish bellied, I.E. deeper in the middle than at the ends.

I don't know how much weight your going to put on here, G3 can be quite weighty so it may pay to go for a box section down the middle, they're all options to consider but given the right choice of material I reckon an inverted U will be sufficient.

If money and fabrication were no object then you could go for something like this, it'd look like a pseudo bridge but probably more than strong enough, again I don't know your full plans or thoughts.



One thought about noise reduction, if you had an Inverted U channel then drill some largish holes in the side members (20-40mm), flip the panel upside down, cover in fine strong mesh and then fill the void with expanding foam, once solid trim off the excess back to the mesh, remove mesh, tidy up any remains and flip the right way up. The foam will never come out as it'll bond to it's self through the holes in the U channel.

Mick D

 

[Overseer]

Don't forget austenitic stainless steels like 304 and 316 have significantly higher rates of thermal expansion than mild steel so are more prone to distortion during welding, especially in thin sections. The only way to avoid it is to heat all the stainless before welding so the temperature difference is lessened. Probably not practical for this but overall heating is used when repairing old cast iron to reduce the chance of thermal shock when welding it. An fine expanded mesh floor would distort less than a flat sheet but would need regularly spaced cross members to prevent it sagging over time.

 

[geoff_nicholls]

I don't know if you've already chosen the rubber underlay, but I noticed damp proofing strips in B and Q yesterday. They have a lozenge shaped pattern which might help keep the ballast in place.

 

[Mike W]

Fair point Oversear, thanks.

Mick - They are professional fabricators so can do most things. They certainly have a bending brake, as do I. They don't have a guillotine (and mine isn't big enough) because they say its as easy to use the laser. I will go to see them next week.

Fellow WT'er Geoff (above) has drawn out the basic track plan for me and superimposed it on a sketch of the garden. Fabricators drew the structures in their CAD but are also happy for me to go there and use their CAD too. Plan is to state the curve radii and lengths, they draw that with slots and tabs etc and cut it, making the joins wherever is most convenient for them. That will leave me with a section about 40ft long which will be near ground level and with a 90ft radius reverse curve for me to make to fill the gap with different material - maybe concrete.

For instance, at one end I want a composite curve totalling 180 degrees, consisting of 15ft radius, with 3ft lengths of 40ft radius at both ends and in the centre (near a tree). They draw that onto the standard baseboard drawing on CAD, cut, weld and I then bolt it together on site and mark the exact position of the supports as needed. Simples - not, but that's the plan!

The ground level part is a different matter for discussion another day, if these baseboards work out first.

Geoff, I was dubious about PVA but have now heard of one G3 and one G1 line which have used it successfully over a number of years to hold the track and ballast in place. It needs to be a waterproof type of course!

Mike

 

[Mike W]

Almost forgot - one factor with additional supports underneath is that of course most of the track is on a curve. Not impossible, but just makes it not quite so easy.

Mike

 

[Simon]

All high tech stuff, not like my bodging!

One thing though, I'd definitely use SBR as opposed to PVA, behaves in the same way but more durable. (And more expensive)

PS Great to meet your brother and see Geoff's layout at Railex over the weekend, hope you enjoyed seeing your father's fantastic work being so well displayed

Simon

 

[Mike W]

Thanks Simon, I'd never heard of SBR, but see what you mean now. As an aside, when visiting Lansdowne Models near to you a few years ago they said they assemble their models with PVA obtained from the USA, because EU specifications mean that PVA sold in the UK breaks down after a few years. May have been hype, or maybe not ...

Yes, you were busy when I passed your stand, but the Risborough club have put in enormous effort to make Aylesbury fit for taking to a show as it was what Dad described as not portable but "a bit too permanent for my liking" and I am sure he expected it to be broken up after he died. Full marks to them.

Now please crack on with "LNWR Wagons Vol.3"!

Mike

 

[Simon]

Hi Mike

I hope to produce it towards the end of this calendar year.

In the meantime vol 2 will very shortly be available as a properly jacketed hardback, uniform with vol 1.

Good luck with the garden line, I'm just in from concreting the yard area at Launceston Road and will now nip out to use the remaining mix to put the base in for the branch bridge abutments

Simon

 

[Mike W]

Work has started on site at last. The workshop floor has been dug down 5 courses of bricks, ready for crushed stone, polythene, insulation and concrete. And if you are wondering what that much soil looks like, here's a picture. The railway will run through the middle of this heap, so it will need to be shaped at some point after the baseboards are in, to make a grassy bank and very short "tunnel" so that folk can cross the line.

Mike

 

[Mike W]

For scale, that heap is about 20ft by 10ft and not yet compacted. I was expecting much worse.

Mike

 

[geoff_nicholls]

It looks very nice, just so long as it doesn't distract you too much from your day job, providing us with kits and bits for our layouts...

 

[Mike W]

It has been a long time, but I am starting to look again at the garden railway plans because Denise says I need to do it now before I get too old - nice of her. Just had some curves cut to aid track laying. You can buy these for Gauge 3, but in only a few radii and they are quite short and aluminium. I well remember some of my late father’s engineering being described by Mum as a “gas woks job” (where he worked) because it was built like a tank. Well, these are 6ft long in 3mm steel! Note the holes down the centre so that I can use dowels to exactly align them to each other, giving a full 90 degree curve in one piece if needed.

They are all 15ft radius except one end of one strip which is 30ft radius and that end has no holes in the centre. It will act as a transition between 15ft and straight sections.

I am undecided whether to make another set of curves slightly larger radius so that the outer of the two tracks can be aligned directly from the inner track, but am not sure it is worth the effort and cost.

All I need now is to get on with making the baseboards.

Mike

 

[Mike W]

I abandoned the stainless steel baseboard design, mainly because the fabricators were unable to make it without distortion from the heat of welding. At least, they were unable to do so without increasing the thickness and therefore weight and cost too. So, this is plan B which seems to work much better.

Laser cut stainless steel strips with holes for crossbars. By varying the pitch of the holes in front and rear fascias the assembly takes up a curve. So, not being clever enough to calculate said pitch mathematically, I drew it all on CAD and let the software sort it out.

A set of parts for 6ft board is: front fascia, rear fascia, 5 x crossbars, 2 x end crossbars. The crossbars are designed so that some can be fixed flat and others vertical, but their top face/edge will always be the same height. That means the boards will sit flat whichever way they are bolted together. All parts are 3mm stainless. End crossbars not yet used on the test frame photographed.

 

[Mike W]

The only problem so far is my inability to countersink the holes in the fascia for M6 stainless countersunk cap screws, in spite of buying new countersink bits for the purpose, so I will have to take those parts back to the laser cutters for them to do. That is why the temporary bolts are visible.

 

[Mike W]

There has been some progress recently with the framework for the raised track. Several sections have now been bolted together, including the first of the four-track sections. Next headache is the two transitional frames which take two tracks to four on the curve with the correct radii and finish up at the correct angle to cross the garden. Err, maybe!

All drawn on CAD and pieces laser cut, so all(!) I need to do now is bolt them together and then it can be laid out on site in the garden Hornby-style. That's about 80ft of framework done now and parts for the supports also cut to size. These are scaffold poles for much of it but stainless poles and posh stainless feet for the area where it will be seen most, and adjustable “feet” at the top..

Fingers crossed for the next week and whether it all fits.

Mike