Bath Spa - High tech modelling: Bringing 1947 into the 21st century. 3d printed scenery, cnc milled track, laser cut baseboards and computer control

[Rabs]

Yes, it was a bit of a squeeze getting all my parts onto 900x600 sheets but it was necessary because that's the machine I have access to.  Fortunately my modular design means that the largest single parts (the box walls) fit in this.  For general baseboard building where some parts might be larger I think I'd go a size or two bigger on the machine.  The motion stages have some impact on the cost, but the main driver is the laser power (length of the laser tube).  60W will cut 6mm ply at about 10mm/s.  I think that this is an absolute minimum - each module here took about 1.5 hours to cut - which would be too slow for a commercial proposition I think.

It might be worth going to a local job-shop and paying for them to cut something for you while you watch and chat to the operator.  That way you'll probably learn a lot about what specs you need for your own machine.

Edited October 16, 2013 by Rabs

[artizen]

10mm/s is slow but I hope to be doing mostly kits using 3mm MDF etc so 60W is feasible. I have had quotes from a cutter in Sydney for some garden scale carriages at only $45 for two kits (six sheets of 300x450mm each) which I consider remarkably cheap and almost too good to be true. I need to redraw the plans before I commit and it will be a good learning exercise to see what I get back. If the quality is good and the cutter is happy to do retail quantities it might be easier and cheaper to go that way instead.

 

But nothing beats having your own tools to do the job - that way you can experiment and learn.

[JeffP]

Can't believe I've never seen this thread before.

 

Interesting stuff.

[Rabs]

Yes, that is much lower than the quotes I got in the UK.  I got quotes in the range of £100-200 per 3 sheets of 900x600.  Needless to say, at that price I would have got my saw out!

[Rabs]

Today I spent most of the morning finding and fixing a problem on my mill and then face milling a sheet to give myself a perfectly level base for doing thin sheet work on. When I finally got it sorted I got down to railway business.  For the most part I'm using fiNetrax track (a new N gauge track offering which is similar to the 2FS Easitrac but with 9mm gauge and some nice features to make the pointwork even easier to assemble).

 

At the board joins I'm taking a leaf out of the 2FS book and milling some little bits of track base in copper clad FR4.  This will allow me to solder the rails down for the last few sleepers before the join and will give a much more robust section of track than the plastic sleepers I'm using everywhere else.  Here is the mill at work:

 

And here is the joining section roughly in place with some bits of rail to check the alignment.  I've milled it so that the copper is removed from the top of the sleepers leaving just an area directly under the rail which matches the chairplates on the plastic plain track.  The part is designed from the templot file and follows the gentle curve of the line in this section.

 

This part is now being stuck to both boards simultaneously with a very thin layer of epoxy.  Once this has cured I will tidy up the burrs on the copper chairplates and solder the rail in place.  Only then will I make the cut to separate the boards.

 

And remember children - Lungs, eyes and ears are for life and not just for Christmas

Edited October 19, 2013 by Rabs

[-missy-]

That looks good Rabs.

 

I hope you are going to straighten up those sleepers though, they are pointing everywhere but straight!

 

Missy

[Rabs]

I'm shocked that you even have to ask! 

Yes - I was just posing it for the photo.

Edited October 22, 2013 by Rabs

[Kelly]

Looks great. Envious of your tools/lab.

 

 

Just no way I'd ever have the room for any sadly... hmm, wonder if I can get a free shed and if the housing association would allow it...

[Rabs]

(cue west country accent) This evening I have mostly been ... making gauges:

And yes, the roller gauge is supposed to be asymmetric.

Edited October 21, 2013 by Rabs

[Colin parks]

Nice gauges Rabs!

 

Are they made of brass? You must have a very finely calibrated feed on that lathe of yours to get the accuracy required to produce such things. But then again, looking through this rather outstanding topic, everything looks very accurate!

 

All the best,

 

Colin

[artizen]

Still thinking of something appropriately witty and nonsensical to stay about the photo of you with all the safety clutter on!

[eldavo]

Some top notch use of the toys Rabs!

 

Will you be using any other form of strengthening, such as brass screws or pins in to the baseboard, with your milled baseplate at the board joint? In my experience the bond between the copper and the paxolin/fibreglass is not overly strong especially when you get down to the size of sleeper or base plate. Maybe I'm just more clumsy when moving baseboards about but I could see the track coming unzipped pretty easily without something more substantial.

 

Keep up the good work.

 

Cheers

Dave

[Rabs]

Nice gauges Rabs!

 

Are they made of brass? You must have a very finely calibrated feed on that lathe of yours to get the accuracy required to produce such things. But then again, looking through this rather outstanding topic, everything looks very accurate!

 

All the best,

 

Colin

 

Hi Colin,

Thanks.  Yes, they are brass.  My lathe is a CNC converted Myford ML10 and my CNC mill is one that I built myself.  Neither are big or powerful but they are accurate.  That said, I did stuff up the first attempt at the roller gauge (forgot to account for the width of the rail - oops!)

 

 

Some top notch use of the toys Rabs!

 

Will you be using any other form of strengthening, such as brass screws or pins in to the baseboard, with your milled baseplate at the board joint? In my experience the bond between the copper and the paxolin/fibreglass is not overly strong especially when you get down to the size of sleeper or base plate. Maybe I'm just more clumsy when moving baseboards about but I could see the track coming unzipped pretty easily without something more substantial.

 

Keep up the good work.

 

Cheers

Dave

 

I'm not sure, thanks for sharing your experience.  I've not done board joins before (or any railway modelling for that matter) but  It had crossed my mind that I've not left all that much area for the copper chairplate pads to bond to the FR4.  I haven't allowed for any other fixing method at the moment but I'm treating this join as an experiment.  That's why I'm only laying track about 30cm either side of it for now and then I'll try abusing it to see how it stands up.  I didn't want to use screws because I couldn't see how to hide the heads.  Pins might be ok though and would make convenient droppers ... hmm

Edited October 22, 2013 by Rabs

[bescotbeast]

Great work Rabs and as others have said I'm envious of your workshop and tools, I like your chosen time period as well very interesting from a historical point of view.

                                                                 Simon

[artizen]

This is your FIRST railway layout? 

[jon_1066]

Well he has to start somewhere. His next one is bound to be a bit more ambitious. (I'm thinking something along the lines of a scale model of Bristol Templemeads carved out of a single block of resin.)

 

Edit - is it just me or does he look a bit like Benedict Cumberbatch in that PPE)

Edited October 22, 2013 by jon_1066

[Rabs]

Yep, I had one abortive attempt on an old door but I never got much beyond early planning with that so I don't think it counts.

I have been known to go to fancy dress parties as Sherlock!

[raymw]

Hi Rabs,

 

I've been trying to think back to soldering to small pads on pcb. (Nowadays most pads are through hole plated, so much 'stronger').I think if you use a lowish melting point solder, and are quick, you'll be OK. Not sure it will work if you need to resolder, however, since I expect the pad will lift off. I like the idea, however, of the pad, sort of like the base of the chair, the solder blob being like the rest of the chair. I think a more robust method would be to have the copper on the bottom of the pcb (mill a slot through the copper in the middle,wide enough for your fixing pins/screws not to short), and drill holes and thread through small brass screws where the chairs need to be, head underneath, soldered in, then put back in your mill and level off the pointed ends and cut your sleeper gaps in the plain side of the pcb. I hope you get what I'm trying to describe. (Or, right sized screw with head on top, rail soldered in screwdriver slot).

 

Best wishes,

 

Ray

Edited October 22, 2013 by raymw

[BG John]

Before he invented Templot, Martin Wynne made 85a Models turnout kits in 4mm, that had all the copper except the base of the chair machined away. I built some in about 1975 that were still in one piece when I scrapped the layout in 2007, although it hadn't run for many years. Yours will be rather smaller though, so I don't know if this helps.

[Rabs]

Before he invented Templot, Martin Wynne made 85a Models turnout kits in 4mm, that had all the copper except the base of the chair machined away. I built some in about 1975 that were still in one piece when I scrapped the layout in 2007, although it hadn't run for many years. Yours will be rather smaller though, so I don't know if this helps.

 

Useful to know, thanks.  I'm going to deliberately try to break the ones I've done to see how they perform.  If they come of I'll drill through the chairplate and put in a brass pin as Ray suggests.

[Rabs]

Track at the baseboard join after laying, cutting, disassembly, getting bashed around a bit to test how strong the rail ends are and finally bolted back together on an uneven floor.  Looks pretty well aligned to me.

[Tim V]

That "N" gauge track looks pretty good, is it chaired bullhead?

[Rabs]

Hi Tim.  Yes, it's the same code 40 bullhead rail used by the 2FS crew on moulded track bases similar to their easitrac but at 9mm gauge.  It's available here: http://www.britishfinescale.com/

So far the range of turnouts is limited (it was only launched a month or so ago) but I'm assured that others will be available soon.  It doesn't matter too much to me as I'll be making my own point work with components from them.

Edited October 23, 2013 by Rabs

[Izzy]

Hi Rabs,

 

I know this is a bit late in the day, but could I suggest that for other PCB track baseboard joints you consider putting chair plate etches on top of the PCB? The reason is that it makes a stronger bond than soldering the rail straight to the PCB, and might help reduce/eliminate the chances of the copper lifting from the paxolin, as the plate absorbs/dissipates a lot of the heat generated from the soldering, reducing the risk of the glue bond rising to/above the levels at which it fails. This is more likely to occur as you have removed a lot of the copper surface which would normally help to do this.

 

Izzy

[Rabs]

Hi Izzy,  thanks for the thoughts.  I understand the point about heat dissipation and you are right, my method has left very little copper to spread the heat around. I'm not entirely convinced that a sliver of brass will have much impact on the thermal mass of the joint, which I would expect to be dominated by the rail.

I don't quite get your point about it being a stronger joint though - why would rail soldered to brass, soldered to copper be stronger than rail soldered to copper? The weakest link will still be the the copper to fibreglass joint, just as you point out.  Or is that what you mean and that my method is more likely to damage this?  If so you are probably right.  I'll see how my experiments go and then decide if a different method is needed.

 

As an aside, these aren't paxolin - I don't think I've ever even seen FR2 board!  It's more difficult to get hold of these days because hardly anyone uses it for electronics any more.

Edited October 23, 2013 by Rabs