Oswestry Works Diorama

[Stubby47]

If you're intending to use the wheels/ rails for pickups, you can't have a full length metal axle for the wheels, as all 8 wheels will connect to all four rails.

 

Stu

 

Edited January 7, 2014 by Stubby47

[alanbuttler]

Cheers Stu, good point! 

 

The axles, pulleys and magnets arrived today.  I ordered short axles in the end so they will be wheeled in pairs to keep them separate.  All being well the traverser deck will be laser cut on Thursday.  The magnets are much stronger than I imagined and were just about able to cope with 2cm of solid wood desk! So hopefully the 8.5mm of MDF should be ok.

 

My only worry is with the magnets being so powerful, will they interfere with the locomotive motor? I'll find out tonight!

 

If all else fails there is Michael's suggestion of hidden slots as a fall back.  Cheers for the input gents, more to come shortly.

[Rabs]

If you find that the magnets are too much and affect the locos then you could use magnets below and just strips of iron or non-stainless steel on the traverser.

[alanbuttler]

Thanks for all the recent advice regarding the trial run of building the traverser gents.   Tonight I've been able to get the deck laser cut and the build tested out.  It's highlighted a few issues, not least how tricky it is dealing with the magnets.  Trying to make any adjustments to the small components with tweezers or a screw driver jolts the whole construction out of kilter before you even make contact....  Also my guides for lining up the layers are metal which also proved problematic!  It will sound ridiculous but not having anything plastic to hand I almost had to resort to dried spaghetti ...

 

On the plus side the magnets helped keep the axles and rails firmly in place and the magnetised rail gives the loco some extra purchase.

 

Magnet challenges aside, I'm pretty happy with the structure, the traverser wheels run freely and the height above the traverser rail is almost bang on the prototype.  

 

Here's some photos, note that the structure isn't glued together yet, I'm waiting on some pickups which I should have by the weekend.

 

 

 

 

 

 

 

 

[Rabs]

Looks good!  Brass will be ok for the rods - only ferrous metals are magnetic. If it were me I'd consider some thin plasticard strip down the long edges to hide the sandwich layer effect.

[alanbuttler]

Cheers Rabs.  I'd made the 2 inner layers slightly narrower in order to fit some plasticard to the front face, though I need to take a bit more off as its not deep enough.  The MDF between the rails is also slightly too wide but that can be sorted easily.  

[alanbuttler]

This weekend I've been able to successfully test the 2 basic concepts of the traverser, pickups via the 4 traverser rails and movement via rare earth magnets.

 

The 6x6 magnets work the traverser deck smoothly when moved, with the wheels running nicely on the rails.  They seem to be powerful enough for a 30-40mm gap, though in the model it will be more like 10mm. 

 

Unfortunately the P-B strip hadn't arrived by the weekend, so to get going I've used copper tape in the mean time.  Taping under the length of the rail and each axle seemed to work well, with the pannier moving straightaway.  The test construction is a bit rough round the edges but it's served its purpose in testing out these ideas.

 

Next step will be building a small test rig with the cnc components, laying 6-8 test roads and hooking up a stepper and arduino...

 

 

 

 

 

[PaternosterRow]

Lovely, accurate modeling there Alan.  Can't wait to see the traverser down in the building and running.  Keep going and don't flag as I really look forward to your updates. This is exciting, heady stuff and it's about time an accurate model of an erecting shop was made.  Great, well thought out workmanship.

Edited January 13, 2014 by PaternosterRow

[stephenashton]

Alan - I'm really enjoying watching this model develop. There's some lovely work here and I really like the way you're using modern methods to create the model.

 

I like the design of the traverser but how are you going to stop it accurately enough? 

[alanbuttler]

Lovely, accurate modeling there Alan.  Can't wait to see the traverser down in the building and running.  Keep going and don't flag as I really look forward to your updates. This is exciting, heady stuff and it's about time an accurate model of an erecting shop was made.  Great, well thought out workmanship.

 

Morning Mike, thanks for your encouraging comments, your modelling sets a high standard for me so I'm really grateful for your input.   I too can't wait to see the traverser in situ, I think another foamboard mock-up might be in order at the weekend  I'll have the internal brickwork completed by then so it will be a good point to get another demo build put together.   If the pre-order update emails from E Hatton's are accurate, there might even be a Dukedog in the frame... ! 

[alanbuttler]

Alan - I'm really enjoying watching this model develop. There's some lovely work here and I really like the way you're using modern methods to create the model.

 

I like the design of the traverser but how are you going to stop it accurately enough? 

 

Hi Stephen, glad your enjoying the updates! It's been a bit of a journey, lots of new technologies to get to grips with but I'm getting there.  

 

My hope with the traverser is that an arduino controlled stepper motor will be accurate enough, using micro steps if needed.   That's the next task after building the test traverser, a 250mm version of the 800mm full size one.  I should be in a position to wire up the motor and start testing code by mid-February I think (waiting for component delivery).

 

There are 3 posts on RMWeb that have been a great source of information for the motorised design:

 

Stepper Motor Turntable Drive by JeffP

DCC Controlled PECO Turntable by Tender

Automated Motorised Traverser by Deev

 

In the mean time I'll carry on with the sketchup model, the internal walls are almost complete so I'll move onto the roof next.  I might pause and build a couple of the outbuildings in the yard though, the laser cutter I have access to is big enough to make all the components for them and it will be good to see something physical periodically!

Edited January 15, 2014 by alanbuttler

[stephenashton]

My hope with the traverser is that an arduino controlled stepper motor will be accurate enough, using micro steps if needed.   That's the next task after building the test traverser, a 250mm version of the 800mm full size one.  I should be in a position to wire up the motor and start testing code by mid-February I think (waiting for component delivery).

 

My worry doesn't relate to the accuracy of the motor, but the reliance on the magnetic field to maintain a reliable and predictable relationship between the magnets in the traverser and the magnets moved by the stepper motor. I really hope it works for you, but it seems to me that:

 

1. the momentum of the traverser will prevent it from stopping precisely enough, and

    

2. the inertia required to start the traverser moving will mean that small adjustments made by the motor won't translate into the movement intended.

 

I hope I'm wrong. I love the concept of the model and the techniques you are using.

[alanbuttler]

I understand your concern now, thanks for explaining - yes very good point!   I've read of people using sensors to slow the motor as it approaches the desired position which might help, but yes ensuring that the magnet held traverser deck is in exactly the right position to the leadscrew table underneath will be a challenge.  

 

I've thought that some kind of home or reset function will be needed prior to operation, to zero the position of the traverser and ensure the deck is aligned correctly with the magnets. I'll be able to experiment a little more at the weekend with the magnets on a small test deck I've built.   

[mjcampbell]

I wonder whether introducing a deliberate offset between the magnets would help - so that those below the board are either wider or narrower than the spacing of those under the traverser deck. They could even be angled. The reason is this should introduce a horizontal aspect to the attractive force, to be in equilibrium the deck should centralise to the magnets below. 

 

If two magnets are aligned there is no horizontal force (just a vertical "attractive" force) until one of the magnets is moved out of alignment, then they will want to re-align - the force will increase with the offset, but there could be a certain ammount of offset until the force is strong enough to move the traverser. Once the stiction (starting friction) is overcome the traverser will "catch up", but as the magnet stops the momentum may not be checked until the magnet has overrun enough to generate a reverse force. This could possibly lead to a jerky motion.

 

By deliberately offsetting the magnets there will always be a horizontal force - so as soon as the magnets start to move they are applying this force to the traverser, and should overcome any stiction or overrun.

 

This is just a theory but could easily be tried in your mock up. If it doesn't make sense let me know and I'll try a diagram!

[alanbuttler]

Good idea Michael, makes sense.  I think the horizontal force will definitely help with alignment, in fact I think it's worth doing this regardless of whether current design works when I come to test it.  I'll let you know how I get on, thanks for taking the time to add your ideas, cheers 

Al

[Rabs]

By deliberately offsetting the magnets there will always be a horizontal force - so as soon as the magnets start to move they are applying this force to the traverser, and should overcome any stiction or overrun.

 

But if you offset the magnets the traverser will just position itself so that the horizontal forces are equal and opposite, with no net force on the traverser - so it doesn't make any difference.  You're still sitting at a position with no net force,.  Ineither case there will only be a horizontal force when the magnets underneath move.

 

If there was a residual horizontal force then the traverser would move until there was not - Newton's 1st law.

[alanbuttler]

From doing a few tests tonight there is a small amount of jutter when the traverser moves without a loco, but once a model is in situ the extra weight removes any sight of jittering.   The real test will be when the leadscrew is fitted, but so far I think things should be ok.   The electrics have stopped working however! There must be a short where I've used the copper tape under the rails on the deck.  It was a bit of a rough job, designed more for testing the magnets than the pickups, so I'll take my time on the next one.

 

Thanks Rabs, Michael and Stephen for your input.  I'll get some video up of the operation once I have something more to show.  That unfortunately will be next week now realistically, work is taking up nearly all my spare time.

 

Here are a couple of progress pictures of the sketchup model which I've been chipping away at during lunch times.  Whilst I'm awaiting the components for the traverser, I'll make a start on the roof and the removable south east wall.

 

 

[stephenashton]

Thanks Rabs, Michael and Stephen for your input.  I'll get some video up of the operation once I have something more to show.

 

No problem. As I said, I hope it works for you. Looking forward to the video.

[mjcampbell]

But if you offset the magnets the traverser will just position itself so that the horizontal forces are equal and opposite, with no net force on the traverser - so it doesn't make any difference.  You're still sitting at a position with no net force,.  Ineither case there will only be a horizontal force when the magnets underneath move.

 

If there was a residual horizontal force then the traverser would move until there was not - Newton's 1st law.

 

Yes of course, that is the intention. But when you start to move the lower magnets I think the lateral force will be stronger and more immediate. Rather like moving something held between your fingers rather than balanced on a finger. But this is just theorising! 

[alanbuttler]

Hi all, not much progress to report at the moment, I've been working in Edinburgh for the last week (where it was good to see Waverley after seeing so much of Waverley West in the realistic photograph thread).   I've had a couple of nights at home and then off to Glasgow tomorrow.   All the components for the traverser have now arrived so once Glasgow is out the way I'll be able to make a start on building the test rig and we'll see whether these magnets are up to the job.

 

So in the brief lull of work I've not done much other than setup a few more mock-up shots, nothing more than an excuse to see the new Dukedog under the camera lens!   

 

38xx outside the works (based on Reg's photograph of 3846)

 

9022 outside the works

 

4MT in the Boiler Shop (another based on Reg's photo's)

 

90xx, 2251 & 4MT in the works

 

9022 on the Traverser

 

4672 & 4560 in the Erecting Shop

 

9022 in the Erecting Shop (based on Reg's photograph)

[Quarryscapes]

This is an excellent thread! So much good Cambrian stuff on this forum, I rather think there's more info on here than in my entire bookshelf! I lived in Oswestry for a while and am now at the other end of the line in Pwllheli considering something like this but maybe not such a grand prototype, I was thinking Pwllheli MPD or Portmadog. 

[alanbuttler]

This is an excellent thread! So much good Cambrian stuff on this forum, I rather think there's more info on here than in my entire bookshelf! I lived in Oswestry for a while and am now at the other end of the line in Pwllheli considering something like this but maybe not such a grand prototype, I was thinking Pwllheli MPD or Portmadog. 

 

Thanks for stopping by and for your kind comment.  There is a lot of inspiring Cambrian representation on here indeed, have you seen Cambrian Coasters Pwllheli layout at all?  I've fancied building a mini-shed scene of Portmadog, I love the residential setting and mix of locomotives that appeared there.   

 

Sadly I've not been able to spend anywhere near as much time on the works as I'd hoped this year, but I should be posting bits and bobs about the traverser over the next month.  

 

Cheers Al

[Quarryscapes]

You're welcome! Regrettably in all the time I was at oswestry, and the amount of times I drove past it, I've never set foot in the works building! 

 

I was think Porthmadog too but I really want a turntable! I may end up with something of a mish mash of bits with a Cambrian flavour, perhaps adding a turntable to Moat Lane or Penmaenpool and going from there.

 

On the other hand, I'm currently reading about the Kerry Branch, and Kerry Station is looking very appealing, as is Dinas Mawddwy! 

[PaternosterRow]

Hi Alan,

 

I'm missing my Oswestry fix!!  It's a shame that life gets in the way of the modeling sometimes.  Very much looking forward to seeing some progress.

 

Mike

Edited February 11, 2014 by PaternosterRow

[alanbuttler]

You're welcome! Regrettably in all the time I was at oswestry, and the amount of times I drove past it, I've never set foot in the works building! 

 

I was think Porthmadog too but I really want a turntable! I may end up with something of a mish mash of bits with a Cambrian flavour, perhaps adding a turntable to Moat Lane or Penmaenpool and going from there.

 

On the other hand, I'm currently reading about the Kerry Branch, and Kerry Station is looking very appealing, as is Dinas Mawddwy! 

 

I agree Kerry and Dinas Mawddwy both really appeal also, definitely a job for a Noch Grassmaster or similar with all that overgrowth!  Quite limited on traction though, but I still like the idea of a few Dean Goods pulling cattle trains into the empty station, or a nice little 58xx on the Mawddwy branch.

 

Moat Lane Junction did have a turntable, between the station buildings and the locomotive shed - I'd recommend Brian Poole's Caersws: Cambrian Vilage for lots of detailed information on Moat Lane.  It's another location on the future to-do list for me.  I think I'd have to make the loco shed interchangeable somehow between the early 1950's wooden version that slowly fell apart and the new version that still exists today