Locomotive driving wheels

[Guest]

First renderings of SECR P class wheel. I'm working from a basic Ian Beattie drawing in RM from years ago plus whatever images I can track down on the 'net. Without getting a drawing of the wheel or going and measuring them myself, this is the best I can come up with from the information I have available. I also suspect I'd get better results with the spokes once I've mastered this mesh-editing lark.

 

 

Cosmetic tyre added for clarity.

 

 

 

For a first attempt, I'm actually okay with them. Whether or not they will print is another question but that's the next stage.

Edited June 25, 2013 by Guest

[flubrush]

First renderings of SECR P class wheel. I'm working from a basic Ian Beattie drawing in RM from years ago plus whatever images I can track down on the 'net. Without getting a drawing of the wheel or going and measuring them myself, this is the best I can come up with from the information I have available. I also suspect I'd get better results with the spokes once I've mastered this mesh-editing lark.

For a first attempt, I'm actually okay with them. Whether or not they will print is another question but that's the next stage.

 

David,

 

Maybe more of a fillet between the front of the spokes and the boss if you can manage - as is apparent on this picture

 

http://www.derekhayward.co.uk/BluebellRailway-1/Locomotives/27/19947660_f45cTG#!i=1676125356&k=4xBF2Sr&lb=1&s=A

 

And I know what you mean about finding your way around 3D CAD . :-)    I reckon a lot of people are thinking that 3D printing is going to make life a lot easier for modellers,  that is until they realise that they will have to get their brains around 3D CAD work,  and that might be the major stumbling block. :-)

 

Jim.

Edited June 25, 2013 by flubrush

[Guest]

Thanks, Jim, that pic is just what I need. I can certainly add an extra fillet at the wheel boss, and it looks to me like the spokes themselves should be wider - certainly near the wheel rim. I think I've got the thickness right, anyway. Also, looking at that image at maximum resolution, the spokes look almost oval (as in train set) as opposed to elliptical. Before I get cracking on some modifications, can anyone comment?

[flubrush]

Thanks, Jim, that pic is just what I need. I can certainly add an extra fillet at the wheel boss, and it looks to me like the spokes themselves should be wider - certainly near the wheel rim. I think I've got the thickness right, anyway. Also, looking at that image at maximum resolution, the spokes look almost oval (as in train set) as opposed to elliptical. Before I get cracking on some modifications, can anyone comment?

David,

 

It might be worth looking for someone on the Bluebell Railway to see if they can give you some help - maybe the owner(s) of the P class tank.

 

Jim.

[Guest]

David,

 

It might be worth looking for someone on the Bluebell Railway to see if they can give you some help - maybe the owner(s) of the P class tank.

 

Jim.

 

No need, I think. Jim. I backtracked that image you sent and found others. This one when shown at X3 resolution shows the spokes in more detail:

 

http://www.derekhayward.co.uk/BluebellRailway-1/Locomotives/27/19947660_f45cTG#!i=1676124809&k=tJx9jF3&lb=1&s=A

 

They look very oval rather than elliptical to me, with a distinct hint of flat surfaces. If not totally flat then the degree of curve is not at all pronounced. I'll do a few experiments on the 3D file and see how it turns out.

 

I looked again at your comment about 3D Printing making life a lot easier for modellers and the effort needed to master CAD packages. You're right, it is not easy learning this stuff. But look at me. I'm hardly what you would call computer literate but I'm getting there. Half the problem, I think, lies in the terminology being used. For example, what the hell is a boolean subtraction? Once you know, it ceases to hold any mystery or, indeed, fear. It may be that I'm lucky in having two sons who are thoroughly conversant in the design package I'm using so help was often just a phone call away. Having said that, I've discovered a remarkable number of tutorials on the 'net which have been of inestimable help. For instance, I was having trouble making multiple subtractions from single primitives. The first would take but the second just wouldn't work. A quick google found the answer: I have to reassign the primitive as Operand A each time I do a boolean operation before subtracting Operand B. Problem solved.

Edited June 25, 2013 by Guest

[flubrush]

David,

 

They look very oval rather than elliptical to me, with a distinct hint of flat surfaces. If not totally flat then the degree of curve is not at all pronounced. I'll do a few experiments on the 3D file and see how it turns out.
 

 

They could well be.  I remember seeing a note on a drawing of a casting which I have which said "round off corners to X inches radius".  I had a look for it again yesterday,  but couldn't find it again. :-)  But it did look as though the drawing office had left the cosmetic points of the casting master to the pattern makers,  which could have led to rounded ends to rectangular spoke cross sections.

 

Jim.

[Guest]

David,

 

 

They could well be.  I remember seeing a note on a drawing of a casting which I have which said "round off corners to X inches radius".  I had a look for it again yesterday,  but couldn't find it again. :-)  But it did look as though the drawing office had left the cosmetic points of the casting master to the pattern makers,  which could have led to rounded ends to rectangular spoke cross sections.

 

Jim.

 

That actually makes things easier, Jim, believe it or not. Back soon once I've wrestled with my computer.

[flubrush]

Just found these pictures on the Bluebell site - in case you haven't dug them up yourself.

 

http://www.bluebell-railway.co.uk/bluebell/loco_news/12/27_removed_buffer_beam.jpg
http://www.bluebell-railway.co.uk/bluebell/pic2/wn-2010/323frames_derekh23may10.jpg
http://www.bluebell-railway.co.uk/bluebell/pic2/wn-2010/323_wheels_derekh_8may10.jpg
http://www.bluebell-railway.co.uk/bluebell/loco_news/09/323_wheels_derekh_23oct09.jpg
http://www.bluebell-railway.co.uk/bluebell/pic2/wn-2009/323_andrews6882_3sep09h.jpg

 

Pictures of wheels during rebuilds are usually the best since you don't get them hidden behind/under bodywork.

 

Jim.

 

[Guest]

Jim, I owe you a pint! Great shots. As you say, not hidden by frames/body so I even get the rear view. I'll get onto it.

[flubrush]

David,

 

 

Jim, I owe you a pint! Great shots. As you say, not hidden by frames/body so I even get the rear view. I'll get onto it.

 

I hope they help.  I'm just hand writing GCode for my Caledonian 498 class wheels at the moment and I'm hoping I might start cutting some trial wheels tomorrow - in styrene in case I've made any mess-ups with the code and the small cutters might survive any non-intended excursions. :-)   If all looks well,  I'll start cutting brass.

 

Jim.

[flubrush]

David,

 

Just found some more pictures by rooting around on the Bluebell site a bit more.

 

http://www.bluebell-railway.co.uk/bluebell/locos/178_wk/178_wheels_derekh_13dec09.jpg
http://www.bluebell-railway.co.uk/bluebell/locos/178_wk/178_frames_derekh_13dec09.jpg
http://www.bluebell-railway.co.uk/bluebell/locos/178_wk/wheels_robf_8nov09h.jpg
http://www.bluebell-railway.co.uk/bluebell/locos/178_wk/robf487_11oct09h.jpg

 

Also dug up this Derek Hayward selection which has got a lot of wheel pictures running through it.

 

http://www.derekhayward.co.uk/BluebellRailway-1/Locomotives/323-Overhaul/21095817_DQkpVp/1677931419_zrxrnk7#!p=1&n=10

 

There's also another photographer mentioned - a Ron Faulkener - but his pictures appear to have been stored on Fotopic and I can't find if he has uploaded them to anywhere else.

 

Jim.

Edited June 27, 2013 by flubrush

[Guest]

Again, Jim, many thanks. I also now have a view of the crank which was not at all clear from the Beattie drawing, good though it otherwise is. There's also a lot of other detail in these pics that will help with the model in general.

 

Meanwhile, back at the Batcave, new design of Boxpok drivers following acquisition of new Boolean techniques.

 

[flubrush]

I've just done a trial cut of all my CNC code in Plastikard.   The basic GCode is generated in Cut2D - that's the wheel,  balance weight,  boss and spoke shapes.  I've then hand written the GCode to cut the flares on the spoke inner and outer ends.

 

First the wheel centre still in place after cutting in sheet

 

 

Then the wheel removed from the sheet and cleaned up a bit,  showing that everything seems to be cutting OK and there are no obvious faults.  The one thing I do note is that the 0.4mm cutter which is used to cut the gap between the boss and the balance weight is cutting into the flare round the boss.  I let it run all round the boss thinking it would be OK,  but it looks as though I will have to work out another way of doing it to avoid this.

 

 

The boss and counterweight are sitting proud by 0.2mm.  I tend to do this so that any marks on the side of the boss when shaping spokes will be removed when turning off the 0.2mm when the work on the spokes is finished.

 

Now to cut it in brass but I might have to order up some 3mm brass sheet to do it,  so might be a day or two till that happens.

 

Jim.

[Hawk]

Very interesting thread!

Any progress since last year?

 

I am going to make some spoked driver centres myself. I plan to have them made in brass by Shapeways. Since all the critical surfaces will be machined on a conventional lathe, It should be possible to sandblast the parts to remove the annoying surface pattern that brass parts from Shapeways unfortunately has.

 

The wheel threads will be machined in steel on a conventional lathe. Hope to be able to show some progress in the not to distant future.

[Guest]

Very interesting thread!

Any progress since last year?

 

I am going to make some spoked driver centres myself. I plan to have them made in brass by Shapeways. Since all the critical surfaces will be machined on a conventional lathe, It should be possible to sandblast the parts to remove the annoying surface pattern that brass parts from Shapeways unfortunately has.

 

The wheel threads will be machined in steel on a conventional lathe. Hope to be able to show some progress in the not to distant future.

 

None, I'm afraid. My time has not been my own over the past year but this is about to change. Having had a couple of test prints done, I'm not entirely certain that 3D Printing is the way to go here. FUD is somewhat brittle and I'm not sure would survive being forced onto an axle - though I'm prepared to be proved wrong. And SWF would probably get over that problem BUT most wheels are tapered along their spokes, and the prints that I've had done in this material  show a distinct "stepping".

 

I'm looking at other possibilities (and not only in wheel production) but don't wish to comment yet. I will post again as soon as I have something viable.

[flubrush]

Very interesting thread!

Any progress since last year?

 

Well my Caledonian dock tank wheels have been completed - brass CNC milled centres and steel tyres turned,  fitted and profiled on the Cowells.

 

 

I've also milled some wheels for a Caledonian 0-4-0 pug using the same methods.

 

 

Jim.

Edited April 20, 2014 by flubrush

[Phil Copleston]

Shapeways print items on-edge, rather than in the flat, if you get my drift, which is a large reason for the surface pattern.

 

3D printing and associated methods generally fall under the heading "rapid prototyping". There is a clue there in the name - it is designed to get a test version (a prototype) produced quickly (rapidly) to make sure that the 3D CAD process hasn't produced something unworkable.

 

It is, nevertheless, a quite useful addition to the modeller's armoury and suited to production of small batches, but there does need to be a degree of hand finishing in most cases. A classic example of this is Jim King's Smokey Mountain Model Works range of kits. Jim produces rapid prototypes for a living, and has used his knowledge and skill to produce patterns or casting, but most of these patterns are subject to a lot of personal attention, prior to moulds being taken off them for casting in resin. The kits are superb: surely there is a lesson there for the rest of us?

 

You are absolutely right, Simon. I have argued this point too: 3D-printing is only a basic production process, requiring quite a lot of subsequent cleaning up. In S scale it is often a lot easier and quicker to simply scratch-build the required parts than spend all that time messing about in CAD. Producing iffy loco driving wheel centres by this method seems to be the hard way to do it! 

 

As for the need for hand finishing off 3D prototypes, Like you, I'm also a satisfied customer of Jim King's Smoky Mountain Model Works kits and can attest to the quality of his finished products. He spends a lot of time cleaning up, filling and adding additional etched details to the 3D printed masters before committing them to resin moulding. The quality of his finish products is renowned!   

Edited April 20, 2014 by Phil Copleston

[Phil Copleston]

Well my Caledonian dock tank wheels have been completed - brass CNC milled centres and steel tyres turned,  fitted and profiled on the Cowells.

 

I've also milled some wheels for a Caledonian 0-4-0 pug using the same methods.

 

Jim,

 

These are superb! Well done. This is clearly the right method to produce ROBUST loco driving wheels. 

[Hawk]

Shapeways print items on-edge, rather than in the flat, if you get my drift, which is a large reason for the surface pattern.

 

3D printing and associated methods generally fall under the heading "rapid prototyping". There is a clue there in the name - it is designed to get a test version (a prototype) produced quickly (rapidly) to make sure that the 3D CAD process hasn't produced something unworkable.

 

It is, nevertheless, a quite useful addition to the modeller's armoury and suited to production of small batches, but there does need to be a degree of hand finishing in most cases. A classic example of this is Jim King's Smokey Mountain Model Works range of kits. Jim produces rapid prototypes for a living, and has used his knowledge and skill to produce patterns or casting, but most of these patterns are subject to a lot of personal attention, prior to moulds being taken off them for casting in resin. The kits are superb: surely there is a lesson there for the rest of us?

 

Please note that I am talking about the brass parts offered by Shapeways. This is one-off brass castings made from 3D printed wax masters. The brass in the castings are easy to drill and machine. So for driver centers in 0-scale where the outer diameter will be machined along with axle holes and counterweight faces, I feel this is a route worth exploring. It depends a bit on the specific drivers, if they have fine detail like lettering  you are likely to loos the finer detail when sandblasting the parts prior to machining.

[Hawk]

Well my Caledonian dock tank wheels have been completed - brass CNC milled centres and steel tyres turned,  fitted and profiled on the Cowells.

 

498Wheel54.jpg

 

Really beautiful wheels!

What sort of CNC mill did you use for those, and what type of brass?

[Hawk]

Interesting: what is the resolution of the wax masters?

Personally, I would rather have the masters for checking before casting, but if the surface finish is good, then there is no need.

 

Can you provide close-up photos?

I have tried to find out what type of wax-printer shapeways uses, and what the resolution is. Unfortunaltely, they seem to be a bit secretive regarding this type of information.

 

I have not yet tried to use the process for drivers, but I made some coupler pockets with it. Here is a close up:

 

 

The square hole in the part is 2mm X 2mm. You can clearly see the annoying pattern. My hope is that sandblasting will remove it. On this part this was not an option, as I felt sandblasting would remove sharpness. 

 

It is quite irritating that Shapeways wax prints is not quite state of the art, I have seen better results using wax prints from Solidscape printers. 

 

If I needed drivers for several engines, I would consider having a silver master made by Precision Wax as they use the Solidscape printer, and then use a casting company for the actual castings. 

[flubrush]

Really beautiful wheels!

What sort of CNC mill did you use for those, and what type of brass?

 

The milling machine is a Seig KX1 with Mach3 controlling it.  The brass I use starts off as clock plate - CZ120 grade.  

 

 

I prefer to use sheet as I can hold the part more easily on the mill and can machine the whole part, just leaving small tabs to keep it in the sheet which then have to be cut through to release the part.

 

The tyres are machined from free cutting steel blanks which I obtained from the SSMRS parts table at a meeting a year or two ago.  John Holden,  the parts officer at the time,  had got a box full of them from someone in the past and I literally bit his hand off to acquire them - saves all the hassle of boring and parting off from solid bar.

 

Jim.

 

Jim.

Edited April 20, 2014 by flubrush

[Hawk]

The milling machine is a Seig KX1 with Mach3 controlling it.  The brass I use starts off as clock plate - CZ120 grade.  

 

498Wheel19.jpg

 

I prefer to use sheet as I can hold the part more easily on the mill and can machine the whole part, just leaving small tabs to keep it in the sheet which then have to be cut through to release the part.

 

The tyres are machined from free cutting steel blanks which I obtained from the SSMRS parts table at a meeting a year or two ago.  John Holden,  the parts officer at the time,  had got a box full of them from someone in the past and I literally bit his hand off to acquire them - saves all the hassle of boring and parting off from solid bar.

 

Jim.

 

Jim.

 

Thanks a lot for the information!

 

I thought about milling the driver centers, but I am afraid that the High-Z CNC router I have access to is not up to the task. It protests loudly when I try to mill brass with it...

[flubrush]

Thanks a lot for the information!

 

I thought about milling the driver centers, but I am afraid that the High-Z CNC router I have access to is not up to the task. It protests loudly when I try to mill brass with it...

 

I am using very small diameter carbide cutters to do the work - typically 1mm and 0.5mm diameter - so I have to go easy to avoid breakage since my spindle top speed is 7000rpm.  I am using 30mm/min at 0.05mm depth of cut at 6000rpm for the 0.5mm mill and 50mm/min at 0.1mm depth of cut at 6000rpm for the 1mm cutter.   These are very light loads for the KX1 and might be OK for your router.   I had been using higher feeds originally - typically 120mm/min - 180mm/min,  but cutter breakage got a bit too high for my pocket and I've got a pile of broken shanks to prove it.   It means that it takes a while to machine one wheel but \at least you can get on with other things while the mill is getting on with it. :-)

 

Jim.

 

Jim.

[Hawk]

I am using very small diameter carbide cutters to do the work - typically 1mm and 0.5mm diameter - so I have to go easy to avoid breakage since my spindle top speed is 7000rpm.  I am using 30mm/min at 0.05mm depth of cut at 6000rpm for the 0.5mm mill and 50mm/min at 0.1mm depth of cut at 6000rpm for the 1mm cutter.   These are very light loads for the KX1 and might be OK for your router.   I had been using higher feeds originally - typically 120mm/min - 180mm/min,  but cutter breakage got a bit too high for my pocket and I've got a pile of broken shanks to prove it.   It means that it takes a while to machine one wheel but \at least you can get on with other things while the mill is getting on with it. :-)

 

Jim.

 

Thanks for the additional info! I have experimentet a bit with Feed/Speed on the High-Z, but do not feel that I have found the "Sweet spot" yet. It i always interesting to hear from people that have found settings that work.