7mm Mickoo's Commercial Workbench

[mickoo]

A new year and a new project.

Another can that needs kicking down the street is the B17 project, I need to have some tangible artifact for this years open day on the stand at CTL Seal.

I'll be honest, B17's never caught my eye but researching and collecting data over the last near year I've found they do get under your skin is a nice kind of way.

Anyway, using the 3D CAD model from 2D outline speeds things up the design no end and the bogie is all bar done; the bogie has been a sticking point information wise since the beginning so I've started there as the rest is bread and butter design wise. The Sandringham society have cleverly decided to engineer a locomotive for the future with several upgrades and alterations under the hood, as have the Clan group.

One of those is the bogie which will now follow very closely the B1 design and incorporate canon boxes for the axles and bearings. If you're going to model historical engines then this is not the best solution.

Fortunately, the B17 bogie is very close to the D49 and one of those does still exist so that plan is/was to get the basics all fleshed out and drawn up.

The bogie etch work is now drawn up (sans corrections and dust shields) and I've started to work on the brass casting masters and other parts, which could be 3D prints or resin from 3D masters.

Yellow will be brass castings and red undecided, I do need to revisit the side control unit and make sure it's profile is as good as information allows, I've also currently streamlined some aspects in that area for ease of building but might revisit later.

Over the weekend I'll draw up the chassis and get it all off to the etchers, whilst that's away I'll work up and brass casting masters I need but the test build will almost exclusively use 3D prints to validate construction and fit.

Until the etches arrive back you're just going to have to suffer with CAD renders I'm afraid but they should give a good idea of what coming ahead.

 

[Tim Humphreys ex Mudhen]

Hi Mick,

Have you given any thought to making it possible for the bogie to be built with working axleboxes in order to provide suspension?

all the best

Tim

 

[mickoo]

Hi Mick,

Have you given any thought to making it possible for the bogie to be built with working axleboxes in order to provide suspension?

all the best

Tim

Tim,

No, that falls under the 'lifes too short to worry about' mandate.

As far as I know no one else does it, so I'm not going to be the first to step off the Lunar lander on that aspect

 

[dibateg]

I've never found the need to spring/compensate bogies in Finescale, maybe in Scale 7 perhaps. I'm looking forward to following this project, I think the B17s had a Gresley elegance about them with that lovely sweep in the running plate under the cab. Mind you - reading Dick Hardy, they were a devil to work on and could be as rough as hell when run down.

Regards
Tony

 

[Hobbyhorse]

Hi Mick,

Have you given any thought to making it possible for the bogie to be built with working axleboxes in order to provide suspension?

all the best

Tim

Hi Tim, Interesting concept, although have used it on all my scratch builds I’ve found the befits to be negligible. My method is to just allow free movement within the horns, primarily not to take any weight away from the drivers, and in this case as it’s a single.
Simon

 

[simond]

I did it on my recent Manor build, in that I attached the bearings to the compensation beams and put coil springs in the shackle of the 3DP leaf springs, the bearings slide up & down in the bogie frame. It was fiddly, and I’m not sure it doesn’t jam occasionally, and I’m far from sure it’s of any benefit. Once I’ve had the chance to run the loco for a few laps, I’ll decide whether it jams or not and obviously fix it, or leave alone. It was fun trying, but not sure I’d repeat the experiment.

and it wasn’t as lovely as Simon’s bogie above.

 

[adrian]

I did it on my recent Manor build, in that I attached the bearings to the compensation beams and put coil springs in the shackle of the 3DP leaf springs, the bearings slide up & down in the bogie frame. It was fiddly,

I've nothing in the pipeline to try it at the moment but again - "air suspension" in the form of the miniature magnets might be easier to install and provide a little light springing to keep the wheels on track.

 

[Ian@StEnochs]

Tim,

No, that falls under the 'lifes too short to worry about' mandate.

As far as I know no one else does it, so I'm not going to be the first to step off the Lunar lander on that aspect

Finney7 Duchess kit has compensation designed into the bogie so precedence set.

 

[simond]

Ditto the MM / JLTRT King


But that’s not suspension…

 

[mickoo]

Ditto the MM / JLTRT King


But that’s not suspension…

MM1/JLTRT, Finney and others all use/used compensation beams or a longitudinal bar at one of the axle centres to form some sort of rudementary compensation.

None can be considered moving axle boxes in the sense of Tims OP

 

[mickoo]

Moving swiftly onward with the B17 project, the main frames and 90% of the NS 0.5 material is all fleshed out; the main frame overlays and front buffer beam overlay are the two last big pieces which I'll do tomorrow as welll as the J hangers, springs and stays that need casting. Still not sure on the conjugated lever support stay, possibly a merger of castings and fabricated etch work.

I also need to add some frame support at the front end in the smoke box saddle area, it'll be simple L stays as they'll all be hidden by the dummy inside cylinder block and steam/exhaust channel castings.

There's a few little nick nacks and brackets to add as well as locating cut outs for the J hanger castings etc; then it's Tetris time to get it all on the sheet, flood fill, tab and off to PPD.

The hornguide castings are place holders from the B1, they're very similar in many aspects and just need the mounting plate profile and fixing locations tweaked to suit the B17.

The original engines had two vacuum cylinders underneath and those trunnions will be cast, the new build only has one set of trunnions and I'm assuming it'll be steam braked or more than likely air braked, the distance between the bearings looks to great for a steam brake cylinder to be honest.

Yellow is brass, light blue is NS but part of the body and not the chassis.

 

[Tim Humphreys ex Mudhen]

MM1/JLTRT, Finney and others all use/used compensation beams or a longitudinal bar at one of the axle centres to form some sort of rudementary compensation.

None can be considered moving axle boxes in the sense of Tims OP

Hi Mick,

Sorry to continue this. On the MOK Standard Tank I built I cut the axle boxes out of the casting in order for them to move up and down.
Could the axle box be a separate casting to the horn guide/spring casting? The rest of any form of suspension could be left to the builder to devise.

Tim

 

[mickoo]

Hi Mick,

Sorry to continue this. On the MOK Standard Tank I built I cut the axle boxes out of the casting in order for them to move up and down.
Could the axle box be a separate casting to the horn guide/spring casting? The rest of any form of suspension could be left to the builder to devise.

Tim

Tim,

Oooer, sounds painful, the short answer is yes it could be done, the long answer is, no I won't

 

[mickoo]

That'll do for today, one last J hanger to draw up and sort out a way to affix the expansion link bracket and that's stage one complete.
Normal service is resumed tomorrow.....

 

[Big Train James]

Jim, in short no.

The inverse actually, I take the 2D outline and extrude to the material thickness and then flip and twist the parts to build a virtual model. I did that with the B1 and it found several errors I'd missed that Mick Roffe (test building) also confirmed as he was going along.

On the S15 it worked really well right off the bat, only a few small tweaks needed, like moving a hole 0.3 mm instead of just filing it which just as easily be done. For me it's a massive game changer and time saver, especially if you then migrate to instructions. I think the B1 in the traditional format (W1 I did a few years back) would have taken 3 months, in this format it was five weeks.

Once I'd got my head around doing that it became much easier/faster to do and the next logical step was to break down the parts and then make images for instructions. In those I tend to add 3D rivets and bolt heads etc, that soaks up a fair bit of time and then I import the 3D masters for castings to fill in the blanks.

In AutoCad you can get an outline from a 3D object but all the edges are exploded so you need to join them all up again, however it will only do outlines for planar objects or more correctly, only planar faces. I can see that being a problem if say for instance you've added a radius to edges, say the nose on an EMD spartan cab in the 3D model, it's only going to break out the flat face and not take into account that half of the radius needs adding on to get the full sheet size; make sense?

So you go the other way instead. I think it could work in either direction though, so I might see what I can do with a frame drawing.

I'll have to check behavior tomorrow when I have access to Autocad at work. Even if you need to join up edges after exploding a solid or copying one of its faces, that's a simple matter of a few clicks.

The point about being limited to planar faces is partly true. You can get the edges of curved surfaces as easily an any other, and you can join them all together as well. What you can't do, and I presume this is what you are getting at, is unroll that surface or its boundaries into a flat sheet suitable for etched artwork. Technically you can, but you would have to get an extension for that capability, it's not a native capability in Autocad itself.

I'm fortunate in that Rhino has a function to unroll surfaces. I've sort of migrated to Rhino as I no longer have Autocad at home (at least for now) and I'm limited for time at work to use it there. I'd much rather be using Autocad, but some planets are going to have to align for that to happen again with any regularity.

I'm confident Fusion has the capability to do at least simple flattening of curved surfaces, and Solidworks can also do it. I actually use several cad software packages because they each have some function that excels over the others, and it might benefit you to do the same with a free version of Fusion, or possibly a better option would be the hobbyist license for Solid Edge. You don't need to use it for most of your modeling, but it's another arrow in the quiver to get desired results. Right at the moment I mostly only use Solidworks for complex or variable radius filleting. Most of my other modeling gets done elsewhere first because I like a direct modeler and a command line.

Either way, having a hybrid approach utilizing both 2d and 3d elements, or deriving one from the other, offers many advantages.

Thanks for taking a few minutes to reply.
Jim

 

[Rob R]

Mick,
Wrong scale for me but I may be tempted......
B17 Tender:-
Group Standard or GER?
Or options for either when purchasing?
R

 

[mickoo]

Mick,
Wrong scale for me but I may be tempted......
B17 Tender:-
Group Standard or GER?
Or options for either when purchasing?
R

Rob,

Tenders are a future issue to resolve, ideally both but then Finney7 do a good GST tender so there's no point reinventing that wheel, not for steam braked engines; there maybe some milage in working up an earlier vacuum braked version but that's a long way off in the future.

The GER one is a more immediate priority and once the engine is all sorted then I'll start fleshing out the tender, I need it for some other projects later in the pipeline anyway as I do some NER tenders.

For clarity, this is not a Finney7 project.

 

[mickoo]

So you go the other way instead. I think it could work in either direction though, so I might see what I can do with a frame drawing.

I'll have to check behavior tomorrow when I have access to Autocad at work. Even if you need to join up edges after exploding a solid or copying one of its faces, that's a simple matter of a few clicks.

The point about being limited to planar faces is partly true. You can get the edges of curved surfaces as easily an any other, and you can join them all together as well. What you can't do, and I presume this is what you are getting at, is unroll that surface or its boundaries into a flat sheet suitable for etched artwork. Technically you can, but you would have to get an extension for that capability, it's not a native capability in Autocad itself.

I'm fortunate in that Rhino has a function to unroll surfaces. I've sort of migrated to Rhino as I no longer have Autocad at home (at least for now) and I'm limited for time at work to use it there. I'd much rather be using Autocad, but some planets are going to have to align for that to happen again with any regularity.

I'm confident Fusion has the capability to do at least simple flattening of curved surfaces, and Solidworks can also do it. I actually use several cad software packages because they each have some function that excels over the others, and it might benefit you to do the same with a free version of Fusion, or possibly a better option would be the hobbyist license for Solid Edge. You don't need to use it for most of your modeling, but it's another arrow in the quiver to get desired results. Right at the moment I mostly only use Solidworks for complex or variable radius filleting. Most of my other modeling gets done elsewhere first because I like a direct modeler and a command line.

Either way, having a hybrid approach utilizing both 2d and 3d elements, or deriving one from the other, offers many advantages.

Thanks for taking a few minutes to reply.
Jim

Jim,

In a nutshell what you write is about the level of it, I'm happy in AutoCAD and to be honest the variable fillet lack of function doesn't really bother me. The areas where I need it are quite small so by tthe time it's been printed and then perhaps cast it becomes a mute point as to whether it's even visible.

For more complex shapes with compound curves then, yes, you'd need something like Rhino or a more bespoke 3D centric software, however, even if you achieve that and the resulatant flat net shape is good to go, you still then need to make that flat shape you've just hived off the 3D model into a complex curved object, that'd probably be harder than getting the net flat dshape in the first place.

To my mind there are two clear paths, that which is flat (or flat that can be folded up to make a planar faced object) should be etched and drawn in the flat, that which has complex curves and shapes should be drawn in 3D and then straigth to output like that, aka 3DP or casting.

All the yellow parts on the model above were drawn as 3D, there are no 2D models or outlines for any of that.

At the end of the day I don't belive there is a right or wrong way, just variations on how people natively feel comfortable in achieving the end result.

 

[Big Train James]

To my mind there are two clear paths, that which is flat (or flat that can be folded up to make a planar faced object) should be etched and drawn in the flat, that which has complex curves and shapes should be drawn in 3D and then straigth to output like that, aka 3DP or casting.

100% agree with this perspective. Just because you can print everything doesn't mean you should. I think the ideal arrangement is much as you do, a hybrid system using whichever methods best suit the parts in question.

I was never suggesting that you would unfold or unroll complex parts like compound curves, rather things with what I'll call 2d folds or bends. The sides of a cylindrical hopper like the Centerflows would be a good example, where the cutouts at the end are easy to 3d model but not as easy to lay out in 2d. Conical boiler sections or the corners of EMD hoods are other good examples.

......to be honest the variable fillet lack of function doesn't really bother me.....

We can revisit this conversation when you start doing driver centers......

 

[mickoo]

100% agree with this perspective. Just because you can print everything doesn't mean you should. I think the ideal arrangement is much as you do, a hybrid system using whichever methods best suit the parts in question.

I was never suggesting that you would unfold or unroll complex parts like compound curves, rather things with what I'll call 2d folds or bends. The sides of a cylindrical hopper like the Centerflows would be a good example, where the cutouts at the end are easy to 3d model but not as easy to lay out in 2d. Conical boiler sections or the corners of EMD hoods are other good examples.


We can revisit this conversation when you start doing driver centers......

On going projects (that also need kicking down the street) which have had an opening for new processes, more later this year