Mickoo's BR modelling

[mickoo]

I don't want to dilute Steves thread any more than necessary so just one final question, does anyone have a rough idea of spring wire needed for 7mm locos, drivers, pony and tender, appreciate weight is an over riding factor but for this situation lets just assume average weight, oh and a source where such spring wire can be acquired. I determined to get at least something to push along the tracks next time I'm at Sudbury!.

To prevent further issues elsewhere I've opened this up in my thread where y'all can pitch in with what ever you wish to add. CSB seems popular with some around here so any feedback, positive or negative is greatly received.

In other news, ordered the bogie and tender wheels and with luck the round bending bar set from Metal Smith to make sure the tender cant rail and front handrail turn ins are uniform and to the correct radius, one of my pet hates is long sections of curvature that are not of uniform radius or have creases in.

Kindest

 

[JimG]

I don't want to dilute Steves thread any more than necessary so just one final question, does anyone have a rough idea of spring wire needed for 7mm locos, drivers, pony and tender, appreciate weight is an over riding factor but for this situation lets just assume average weight, oh and a source where such spring wire can be acquired. I determined to get at least something to push along the tracks next time I'm at Sudbury!.

Mick,

I missed your re-direction so see my very recent message on Steve's thread with URLs to spring steel wire sources. As Steph has indicated, Russ Elliot gives spreadsheets, graphs and formulae on the CLAG web site which work out fairly accurate as I've found in my own experience. And fine tuning is quite easy by changing wire gauge for larger or smaller gauges provided you build easy changing into your design.

Jim.

[Edit] The link to Steve's thread is here.

 

[mickoo]

Jim,

Thanks, I follow the theory, just but fail to understand why the fulcrum points between the axles are not equi distance from each axle? All the examples leave the intermediate axle lighter loaded, I presume there must be a reason for this but as an engineer I'd of thought you'd want all axles to have the same weighting.

I'll take Steves process and use handrail knobs as the fulcrum points and the spring will run either direct on the axle or on the slaters horn guides, need to work those issues out when the parts arrive.

 

[mickoo]

Right, just done some basics here and possibly running into problems, assuming a weight of 300g for the tender I'm already up to 20thou if I maintain a 0.5mm deflection. The kit alone in the box, with no motor or wheels etc weights 1.925Kg, so I think 300g for the tender may be too low, in which case I'm going to run out of wire gauge, from what I can see?

 

[JimG]

Right, just done some basics here and possibly running into problems, assuming a weight of 300g for the tender I'm already up to 20thou if I maintain a 0.5mm deflection. The kit alone in the box, with no motor or wheels etc weights 1.925Kg, so I think 300g for the tender may be too low, in which case I'm going to run out of wire gauge, from what I can see?

Mick,

If you need larger gauge wire, look for music wire, or piano wire. It is better to get it in straight lengths for springing. It is almost impossible to straighten a coil unless you put it under a lot of tension. I found one supplier with a fair range of gauges here.

Jim

 

[Steph Dale]

Thanks, I follow the theory, just but fail to understand why the fulcrum points between the axles are not equi distance from each axle? All the examples leave the intermediate axle lighter loaded, I presume there must be a reason for this but as an engineer I'd of thought you'd want all axles to have the same weighting.

If the fulcrum points are equidistant the centre axle carries double the load of the outer axles. Look at the CSB as a pair of levers running outer axle - fulcrum - middle axle. Each of the levers so described put half the force through the centre axle as they do the outer axle. With two beams in place all axles end up equally loaded. Get out those mechanics books and you'll see what I mean - moments of inertia will give you a method you can work through on paper if you need to see what's going on.

The centre axle will help to be more lightly loaded to reduce the tendancy for it to rock around the middle axle on rough track. I'd suggest something like a 10% variance in loading would be fine.

I'll take Steves process and use handrail knobs as the fulcrum points and the spring will run either direct on the axle or on the slaters horn guides, need to work those issues out when the parts arrive.

I suggest you fit a handrail knob to the top of the hornblock and use that - the theory (and spreadsheets) will fall over somewhat if you don't aim to have as small a contact at each point as possible.

Right, just done some basics here and possibly running into problems, assuming a weight of 300g for the tender I'm already up to 20thou if I maintain a 0.5mm deflection. The kit alone in the box, with no motor or wheels etc weights 1.925Kg, so I think 300g for the tender may be too low, in which case I'm going to run out of wire gauge, from what I can see?

Allow much greater deflection than that for normal running. In 0 I use something like 1 - 1.5mm to get stability and allow for any slop in the wire-handrail knob fit. For example, running a 0.5mm wire in an 0-gauge handrail knob you'd lose 0.5mm in height even if the wire was perfectly rigid.

All of this is actually in the CLAG pages. It'll be worth you having a careful read through of all them - and there are a few. It's also probably worth building a simple 0-4-0 or 0-6-0 chassis as a test. The issues aren't all that easy to describe, but it may help to have a real-world example to work through. There are a lot of clearances to work out in the frames, such as where to drill spacers, vertical clearances and whether you need to put the end fulcrum somewhere in front of the buffer beam.

After all the complications it can be worthwhile though. One example is that it's possible to put your locos centre of mass exactly where it is in the real thing, which will help give a realistic appearance to its running.

Just checking one thing though - you have built a conventional compensated or sprung chassis before haven't you?

Steph

 

[adrian]

I'm building my 4MT with springy beams, http://cherryclan.com/locos/standard-4/chassis.html .

I've used the CLAG spreadsheet to calculate deflections and wire gauge etc. and ended up with, under load, 0.9mm deflection on the outer axles and 1mm on the centre axles using 20swg spring wire.

Unfortunately the guitar/piano wire sources didn't seem to have any suitable wire, most wire was too thin.
I ended getting mine from Folkestone Engineering Supplies. http://feskent.co.uk/cataloguemenu.htm . Although other model engineering suppliers may stock spring wire.

Not much to add to what has already been said, although for the fixed fulcrum I've sometimes adjusted the settings to match the frame spacers, or moved the frame spacers slightly and just drilled a hole in the frame spacer to act as a fulcrum point. I would get too concerned about getting the precise wire gauge right. One of the nice things about the system is that you can swap wires quite quickly so it is easy to adjust the spring rate. My first effort was a Jinty
http://cherryclan.com/locos/jinty/chassis.html . Initially I used a wire spring too small, it worked fine on the chassis but by the time I'd finished the body it became apparent the springing was too soft. It was a 10 minute job to replace the wire with a thicker gauge.

 

[mickoo]

Steph, I was actually thinking of four fulcrums, two outers and two between outers and intermediate, four fixed hand rail knobs and three moving, that way the centre axle should be presented with 1/3rd of the load and A J Turners spreadsheet confirms this. Ok on the center axle being 10% less, makes perfect sense.

Deflection, cheers that's the sort of info I was looking for. I hadn't thought of handrail knobs for the axle boxes, was just going to round off their tops so that there was only one point of contact and not two from each corner, then file a small groove to prevent the wire from moving laterally, but hand rail knobs would be a better 'engineering' solution.

CLAG, yes it is, but I've read most of it twice now and am now having to sit in a darkened room to help my headache go away LOL, the first 'experiment' is the Jubilee 4000gal tender, no traction loads, no motor, just three axles all of uniform spacing/ weighting and frame spacers can be put where ever to avoid the CSB, simples! Yeah right LOL, it does have dummy inside frames so if it all cocks up then new ones can be easily made up. Centre of mass will be in the next 'project' of the Jubilee main loco frames where bogie springing will come into play...uuuggh.

Compensated yes, sprung no, which is why I'm going down this road of pain ATM

Kindest

 

[mickoo]

I'm building my 4MT with springy beams, http://cherryclan.com/locos/standard-4/chassis.html .

I've used the CLAG spreadsheet to calculate deflections and wire gauge etc. and ended up with, under load, 0.9mm deflection on the outer axles and 1mm on the centre axles using 20swg spring wire.

Unfortunately the guitar/piano wire sources didn't seem to have any suitable wire, most wire was too thin.
I ended getting mine from Folkestone Engineering Supplies. http://feskent.co.uk/cataloguemenu.htm . Although other model engineering suppliers may stock spring wire.

Not much to add to what has already been said, although for the fixed fulcrum I've sometimes adjusted the settings to match the frame spacers, or moved the frame spacers slightly and just drilled a hole in the frame spacer to act as a fulcrum point. I would get too concerned about getting the precise wire gauge right. One of the nice things about the system is that you can swap wires quite quickly so it is easy to adjust the spring rate. My first effort was a Jinty
http://cherryclan.com/locos/jinty/chassis.html . Initially I used a wire spring too small, it worked fine on the chassis but by the time I'd finished the body it became apparent the springing was too soft. It was a 10 minute job to replace the wire with a thicker gauge.

Adrain, yes I'd seen that....after researching ball races for locos etc, though might have another solution to attaching ball races to slaters axle boxes, wont work if you can see inside the frames, for hidden axle boxes it might have some mileage.

Anyway, I'd seen your work but no reference to what wire you ended up with and what weights you'd input for the 4MT. Now I have some ball park figures to aim for I'm sure the rest will drop into place, as you say changing the wire should be simple enough if it all goes wrong. I've punched in your values but cannot get the deflections you note unless I put a really high weight in, obviously I'm still missing the plot

After more math I'm looking at between 300 & 500g for the tender which with 17thou (27swg) will give a deflection of between 1.1 & 1.6, unless I'm still all to cock.

Kindest

 

[adrian]

might have another solution to attaching ball races to slaters axle boxes, wont work if you can see inside the frames, for hidden axle boxes it might have some mileage.

Now that I'd be interested in.

Anyway, I'd seen your work but no reference to what wire you ended up with and what weights you'd input for the 4MT. Now I have some ball park figures to aim for I'm sure the rest will drop into place, as you say changing the wire should be simple enough if it all goes wrong. I've punched in your values but cannot get the deflections you note unless I put a really high weight in, obviously I'm still missing the plot

After more math I'm looking at between 300 & 500g for the tender which with 17thou (27swg) will give a deflection of between 1.1 & 1.6, unless I'm still all to cock.

You're probably about right there, the 4MT is a heavy loco as there are a lot of brass castings. I weighted the castings alone at 1.6kgs, so I was reckoning close to 2kg by the time I've added all the nickel platework. Anyway I ended up estimating 650g per axle which is probably too high. However I thought I'd get the loco finished so I can get an accurate weight before worrying too much about the wire diameter.

 

[mickoo]

Adrian, hmm still don't get your values LOL, but your right, wire size can be determined at a later date once its all weighed in.

Bearings, ok a quick whizz in Max8, actually took longer to remember how to use Max than to do the bits LOL, I think the images are self explanatory but just for completeness you solder a brass tube whose ID matches the bearing, you file one wall away to give more solder surface if possible, I've got the bearings already but not the slaters axle boxes so it's all theory ATM. You pass the axle through the bearing and axle box, solder on the sleeve, then withdraw the axle and bearing and then ream the axle box to give clearance so the bearing takes the load, alternative you could ream the axle box and then sleeve the axle to align everything. Whether it will work with heavy loads I don't know or if its even practicable but the theory has some mileage and if it does work can be retrofitted to locos providing there's enough clearance inside the frames.

 

[Steph Dale]

Mick,
That looks workable. FYI IIRC FR156zz are 5/16" od, the Slater's square bearings are 8mm af, so that's pretty much a match...
Steph

 

[mickoo]

Mick,
That looks workable. FYI IIRC FR156zz are 5/16" od, the Slater's square bearings are 8mm af, so that's pretty much a match...
Steph

Steph, Indeed I have ten FR156zz here on my desk after your advice last year LOL.

I didn't know the bearing af but looking at Adrians blog it looked feasable, the only issue might be a twisting force on the axle box in the guides as the bearing is quiet a way inside the frames, you may be able to thin the axle box and get the bearing closer? By retaining the top and bottom 'fingers' on the brass tube you might be able to go quiet a way onto the axle box, even right up to the front face, so long as the frame is cleared above the bearing combination at full deflection.

Of course I could now render this in exact scale and see how much tube wall is present to create the 'fingers' and you may be able to retain the side ones as well, I cut them off as it looked too difficult to cut where the corners would be in practice.

Thinned axle box with extended fingers to give more support


Four fingers to give even more support


Four fingered collar split to show problems machining to give four good tight fitting, a mill would be about the only sure way to achieve this, two fingers could be done with vice and file + patience.


I've also amended the fingers to reflect the adjusted ID to match the af of the bearing.

Kindest

 

[Steph Dale]

Mick,
I agree with your thoughts above, but on refection I wonder if there isn't a simpler solution;
If the front of the square bearing was drilled 5/16" the flange on the ball race could act as the front flange on the square bearing. Less machining and greater stability in the finished model as the bearing surfaces would be further apart.
Still doesn't solve the sideplay issue though...
Steph

 

[mickoo]

Steph,
I think I see where your coming from, but as Adrian has found drilling the bearing can be very tricky and I'm not sure what the distance is between the rebates that fit in the frame guide, you could end up with no wall there at all. Judging by the photos on Adrians blog there is just enough meat to bore the inside face of the bearing to accept the 5/16th dia ball race and that bore depth stops short of the frame guide recesses as it is to great for that area, does that make sense or have I misread your comments?

Side play could be an issue as could springing which depending on where you put the bearing fulcrum could induce a force on the axle box that may twist is, the spring fulcrum would need to be over the ball race and not the original axle box. Not sure on side play as yet, hence my questions elsewhere, I'm hoping there will be enough play in the frame guides and axle box across all three axles to allow transition through most point work.





Kindest

 

[28ten]

My slightly luddite method is to use equidistant fulcrum points and buy lots of piano wire to play about with

I cant help thinking that rather than machining bits for the hornblock, would it not be easier just to make a new one slightly larger? fewer parts equals fewer errors in my world I guess im just saying dont over engineer it and having built several CSB locos I would second Stephs suggestion to try an 0-6-0 first as a test piece

 

[mickoo]

My slightly luddite method is to use equidistant fulcrum points and buy lots of piano wire to play about with

I cant help thinking that rather than machining bits for the hornblock, would it not be easier just to make a new one slightly larger? fewer parts equals fewer errors in my world I guess im just saying dont over engineer it and having built several CSB locos I would second Stephs suggestion to try an 0-6-0 first as a test piece

Yes and no, to make new horn blocks will require a mill, I don't have a mill so what ever solution is arrived at has to be accomplished by a lathe, hand tools or materials already suited for the task in hand, I did a simple drawing in another thread but will whizz up a 3D shortly of what I initially thought might work.

Thanks for the tips on the CSB, I am coming to that conclusion myself LOL but needed some basic ball park ideas first, it was your Mk1 bogie at Reading that sparked all this off LOL.

My first 'project' (post #68) is a six wheel tender, all with equi distance fulcrums and weightings, though with a planned 10% reduction on the intermediate weightings. No motors, no coupling rods, just three axles, how hard can it be

Addendum, the original plan to incorporate ball races, simply put, flat bar of appropriate size (Eileens Emporium F12018G 1/2" x1/8"...12.5mm wide and 3.4mm thick), bore hole of 5/16", add angle guides to inside of frames to suit flat bar width. Flat bar is a tight fit on bearing, bearing is a tight fit on axle. Note there is no reverse face to the axle box guide and thus side play is only controlled one way, the other axle box on the other side will control the amount of side play that way. Accurate placement of the ball race on the axle will fix the new flat bar and allow for intended side play.

That's about as simple as I can make the whole assembly without resorting to machine tools like mills and lathes etc.







The brown bit is the flat bar cut to size and bored 5/16th", the slot in the frames is also a fraction over 5/16th" but could be the same to give some longitudinal accuracy. The green sections are angle to stop the flat bar and thus the outer race rotating. The green knob is the fulcrum point.

Hope that makes sense.

Kindest

 

[adrian]

As already mentioned boring (drilling) the brass hornblock for the bearing was very border line, the brass was cracking in a couple of places. I'm glad it worked but it was more by good luck than machining ability. I'm not sure I'd do it in the same way next time, as the Guv' mentioned I think making new ones from scratch would be easier than messing about with existing ones.

My current thoughts are to use a bit of 3/8 brass square bar, drill through slightly over 3/16" to give a clearance to the axle. Then bore out at 5/16" to provide a friction fit for the bearing. However although the width of the bearing (less flange) is 3/32" I'd only bore out 1/16". Then press fit the bearing in will leave it slightly proud, the 1/32" gap between the bearing flange and the brass bar being the slot to go in the frames. Anyway that'd be the next project so just an idea at the moment. I wish I had the skills with drawing packages that other have as a quick drawing would have been much easier than trying to describe it. (I might have a try later but'll probably take me most of the evening.)

Anyway doesn't starting with the tender count as the 0-6-0 test piece?

Edit: took so long to write this Mick already got in the 0-6-0 test piece!

 

[adrian]

Yes and no, to make new horn blocks will require a mill, I don't have a mill so what ever solution is arrived at has to be accomplished by a lathe or hand tools

I think my virtual solution is feasible with out a milling machine. Besides if you've got a lathe add a vertical slide and you'll have a mill. That's my Christmas project, fitting a vertical slide to my mini-lathe.

 

[mickoo]

I think my virtual solution is feasible with out a milling machine. Besides if you've got a lathe add a vertical slide and you'll have a mill. That's my Christmas project, fitting a vertical slide to my mini-lathe.

Then I shall follow you very closely LOL, I've considered a vertical slide but its not quite the same as a proper mill is it LOL.

Ok I follow your ideas, similar to the amended above except the top hat is outside the frames and forms the other side of the frame guide, I'd probably drill right through at 5/16th" and make the bar a tight fit, one way to ensure a tight fit is to copy Slaters and make a small nick on the end in the brass where it slides over the bearing, or if interferance fit then loctite should suffice.

I had considered using the top hat as one side of the frame slot guide but was concerned that it might push the frames too far away from the drivers and pose issues with brake hangers and the such like. The top hat is only about 1mm thick, probably perfectly fine if you kept FS frames as the top hat would take up the slack from the S7 axles, not knowing how the kit is designed and how the brake hangers are means I'm guessing a bit here, that's the downside of no experience in these kits, a lot of best guess work going on with an eye on not boxing yourself in a corner at a later date.

I've whizzed it up quick in Max what I think your on about




The new flat bar will certainly place the bearing in line with the frames or very close this way round, the previous way places the bearing centreline more inboard.

Regarding drawing skills, you keep to your skills you already have LOL, trust me, you don't want to loose the amount of time it takes to learn these packages these days took me years to learn Max well enough and at great cost to my other activities, especially the family.

Kindest