North Sunderland stock

[davebradwell]

David, I've read through your very thorough notes on the chassis construction and can't help thinking that your vernier, adjusted to the required setting and locked would be a direct replacement for your carefully made hornblock gauge. After clamping the hornblock you might do a crafty check of the dimension over the outside before soldering.

Still pondering on how best to locate the brass cranks in the plastic wheel as, of course, the axle holes will be slightly different sizes. Not sure about your cocktail stick but must admit I can't come up with anything better unless you assemble them while pressing on axle which would probably make filling the boss very awkward.

As for size of drills, there's only one thing that's certain - the drill can't make a hole smaller than itself but you might be amazed how large a drill can cut when the wind's blowing the wrong way. Remember, also, that the second decimal place on your vernier is subject to some uncertainty - about +/-2 at very best.

It's looking like a fine thing, though.

DaveB

[Daddyman]

Thanks for those thoughts, Dave. You're of course right about using the vernier instead of the gauges, but it (the vernier) makes just one more thing to hold - and one more hand than I've got (one for the vernier, one for the hornguide, one for the iron). I've never found anything that holds the piece to be soldered as well as the human hand, not even an RSU probe. Everything I've tried either shifts during clamping, or obscures the view so that I can't see whether things have shifted. Keen to be enlightened on this. But in the absence of that, being able to tack-solder the gauges to the frames frees up a hand.

Point taken on the drills. But I've always thought broaching was a no-no - is it?

Ensuring centrality when fixing the cranks is tricky, yes. I just have to rely on that 6 minutes' grace with Devcon and magnification - or use the axle. I glued two cranks with the wheels on the axle, and the Devcon still allows the wheels to be turned on the axle for quartering. Things are looking good, but I won't know if it all works until I get the chassis set up.

But that's been is being delayed by my trying to work out a torque reaction set-up. The gearbox is horizontal, and the motor vertical in the firebox. I've looked at the Clag site, and Mike Clark showed me some of his designs the other night, but nothing quite fits my application. After staring into space for most of Thursday (and for a year on the Y7), I came up with the idea below - no idea if it will work in practice. I think the 10-thou vertical tag on the gearbox is able to flex too much at present, so I will need to lower the "slotted arch". But I'm worried the motor will be able to chatter against the "arch". But I can't put it any further forward as ahead of that it's all boiler, and within the frames it's all inside motion. If it went aft of the motor there'd still be little clearance within the firebox. Has anybody ever just trapped the motor within vertical "walls" for and aft? There'd have to be clearance for it to slide up and down, obviously, and perhaps some sort of thin padding within the walls.

20230430_101545.jpg

[davebradwell]

Clamping hornguides while they are tweaked to the correct position is a challenge as the clamps prevent access for measuring and soldering. I might be inclined to try clamping it all to the bed of your drill-thingy on a piece of MDF or Tufnol using a proper clamp set in the T slots. Technique would be to measure, then tweak position by twisting a screwdriver blade against hornguide - we used to set mirrors very accurately this way. Tweak and measure - it may take some time. You'll see why I use a more direct approach with the mill. Spent much of my working life eliminating the need to adjust things.

The vertical tab in a slot is about the best way I've found for restraining gearboxes as it allows the appropriate freedoms, in particular for the axle to rock. Ideally it should be vertically above the axle but sometimes this just isn't possible. I've gone on about it on this forum several times. I make my slot with a bridge shape of 0.45 fret waste about 2mm wide. The slot is formed with a staple of wire soldered to this strip. To avoid rattle I line both sides with slit ptfe tube as used when modifying pc boards so is available from electronic places. Yes, cradling the motor should work just as well. I usually make the tab as long as can be fitted within the body.

I trust your chassis will have a cylinder rear face, motion bracket and weighshaft between the frames.

DaveB

[Daddyman]

Thank you, Dave - all clear.

I trust your chassis will have a cylinder rear face, motion bracket and weighshaft between the frames.

Yes, and a cylinder front!

[Daddyman]

Some progress to report - cab roof, cab, and boiler/smokebox/tank units not attached to each other at the mo:

20230501_170756.jpg

20230501_171018.jpg

Typically, I bolt rather than solder as many sub-assemblies to the footplate as possible, using baseplates for things like cabs or cab/tanks, where appropriate. On this loco, footplate, cab/bunker, tank top, tank base/boiler bottom and smokebox are all separate assemblies - especially important if I'm going to be the one lining it! This photo shows my standard baseplate (15 thou) and holes drilled for something-BA nuts and bolts (14 BA?) while the baseplate is tack-soldered to the footplate:

20230430_120122.jpg

All very clever, but it never works - I've always had to ovalise the holes in the footplate!

This is the cab from underneath, showing the hollow firebox, which will house the motor. I made the firebox in two pieces - the main part and the "slice" in front as I didn't fancy trying to cut the aperture in the cab front for a one-piece firebox to fit through; this way, the aperture can be finished off after the fact, trimmed to the firebox. I filed up both ends to the firebox tacked together then flipped them and tacked them again to make sure they were exactly the same shape and the curve even. Cosmetic backhead will come later, probably in plastic.

20230502_123440.jpg

I was sick of it by Sunday night so made a start on the second model of the NER saloon. The one upthread is in 1951 banjaxed condition, but for Bamburgh's lifespan it needs to be a bit less run-down, carry different lettering, and also have lamp tops on the roof - hence the second kit. It still has the door mods, though:

20230502_173503.jpg

Problems
I thought I was out of the woods with the splashers, having got their height worked out. Turns out it's width that's going to be the problem. The MRJ drawing of Bamburgh (MRJ 8) shows the outboard edge of the real thing as being 11mm from the loco centre line. However, my wheels are 21.80 over, and my splashers are 0.8mm from their rear faces to their front edges (because of the inset character of the splasher faces). So that means I need at least 23.6mm, and that's not allowing any clearance between wheel face and splasher rear. How much clearance should I allow? Clearly, the designer of the RT Models K Class kit from MRJ 272 and 273 felt the need to move the splashers way outboard of where they should be:

20230502_171746.jpg

Any thoughts?

I'm also having a crisis of faith with my wheels... I think the tyre is too thick. I keep telling myself that the preserved Ls have thick tyres, and that Bamburgh's only appear thin because they're lined black-white-green. But I'm not convinced. I'm thinking of doing it all again with 3'8" tender wheels as opposed to 3'6", and getting Jeremy with his lathe to turn them down to 3'6", the excess coming off the tyres. I'd be very happy to be talked out of this....

[bécasse]

I have tried creating a scale drawing of the wheels over the side on image of the prototype reproduced above. On the assumption that the wheel diameter (i.e. over the tyre) is 3' - 6", the drawing produces a diameter inside the tyre of exactly 3' - 0", a flange depth of a little over one inch and a central boss diameter of perhaps 11⅜".

BamburghWheels.png

[Daddyman]

That's very helpful, thank you - and confirms my suspicions: the tyres on the model seem to be about 1.2mm so 4 inches-ish. And the wheel inside the tyre is 11mm-ish. My plan of taking 3'8" wheels down to 3'6" would lose me the requisite inch on the tyre radius and give me a bigger centre... Jeremy, are you there?

[davebradwell]

Before leaping, it's worth considering that the wheels in the photo may be worn by an inch or more. Suggest try scaling from the wheelbase first and see where that puts your sums. You might get as near by skimming the ones you have.

You can help the splasher clearance issue by thinning wheels to 1.85 which is more like the prototype figure (for main line engines!).

DaveB

[TonyMont]

I must say your loco looks first class, very clean work. A big thumbs up from me.
Regards Tony.

[bécasse]

Scaling up from the wheels in my superimposed drawing, the wheelbase comes out at almost exactly 10' - 9", so we can assume that the wheels in the photograph hadn't worn down - and that is logical because 3" was typical for such a dimension with unworn tyres.

[Daddyman]

Before leaping, it's worth considering that the wheels in the photo may be worn by an inch or more. Suggest try scaling from the wheelbase first and see where that puts your sums. You might get as near by skimming the ones you have.

You can help the splasher clearance issue by thinning wheels to 1.85 which is more like the prototype figure (for main line engines!).

DaveB

I hear you, Dave, and these are the kinds of things I've been telling myself for a few weeks. But bécasse's figures only confirm what I'd been feeling - my wheels are Wrong. It's not that hard to start again. And it would have the advantage of letting me lengthen the crankpin bushes so that I could thicken up the coupling rods to 3 ply, giving 2 ply at the joint - for reasons you have gone into with me (I forgot to do it first time round). While getting the tyres turned down I could also get them thinned to 1.85 as you suggest. What would you allow for clearance inside the splashers?

I must say your loco looks first class, very clean work. A big thumbs up from me.
Regards Tony.

Why, thank you!

[davebradwell]

Yes, I've gone through the dimensions and it does look as if the 3-8 wheels would give you a more accurate wheel rim. Bamburgh seem to have worn 3-5 wheels with a rim inside dia of 35 and a bit inches which scales to 11.8mm. I don't have any Gibson 3-8 wheels but Ultrascales come out at 11.4. Gibson 3-8 tyres have a main bore of 12.7 so could be turned to 3-5 with care.

Splasher clearance. Half a mm will look pretty close but before you can start heading that way you have to face some realities. How symmetrical are your body and chassis - are the wheels spaced the same from the frames each side? Crucially, how much sidepaly is there? Remeber the axle can rock so this won't help. When you fix body to chassis how accurate is the location or can it just drift about on the screws? You'll see that to get narrow splashers you have to tighten up the whole build.

DaveB

[Daddyman]

Yes, I've gone through the dimensions and it does look as if the 3-8 wheels would give you a more accurate wheel rim. Bamburgh seem to have worn 3-5 wheels with a rim inside dia of 35 and a bit inches which scales to 11.8mm. I don't have any Gibson 3-8 wheels but Ultrascales come out at 11.4. Gibson 3-8 tyres have a main bore of 12.7 so could be turned to 3-5 with care.

Splasher clearance. Half a mm will look pretty close but before you can start heading that way you have to face some realities. How symmetrical are your body and chassis - are the wheels spaced the same from the frames each side? Crucially, how much sidepaly is there? Remeber the axle can rock so this won't help. When you fix body to chassis how accurate is the location or can it just drift about on the screws? You'll see that to get narrow splashers you have to tighten up the whole build.

DaveB

Thanks, Dave. Gibson inside rims are 11.8, though I don't have any bogie/tender wheels at the moment, only drivers.

Thanks too for the reminders for the continuing need for tightness. I'm planning a gauge to get the splashers spaced equally off the footplate edge. I can create gauges to set some spacers accurately under the footplate, to go between valance rears and frames, so that the frames are held absolutely central; they'll also help position the valances consistently off the footplate edge, which is a job I've been dreading.

I don't know about sideplay yet - nothing on the outer axles and 0.5 (total or each side?) seem to be figures bandied about. So that would mean 21.5mm over the wheels (if turned down to 1.85), plus 0.5mm for sideplay, plus 1mm for clearance, plus 1.6 splasher rear face to splasher front face. Looks like 24.6 is the best I can do, leaving 1.7mm each side between footplate edge and outer edge of splasher, rather than the prototype's 3mm. Disappointing... The "walls" of the splasher faces are 0.5mm thick - I suppose I could shave and save something there, and I might get the footplate-to-splasher dimension out to 2mm, but it would probably be high risk. But then I have 8 splashers - Jeremy provided some spares - and only 4 need be life-or-death (2 are almost hidden in the cab).

[davebradwell]

You shoud be able to chip away at those numbers but your zero sideplay will be an underestimate but will probably avoid the need for extra play on the centre axle. Perhaps aim to halve the error.

DaveB

[Daddyman]

You shoud be able to chip away at those numbers but your zero sideplay will be an underestimate but will probably avoid the need for extra play on the centre axle. Perhaps aim to halve the error.

DaveB

Thanks, Dave. I'll do that.

3'8" bogie wheels ordered from Gibson, so it's happening...

[Will L]

...I don't know about sideplay yet - nothing on the outer axles and 0.5 (total or each side?) seem to be figures bandied about. ..

Actually, given the short wheelbase of this loco there may not be much need for side play on the centre axle. The following graph was produced to illustrate a discussion on the need for Gauge Widening, but it can also be used to illustrate if you need side play on a centre axle to go round a curve of a given radius (given the usual P4 Standards for BtoB - these figures were calculated using the maximum P4 BtoB {17.75}) The bottom axis in the radius of curvature and all the curves deviating from it show the point when a loco of a given wheelbase and no centre axle side play would start to need gauge widening (or centre axle side play) to get round it (everything else being equal)

[Daddyman]

...I don't know about sideplay yet - nothing on the outer axles and 0.5 (total or each side?) seem to be figures bandied about. ..

Actually, given the short wheelbase of this loco there may not be much need for side play on the centre axle. The following graph was produced to illustrate a discussion on the need for Gauge Widening, but it can also be used to illustrate if you need side play on a centre axle to go round a curve of a given radius (given the usual P4 Standards for BtoB - these figures were calculated using the maximum P4 BtoB {17.75}) The bottom axis in the radius of curvature and all the curves deviating from it show the point when a loco of a given wheelbase and no centre axle side play would start to need gauge widening (or centre axle side play) to get round it (everything else being equal)

Thank you, Will. I'll study that carefully.

[bécasse]

Doesn't wheel diameter come into that equation too? Surely the smaller the diameter of the wheel the shorter the length of flange which lies below rail head level - and therefore potentially requires gauge widening and/or increased sideplay.

[Julian Roberts]

There's quite a bit of play within P4 track, without any added centre wheel sideplay, if things are really getting down to fractions of a mm on the splashers! You can shave vital hundredths by going for 17.67 back to back and no wider. Then supposing you have the widest flanges in the spec, 0.4, that gives you 18.47 over the flanges, leaving 0.36 of play, 0.18 each side. You might have the narrowest flanges (0.35) so that's 0.23 each side.

Of course that's without the wheel diameter and flange length below rail level taken into consideration - but these wheels are tiny are they not...

Obviously depends what your minimum radius is, but if it's a challenge you can always put in prototypical gauge widening which is that tiny bit more than the triangular gauge gives...

Not wishing to resurrect an old argument, but here is the table. The first two columns give radius in feet, then mm, the blue column in theory what the triangular tool gives, and the last column the prototype amount. Have you got those handy little track gauges that give 0.1, 0.2 and even 0.3 widening? Not sure they're available now.

I'd say it's easier to adjust the track (if necessary) than these splashers!

[davebradwell]

I don't think the diameter of the wheel applies here as the flange tapers away from the rail so isn't touching anything. Once the flange tip touches the rail I believe that's a minimum radius condition and it will climb and you're done for - wheel diameter would certainly come in here. Recall Russ Elliot did a paper on this in Snooze a very long time ago. There's since been some troubling stuff with a square flange which we don't have and therefore isn't relevant - one of a number of attempts to prove P4 doesn't work. It doesn't with square flanges!

DaveB

[Julian Roberts]

Oh yes of course, Dave; I was muddling up with the vital part the wheel diameter + length of flange below the rail plays in checking on a diamond crossing.

Obviously it' s the wheelbase and radius that defines the play needed - here's a Russ snippet

[Will L]

Doesn't wheel diameter come into that equation too? Surely the smaller the diameter of the wheel the shorter the length of flange which lies below rail head level - and therefore potentially requires gauge widening and/or increased sideplay.

We have had that discussion and you can read all about it on the Track Gauge Widening, All You’ll Ever Need to Know thread which is where that graph came from. Read on to the end of page 1 and page 2 and you will see we got round to wondering about the implications of flange contact the rail side on curves. And the answer is not unless you're planning to run models of 30ft wheel base vehicles round 3ft radius curves. Don't think that applies to David's little loco.

[davebradwell]

Will, you might like to find the article I'm referring to - it's about minimum radius for 4w vehicles. It wasn't long after Russ's articles on stability of compensated rolling stock so a long time ago. I'm pretty sure that contact of flange tip with rail was regarded as terminal. It's also a reminder that flange shape is important.

DaveB

[Will L]

Will, you might like to find the article I'm referring to - it's about minimum radius for 4w vehicles. It wasn't long after Russ's articles on stability of compensated rolling stock so a long time ago. I'm pretty sure that contact of flange tip with rail was regarded as terminal. It's also a reminder that flange shape is important.

You mean this one Dave? It was well picked over in the "All You Need To Know" thread referred to above and leads to the to the conclusion about 30ft vehicles quoted above. The article is short a useful graphs which can be found on Keith's web pages http://www.norgrove.me.uk/history_files/Mar73/Diagm.gif. Anybody wanting to know how to use it should read the thread.

[davebradwell]

It isn't the one I remember, Will but it has a similar message. Tried to find the other using Index but no show. Yours fits in with what's required for this thread although the diagrams are rather difficult to follow - wheel diameter doesn't matter until it does! Even a 2ft radius won't trouble the wheel but I haven't checked the versine and it would look awful, anyway.

We'll leave David to juggle his numbers - splasher fronts would be high on my target list.

DaveB