Accepted side play in long wheelbase locos

mickoo

Western Thunderer
Hi guys, just going through the thought/research process for bearings on locos, mainly six coupled and larger, is there an accepted limit to side play on drivers or tender wheelbases to allow stock to pass through an average set of points etc, I'm talking normal station crossing angles, not dock side or small depots.

If so, how does one then translate that to locos fitted with ball race bearings which by nature require an interference fit on the axle and horn block and thus no side play. Are S7 tolerances enough to allow a six coupled loco to pass through complex track work where no side play exists?

By six coupled I'm looking at Pacifics and 4-6-0 types with their associated tenders.

Thanks in advance.
 

3 LINK

Western Thunderer
Hi Mick,

What I did on the last 2-8-0 I built was to have literally no side play in the first axle, as the tolerance between the slide bars and the coupling rods was around two fag papers width. The next axle I allowed as much side play as possible, the third axle was driven so no side play but saying that if you have enough room with certain gearboxes you can still "float" the axle. And the last axle was also allowed as much side play as possible, what you will have to do is use packing washers and it will be a lot of trial and error so to speak. Hope this is of some help.

ATB,

Martyn.
 

Dog Star

Western Thunderer
The simple and quick answer is to follow what the prototype did... as long as the frame and wheels are scale equivalents of the prototype and provided that the track has a scale equivalent equal to or greater than the "dead slow" figure shown on the relevant engine diagram then there is generally no trouble... although you can ease the situation by incorporating gauge widening (which is what was done on the prototype).

What will impact upon the result is the relationship between the Walshaert's gear, leading coupled wheelset, bogie control and prototype shaping of front of the frame plates.

If you use frame plates of scale thickness, set at scale separation, then the sideplay ought to be deducible from a
(you may have to estimate the thickness of the axlebox flanges).

Our A4 has b****r all, roughly the thickness of a fag paper, for the leading cpipled wheelset - for reason see my reply to your earlier query. There is about 1mm total clearance on the centre wheelset and a nat's whisker on the trailing coupled wheelset. The trailing wheelset has a radial axlebox with side control sprining. The bogie has about 4mm total side movement controlled by springs. At the moment, the frame negotiates something like B7 with prototype gauge widening throughout the closure rails.

regards, Graham
 

Dog Star

Western Thunderer
Martyn and I were typing at the same moment... clearly we have been to the same school where the engineering units are measured in Rizlas and calculations done on the back of the proverbial fag packet!
 

Steph Dale

Western Thunderer
Mick,
It's entirely possible to calculate out the sideplay required for any rail vehicle, it just depends on how good your trigonometry is! I tend to work through a sideplay calc for any loco I build in order to maximise the frame spacing. This is particularly useful for locos with inside motion.
You're also just touching on the reason I don't use ball races in the bearings of my models. I prefer to use plain bearings which take the sideplay better and I also use split axles wherever I can.
Ball races have their place in gearboxes, but I'm far less sure about putting them in a chassis. To be honest this topic and many of the pros and cons have been covered on this and other fora, at the end of the day it's down to personal beliefs.
If it'll help, I'll stick a calculation/diagram up here at the weekend to show how I work through sideplay on an example loco. Simple it isn't, or at least, not when real-world conditions come into play.
I'll also be at Kettering show next year if you want to go through an example...
Steph
 

JimG

Western Thunderer
Mick,

There's a quick and dirty way of working out sideplay which I have used, and that is to use a CAD program and draw an arc or circle to the minimum radius you want to use then superimpose your wheelbase as a straight line on that circle/arc - either with the ends of the line (as a chord) on the circle and measure the distance between the line and the circle at the axle centre(s) between the ends. Or place an inner axle centre on the straight line as a tangent to the circle and measure the distance between the circle and the outer points. For four or five axle locos, there are several permutations of tangent or chord placements. On a good CAD program you can probably measure to four places of decimals, or better. :)

Jim.
 

Scale7JB

Western Thunderer
I have found on the K2 and L1 that it is possible to have side play on a roller bearing axle, just a case of gently taking down the axle with some wet and dry until it slides freely, !!but without movement!! inside the bearing.. There is usually just a little movement in the etched beading carrier too which will help..

JB.
 

mickoo

Western Thunderer
Thanks for all the input guys, very informative but I may have misled with my initial question, so to tighten this and not run the risk of repeating what's been covered before.....unless this has been covered before....to which I am obviously oblivious :)

I'm hoping to explore the possibility of battery control in S7, to maximise battery life it makes sense to reduce load and thus by default ball races would logically be a good place to start, not just gear boxes but all major axles with the heaviest loading, tenders and drivers, ponies and trailing trucks would fall outside that area I feel.

So going one step further would all ball race axles be of benefit? if yes then by default you will have little to zero side play, if you then have no side play, will that piece of rolling stock then become pretty useless at travelling through most point work. If it were my layout I could simply fudge the track to accept the rigid wheelbase, but if wishing to use it elsewhere it obviously ain't gonna work! Zero side play also negates the need to split main rods to allow for lateral play, though that may fall outside of the 'S7 ethos'.

If ball races make next to no difference to load then plain bearings is the simple and obvious choice and the debate is closed....bearing in mind the advice given above for plain bearing chassis with regards to side play.

If however, ball races give a marked improvement, then I feel the possibility (personal) to explore further should be taken up, even if it's all theoretical. One way I can see around this would be to fit the leading and trailing axles with ball races and zero side play and the intermediates with plain bearings that allow controlled side play.

This then leads onto springing and is it really required, ball races lock lateral play but could also lock vertical play, vertical play could be introduced on the intermediates only. One could also argue that horn blocks may not be needed, the ball races simply sitting in oval holes, it might pay to add NS to the inside of the frames to strengthen the oval holes to match the brass holes in the frame....actually thinking further left field why oval holes, simple slots would suffice. This would increase the surface area in contact with the bearing and thus reduce the wear imparted by the steel ball races. Some method to stop the outer race rotating would need to be employed.

In short, I'm wondering if we are over engineering models because of past habits or practices introduced by older technology, yes it is very nice to be fully sprung in authentic horn guides (I accept they are mandatory if you can see inside the frames, ala 9F) but is it really necessary if your PW is up to scratch. Eastsidepilot has already show that horn guides simply fitted to frame slots with out the rest of the package work just fine, are there other simplicities that can be used or explored?

Sorry just spent all day in an out of the box management thinking course, can ya tell LOL

Kindest
 

3 LINK

Western Thunderer
Hi Mick,

As you can see from the other replies I am more of a "nut and bolt" type of guy for want of a better word :rolleyes:. The one thing I would say is that the loco in question can transverse a standard Peco turnout without any flange squeal ( tight spots ), not that I know the radius of the Peco point or what it would equate to in PW scale.

Martyn.
 

Dikitriki

Flying Squad
Hi Martyn

You are yet to discover that being able to negotiate a Peco turnout does not mean that it will negotiate a 31.5mm turnout to the same radius.......

On that cheerful thought.

Yours

Richard
 

mickoo

Western Thunderer
I have found on the K2 and L1 that it is possible to have side play on a roller bearing axle, just a case of gently taking down the axle with some wet and dry until it slides freely, !!but without movement!! inside the bearing.. There is usually just a little movement in the etched beading carrier too which will help..

JB.
Interesting, what horn blocks did you use if any to accept ball bearings.
 

Scale7JB

Western Thunderer
Interesting, what horn blocks did you use if any to accept ball bearings.

Hi Mick, on the k2 I used the hobby holidays hornblocks and bearings, but I couldn't bare the thought of paying whatever the cost is these days, so a while back I bought a couple of (long) lengths of 1/8" x whatever the other dimension is and mill these myself, I then etched up my own bearing carriers, and they work perfectly in the L1 and G5.

The added bonus is that you can mill the slots and get exactly correct centres if you mill the entire side be it a 2/3/4/5 coupled engine on one piece of brass stock. I then solder the whole length to the inside of the etched frame side and chop out the bits between.. Works perfectly, though only if you can't see between the frames.

I would go the old fashioned route if I were doing something like a J15 or J39 with plenty of space between the boiler etc..

Regarding friction and draw on the battery, do remember that decent gearboxes such as ABC (other gearboxes are available, but I wouldn't touch them now) are far more efficient with more torques than the standard fold up jobbie as supplied in most kits.. Expensive, but I'd you can afford it, well worth it..

JB.
 

Scale7JB

Western Thunderer
As an aside, on the 2nd K2 that I'm building, I have used the same method with my homegrown Hornguides and bearings...

I put the engine on the laminate floor in the flat, and it rolled off under its own momentum.. The floor in the flat isn't level!!

Impressed with the engine, not so much with the floor!

JB.
 

Tim Humphreys ex Mudhen

Western Thunderer
When calculating clearance we also remember that the real thing had much more flexibility in its frames than we allow, witness wheel marks on frames. Also the track is floating and so also flexible; my philosophy is build as much side play as is practical.
Tim
 

3 LINK

Western Thunderer
Hi Martyn

You are yet to discover that being able to negotiate a Peco turnout does not mean that it will negotiate a 31.5mm turnout to the same radius.......

On that cheerful thought.

Yours

Richard

Cheers Richard,

That's me in for a sleepless night then :eek::confused:.

The plasterers have been in today so starting to look like a "playroom" ;).

Regards,

Martyn.
 

mickoo

Western Thunderer
JB, ah yes, Hobby Holidays, as you say a trite expensive these days! Follow all that but out of my reach at the moment due to lack of a milling machine, though it might be possible to attach the work to the lathe saddle and mill the slots that way....difficult to set up I appreciate.

Being forced into self build mode I'm toying with the below, use existing horn guides to allow the bearing to pass through or open them out into slots, for a tender exact tolerance is not required, probably not for drivers either. Form new thin axle boxes from machined flat bar....size determined by what's available on the market and will accept a 5/16" OD bearing and thus no machining required except boring of bearing hole. Axle box is a tight fit on bearing, bearing is a tight fit on axle, the position of the bearing on the axle can be as tight within the chassis plates or as loose as required to give amount of side play. Being as the bearing is tight on the axle then the axle box and bearing will float laterally within the frames. Axle box rotation is limited by soldered angle plates or solid bar, this being the only critical area to allow free vertical movement and allow for the small lateral movement, ideally these plates could be milled but again no access to mill so accurate soldering required and I suspect a lot of fettling.

The tender inner dummy frames are an ideal playground to try this out before tackling loco main frames as if it goes pear shaped can be recovered with out visual impact.

One final alternative to prevent the outer race from rotating would be to file flats in the top hat of the bearing and eliminate the thin axle box, move in the angle guides to suit and retain movement required, knowing my filing I doubt I would end up with parallel flats LOL.

I hear you on ABC gearboxes and that's a given after Sunday at Sudbury, they are very smooth and allow wheels to back drive the motor.

Image3.jpg
 

Scale7JB

Western Thunderer
There is someone in the group who may be able to help with a milling machine with DRO's so long as you don't mind helping with spiking of track for his layout..

JB.
 
S

Simon Dunkley

Guest
The simple and quick answer is to follow what the prototype did... as long as the frame and wheels are scale equivalents of the prototype and provided that the track has a scale equivalent equal to or greater than the "dead slow" figure shown on the relevant engine diagram then there is generally no trouble... although you can ease the situation by incorporating gauge widening (which is what was done on the prototype).

If you use frame plates of scale thickness, set at scale separation, then the sideplay ought to be deducible from a
(you may have to estimate the thickness of the axlebox flanges).
I don't wish to be contentious, Graham, but this is one area where the "everything copied from the prototype" advocates get it wrong.

In steam days, locomotive wheels had different flange profiles depending on where in a wheel arrangement the axles were. In some positions it was merely a case of a wider or narrower tread (i.e. between 5 and 5.5") to allow a bit of extra clearance (for example, driving wheels behind slidebars and crossheads) but there were also variations in the width of the flange - both from the front and the rear of the flange. The idea that the "check gauge" on wheelsets (i.e. back of one flange to rubbing face of opposite flange) is constant is a model railway myth: the British Standards focussed on the centres of the flanges. I have a copy of the drawings somewhere, but as I am at work there is even less likelihood of my laying my hands on them immediately than if I had to rummage for them at home, so I cannot give precise information.

The important point here is that the flanges on centre driving wheels were thinned from both sides, as typically this axle was the one driven by the connecting rods, and sidely play needed to be minimised. We may be talking in terms of 1/16" of an inch here - maybe even less - but it would still scale up to a few thou of sideplay.Setting the plate frames to be dead-scale requires a little thinning of some flanges, if problems such as those described by Geoff Holt in MRJ Compendium number 2 are to be avoided.

Personally, I would just set the frames 10 thou closer together - in S7 this is about 5/12 of a scale inch - and allow a little bit of sideplay.

The other issue is that various forces, when scaled down, may be reduced by the square of the scale and others by the cube, leading to frames being 43 times less flexible (if made of like-for-like materials) than may be desired.

It is also impossible to scale down tolerances, as the consequence would be to make running fits into interference fits.

A little bit of tolerance is required in dead scale modelling, is all I am saying - and I am being purposely ambiguous in saying that!
 

Eastsidepilot

Western Thunderer
............This then leads onto springing and is it really required, .................

Mick,
You will need springing on all stock and loco's in S7 regardless of how good your trackwork is, the flange depth is quite shallow in this scale and the springs will help keep the wheel in contact with the rail head, it will also help towards any tendency fot the wheels to ride up on to the top of the rail when negotiating tight curves.
Another point is bogies and pony trucks, fit sprung axle boxes to these and side springing, this helps guide the loco through the curves and stops derailment.
The stone age method of placing a spring around the vertical pivot on bogies is a no no as far as I'm concerned ( clockwork, coarse scale technology):eek: , I prefer a solid bearing surface (bolster) which helps support and spread the weight of the loco across all axles, all this adds up to a smoother better running loco.

With regards side play I allways try to introduce this on the drivers ( not behind slide bars) especially on long wheel base loco's (0-8-0, 2-4-2 etc.), but not on the bogie and pony truck wheels.
I also believe that it helps to provide gauge widening through point work and tight radius curves, this will help aleviate the amount of sideplay you build into your loco's.

ATB, Col.
 

richard carr

Western Thunderer
Mick

If you want to use battery power, there is no need for ball races for any of the axles. I have 12 locos fitted with red arrow control, none of which have ball races and all will run for several hours on one charge using AAA batteries, 8 in the smaller locos ten in the larger ones, including 2 DJH V2s. The only down side of red arrow is that there is no sound which is why I have swapped back to DCC for my diesels.

There are of course other good reasons for using ball races, you can get some real momentum effects if you do, but they are not necessary for battery operation.

Richard
 
Top