EM Gauge Locomotive Drive Wheel Side Play - Optimum Values

[NFWEM57]

I have embarked on converting a number of RTR OO locomotive of varying ages to EM gauge with finer scale wheels in most case.  Whilst here is plenty of information on most topics there is next to nothing regrading the appropriate side play to allow for on a say a 4-6-0.  I have researched a little but found nothing concrete but 0.4~0.5mm per side with an extra 0.2mm per side for the centre axle seems to be suggested.  The minimum gauge on a test track I have built is 32 inches ( 810mm) with a test siding with reducing radius from 30 inches to 26 inches (760mm to 660mm).   I intend to set the back to back at 16.5mm.  Using the the versine calculation gives too small a value I think. 

 

Is 0.4mm per side plus an additional 0.2mm per side for the centre axle about right?

 

Grateful for any guidance and advice. 

 

Patrick

[dhjgreen]

I think the issue here is your minimum radius.  I do not know any EMers who would go down to 26" curves, my min is 4'.

[NFWEM57]

27 minutes ago, dhjgreen said:

I think the issue here is your minimum radius.  I do not know any EMers who would go down to 26" curves, my min is 4'.

Hi, the 30 to 26 reducing curve is to test limits.  The main test track minimum is 32 inches and the planned layout will have minimum 36~48 inch radius for the helixes, otherwise far larger radius for the scenic section.

 

Space limited for the test track.

 

Patrick

[NFWEM57]

I think I have managed to work it out.

 

On the outer rail of a curve, the flanges on the far edges of the  front and rear driving wheels are in contact with the rail.  Assuming a 15ft wheelbase and 5ft 8 inch wheels, the scale versine is 1.1mm. But the centre wheel has a versine too, 0.1mm, which is subtracted from the first versine so the required side play is 1mm.   So with 0.4mm suggested for the front and rear driving wheels and 0.6 suggested for the middle driving wheels the total is 1mm.  To be on the safe side, a total side play of 1.1mm would seem appropriate to allow for the fact that the flanges of the front and rear driving wheels are in contact with the rail adjacent to the centre driving wheel.  

 

For a minimum track radius of 48 inches the side play is approximately 0.75mm and 1.1mm for my minimum (working) test track radius of 32 inches.

 

 

Hopefully this all makes sense and I have worked it out correctly..!

 

I have ignored gauge widening.

  

Edited January 26, 2021 by NFWEM57
added additional info

[Miss Prism]

Don't forget that sideplay has to be applied on both sides of the chassis.

[NFWEM57]

28 minutes ago, Miss Prism said:

Don't forget that sideplay has to be applied on both sides of the chassis.

Indeed...

 

[97xx]

I understand your versine calc approach, and I too looked at this when working out how much sideplay I needed on my 00 gauge Q1 chassis.

 

What I initially overlooked - which resulted in what would have been unnecessarily excessive sideplay - was the fact that the wheelset has sideways play on the railheads at the outset.

 

In other words if you took a fully fixed say 0-6-0 chassis with zero sideplay in any axle, you would in fact be able to rotate the chassis about the middle, and/or slide it side to side on the track as the flanges on any one axle are never both in contact with the railhead.

 

Another way of saying this would be that this same totally rigid no-sideplay chassis would be able to negotiate at least some large radius of curve on its own.

 

Hope that makes sense?

Edited January 26, 2021 by 97xx

[97xx]

1 hour ago, NFWEM57 said:

 

 

 

  

I'm sure your diagram is indicative only, but needless to say the front and rear wheels won't actually touch the rails where you've indicated - which looks like the flange point at axle level.

 

Yes, there will be a very small displacement of flange contact away from bottom dead centre, but not sure it's significant?

 

Again just saying this as if you took your diagram literally you'd be necessitating even more sideplay.

Edited January 26, 2021 by 97xx

[NFWEM57]

Somewhat exaggerated.  In practice, more side play allowed on the centre axle.

 

.

[97xx]

5 minutes ago, NFWEM57 said:

Somewhat exaggerated.  In practice, more side play allowed on the centre axle.

 

.

FWIW, my Q1(32mm + 34mm wheelbase) in 00 has:

1. rear axle (driven): zero sideplay
2. middle axle: 1.5mm sideplay
3. front axle: 1.2mm sideplay

Even on a bit of flexible track bent to 21" radius (TrackSetta jig) there is a surprising amount of movement 'left' between flanges and railhead - which surprises me.

[Izzy]

With outside cylinder locos in many cases there needs to be minimal sideplay on the leading driven axle to provide adequate clearance behind the cross heads. Recessed crankpins are often used as well.  Any sideplay thus has to be confined to the other coupled wheels. Generally 36” is a good minimum radius to aim for with the wider 4mm gauges, EM & P4. I’ve always used A5’s in crossover formation as the test benchmark. Often it’s the angle the leading edge of the flange meets the inside of the rail head that becomes the deciding factor and will of course vary with wheel size and tyre profile. 

[NFWEM57]

6 hours ago, Izzy said:

With outside cylinder locos in many cases there needs to be minimal sideplay on the leading driven axle to provide adequate clearance behind the cross heads. Recessed crankpins are often used as well.  Any sideplay thus has to be confined to the other coupled wheels. Generally 36” is a good minimum radius to aim for with the wider 4mm gauges, EM & P4. I’ve always used A5’s in crossover formation as the test benchmark. Often it’s the angle the leading edge of the flange meets the inside of the rail head that becomes the deciding factor and will of course vary with wheel size and tyre profile. 

Hi,

 

Than you for the advice.  First conversions are two old Bachmann split chassis locos, a 4MT and a GWR Mogul.  Brass sub axles with AG wheels so shorting straps needed.  For the 4MT, trying to get a 16.5mm back to back whilst still being able to fit the stub axles in the chassis has proved interesting and testing..!  You get what you are given unless you resort to fine tuning.  In the end I reduced the thickness of the chassis slightly around the axle area by about 0.1mm (front) to 0.3mm (rear) each side.  Removed shorting strap from hub of centre wheels and used silver conductive paint to bridge strap to hub and achieved more side play.  Each axle a bespoke fit to a particular location.

 

In contrast, the Mogul chassis is thinner so plenty of side play already.

 

Good idea to use an A5 with 33 inch radius but none to hand..!  I am of the EM ready made track breed..!

 

Patrick

[NFWEM57]

17 hours ago, 97xx said:

I'm sure your diagram is indicative only, but needless to say the front and rear wheels won't actually touch the rails where you've indicated - which looks like the flange point at axle level.

 

Yes, there will be a very small displacement of flange contact away from bottom dead centre, but not sure it's significant?

 

Again just saying this as if you took your diagram literally you'd be necessitating even more sideplay.

Hi,

 

Agreed. And the face to face size dimension with a B2B of 16.5 is around 21.2mm for the wheel set I have with AG wheels. The AG wheels have a tyre width of 2.3mm with a 0.6mm flange width so 17.7mm flange to flange.  Thus there is 0.5mm of side play available with the rail meeting the tread 0.25mm out from the flange outer edge and 1.45mm of tread to the wheel face and, with the 0.8mm wide rail on my track, 0.65mm to the tread edge from the outer rail edge.

 

So I could reduce the side play required by 0.5mm to 1.5~1.6mm.  I have 0.05mm on the front, 0.8mm on the middle  and 0.45mm on the rear.  Versine for Front to Middle is 0.83mm so that is achieved.  Middle to Rear is 0.96mm and with the middle wheel at one extreme and the rear wheel at the opposite extreme the total is circa 1.2mm, above the versine calculation.   Factoring in the 'free' side play in the first paragraph, that is a total of 1.7mm.

 

Next is to allow enough 'play' in the coupling rods to be able to cope with the side play.  There is enough flexibility in the rod so the crankpin holes must have sufficient clearance.  Mine are currently 1.65~1.70mm for a 1.5mm bush; about 6~8 thou.  Iain Rice, in his book Locomotive Kit Chassis Construction,  suggest that 6 thou, or slightly more for flexichas arrangements, as about right so my 8 thou seems adequate as as starting point for any fine tuning.

 

Patrick

Edited January 27, 2021 by NFWEM57
typo

[97xx]

What I've done with the rods is drill undersize and then to use a 5-sided taper broach to ease open. I found this makes a neater job of making really smooth really round holes, plus by definition has a very slight taper to each outside edge which all works in the right direction with the relative wheel movements.

[PenrithBeacon]

This method will produce round holes but not cylindrical ones.

[Jeff Smith]

If I remember correctly Ian Rice suggests wheels for an 0-6-0 chassis should be a running fit (minute sideplay) for the outside axles and about 1mm total for the centre axle for a 24" radius.

[NFWEM57]

26 minutes ago, Jeff Smith said:

If I remember correctly Ian Rice suggests wheels for an 0-6-0 chassis should be a running fit (minute sideplay) for the outside axles and about 1mm total for the centre axle for a 24" radius.

Hi Jeff,

 

Well, I am about that on the front and, if required, an 8mm OD, 6mm ID, 0.2mm thick washer on either side of the rear axle will give me running clearance.  But, I'll see how it runs as is first..! 

 

Patrick

[KeithHC]

Thanks Patrick for the insight and The info. Maybe someone will start a small business doing EM conversions. At the moment  when I come round to it I have currently only two steam locos to do. Both 0-4-4T an M7 and an H class. The M7 has conversion kit available but there is not one for the H class.

 

Keith

[NFWEM57]

2 hours ago, KeithHC said:

Thanks Patrick for the insight and The info. Maybe someone will start a small business doing EM conversions. At the moment  when I come round to it I have currently only two steam locos to do. Both 0-4-4T an M7 and an H class. The M7 has conversion kit available but there is not one for the H class.

 

Keith

Keith,

 

Thank for the interest and feedback.  Why not convert the M7 and then see if you can buy the separates to convert the H class?.

[KeithHC]

9 hours ago, NFWEM57 said:

Keith,

 

Thank for the interest and feedback.  Why not convert the M7 and then see if you can buy the separates to convert the H class?.

Hi Patrick,

That’s my thoughts just need to find suitable 5ft 6ins wheels the M7is 5ft 7ins. Keep up the good work.

 

Keith

[97xx]

21 hours ago, PenrithBeacon said:

This method will produce round holes but not cylindrical ones.

Precisely - which I found helps given that the wheelsets move relative to each other - which forces the rod out of parallel.

 

I suppose a bit like we need a minute amount of pitch away from the vertical in horn guides as the axles ride out of parallel with the rails.

[1466]

Re the suggestion of a small factory doing EM conversions . I recall a couple of people offering this service in the EM Gauge Society newsletter. I’ve used John James some years back and can recommend his services- no connection except as satisfied customer . He did an M7 for me amongst others .

One small point about the M7 . The “effective “ wheelbase is from the leading driver to the centre of the trailing bogie which is longer than many six coupled engines . My M7 will traverse 36 inch radius track but I think will grind on anything smaller.

Hope this helps .

 

[PenrithBeacon]

3 hours ago, 97xx said:

Precisely - which I found helps given that the wheelsets move relative to each other - which forces the rod out of parallel.

 

I suppose a bit like we need a minute amount of pitch away from the vertical in horn guides as the axles ride out of parallel with the rails.

Perhaps I should have continued my post by pointing out that a broach will produce holes that are round but also trapezoidal in section (perhaps even a double trapezoid! with the high point somewhere in the middle of the hole) and this will wear much more quickly than a cylindrical hole. Best not to use broaches  for a bearing.

[Jeff Smith]

When you say broach you really mean taper reamer.  Broaches are used for making key hole slots, splines, etc.  Some taper reamers are mis-named broaches.

[97xx]

19 minutes ago, Jeff Smith said:

When you say broach you really mean taper reamer.  Broaches are used for making key hole slots, splines, etc.  Some taper reamers are mis-named broaches.

 

I would see a tapered reamer as a rather coarse fluted thing often with a T handle, and a pretty steep taper.

 

Perhaps I’m using the wrong term for modelling, but I can assure you that in watch and clock making they are called broaches. There are cutting (5-sided so they cut concentrically but are hardened steel and have no ‘blade’ or ‘teeth’) and smoothing (smooth hard circular), both tapered obviously. The taper is small - around 1 in 100 so you need a lot of them.

 

When broaching out holes for shafts in clock/watch plates the process creates a hole when done correctly from both sides is tapered to a minimum in the middle, thus having lowest friction for the pivot that runs in it.

 

I appreciate that in machinists’ circles a broach is a toothed straight job for doing as you say. And expensive too...!

Edited January 28, 2021 by 97xx