Sign my





Steam locomotives, or put in more general terms rod driven locomotives be it steam, diesel or electric, have their rods on either side offset by 90 degrees. For steam locomotives this is true for two and four cylindered engines. A model of a steam locomotive should depict that offset of course. It is not so important that the offset is exactly 90 degrees. If it is 89 or 91 no one can tell, really. On a real steam locomotive that would make a difference but not on a model. What is more important for a model, is that all coupled wheels have the same offset. If not the uneven positions of the crankpins will cause binding on the coupling rod which will make your precious model run uneven or in a bad case not run at all.

The  Romford way

When you buy a kit with Romford wheels and axles you are in for an easy job: the axle ends are squared and the wheels have a corresponding square axle hole. Quartering is almost automatic if only you observe the correct position of the crankpins.


Being on the subject of crankpins there are two options:

When the chassis is viewed from above from the driver's cab towards the front, the crankpin of the right hand wheel is in the fully forward position and the left hand crankpin is at the top. This means that the right hand wheels are leading the left hand wheels by 90 degrees. This is the norm for most British locomotives.

The second option that the left hand wheel is leading of course ;-).


Glue and slide quartering

When your kit comes with plain axles you must ensure that all wheels sets have the same quartering angle. When building my NGG16 I used a very simple and rather crude method. If you build accurately this method might work for you as it did for me.

First finish the coupling rod on one side completely without bothering about quartering. One rod alone will never bind unless there is something terribly amiss with your crank dimensions (if so solve that first, going ahead with quartering will only cause more trouble).

On the photo I have already completed the right hand side (believe me or read the NNG16 page on the subject).


Next I attached the most forward fly crank as a datum (fixed reference point) which I approximately quartered by eye. Again: exactly 90 degrees is not paramount. I reamed the flycrank to a tight press fit and fastened it with epoxy glue.

Then I reamed the other two flycranks with just a little play so they can slide easily over the axle without wobbling. Ream no more than just the first hint of sliding over the axle after achieving a press fit. Apply expoxy glue which will set in about five minutes, ample time for quartering. Offer the fly crank up to their axles and quarter them by a rough approximation. Add the coupling rod with its bushes in the very way as it will later be mounted. Now slowly move the locomotive up and down a strech of rails. The already fixed datum axle will communicate its position via the coupling rod and draws the other two fly crank into their correct positions. Let the glue cure and it is done. It worked without failure for me.


Two concluding remarks.

  • Above I described a loco with outside frames and fly cranks. For a inside frame loco it works exactly the same.
  • This method is especially useful if you must quarter the wheels within the frame and cannot use a quartering jig as described below.

Quartering jig

You can use a quartering jig if you can assemble the axles separately from the loco and mount the completed and quartered axle in the loco frame as a whole. Quartering jigs have the advantage of getting an almost 90 degrees offset and of repeated accuracy. Quartering jigs are commercially available and they vary widely in price from £8 for a simple brass thing by Cooper-Craft to the NWSL Quarterer II with its hefty price tag of around $140.

DIY jig

I thought I might have a go at making a quarterer myself. I had some 0.75 mm brass sheet lying around and sat down for some serious sawing, filing and drilling. A few ends of M3 threaded rod and 24 M3 nuts completed the set.

Alls four brass sheets should be made identical. Sheet thickness is 0.75 mm

A diagram to illustrate the approximate measures of my quarterer. Most dimensions are not critical, not even the 90 degree angle. The most critical thing about it is that all four plates should be identical. I achieved that by sawing them approximately to size, spot soldering them together and then file them as one whole to the desired measures and drilling the holes.

Drilling through 3 mm of solid brass takes patience and care. I drilled a 1 mm pilot hole and then redrilled it whith a larger drill stepping up 0.5 mm each time. Adding cutting oil helps greatly.

The 3mm holes should accomodate the M3 threaded rod as accurately as possible. Assemble the jig on absolutely flat surface, like a sheet of float glass.

An accurate 90 degrees quartering will not be achieved with this type of quarterer unless your axle and crankpins have the same diameter. The diagram shows the basic flaw of this design. Even if you get the "legs" of the quarterer to an accurate 90 degrees, the difference in size of the axle and the crankpins will cause the resulting angle to be greater than 90 degrees. This is not a problem for any other than live steam powered locomotives. It won't be visible by the naked eye, and the quarterer ensures an exact identical "wrong" angle for all axles.


A simple solution would be to fit pieces of tube over the crankpins with the inside diameter of the crankpin and the outside diameter of the axle. These tubes should be custom made for every combination of axle and crankpin diameter.


A Fairlie wheel set was the first guest in my new home cooked quarterer. This quarterer can be adpated to serve various gauges

Quartering is now easy. Fit the axle with all its components in the jig and adjust the wheel position until both cranks touch their respective sides without lifting the axle from the middle bars.

Total costs

Stock purchase Price You need Costs (rounded)
Brass sheet of 200 x 200 x 0.75 mm € 9.20 slightly more than 1/10 € 1.10
Pack of 80 M3 nuts € 3.19 24 € 0.96
Threaded rod M3 of 1000 mm € 1.79 3 x 40 mm € 0,24
    Total costs € 2.30


That isn't a very hefty price to pay, is it?

Quarterer 2.0

I couldn't help keeping thinking about that angle not being quite 90 degrees. Sorry, but I am a perfectionist, that is why I became a modeller! When trying to wriggle the geared axle into the jig I found I had no room to fit the innner plates of the quarterer between the frames and the wheels. The only way was to put the axle boxes on top of the plates. This however would aggravate the angle problem, because the axle moving upwards would cause the angle of the crankpins to become even shallower.

Inversely if I would make a square in the inner plates of the quarterer than the crankpins would move together and the angle would get smaller. This also would provide a means to position the axle boxes firmly in the inner plates using the slot provided in the axle boxes. But how deep would the axle boxes have to be lowered to achieve an angle of approximately 90 degrees? I had no idea. So I drew the starting stuation in Sketchup and first measured the angle that would result from putting the axle boxes flat on the jig.

The angle was 142, degrees!!


Then I lowered the axle box vertically, overdoing the situation a little. Lifting it until the top of the axle in the axle box was in plane with the top of the jig resulted in an angle of 90.3 degrees. Good enough.

The depth that needs to be cut out from the plate is 3.6 mm. Add to that the radius of the crankpin (0.5 mm) and you have exactly the distance between the middle of the axle to the middle of the crank pin. Coincidental or a mathematical causal connection? I don't know.


Next thing is to cut it out an try it live!