5" Burrell Gold Medal Tractor

Monday 21st April 2008: I have been searching for some time for a new project and I am coming round to the idea that I would like to scratch build a traction engine.  I though I wanted to build a track locomotive but I don't really want the resulting model.  I believe I have proved to myself that I could do it having built a boiler for a 5" rail motor.  There is a published account of my exploits starting in the May 2008 issue of Model Engineer.  I plan to finish off the steam section of the Rail Motor ie the smokebox, supper heater and boiler fitting and pass it on to my brother for completion.

I don't know if I will ever finish this project as it is way beyond anything I have tackled before.  I have decided to start a diary as I enjoying witting and I find it helps focus my thoughts.   I have started writing straight away as if the project takes off it would be very difficult to capture my early thoughts.  The working name for my project is Goldie and I don't plan to make the Goldie website live for some time.

Choice of Project: When I first saw the MJ Engineerings 2" Gold Medal Tractor I was smitten and have at various times considered building it.  With experience I have gained over the years I know that at 2" scale I would find it hard work on the rally field and too big to display on the for the mantelpiece.  I have been harboring the thought of going small and modeling in either 1" or 3/4" but my current thing is that I will need a replacement for "Little Beastie" in a few years time.

The obvious scale to go for would be 4" but as the Gold Medal is a very small prototype the resulting model would be small and to all intense and purposes behave as a 3" model.  If I am going to make a worthy successor to "Little Beastie"I am going to have to go for a larger scale.  I have seen a 6" Gold Medal Tractor at the 2006 Whissendine rally and while I loved the engine I judged it a bit big to live with.

6" Gold Medal Tractor Built by Dave Bennion

I am very sensitive to weight as I want to continue to tow using a mid size family car.  While my current car is plated to tow 1.3 tonne for various reasons want to keep the loaded weight of my trailer below 1 tonne.  Allowing for the weight of the trailer and accessories the engine needs to come in at about 1/2 tonne.  To try and home in on a scale I have produced a table to investigate possible sizes and weights.  

AT present I have very little hard data to go on.  I have the leading particulars for the MJ design which gives the L,W & H at 26,12 and 19 3/4 inches respectively and a wight of 80 lbs.  At present weight is my chief concern an I am fortunate that the Whissendine programme  gives the weight of the 6" as "Just over 3/4 ton".  I am not sure if this is a metric or an imperial tonthought the difference is only 35lbs.  Harder to guess is what just over means.  I plumped for a round number  of 1800lb which is 1 cwt over 3/4 over an imperial ton .  Another weight I can guess is that of the full size engine.  I will take this as 5 imperial tons thought I believe actual examples are slightly over this figure.  Knowing that weight scales according to the third power of scale I can guesstimate the weight of a gold medal tractor at various scales.

My figures confirm that a 4" engine is going to be too small.  With guesstimated weights ranging from 415 to 640 lb it would be about 1/2 the normal weight of a 4" engine.  The next thought is to consider 4 1/2" as this is an accepted scale but again the weight ison the light side.  The next larger "standard" scale is 6" but at 1800 lb would make a model that would be too heavy to tow comfortably.   The length is also a problem since at 7 feet is too long to fit in my engine storage space.  I have thought long and hard about 6" scale as Dave is currently advertising his patterns in Old Glory.  

As with Goldie Locks, 5" scale  is about right neither too heavy or too light and at 70" much the same length as "Little Beastie".  As the scale gets smaller the achieved weight is greater than predicted by the cube law which is why the 2" engine comes in at 80 and not the 52 lbs scaled from the full size weight.  This is because plates have to be made oversize to ensure sufficient strength.  Scaling this back up to 5" will give an estimate that is high where as scaling down from either fullsize or 6" will give an estimate that is low.   If I were to use the same plate thickness as the 6" but scale the areas down to 5" I could expect the weight to reduce by a square law (scale factor [5/6]² = 0.694).  Using this scale factor I estimate the weight of a 5" engine at 1250 lbs which surprisingly is the same as estimated by cube law from the 2" weight.  I believe I can be confident that the weight of my 5" engine will not exceed 1250 and will not be less that 1000lbs.  Thus I believe that a 5" Gold Medal Tractor is the right scale for me.

Boiler:  Thus far I have considered weight as my "goodness factor" but what I am really interested in is having a model that is large enough to steam easily.  My reference standard is "Little Beastie" as I find it extremely easy to maintain a good head of steam provided I don't do anything daft.  Even thenI often find there is reserve capacity to enable me to get out of jail.

I don't have much data to go on but I believe the distance between hornplates of the 2" is 3.844".  This would make the width of the 5" boiler 9.6 inches.  The length of the outside of the fire box is even harder to estimate but I think it will be just over 13" at 5" scale.

"Little Beastie's" hornplates are 9.84" apart and the length outside of the firebox is 13".  This is very good news as the 5" Gold Medal Tractor will have pretty much the same firebox and grate dimensions and I can therefore expect good steaming characteristics. 

Next steps:  Before I can go any further I need more information.  I shall be attending Watham Abbey rally over the 17th and 18th of May and I shall take special interest in any Gold Medal Tractors present.  Hopefully I will get a chance to measure some leading dimensions.  I particularly want to get an accurate handle on the boiler dimensions.  In the mean time I have order a set of drawings for the 2" engine from MJ Engineering.

Friday 25 April:  I have spent a lot of time on this project.  I have mixed feeling of excitement and dread - can I actually design and manufacture a 5" engine.  My biggest worry is that many of the components are way beyond the capacity of my workshop.  Another concern is that I will have to make my on patterns and get these cast.

I got lucky when I visited my club on Wednesday.  I had a faint recollection that a construction series was published Model Engineer but Jack found them in Engineering in Miniature.  The articles start in 1981 and go on for 5 years. This gave the project a tremendous boost as I was able to start firming up on the dimensions of my 5" version.  I am finding that Geoffrey Gilbert's book Traction Engine Design and Construction is of great help in sorting out the design details.  Surprisingly this book has more relevant information than his book Burrell Style.

Outline design for hornplates (links to drawing)

I have now established the rough dimensions of the boiler and hornplates and have begun working on my design.  I am proposing that my hornplates will be 9 3/8" apart.  I have decided to maximize the size of the firebox by doing away with the dummy throatplate and have the real throatplate against the belly tank.  Looking at the picture of David's 6" engine the throat plate wont be visible anyway.

I can now understand a point by Steve Baldock that I need to establish what sizes of boiler tube are available size many of the dimension stem from it.  For now I am taking the diameter of the barrel as 9 1/4 inches.  The width of the fire box outer is the same as this.  I plan that the hollow stays for hornplate fixing are slightly proud on the fire box sizes so that they may be machined to give my nominal hornplate separation of 9 3/8".

By doing away with the dummy throat plate the length of the fire box is 12 inches.  Working up a typical foundation ring inside this gives a grate dimensions of 9 x 6.5".  The resulting grate area of 58 square inches is a little less that I had hoped.  As I work up the design I shall see if it is possible to borrow a little elsewhere but I think 58  sq" will be OK.

The first daunting challenge I have uncovered is how to make the left hand hornplate.  This is dished to accommodate the low pressure eccentrics.  At full size this would have been made from a single piece of red hot boiler plate and a giant press and form tools.  This is way beyond anything I can attempt at home. My initial thought was that I might be able to fabricate by welding tow partially flanged plates but I am suspicious that the welding would introduce distortions and my not be strong enough.

One of the skills I am going to lean is pattern making and to support this I have obtained three books on pattern making.  Of the three books "Wood Pattern Making" by J McCaslin looks the most useful.  This seems to cover the sort of complex shape I will need to case but as it was written in 1923 may be out dated.  The other two books are more recent but may be too superficial.  "Foundry Work for the Amateur" looks like it may be a good primer so I shall read this first.  The final book I have is "The complete handbook of sand casting" by CW Ammen.  I hope this book will give me an insight in foundry techniques.  I wonder what I will think of these books once I have had time to study them!

I have just received a set of 2" drawing from MJ Engineering.  While these contain much the same drawings as accompany the article they are easier to read.   From my studies of the articles I have uncover a few errors and I was pleased to see these have been corrected on the drawings.  I have seen drawing for several traction engines and these drawing are the best compressive drawings I have seen.

I can only go so far using the 2" drawings as I suspect they include compromises to achieve a working model in 2" scale.  I cannot finalise anything until I have had the chance to crawl over and measure a full size example.

Tuesday 29th April: I have spent the last few days getting to grips with the involute gear tooth form.  This investigation was instigated to discover if it would be practical to design working rear suspension.  At fullsize my researches revealed that only about 3/8" total movement was possible.  This is because the suspension works across the final drive gears and any movement of the rear axle causes the gears to move in and out of mesh.  With a cycloidal tooth form proper constant velocity operation is only achieved at one depth of engagement thus the suspension needs to be adjusted so that in service the back axle is in its correct place respective to the 3/8" travel.  I also presumed that the cycloidal tooth form limited the designers to such a small range and I wandered if I could allow proportionately greater movement using an involute tooth form.

As I intend to get the gears water cut using CNC I am free to adopt a non standard tooth form.  Normally involute are 2/DP long (plus a bit more depth to provide clearance).  If I can make the teeth longer the extra can be "given" to the suspension.  So if I could add 1/2"  inch on to the teeth I could get a useful travel on the suspension.  Plus I would have the benefit of the involute form giving me constant velocity irrespective of travel.

If only it was that simple.  The first task was to establish how big my teeth would be, namely the DP.  The driving pinion has to have fewer teeth as the teeth get bigger in order to keep the final drive ration constant.  Unfortunately as the number of teeth fall to the 20-10 region funny thing start happening to the involute tooth form requiring severundercut to provide clearance for the driven gear teeth.  I concluded that my DP would have to be in the range 4-5 with a final drive pinion of 13-20 teeth.  This gives a starting tooth height in the region of 1/2".

Pressure angle 14.5 dp=4 Tooth length 1"
Driver n=13 Driven n=62
 
Pressure angle 25 dp=4 Tooth length 0.65"
Driver n=13 Driven n=62

Trying to make the teeth 0.5" longer I soon discovered that the tooth form goes up to a point and beyond.  Also tips of the driven teeth pass through the base of the driver.  It would be possible to undercut the tooth to provide clearance but this would make the teeth even weaker.  Strength of teeth is going to be a major consideration.

In an attempt to move towards a more practicle tooth form I restricted the added length to 0.15" and increased the pressure angle to 25 degree.   The nick shown at the base of the blue driver tooth is the bottom of the involute curve and as a result no contact should occur below this point.  Watching an animation of the gears turning I believe there is sufficient clearance to include a generous fillet in this region to strengthen the tooth.

If I go for a fixed back axle and no suspension it begs the question is it worth putting in the detail to parts which will never be seen.  My inclination is that if a part is non working and unseen it might as well be simplified to ease construction.

And then....
I decided not to built a 5" GMT but a MAC