Just checked the stats on my prototype's locos, and found something very interesting - not a single one of them is an odd size!
Not until you get into ancient locos imported from places that didn't use the Metric system do you find any off-sizes - and they were disposed of long before the later-built standard types.
OTOH, the track gauge, 1067mm, i distinctly odd in metric. It's much more sensible when stated as 3 feet 6 inches.
Chuck (modeling Central Japan in September, 1964 - in 1:80 scale)
They are even because you have chosen a very small unit of measure, the mm. The proposal is that drivers should be some large fraction of a foot.
If you express your driver sizes as a larger unit of measure they don't look so even:
Or convert them into English units:
Still no organic "even" multiples.
Dave H.
Dave H. Painted side goes up. My website : wnbranch.com
concretelackey wrote:So it is true...the size of the booster rocket segments (for the space shuttle) are based on the width of 2 roman war horse butts!
The solid rocket boosters are about 12 foot in diameter and each one is made of 4 segments. Since each booster has about 1,200,000 lbs of propellant, each segment has about 300,000 lbs. Add in the weight of the case and shipping container and each segment weights in at about the max gross weight for an 8 axle heavy duty flat car. So the dimensions are more driven by weight than clearance.
dd
ps - there is at least one route between Utah and Florida that has no tunnels - but there isn't a route that has no bridges. Bridges are the bigger issue.
The one diameter one would try to avoid is 74.5 inches as that results in a circumference that is an exact multiple of the 39 foot rail length in the steam era. The vertical dynamics of the rods and counter weights - coupled with the fact that rail joints tend to sag a bit - could cause all kinds of weird oscillations to occur.
dldance wrote: The one diameter one would try to avoid is 74.5 inches as that results in a circumference that is an exact multiple of the 39 foot rail length in the steam era. The vertical dynamics of the rods and counter weights - coupled with the fact that rail joints tend to sag a bit - could cause all kinds of weird oscillations to occur.dd
And yet, the first NYC Niagara had 75 inch diameter drivers! I wonder how long it would take to wear off 1/4 inch of tire thickness...
Later Niagaras had 79 inch drivers. How much of that was the 39 foot problem, how much was a desire to standardize driver diameter with the Hudsons and how much was the ability of the combination of boler pressure, cylinder bore and stroke to slip the smaller driver?
Come to think of it, didn't the Boston and Albany Hudsons have 75 inch drivers?
Chuck (modeling Central Japan in September, 1964)
markpierce wrote: Can anyone explain why almost all common diameters of steam locomotive drivers are odd sizes? I know there are reasons for different wheel diameters, but why not have drivers 48, 54, 60, 66, 72 and 78 inches in diameter rather than the almost always odd diameters such as used by the Southern Pacific which were mostly as follows:0-6-0s: 51 and 57 inches2-8-0s: 57 inches4-4-0s: 69 and 73 inches2-6-0s: 63 inches4-6-0s: 63 and 69 inches4-4-2s: 81 inches4-6-2s: 73 and 77 inches2-8-2s: 63 inches2-10-2s: 63 inchesArticulateds: 57 and 63 inches4-8-2s: 73 inches4-8-4s: 73 and 80 inchesThe actual driver diameters most often appear 3 inches larger than the "even" size except for the largest sizes. I have a guess why, but I don't want a "guess" answer. Thanks.Mark
Can anyone explain why almost all common diameters of steam locomotive drivers are odd sizes? I know there are reasons for different wheel diameters, but why not have drivers 48, 54, 60, 66, 72 and 78 inches in diameter rather than the almost always odd diameters such as used by the Southern Pacific which were mostly as follows:
0-6-0s: 51 and 57 inches
2-8-0s: 57 inches
4-4-0s: 69 and 73 inches
2-6-0s: 63 inches
4-6-0s: 63 and 69 inches
4-4-2s: 81 inches
4-6-2s: 73 and 77 inches
2-8-2s: 63 inches
2-10-2s: 63 inches
Articulateds: 57 and 63 inches
4-8-2s: 73 inches
4-8-4s: 73 and 80 inches
The actual driver diameters most often appear 3 inches larger than the "even" size except for the largest sizes. I have a guess why, but I don't want a "guess" answer. Thanks.
Mark
Why not? What's sacred about even numbers? Besides, SP often used 1/4" thicker tires, which ended up giving its GS-1's, 2's and 6's driver diameters of 73.5" rather than 73" (source "The Daylight Locomotives" by Robert Church). IIRC, the same was done with the Mt's, P's, F's (63.5") and a few others. Driver tires wear and a nominal 69" driver loco would have an actual driver diameter of 68" after 1/2" of wear.
In any case, SP's 4-4-2's had drivers of various sizes (84, 81, and 73 depending on the class, its age and whether it was assigned to Pacific Lines or T&NO). The original A-1 4-4-2's were Vauclain compounds with 84" drivers. Some of the T&NO A-1's were simpled and rebuilt with 73" drivers (not to mention boosters). At least one Pacific Lines A-3 had 73" drivers (or more likely 73.5" fresh from the shop) at one point (#3041, IIRC). Factoid: an 81" driver A-3 Atlantic was rated at 2400 tons in freight service in the San Joaquin Valley.
SP's 2-8-2's had both 57 and 63" drivers, the smaller drivers being used on classes Mk-2 and Mk-4 (the earliest built). The same situation obtained on the UP.
SP's TW-8 4-8-0's had nominal 54" drivers. IIRC, the TW-4's had 56" drivers. These were the ones built in Sacramento Shops in the early 1880's. Other classes had 51" drivers.
The 4-4-0's were all over the map from about 56" (some of the earlier ones even some that made it to the 1901 re-numbering) up to 73 (73.5) inches.
Not all SP 4-6-0's (or even most given the wide variety owned by SP) had 63" or 69" drivers. This was only true of the Harriman Standards.
Andre
Andre,
So, there is only chaos? And you see no emerging pattern? That's OK. I'm fine here.
markpierce wrote: So, there is only chaos? And you see no emerging pattern? That's
Just because they don't use even fractions of a foot, that doesn't mean its "chaos". It is (as it has been explained several times in this thread) the exact opposite of chaos. The driver sizes are carefully planned for specific applications.
Although I do very little design any more - when designing, it is convenient to use round increments of the ruler where possible. It also improves the accuracy of the resulting product. For example, it is much easier to measure 12 inches than to measure 11 13/16 inches. If 20 people measured and cut something 12 inches long, the average error in the result is about 1/32. However, if 20 people measure and cut something 11 13/16 inches long, the average error will be about 1/16. Twice as much.
If I were designing in metric, I would use the nearest convenient round increment of centimeters. So in lieu of other requirements, I would design something at 10 cm (25.4 inches) rather than 2 feet (24 inches or 9.45 cm).
dldance wrote: If I were designing in metric, I would use the nearest convenient round increment of centimeters. So in lieu of other requirements, I would design something at 10 cm (25.4 inches) rather than 2 feet (24 inches or 9.45 cm).dd
Keep designing in inches etc dd! 1 inch = 2.54 cm, 1 foot = 30.5 cm (30.48 to be exact).
greetings,
Marc Immeker
marcimmeker wrote: dldance wrote: If I were designing in metric, I would use the nearest convenient round increment of centimeters. So in lieu of other requirements, I would design something at 10 cm (25.4 inches) rather than 2 feet (24 inches or 9.45 cm).ddKeep designing in inches etc dd! 1 inch = 2.54 cm, 1 foot = 30.5 cm (30.48 to be exact).greetings,Marc Immeker
Marc,
I believe that dd's point was that designers will tend to use the closest round number in the units being used, be that English or Metric units, whether it is an odd or even number is not important. But it is desirable to avoid using .xxxx in measuring, regardless of the unit of measure used.
The designers are trying to achieve the best possible combination for the purpose intended. Working in round numbers to ease production, be they odd or even numerals, English or metric is just for the men building and maintaining the equipment. The Locomotive doesn't care what arbitrary numbers are associated with the design, so long as the actual size is suited to it's intended use.
Doug
May your flanges always stay BETWEEN the rails
My point exactly - now you see why I don't do much design anymore ;)
challenger3980 wrote: I believe that dd's point was that designers will tend to use the closest round number in the units being used, be that English or Metric units, whether it is an odd or even number is not important.
Obviously nobody making this argument has ever actually seen a set of plans for engine or mechanical device or they would realize how wrong it is. For example if the size of a piston is an even diameter then the cyclinder HAS to be an odd size by a couple thousandths of an inch to allow for the parts to move without binding. The people who build locomotives are professional machinists. They are paid to make things to odd sizes.
Here is a link to PRR standard plans. Pick any plan. Look at the dimensions. Yes,some of the things will be to even feet and inches, but look at how many are to odd 1/16ths of inches.
http://prr.railfan.net/standards/standards.cgi?plan=&rr=&type=TRACKWORK
Here is a link to plans for a RDG steam engine. Yes, some of the dimensions are to even inches, but look how many are to halves, quarters or eights of inches.
http://gelwood.railfan.net/rdg/pr-30293.gif
Here are brake gear drawings, once again look at how many non-even inch dimensions there are.
http://gelwood.railfan.net/rdg/pr-40139.gif
Dave,
I agree with you on the odd sizes for cylinder dimensions, and other part to part tolerances, but that was not the question being addressed. The OP's original question was regarding driver sizes often being in odd number increments, rather than even numeral increments, which was the question being addressed. The designer would chose the closest inch size to what would be the best suited for the application, there not being a need for .xxx dimensions such as in piston/cylinder tolerances. That is not even to say that the blue prints didn't specify that level of tolerance, but the even inch size (not odd/even numeral) was what was commonly referred to, so maybe it was specified for say a 72. 8675 inch driver and just commonly called a 73 inch driver. Your comment about the odd sizes in many items IS entirely accurate, just not relevant to the question asked which was Driver size specific. Not disagreeing with you in any way, just clarifying my answer.