how to figure out the scale speed of your train. (distance traveled in inches) divided by (time used to cover the distance in seconds) divided by (.11) = scale mph example: (one 4' foot ntrak module)/(10 seconds time)/(.11)= (48")/(10)/(.11)=43.6 scale mph usual i keep my 100+ car coal drags around that speed
Me too, although my guests get bored at 40 smph, as it takes 40 minutes to run a train around the loop.
Bored...your guests get bored! Bored is operating a coal train at the loader, under 1 mph average for 8 hrs!
strikefour; Welcome to TrainBoard! I was watching a video a few days ago. About a notable N scale layout. The operator carefully ran his trains at a speed which appeared more like a prototype. It made a huge difference in my perception of his efforts. Boxcab E50
60 mph = 88 ft/sec 60 mph = 88 x 12 in/sec 60 mph = 1056 in/sec 60 mph (scale) = 1065 / 160 in /sec 60 mph (scale) = 6.66 in / sec how long to cover 3ft if going 60 scale mph? 66in/sec = 36in/x sec x = 36/6.66 x = 5.4 seconds
Jerry, I had one mule, the old Atlas FA-1, with a decoder on it, so the body doesn't fit, that topped out at 20 smph and would easily run 5 smph. That engine had a huge motor in it, but weighed nothing since the motor took up all the space needed for weight. With the huge motor I could limit the top speed. It ran smoothly, but pulled maybe two cars. It took almost 90 minutes to complete the loop, as it slowed on the long uphill grades. I scavenged the decoder about a year ago for a steamer, but I'm still thinking it would be the ultimate slow speed machine with a little tinkering--and more weight.
I really run by what looks good. Different models run different speeds. I can put the Tech II controller to 45mph and different engines will run at different speeds. When I am operating by myself, I tend to run trains slower, to appreciate the detailing. If I am making a video for Youtube, I tend to run faster, so the train will cover the distance quicker so downloading to Youtube don't take as long.
From the old NTrak days and the four foot modules, we used eight seconds for four feet for sixty scale mph. The distance was easy to see and you just counted out the four seconds to see if you were going too fast or too slow. Eight seconds for four feet is a long time. Try jagged ben's calculator to see actual scale speed of four feet in eight seconds.
The way in which most modelers scale speed actually has little do do with physics, and is mostly built around the intuitive (though probably incorrect) concept that time does not scale. Some simple physics, however, goes counterintuitive to both the common wisdom, and your observation. Indeed, it actually leads to some startling and counterintuitive results. Being a physicist, I naturally had to give a couple thought experiments a try. The first one is in relation to superelevation. If you have a perfectly designed superelevated curve, when the train is going at the curves design speed a passenger in a coach would only feel a "downward" force, because the curve angles the coach such that it balances gravity and the centrifugal forces. This, of course, keeps ones coffee that is on the table from spilling while going around the curve. If you make a perfect scale model of the curve, including the same superelevation angle, and drive your train around it at the scale speeds that common wisdom gives, a tiny cup of coffee would spill all over your scale passengers because the forces do not balance. It turns out that similar problems occur with keeping time via scale model grandfather clocks and with scale model trains running off of scale model cliffs. In all instances, you get one coherent result: If a scale distance is derived from the prototype distance by multiplication of a scale factor m (which is 1/160=0.00625 for N-scale), then scale time is derived from real world time by the square root m (which is sqrt(1/160)=0.0791)! Combining these two transformations with rate*time=distance and scale-rate*scale-time=scale-distance, it is easy to find that the speed scales according to the same factor as the time. Thus for an N-scale model, it is much more scientifically accurate to run the model at a counterintuitively large 7 ft/sec to model a prototypical 60mph, than to run it at the standard 0.55 ft/sec that is well documented! So, as strange as it is, in order for the physical forces to be correct on the model, you have to run it nearly 13 times faster than the common wisdom holds (but if you filmed it you'd have to slow it back down by a factor of nearly 13 in order for it to look right to human perception!) It's fun how a little physics can turn your world up-side-down. (Now I hate to plug my own blog because it makes me look self-important, but I've been looking around at this issue due to a conversation a friend and I had, and I posted a quickie blog article for him that has more details on this topic than I have room for here. If you care it is at blog.PaplooTheLearned.net: Essay: Scale Railroading (Part 2: Time) ).
Wow, I actually followed that. Not that I'm going to run my trains faster. I would imagine they know this trick in Hollywood, no? (Or at least they used to know it before everything went CG...)
Velocity is d/t. It is not d/[t^2]. Distance is directly scaled to 1/160th [or whatever you are running] of the actual distance. Speed is nothing more then a the measure of distance per a unit of time. The physics do not hold true for forces because atoms and density does not scale out so well, and then neither does momentum. Mass is not a quanitiy that is scaled down; we accept a weight that will stay ont eh rails, and leave it at that. Only distance is actually scaled down in our models. Yes, most modelers run their trains at ridiculously slow speeds. They then commit a gross educational moment when they tell the public they really move that slow. In reality, most trains at least around here are moving very fast - they blaze through town at around 45-50 mph. That's In town, and they DO appear to be moving like bullet trains! They only seem slow along highways because you are in a speeding car that is moving at 85-90 mph...and the train is only moving between 60 and 75. But most modelers insist on moving at 45 mph...or 25 mph, and insist the real ones run that fast on a daily basis. If the real ones ran 25mph over their entire routes, they would have lost all traffic a long time ago to trucks. Sorry Folks. Someone has pulled your leg. And yes, running slow Is the only way to make your layouts seem bigger then they actually are.
This thread is getting somewhat obscure. The original title is Scale MPH vs. Scale Speed. Those terms are really the same thing and in their simplest form they can both be interpreted as scale distance divided by scale time, as Benny states. By definition, N scale distance = actual distance*160, and for most people, scale time = actual time (unless you use a fast clock). So, for most people, scale speed = actual speed*160. That's it. The question of how forces scale is completely different from how speed scales. It's true that the centrifugal force felt by a body moving around a curve at speed v scales like v^2, as paploo states, so super-elevation physics is different for scale models than for the prototype. But that does not imply that scale speed is more complicated that the simple linear scaling noted above.
I just measured the straight section between the curves on my door mounted Unitrak test track. It is just a little over 4 feet. I got a train running at what appeared to be about the speed I like to run on an NTRAK layout. I timed it at a little over 7 seconds. According to the calculator, that is about 60 MPH. Works for me.
Ahem! Actually some trains do actually go 45. Not everyone models a high speed class I mainline. In the rural midwest, where I live, many of the tracks speeds are in the 45 mph category. Some are even slower, especially through towns (some as low as 10 mph). Even the mighty Omaha Borg still has jointed track in this area. And they do run at those speeds on a daily basis.
Near as I can tell from this thread is that people run trains based on: What looks good What "feels" right What is the most entertaining All of which could be the same thing. The seem to have a complete disregard for physics.
Physics aside. It's a hobby folks...not rocket science ! My original layout is on a 32HC door. Track is set 2 inches from the edge OC. By my calculations the radius is 14 inches. I dont know what that translates to proto...but at full throttle I cover 4 feet in -4 seconds...and that train is haullin !!! I can sustain that speed till it burns up...and never throw a car! I would love to see a proto take a curve at whatever the 'real' radius that translates too at 100+ MPH !:tb-wink: I'll bring the popcorn ! If it feels good...if it looks good...and your happy...its all good ! My footer says it all...LOL :tb-biggrin:
Before trucks took away a lot of their traffic, many mainline trains were that slow. And 25 mph sounds about right for a branchline train.
It depends on your goal... I wouldn't either. Because what is more important to most is not balancing of gravity (which doesn't scale), but the perception of scale speed that our brain gives us. And for most people, it works out best to scale space but not time. Of course, this is because the perception of the passage of time is very engrained into our brains and is non-debatable, while our brains have no troubles with the concept of scaled distance. Indeed, both when learning (and teaching) special relativity, time dilation was the most difficult concept for students, while length contraction wasn't a problem (relatively speaking). Indeed, the famous twins paradox is such a problem *because* time running at a different rate is so difficult for our brains to comprehend, and the solution is usually pretty obvious when it is explained in terms of length contraction! So that being said, the way that one normally does it is fine and I'm not trying to say everyone should switch. Instead, the point is that there is more than one way to do it depending on your goal.
Yes. I don't think anyone said that it was. v=d/t is the equation that I used as well for all my computations. I'm not quite sure where you got v=d/t^2 from, as the units don't even work out (dimensional analysis sayds that [v] = mph, [d]=mi, [t]=hr, so [d/t^2]= mi/hr^2=mi/s/s, which is a measure of acceleration, not velocity.) Could it be that you misread an equation where, under gravity, an object falls a distance d base on the equation d=(1/2)*g*t^2, which would give the acceleration of gravity as being proportional to d/t^2? (This relationship being related to the same root cause of all of my time scaling results, which is that gravity does not scale, and thus time has to run at a different rate to make the physics work out.)
For some rocket science is a hobby, though I do agree with your sentiment. My point, which was spured by the original poster asking a question about the validity of the speed scaling equations, was just to make people think about why we do it the way we do, if they think it is interesting to do so. I realize that not everyone finds physics a fun hobby. But that is beside the reason why I'm replying. In N scale, you can get the prototypical radius by multiplying the scaled radius by 160, which gives 2,240 inches, or about 187 feet. One can then find the curve degree to good enough accuracy by dividing 5730 ft-degrees by 187 feet to get a 30-31 degree curve. To put this in perspective, an article I read in model railroader says that mainlines are usually 1-2 degrees, though mountainous terrain pushes curves of 5-10 degrees, and the limit for a 4-axle unit with rolling stock is 20 degrees. The reason why we can have such prototypically insanely tight curves in our models, then, is related to the same physics as the super-elevation stuff I presented above. More specifically it is because we run our models 13 times slower around the curves than the forces would balance at, so we can afford to have tighter curves without causing a problem.
