Critique my train design please

Okay. I'm not looking for "the G forces will tear it apart" or anything like that. I've done all the calculations and except for the pivot on the 0 car, everything holds with a factor of safety of at least two. The pivot on the zero car actually is dfferent a bit, but it got late and I was tired of modelling for the night.

This is just chasis but seat design isn't really that interesting anyway.

I would like the know what peole's thoughts on train dynamics are. I think I have everything more or less correct, but there are those with much more experience in these matters than I.

On with the pics:

Wheel assembly. This sits on tapered bearings on either end.

View one View Two View Three

Axles

View one two three four five six

Single car (not the zero car)

view one two three four [url="http://www.andrew.cmu.edu/~jjb2/carassembly0006.jpg" target="_blank">five

And the train

one two

If anyone can catch errors like the ones on the horrible trains on RWB, that would make me pretty happy. For size reference the wheels are six inchers and the major tubing is 2.5 inches square.

And so there's no confusion, that is wooden track, seven laminations.

Thanks

Coma

It has upstops. Look at the position of the lower member relative to the guide wheels. The upstop rides an inch and a half under the lip of the track. Or did you mean upstop wheels?

*** This post was edited by Comatose on 5/29/2002. ***

Very helpful, except not. WHAT am i missing that has anything to do with train DYNAMICS? Please don't say "seats, chain dogs, anti-rollbacks,brake fins, floor, lap bars, pretty paint job."

First attempt at building a roller coaster? Why yes, it is in fact. When was yours?

*** This post was edited by Comatose on 5/29/2002. ***

How in the heck does it go through turns? It looks so ridgid, no moving axels or anything... oh wait... you're trying to make RWB's trains? Excellent start!

-----------------
Jes
Jes's Roller Coasters DJ Jes MCS Please, Feel Free To Call Me Jes!
"Thank You Jeff Putz"

Sorry. I should have put more detail in the stock components, but i don't design them i buy them. The wheels have bearings built in; they turn. The conectons between the cars are trailer hitches, so they have three axis rotation. The wheel assemblies pivot on along the vertical axis, the sway bar keeps them both aligned.

The axles do move. it's just they aren't solid axles. The wheels orient themselves to the track and the cars also orient themselves to the track, but the wo motions are independent. So that's where yaw freedom comes from. Pitch and bank freedom are a function of the trailer hitch. It SHOULD be very flid, at least much more so than most current wooden trains. Of course we will see.

Antuan: I hate Pro-E with a passion that I can't even explain. It's just so user-unfriendly. I actually did this visualiation model in Lightwave. I did some of the analysis in Ansys and worked with Solidworks for some if it, but for many parts the time to set up a computer model isn't worth saving you the 5 min of hand calculations, so I haven't a complete model of it in a real engineering package. Lightwave I'm fast with though.

Keep the comments coming, I'm off to machine, but will definitely look around lunchtime to see if I made a cataclysmic error and need to rework something. And obviously won't finish anything today.

-----------------
Posting, "Me too" like some brain dead AOLer. I ought to to the world a favor, and cap you like old yeller...

As for the basic chassis design, it looks reasonable enough, but I have one concern for the basic configuration: It looks like either all the cars are running downhill from axle to hitch except for the lead car, or the lead car is running uphill with respect to the rest of the train. Or is that merely an illusion?

With the dual guide wheels, you could probably safely dispense with the tie rod. Watch it carefully on curves, because with the tie rod you have to make sure the left and right rail are always in sync with one another; without the tie rod the wheel sets will track independently. Although that may not really be an issue. As long as you have the hitch in the right place (which you do), you could probably safely eliminate the wheel set articulation on all but the zero car, using only one guide wheel. It would be slightly less ideal, but a lot simpler, and would probably work just as well.

How about weight handling? Can the hitch handle half the weight of the car? Can the axle handle the full weight of the car (actually half the weight of two cars)? Would a tandem axle be desirable?

Finally, are the dual guide wheels far enough apart? To optimize axle steering, you might want to position the guide wheels further forward and backward of the road wheel.

Can you put together a set of these frames to set under the Raging Wolf Bobs train? :)

Actually the axle set looks a little like LoCoSuMo. :)

--Dave Althoff, Jr

*** This post was edited by RideMan on 5/29/2002. ***

I like what I see here. For starters, I've got to assume that this is for a wooden coaster train.

Is it just me or do the guide wheels look to be inset too much from the road wheel? Maybe a larger diameter guide wheel would work along with what RideMan said about moving the wheels outward parallel to the rail. Take a look at the wheel asseemblies of Big Bad Wolf or Iron Dragon, then look at Top Gun at PKI. That's a perfect example.

Another concern is with the chassis and the rail during a turn. I think it's ok, but will the frame make contact with the rail during a sharp turn. Maybe another model will reveal this.

Finally, I think that upstop wheels are a must. Looks like you could easily put a set off the bottom of the axle frame that extend from your current configuration. Without wheels, you could loose a lot of energy if that square stock starts riding the bottom of the rail.

Nontheless, keep up the good work and let us know what the next revision look like.

-----------------
Scott W. Short
scott@midwestcoastercentral.com
http://www.midwestcoastercentral.com

--Dave Althoff, Jr.

I checked the turning radius, and it'll just barely clear a 20 foot in diameter circle.

Moved the bracing in a bit, v2 clears 15 feet. I don't particularly want to be in a turn less than 15 feet in diameter, though the potential to slap hands with the people in the back of the same train does hold a twisted sort of intriugue.

Rideman: you are correct, the rear trains angle down and the front does not. It juat makes attaching the zero car easier. Beyond looking kinda finky, does it have a detremental effect on the ride of the train?

The tongue and hitch will hold: the adjusted weight of a car is 14400 pounds (800 pounds of rider, 400 pounds of car, going through a 3g maneuver with a factor of safety of four) and the eight distribution is such that roughly 1/3rd of the weight is on the hitch. The hitch is rated at 5000 lb.

Looked into not articulating the wheels: would be fine on a 60 foot turn, but really not good on a 20 footer.

The spacing between the guide wheels was jsut a pracical consideration: they're attached with 2x3x.25 angle, and two inch base angle doesn't come usually longer than three inches of protrusion. Could be extended, but it'd be more work. I like the one extended guide on LoCoSuMo, but i was designing this project well before that ever went public and I'd rather not start blatantly ripping ideas off. Either way, what sort of separation would be appropriate, curently it is seven inches between points of contact.

The axles will hold. I can put up some or all of the engineering behind the design if anyone is interested.

The only thing I'm worried abut is my design rdes totally above the rails, whereas a more conventional design laces their riders feet at least between them. Will the higher center of gravity have ill efects, besides a bit rougher transitions?

Coma

I'll get pics of v.2 and really bad (webcam) really early (parts) sneak pics of v.3 up tonight if anyone is interested in either.

--edit-ShiveringTim: The guide wheels do look inset quite far, there is a reason. The track is more or less standard woodie style. The guide wheels ride the top third of the track, which is pushed out past the main section to form a lip for the upstop. However, the main wheels ride on the tich section beond the lip. Maybe some ascii would help

|M|

|A|

|I |

|N|

====== |guide wheel|

==== |Upstop|

======

wierd formatting, but maybe that helps. Doing it that way keeps a lot of shear force off of the lip of the track (where it's weak)

*** This post was edited by Comatose on 5/29/2002. ***

---

RideMan said:
smartass
Looking at it again, I think you need rounder wheels.
/smartass

--Dave Althoff, Jr.

---

I disagree, if poor train design isn't causing the train to jump and bang around everywhere, how besides polygonal wheels do you propose to get that classic woodie feel?

---

Comatose said:
"I checked the turning radius, and it'll just barely clear a 20 foot in diameter circle.

Moved the bracing in a bit, v2 clears 15 feet. I don't particularly want to be in a turn less than 15 feet in diameter, though the potential to slap hands with the people in the back of the same train does hold a twisted sort of intriugue.

Rideman: you are correct, the rear trains angle down and the front does not. It juat makes attaching the zero car easier. Beyond looking kinda finky, does it have a detremental effect on the ride of the train?

---

I can't think of any particular problem with it aside from being a little odd. You could even it out, I suppose, by dropping the hitches on all but the first car just enough to put the chassis level. The only thing I can think of that might be a consideration is the amount of vertical clearance for the hitch arm.

---

The tongue and hitch will hold: the adjusted weight of a car is 14400 pounds (800 pounds of rider, 400 pounds of car, going through a 3g maneuver with a factor of safety of four) and the weight distribution is such that roughly 1/3rd of the weight is on the hitch. The hitch is rated at 5000 lb.

---

Hmmm...and you're loading it with about 4,800 lbs. I suspect that is the weak point of the design right there, and I think if we were to research it (actually I am hoping to do exactly that...) we might learn that the reason PTC did their trailers the way they did was to reduce the load on the hitch, as they put almost the total weight of the car on the axle. It's close, but it ought to work...and you're more of an engineer than I am. If the hitch proved to be a problem, I suppose the obvious thing to do would be to do as Boodley did with his train and use a spherical bearing instead. Some of those bearings have ridiculous load ratings and require no lubrication...

---

Looked into not articulating the wheels: would be fine on a 60 foot turn, but really not good on a 20 footer.

The spacing between the guide wheels was jsut a pracical consideration: they're attached with 2x3x.25 angle, and two inch base angle doesn't come usually longer than three inches of protrusion. Could be extended, but it'd be more work. I like the one extended guide on LoCoSuMo, but i was designing this project well before that ever went public and I'd rather not start blatantly ripping ideas off. Either way, what sort of separation would be appropriate, curently it is seven inches between points of contact.

---

You're the engineer here. :) As I see it, the idea behind the dual guide wheel is so that both wheel sets will be steered by the curve of the track. I'm not sure what the effect is of increasing or decreasing that distance because I haven't really studied it. But another thought is that since you have the tie rod, you could probably dispense with the dual guide wheels altogehter, provided that the remaining guide wheels are ahead of the road wheels...instead of steering the inside wheel carrier with its back guide wheel, you steer it with the front outside guide wheel. Hmmm...perhaps that's the key to the guide wheel spacing: the turning radius of the wheel carrier.

---

The axles will hold. I can put up some or all of the engineering behind the design if anyone is interested.

The only thing I'm worried abut is my design rdes totally above the rails, whereas a more conventional design laces their riders feet at least between them. Will the higher center of gravity have ill efects, besides a bit rougher transitions?

---

Note that virtually every coaster train design on the market places the entire passenger compartment including the footwell above the axle. The exception is the Arrow looping coaster train, with the axle sets between the cars. Oh, and the PTC trains have the floors set a little lower...but it's about even with the track, not below it. With that in mind, I doubt that it would be a problem.

--Dave Althoff, Jr.

---

RideMan said:

---

The tongue and hitch will hold: the adjusted weight of a car is 14400 pounds (800 pounds of rider, 400 pounds of car, going through a 3g maneuver with a factor of safety of four) and the weight distribution is such that roughly 1/3rd of the weight is on the hitch. The hitch is rated at 5000 lb.

---

Hmmm...and you're loading it with about 4,800 lbs. I suspect that is the weak point of the design right there, and I think if we were to research it (actually I am hoping to do exactly that...) we might learn that the reason PTC did their trailers the way they did was to reduce the load on the hitch, as they put almost the total weight of the car on the axle. It's close, but it ought to work...and you're more of an engineer than I am. If the hitch proved to be a problem, I suppose the obvious thing to do would be to do as Boodley did with his train and use a spherical bearing instead. Some of those bearings have ridiculous load ratings and require no lubrication...

---

No no. I should have been more explicit. The concept of a factor of safety: design for no faulire on an overlaod of that amount. I am working with a factor of safety of four on everything critical. The actual tongue weight when the car is stationary is 400 pounds. Multiply that by four to get the "padded" tongue weight gets 1600 pounds. Max design spec for this car is +3g, so multiply by three to get the "design" tongue weight of 4800 pounds.

It's easier to just pad your weight figures at the beginning (by twelve in this case) than do that at the end of every calculation, hence "adjusted" weight. The tongue will not fail catastrophically. It might bind and be obnoxious, though. I'll peek at spherical bearings.

John

--edit--

I rather doubt that they did it to reduce the weight on the trailer hitch. My theory:

With the train balanced on the axle completely, there is no moment (force cross distance, linearly related to stress) due to bending exerted by the trailer hitch. In my design the moment due to bending is nearly 350,000 inch pounds. Nearly half the projected cost of the train is due to having to resist that moment, in the form of thicker, wider spaced main chasis tubes. So if I cold think of a good way to balance on the axle and keep the train dynamic good, the cost of the train (materials) would go down drastically. From a stress analysis point of view, balancing on the axle is te only thing that makes any sense.

*** This post was edited by Comatose on 5/29/2002. ***

---

RideMan said:

---

Looked into not articulating the wheels: would be fine on a 60 foot turn, but really not good on a 20 footer.

The spacing between the guide wheels was jsut a pracical consideration: they're attached with 2x3x.25 angle, and two inch base angle doesn't come usually longer than three inches of protrusion. Could be extended, but it'd be more work. I like the one extended guide on LoCoSuMo, but i was designing this project well before that ever went public and I'd rather not start blatantly ripping ideas off. Either way, what sort of separation would be appropriate, curently it is seven inches between points of contact.

---

You're the engineer here. As I see it, the idea behind the dual guide wheel is so that both wheel sets will be steered by the curve of the track. I'm not sure what the effect is of increasing or decreasing that distance because I haven't really studied it. But another thought is that since you have the tie rod, you could probably dispense with the dual guide wheels altogehter, provided that the remaining guide wheels are ahead of the road wheels...instead of steering the inside wheel carrier with its back guide wheel, you steer it with the front outside guide wheel. Hmmm...perhaps that's the key to the guide wheel spacing: the turning radius of the wheel carrier.

---

Well, just a few thoughts on guide wheels: Guide wheels approximate a tangent to the track, the farther apart the less accurate. Now whether absolute accuracy on a track with the tolerances of wood is desirable is not something I can coment on much. The farther apart the guide wheels are the greater the moment exerted on them, which means that the carriers have to be stronger, but also that they'll steer properly with less lateral force. Too short may not overcome friction and steer at all. Too long will get jammed in a transition to a curve. Given my druthers, i'd want longer and spring-loaded, too. Actually, if the tie bars were spring loaded, the wheels would steer with a relatively short spacing. My guess is that the're placed widely to deal with track irrgularities and imperfections, as narrow placement would make for a rather rough ride.

I don't see how you can articulate the wheel sets on a trailer train with respect to yaw and not have dual guide wheels. What would stop the wheel set from rotating beyond the point where the main wheel is tangent to the curve? It's very possible I don't understand properly.

Coma

--edit-- never buy $8 officemax keyboards, not even as a backup.

*** This post was edited by Comatose on 5/29/2002. ***

---

Comatose said:
I don't see how you can articulate the wheel sets on a trailer train with respect to yaw and not have dual guide wheels. What would stop the wheel set from rotating beyond the point where the main wheel is tangent to the curve? It's very possible I don't understand properly.

---

I haven't given it much thought in terms of the trailer. But it's exactly what PTC did with their junior cars. The flanged road wheels are mounted so that they pivot on a point located rearward of (for the front wheels) or forward of (for the back wheels) the axle, then they're tied together with a tie rod. Switching to your axle set, if you have a single guide wheel located ahead of the road wheel, the outboard guide wheel will hit the edge of the curve and steer the outboard wheelset inward. Trouble is, at this point, without the rear inboard guide wheel, the inboard wheel set will plow straight ahead and ultimately end up at a tangent to the curve. That's no good. But that problem is fixed by adding the tie rod. Exiting the curve should steer the inboard guide wheel in the same way, resetting the wheel assembly. I think. :)

I gotta go home. I'll see this thread again tomorrow night, but then I'm at Stark Raven Mad over the weekend, so I may not say much. This is an interesting conversation, though. I think I'm getting something out of it anyway... :)

--Dave Althoff, Jr.

Just two webcam (world's worst webcam) mega early pics of v3.

One super secret thing

Several super secret things

Only about 80 super secret things to make yet before i move on to top-super-secret things.

I'm excited. I don't expect any of you to share said excitement. yet.

*** This post was edited by Comatose on 5/29/2002. ***