Not signed in (Sign In)

Discussion Tag Cloud

Categories

Vanilla 1.1.9 is a product of Lussumo. More Information: Documentation, Community Support.

  1.  
    Archy's Friction List


    (vout^2 - vin^2)/1961.33

    vin = the speed at the beginning of the colored section in mps
    vout = the speed at the end of the track in mps

    With thanks to Wardaide/mgizozlodu for his formula, source
    If you would like a utility to help with this, download ROLLER97's calculator here.

    For FVD++
    (newton2 value) x 9.81

    FVD++ now uses same values as Newton2.


    ---
    Bolliger & Mabillard:

    Hyper
    -with 8 cars =0.0314 FVD++=0.308

    Inverted
    -with 8 cars use =0.0313 FVD++=0.3071

    Twister/Floored
    -with 6 cars use =0.0471 FVD++=
    -with 7 cars use =0.0392 FVD++=
    -with 8 cars use =0.0313 FVD++=0.3071

    Floorless
    -with 8 cars use =0.0313 FVD++=0.3071

    Standup
    -with 8 cars use =0.0337 FVD++=


    ---
    Gerstlauer:

    Eurofighter II
    -with 1 car (obviously) use =0.0280 FVD++=0.27468


    ---
    GCI aka Great Coasters Int.

    Millennium Flyer (Wooden, Trailored 2-Seat)
    -with 12 cars use =0.0157 FVD++=


    ---
    GG aka The Gravity Group

    Wooden Coaster (Classic 4-seat)
    -with 6 cars use =0.0157 FVD++=


    ---
    Intamin:

    Standard sitdown
    -with 5 cars (ie, accelerator) use =0.0313 FVD++=
    -with 8 cars use =0.0295 FVD++=


    ---
    Maurer Söhne:

    X car
    -with 1 car use =0.0471 FVD++=
    -with 2 cars use =0.0316 FVD++=
    -with 3 cars use =0.0157 FVD++= 0.154017


    ---
    Schwarzkopf

    Looping Star/Silverarrow/Shuttle Loop (lap bar)
    -with 6 cars use =0.0314 FVD++=


    ---
    Vekoma:

    Flying Dutchman
    -with 1 car use =0.0469 FVD++=
    -with 6 cars use =0.0313 FVD++=0.3071

    Suspended Looping (SLC)
    -with 10 cars use =0.0321 FVD++=0.3149

    Sitdown
    -with 7 car, use =0.0313 FVD++=


    ---
    post your own and i'll update the wall.
    •  
      CommentAuthorRNB
    • CommentTimeJan 26th 2012
     
    Is it correct that if I want a Vekoma Dutchman with 1 car I can simply divide 0.0531 by 6? (you don't have to calculate it if I'm wrong, this is just an example)
    • CommentAuthorwardaide
    • CommentTimeJan 26th 2012
     
    Awesome, and thanks for the recognition Archy (I used to be mgizozlodu, but after leaving the school that name was based on I created a different username). I'm glad someone still uses that thread besides myself every now and then. I just wish I could finally commit to a project and finish it. But anyway, I thought I would mention that because NoLimits' physics engine uses a velocity-dependent friction value, you also need to take into account the speed of your ride. So it would be best to estimate....using basic energy formulas....the top speed of the ride and use that upon testing for the friction coefficient for newton. Then you would have to use trims to adjust the speed if need be during some of the slower sections of your ride to compensate for the difference in friction values between NoLimits and Newton. Just thought I'd point that out. But thanks for posting what you've found, I think it will definitely prove useful, especially so that people won't have to do a new test for every ride they build.
  2.  
    Posted By: The_ArchitectSchwarzkopf

    Lap Bar/non inverting
    -with 6 trains use 0.1127


    The capacity must be amazing on that one!
  3.  
    Posted By: RNBIs it correct that if I want a Vekoma Dutchman with 1 car I can simply divide 0.0531 by 6? (you don't have to calculate it if I'm wrong, this is just an example)


    no it will be 0.0469, i've updated the list to accommodate this.
    • CommentAuthorwardaide
    • CommentTimeJan 26th 2012 edited
     
    Posted By: RNBIs it correct that if I want a Vekoma Dutchman with 1 car I can simply divide 0.0531 by 6? (you don't have to calculate it if I'm wrong, this is just an example)


    No, there would be a rather big difference between 1 car and 6 cars but it would not be by a factor of 6. I don't know exactly what it would be, and quite frankly I don't feel like doing the test right now to find out, but it wouldn't be that.

    Edit: Damn you Archy, you beat me to it....haha.
    •  
      CommentAuthorL Bosch
    • CommentTimeJan 26th 2012 edited
     
    Gerstlauer Eurofighter II

    0.0280

    But this is no exact value, just tryin'...
  4.  
    Posted By: devilsrule911
    Posted By: The_ArchitectSchwarzkopf

    Lap Bar/non inverting
    -with 6 trains use 0.1127


    The capacity must be amazing on that one!


    That's wrong, i've changed it. It's acutally 0.0314. The rest are all cool. Not 100% nail on the head perfect friction, but good enough to make a ride in Newton 2 with. I always make rides with a sample rate of 10,000 hz btw.
    • CommentAuthorwardaide
    • CommentTimeJan 26th 2012
     
    ^I think he was referring to the fact that it says "with 6 TRAINS" instead of "with 6 CARS".
  5.  
    Posted By: L BoschGerstlauer Eurofighter II

    0.0280

    But this is no exact value, just tryin'...

    Thanks, added :D
    -if people use these and they are wrong and you have a fix value, post it here!


    Posted By: wardaide^I think he was referring to the fact that it says "with 6 TRAINS" instead of "with 6 CARS".

    Herp derp, fixed :D
    •  
      CommentAuthorLing
    • CommentTimeJan 26th 2012
     

    OMG, I have been waiting for something like this. Thanks Archy!

    •  
      CommentAuthorLiMiT
    • CommentTimeJan 26th 2012 edited
     
    Can someone please tell me how you figure this out? I take physics and this might be useful. So what values would I put in?
    • CommentAuthorwardaide
    • CommentTimeJan 27th 2012
     
    ^Follow the link he provided, the whole thing is explained there.
  6.  
    There is a much more simple to measure friction, just build a 100m long segment in both newton and in NL.
    Change the coastertype and the number of cars to what you want to use.
    Add a transport segment before the 100m segment and change the launch speed until the train get's as close as possible to the end of the 100m segment.

    Now set the start speed in newton to the same speed used in NL.
    Change the friction until the exit speed in newton is as close as 0 as possible.

    You've now got exactly the right friction value, start build your coaster.

    I hope it didn't get to complicated.
  7.  
    That is a good method, sounds a bit tedious but then so is the existing method. Post any values you find that work!
  8.  
    On my macbook right now, no newton until sunday.

    But you might test it yourself to see if it work.
    •  
      CommentAuthorLiMiT
    • CommentTimeJan 28th 2012 edited
     
    X-Car 2 cars - .00316

    It worked, but check yourself first in case I screwed up somewhere.
  9.  
    ^^I like your username. And yes, I get it. :)

    Thank you so much for creating this discussion, it's helping me already!
  10.  
    Posted By: LiMiTX-Car 2 cars - .00316

    It worked, but check yourself first in case I screwed up somewhere.


    Thanks, updated! If i had to check every one of these myself then it wouldn't be much of a help you guys posting them, lol. So i'll put them up and i've got my disclaimer at the bottom. If anyone tries these and they don't work, post your fixed version.
    •  
      CommentAuthorLiMiT
    • CommentTimeJan 28th 2012 edited
     
    Posted By: The_Architect
    Posted By: LiMiTX-Car 2 cars - .00316

    It worked, but check yourself first in case I screwed up somewhere.


    Thanks, updated! If i had to check every one of these myself then it wouldn't be much of a help you guys posting them, lol. So i'll put them up and i've got my disclaimer at the bottom. If anyone tries these and they don't work, post your fixed version.

    That works. It's a good system.
    •  
      CommentAuthorLiMiT
    • CommentTimeJan 28th 2012 edited
     
    Inverted 4-across 8-Car - .0313

    Sorry for double post, but also I noticed the X-car 2 cars has an extra zero I put in by mistake. You might wanna change that. It's supposed to be 0.0316, not .00316.
  11.  
    ^^ added, thanks. Also added twister 6 seat (my own).
    • CommentAuthorbigjoe97
    • CommentTimeJan 29th 2012
     
    I've found that with a 7-seat Intamin Hyper, the Newton default works pretty well.
    • CommentAuthorMaverick
    • CommentTimeJan 30th 2012
     
    Just so no one is blindly plugging in numbers..

    (vout^2 - vin^2)/1961.33

    Is this equation:

    Vf^2 = Vi^2 + 2 * a * d

    Where you are solving for PART of the acceleration (a) function, and the distance (d) is predetermined along with the acceleration due to gravity (part of a) and the initial velocity (Vi). What we really need to get an accurate friction coefficient (whilst ignoring rolling friction, wind resistance, humidity, temperature) is to find out from Ole how the software determines acceleration/deceleration of the ride. The way I see it, in the simulator, if your ride is designed correctly and safely then a 10% difference in velocity will not dramatically affect the riders experience.
  12.  
    Someone should probably make this a sticky in case this somehow gets sunk down the forums.
    •  
      CommentAuthorLiMiT
    • CommentTimeJan 30th 2012
     
    ^^I remember that equation from Physics. When you rearrange the equation, you get (Vf^2 - Vi^2)/(2ad). So that explains what the equation was derived from.
    • CommentAuthorMaverick
    • CommentTimeJan 30th 2012
     
    Yeah, most people should remember it from physics class... but it's better to understand the math than to simply plug in numbers (i.e. Newton...)
    •  
      CommentAuthorLiMiT
    • CommentTimeJan 31st 2012
     
    Well, that's good to know. I never knew the purpose of that value in Newton.
    • CommentAuthorROLLER97
    • CommentTimeJan 31st 2012
     
    Intamin hyper with 8 cars use 0.0623
  13.  
    Intamin Hyper: For 4 cars (Mega-Lite), use 0.0157.
  14.  
    Does friction in NL vary with respect to forces and speeds as it does in real life? I would assume (hope) so, but I haven't tested myself.
  15.  
    I wouldn't think so, but I could be wrong. Pumps would start really slowing down coasters.
    • CommentAuthorbigjoe97
    • CommentTimeFeb 1st 2012
     
    I'm pretty sure it varies according to speed. Not sure on force, though.
  16.  
    added the above 2 for intamin, thanks guys! This is going well, the wall is slowly getting full up. Must admit the 6 row B&M twister didn't work out too well, needs updating.
    •  
      CommentAuthorBBSpeed26
    • CommentTimeFeb 1st 2012 edited
     

    I'm quite certain that friction in NL varies based on speed and forces, which is why the "launch it down a straight track, measure the start and finish speeds, and plug it into a formula method" doesn't work (or at least, doesn't work accurately enough for my tastes). It's also why there is no such thing as a "correct" friction for a given train style / length, because newton factually does not handle friction the same way NL does.

    The numbers in this thread may get you close, but you'll still find speed and G variations as your track gets longer, or if your track is on average faster/slower than the track that was used to develop these values, or both. I like where your guys' heads are at, but the numbers in this thread won't always work.

    Source: I've done this more times than I care to count.

    •  
      CommentAuthorLing
    • CommentTimeFeb 1st 2012
     

    Here's hoping NL2 makes this way easier on us accuracy freaks.

  17.  
    Oops... I'm sorry, I made a mistake in the calculation of the 4-car Intamin Hyper. It is actually just like a 5-car Intamin Accelerator, 0.0313. Apologies for any confusion made by that mistake.

    I would also like to note that 0.0157 is actually the friction value for a 12-car GCI Millennium Flyer (Wooden, Trailered 2-Seat).
    • CommentAuthorbigjoe97
    • CommentTimeFeb 2nd 2012
     
    That sounds right to me - woodies are ridiculously fast in NL.
    • CommentAuthorwardaide
    • CommentTimeFeb 2nd 2012
     
    The friction value you use in Newton works the same as in NL as if the NL environment were in a vacuum, thus eliminating any drag force. So in reality my method works fine so long as you launch the train at what would be the maximum speed your ride will achieve which, admittedly, is something I forgot when I wrote that. So you should also include what the input speed you used to perform your test, as this will make a difference.

    And again, this value is the coefficient of friction, which NEVER changes. So if the value is .0295(standard value) at 1g it will be .0295 at 4g or 1,000,000g. The only thing that will change is the amount of force is applied against the direction of the train, NOT THE COEFFICIENT OF FRICTION. So, this method, while inherently flawed due to the fact that Newton does not account for any type of drag whereas NL does, does exactly what it is intended to do. You just have to realize that as the speed of your ride decreases in Newton, you must lower the forces to accommodate the fact that the true speed in NL will be higher, thus increasing the forces as compared to what they were in Newton.
    •  
      CommentAuthorBBSpeed26
    • CommentTimeFeb 2nd 2012 edited
     

    You just have to realize that as the speed of your ride decreases in Newton, you must lower the forces to accommodate the fact that the true speed in NL will be higher, thus increasing the forces as compared to what they were in Newton.

    Doesn't that mean that this method doesn't work? Sure it gets you a number, but it gets you a number that will be wrong as soon as your track differs from the flat, launched test track. The longer you go, the more wrong it will be. I guess that's predictable, but it's not particularly useful if you're building a long ride...

    The problem is with how you're measuring it. A flat, straight track with speed held at 1G is a terrible approximation of a coaster. Your calculated coefficient with this method will always be too low, because the test track will always experience fewer Gs (and thus, less friction) than an actual coaster layout would. I don't know about you, but I'd rather set the "correct" part somewhere in the middle, whereas this method always, always will result in dead-on newton behavior at the beginning that gets progressively more wrong as you build. Always. In some cases, it will get so wrong that you'll get lats in flat turns or the train will valley in the sim because, with this method, newton will always think the train is going faster than it actually will be.

    To reiterate - the list is a noble cause, but even if you prescribe a number for a whole range of max speeds for every possible train and car combination, your calculated coefficient of friction will always be too low.

    For my money, the best method is still Buster's AHG method:

    Sketch your whole layout out by hand (it can be extremely rough, that's fine), run the coaster through that section and note the start/exit speeds. Plug this section and those speeds into the AHG, and the AHG will give you the friction value. Obviously your track isn't going to match up perfectly your final newton track, but it doesn't need to.

    This method does take into account the Gs experienced throughout the ride, and the resulting friction coefficient means that you'll be slightly wrong at the beginning of your track, dead-on accurate in the middle, and slightly wrong at the end. This makes for the least possible newton-induced error in track shaping and G control throughout a ride (even a long one). I did this with Durango (which is stupidly long), and was never off by +/-0.5 G between the target in Newton and what the results were in the sim.

    Just saying. I'm trying to save you guys the pain of using a number off this list only to be frustrated by how far off the track is as you build more of it. Nothing sucks more than building half a track only to realize that your coefficient of friction is so far off that correcting it requires a total rebuild of the track.

    •  
      CommentAuthorrattler
    • CommentTimeFeb 2nd 2012 edited
     
    By: crazydud619dxSomeone should probably make this a sticky in case this somehow gets sunk down the forums.


    I agree; this is extremely useful information. It would be a real pain to have info like this go away.
    • CommentAuthorwardaide
    • CommentTimeFeb 2nd 2012
     
    Posted By: BBSpeed26Your calculated coefficient with this method willalwaysbe too low


    Actually, the coefficient with this method will always be too high, because air drag increases with speed. This means that when performing your test in the sim the faster your test speed is the higher of a coefficient you will get from the test. But that's irrelevant, as I think we can both agree that unless Newton gets it's physics updated, any method will be inaccurate. One method may be better than the other, and I will not try to say which method that is. I think my method works fine as long as you don't mind thinking a little for yourself, while Buster's works if you want to do the basic hand work first and then not think while using Newton. I prefer my method due to the fact that I tend to design on the fly and am way too much of a perfectionist to roughly draw out my track in the editor and then redo it in Newton. So it really does, as it always does, come down to preference.
  18.  
    Am I the only one that just leaves the value alone? I think it doesn't really affect much, and MCBRs will always "refresh" your speed in Newton so the accuracy can stay relatively good. I never get wild G's from my Newton rides, but I guess if you're really designing "at the limit" you'll get some things like that.
    •  
      CommentAuthorBBSpeed26
    • CommentTimeFeb 2nd 2012 edited
     
    @Wardaide

    Preference aside, might I suggest a slight revision to your process? Bank the whole thing at 90 degrees and make 100m of oscillating -1 - 4G track. You'd still get a friction value, but your number should be a little better*. And to reiterate your point from before so it's not lost in the wall of text...

    If you use wardaide's/archy's method, the max speed of your ride needs to be the number you use to calculate the friction. There will not, for example, be a one-size-fits-all friction coefficient for a 2-car Eurofighter.

    *disclaimer, I have not yet tested this
    •  
      CommentAuthorbaadrix
    • CommentTimeFeb 3rd 2012
     
    Posted By: AJClarke0912Am I the only one that just leaves the value alone? I think it doesn't really affect much, and MCBRs will always "refresh" your speed in Newton so the accuracy can stay relatively good.
    Nope, I only increase the given coefficient slightly (0.002-0.005), if the train is very long (e. g. 9 cars at a 2-seat Hyper coaster).
  19.  
    Posted By: BBSpeed26
    If you use wardaide's/archy's method, the max speed of your ride needs to be the number you use to calculate the friction.


    Good shout, i'll remember that next time.
  20.  
    list is updated guys.
    • CommentAuthorROLLER97
    • CommentTimeJul 5th 2012 edited
     
    Hi everyone! I am learning java and I wanted to write a "simple" program where you could calculate your friction (using the formula on mgizozlodus thread) and save the data to a file. If you want to use it to calculate your friction, go ahead, I'll leave you a download link below.

    Please remind that this is my first java program, so I don't have much experience yet (except for some Minecraft coding). Don't expect anything fancy please...

    Download moved to next page because of an updated version
    Disclaimer: This may not be a final full working version. Please report any errors or bugs.
    •  
      CommentAuthorxmantoo00
    • CommentTimeJul 5th 2012 edited
     
    THANK YOU SO MUCH! Now get it to work on Mac, and you are my god.
  21.  
    What is the benefit of knowing the friction?