I'm here again, this time I start discussion from brakes, which oddly enough I have never seen much of indepth realworld data or formulas. Here I have what I have for my Volvo car at the moment. Code: [U]BrakeDiscRange=[/U](0.01285, 0.000, 1) // disc thickness. Volvo greenbooks specify 12.85mm and 11.2mm minimum thickness. BrakeDiscSetting=0 BrakePadRange=(0, 1, 5) // pad type (not implemented) BrakePadSetting=2 [U]BrakeDiscInertia=[/U]0.650 // inertia per meter of thickness [U]BrakeResponseCurve=[/U](-350,80,500,850) // cold temperature (where brake torque is half optimum), min temp for optimum brake torque, max temp for optimum brake torque, and overheated temperature (where brake torque is half optimum) BrakeWearRate=1.20e-013 // meters of wear per second at optimum temperature BrakeFailure=(0.011,7.21e-004) // average and variation in disc thickness at failure BrakeTorque=1655.0 // maximum brake torque at zero wear and optimum temp [U]BrakeHeating=[/U]0.00135 // heat added linearly with brake torque times wheel speed (at max disc thickness) [U]BrakeCooling=[/U](0.00065, 0.000645) // minimum brake cooling rate (base and per unit velocity) (at max disc thickness) BrakeDuctCooling=0 // 0.9e-004 // brake cooling rate per brake duct setting (at max disc thickness) One of the problems I meet quite often is brake torque, problems come when I use realworld value of 2.1kN, which is from my understanding BrakeTorque=2100. That would result brakes that do lockup instantly when one touches brake pedal, well almost at least. With tires that are not locking up instantly car stops inhumanly quick. Something must be wrong then, but what it is, I have not found out. BrakeDiscRange= is meters, disc thickness when disc is new, that was probably easiest in whole hdv, even weight is more of challenge. BrakeDiscInertia=0.650, now this one was something I learned to do, what I have heard is that this is often overlooked and way too low value. Can't even remember if this is my latest value, but Chronus told me once how to calculate brake disc inertia, I quote his writeup here (source http://www.nogripracing.com/forum/showpost.php?p=1519755&postcount=13 at bottom part of his post) : These three are some that I have adjusted only by testing and experimenting, knowing real vehicle behaviour and that 800C is not completely unheard of even for street brakes, I have tried to replicate fading and general 'bite' of brakes to meet what I have experienced in real life, brakes in car are like heat storage units, 5 laps and brakes are gone, need to cool them off. BrakeResponseCurve= BrakeHeating= BrakeCooling= For brake wear rate, I believe there might be some formula, but haven't found anything, I doubt that one could wear out brakes before boredom, it takes more than 10 years on road daily usage, pads are of course renewed few times before discs, but in rfactor pads are not one wearing off? Also pad types are not implemented, maybe in rF2 at some point those become implemented. I'll leave brake duct cooling for fancy race cars, don't know anything about those. So there are few holes to this, mostly that brake torque, which is causing issues if I use real world settings, I imagine there must be more to it than what I imagine knowing. 1800 to 2500 is reasonable for range of street cars from my experience, don't know american cars though.
I swear you've been stalking me for the last 8 months! I'm amazed at how many of these topics are the very things I just got through looking at or are currently looking at. RE:Brake locking I think that is a product of two issues; Inertia, and tire decel grip. If you look at the rf panoz, you will see it breaks away from previous rf standards and uses a much higher number for inertia. It is close to your 8 number IIRC I would look suspiciously at the tire. I've been increasing just about every inertia value for what I'm working on, and this is no big lunky American Muscle car. If you want to implement pad compound, you can do it with upgrades. Simply take your brake torque and divide by the coefficient of friction. Then add something like "HDV=BrakeTorque*=0.49" to section in the upgrades (Where 0.49 is the coefficient of friction).
Volvo is under 1100kg small vehicle, at MOT/inspection they measure brakes and it has got 2.1kN for fronts. Now with rF2 we have completely new tire model which has not been spoken much by wizards yet, there are so many parameters that building understanding from it might take time, I'm constantly ending up tires with too little longitudal grip and too much lateral, also tires have temporary parameters until tire model is completed, so I don't really know how close to real like behavior of street tires one can get at the moment, nobody does know I think. When I managed to almost start to understand TBC it was thrown away and I start learning from beginning Anyway for tires, I posted here what I currently think that I know, which can be less than what I really know, but I like to keep tire stuff on single thread and shamelessly I have found myself abusing that topic, but it might be good at point where we start pulling bits and pieces to rF2 Wiki: http://isiforums.net/f/showthread.php/2796-The-tgm-and-ttool?p=83771&viewfull=1#post83771 I try to take pics of brake discs today, I have pair of unused ones that wait to go to car, then I should be able to do more exact inertia calculation and see if that helps to reach correct values without insane locking. Of course one thing at play is our controllers, brake pedal response is nothing like in reality, but I like to compensate that lack of resistance with setting brake pressure to 80-85%, which now reminds me that I did calculate that 20% off from brake torque as car should not have adjustable brake pressure, that was why I had 1680 there, but I'm now bit of two minds, to have pressure control or to have it hardcoded to car, both are unrealistic in their own way.
First of all, BrakeTorque is expressed in, well, torque So a value of "2100" means 2100Nm of max torque. I will write more on that later and explain what is my attempt to do this thing properly (most of us might not like it though ).
At inspection brake dynamometer measures also this torque from my understanding. Will be interesting to see your method
I was hoping you'd stop by here. I just tried to send you a pm. Your box is full. I know this is OT but I posted some findings to your Caster thread. http://isiforums.net/f/showthread.php/1342-Caster-in-pm-file-vs-HDV-setup?p=86955#post86955
i'd say your brake dics inertia i a bit off!? it seems like those values should be calculated in SI-Value, so meter and kg. that makes your density = 7850 kg / m³ your inner radius maybe something like 0.07m (just guessing that your inner diameter could be 14cm) and your outer radius 0.12m (presuming outer diameter of 24cm) so your equation is as follows: pi*0.5*7850*(0.12^4-0.07^4)*thickness = 2.261 per m thickness i never did any car-modding, so i am just guessing, but this is the right way to calculate it and the way you quoted it. how did you calculate the 0.65?
You mentioned values in kN (kilo Newtons) which are about force, not torque. For the purpose of this thread, we both should operate with the right units and talk about the same thing which in this case, is torque expressed in Nm (Niuton*meter). And torque [Nm] = force [N] * radius [m]. -------------------------------------------------------------------------------------------------------------------------- Let's focus on the main parts of the whole brake system: - brake pedal (acting as a lever with a known ratio), - master cylinder(s) (with known piston diameter), - calipers (with known piston(s) diameter), - brake pads, - brake discs. Driver pushes on a pedal which transfer the force through leverage onto master cylinder's rod. Difference in diameter between pistons located in calipers and master cylider create multiplier for driver's force at MC's rod and by the same ammount reduce piston stroke on the calipers side. Then, that force is applied through brake pads (their surface doesn't matter, by the way) on a brake disc at a known radius and becase there is some friction between brake pads and discs, we have our braking torque. And that's exactly, how I calculate that parameter for all my cars. Of course, you have to take efficency into account - brake lines expand (even those wrapped in steel armor), copper lines as well but of course not that much. An example (a simple one, without including any losses). A Race car with separate brake loops (so, two master cylinders connected with brake bias rod to a brake pedal). - force applied by a driver: 80kG, - brake pedal ratio: 4.0:1, - MC piston diameter: 0.02m (20mm), - piston set per caliper: 2x (0.025m + 0.03mm + 0.035m) = 0.18m, - radius for center of forces applied on a brake disc: 0.15m (340mm disc, forces act 20mm from its edge), - CoF (coefficent of friction) between brake pads and brake disc (already bedded in): 0.4 80kG = 784.8N 784.8 * 4 = 3139.2N (2x 0.18) / 0.02 = 18 [2x0.18, because we have two calipers per each loop and only one master cylinder] 3139.2N * 18 = 56505.6N now, CoF takes into account: 56505.6N * 0.4 = 22602.24N And finally... 22602.24N * 0.15m = 3390.336Nm That is our calculated BrakeTorque value. Notice the total force multiplier, which in this case is 4x18 = 72.0:1 As I said earlier, that have influence on pedal travel. Usually, the distance between brake pads and discs is very small, like 0.3mm. With our multiplier and conditions, that gives us 21.6mm of pedal travel when driver applies 80kG force. Theoretical of course, because brake lines do expand and effective pedal travel would be longer. Brake bias is a separate thing and rF deals with that internally. That means, if the above BrakeTorque would be valid for all 4 brakes and we have brake bias set at 50% in our setup, then the maximum available braking torque per wheel will be half of the calculated one (again, rF does that by itself so you don't have to worry about it at all). As for inertia - guys, DON'T take just material density, until you are dealing with solid brake discs (not vented and/or drilled). Also, you have to take into account disc bell.
That is great post LesiU, thanks It is confusing those Newton units, they all look just the same, but indeed very important to know difference, thanks from expaining that. In my case, as measured force was 2100kN and radius is 0.288m I can calculate that at test my car's brakes resulted 604.8Nm of torque when wheels started to lock up on brake dynamometer's rollers. If I account tire deflection when on level ground that becomes to ~585Nm and at rollers we can only guess deflection, maybe more severe as tire likes to deform under stress. I believe that is limit of tire's grip, that much torque tire can transmit to rollers, road might be different, I don't know that machinery so well that I could say if weight transfer has been attempted to be compensated with position and groove of rollers etc. What I can see is 3 rollers, tire is in kind of pothole, so I would imagine there being some compensation, but that is only a guess. However that should provide some kind of view just how much braking one actually can use before tire locks up. I think that out of 10 vehicles only one has vented disc brakes, some have drums at all corners which is bit different of course, but I don't see why that would not be possible to calculate similar way. With inertia it is indeed very important to know where mass is. With vented discs I have been thinking vent size and how many % it is from volume of disc, then I think that much of volume should be removed from thickness of disc as it is symmetric along the area where most of inertia comes from. Bell contibutes quite small amount to inertia I believe as it has less mass than actual brake disc surface and it is near center. From my understanding if we have 300mm dia disc, 1 gram at 75mm radius from center has 50% of effect to inertia compared to 1 gram that is at 150mm radius of center, but it is only guessing, with air resistance it would be 25% as that is squared, not so sure about inertia if that did work squared too. Need to do some measurements and calculate inertia and brake torque, then I can post up my results and finding here. For brake heating, I accidentally bumped to some study that has curves that show deceleration, brake temp and speed, might be useful, there was also formulas, but I have not read it completely, it does focus on buses however, but only scale is bigger compared to smaller vehicles: http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0100-73862001000100007
Disc bell can weight even 3-4kg in typical road car. Of course in race cars, it weights much less (about 1-1,5kg) but still, I'd suggest to include it in the calculations because in the end, everything matters, even those small things. Drum brakes can be calculated in a similar manner but I simply don't have any data about them, because none of my cars have such brakes Thanks for the document. Will take a look at it.
Yes, it must of course be included, but it is good to know what is it's effect to whole brake disc so one ending up with higher results from bell than outer disc can spot errors, however it is not always very simple or clear because shapes and sizes varies. My guess is that it is 10-25% of total brake disc inertia, but haven't really done much calculating of it yet. Anyone making Porsche 928? Some brake stuff from it: http://www.928motorsports.com/parts/gt1_brake_kit.php Also anyone not having access to brake discs can use this site, it has drawings, dimensions, weights etc: http://bremboaftermarket.com/En/Car_Disc_Catalogue/Catalogue_Search.aspx
Haha, true, I forgot to type few words there, as I meant vehicles I'm currently working on, missing few words changes meaning quite easily
I realized now that *thickness = 1, I guess there was error as for thickness I have been using disc thickness of 12.7mm. These are real values for rotor, which are very close to what you guessed: Disc thickness 0.0127 Inner radius 0.0695 Outer radius 0.1195 Then there is need to calculate bell too and when using this kind of formula I would need to account edge area somehow as it has 20mm thickness and difference of outer and inner radius of only 6.4mm, but rest of 18.6mm of bell there is thickness of 6.4mm with inner radius of 35mm and outer radius of 53.6mm. If I understand correctly, rFactor uses disc thickness and inertia per meter to get final inertia for brake disc, which would be 2.227 x 0.0127 = 0.028281153 in my case if I omit bell area. For bell area, I figured out that calculating volume is in order, for edge section I managed to calculate volume of 0.00009393613 cubic meters and weight for that is volume * 7850 = 0.737399kg. For rest of bell area I managed to calculate 0.00003313420 cubic meters and weight for that is volume * 7850 = 0.260103kg. Bell area weights in total of 0.997502kg Whole disc weights around 4.14kg, maybe bit less because of bolt holes, so bell weight is around 1/4 of total disc weight, but it is naturally less from total inertia, I randomly guess that 10% would be needed to be added in my case. For volume calculations I used this calculator: http://www.calculatorsoup.com/calculators/geometry-solids/tube.php I did take last V value as it is volume of tube wall, which brake disc in our sense really is, not disc, but tube. Sometimes calculator speaks engineering, like V = 2.33432901E-5 m3, to get plain simple guy numbers I pasted number bit including E-5 to excel and set cell type to be number, then added enough decimals was much nicer to compare numbers that way. I think my solid disc as if it was made with three tubes, rotor part, bell edge part and bell area from center hole to near edge (edge - material thickness), it makes it easier to 'see' and calculate. That will get me much closer with inertias, next up is brake torque, I need to measure again as I forgot to measure when I was in garage, I forget lot these days, LOL. edit: I have posted way too many times in a row already, so I just edit stuff here. There is only one cylinder per piston, diameter is 48mm Master cylinder piston diameter is 20.5mm Center of forces I figure out to be 239mm / 2 - 25mm as I have 50mm brake pad contact area and disc diameter is 239mm. I remember reading that 50kg is needed to lock the brakes, can be more of course. Brakes are vacuum assisted, ratio is 3.1:1 which is probably just additional leverage. Pedal leverage I must actually still measure and calculate, also do more research for brake pad friction for non racing purposes, but I used LesiU's values for those, need to get calculations correct first. Step 1, 50kg becomes 490.5N Step 2, 490.5N * 3.9 = 1912.95N (3.9 is pedal leverage) Step 3, (2*0.048m) / 0.02064m = 4.651162791 (caliper pistons to master cylinder piston ratio) Step 4, 1912.95N * 4.651162791 * 3.1 = 27582.06977N (3.1 being vacuum servo assist multiplier) Step 5, 27582.06977N * 0.4 = 11032.82791N (0.4 being Coefficient of Friction for brake pad) Step 6, 11032.82791N * 0.095m = 1048.118651N brake torque for HDV (0.095m being center of forces) So as I know that 585N is somewhere in range where wheel lockup occurs in reality, that 1048N should be still plenty for locking up wheels when brake balance is taken account to. Of course there are still rubber brake lines that reduce efficienty with unknown factor, leverage of brake pedal and also friction of street car brake pads to get right until it is proper value to be used in rFactor. My spreadsheet seem to be outputting same values when used similar values as in LeisU's example, so it should be usable, it is rather rough tool and needs to be beautified still, inertia section at sheet 1 might be even unusable for most at this stage, also rear brake calculation is not done and number of caliper's cylinders is not working, one must calculate all cylinders to one by himself, but must refine that soon, it is just rough start of tool. Another edit: I found something from brake line expansion. Source: http://www.modified.com/tech/modp-0909-brake-lines-upgrade/viewall.html For me that 0.290cc/ft looks rather small expansion, when there is not very much of that rubber line, I wonder what kind of percentage loss that could be? Under 5% for sure? With drum brakes, I wonder if piston diameter should be manipulated as single cylinder has two pistons at opposite directions and fluid pressure moves pistons away from each other, pushing two brake shoes against the drum. Shoe design has also some level of self assisting element. Generally drum brakes are considered to be less efficient than disc brakes, also more prone to heat issues. I get rear wheels locking first, F70:30R bias and when using same values for front and rear torque, weight distribution is 53:47, braking distance is bit too long and brakes are quite hard to lock at front, so there is still some work to be done. However I found out my pedal ratio was 3.9:1 so with that there is hardly chance to lock front wheels. Over 60 meters from 100kph to 0 with 1075kg and hardly tires locking, while not having losses in, there must be something at fault, maybe that 50kg that I have just heard. Then there is brake tests under 600N to lock wheels, of course there is no weight transfer which changes things dramatically. In reality, I could lock the brakes, but not sure about force needed, pushing bathroom scale against wall while attempting to be at same position as sitting in car gave me results anything from 43 to 79kg, so that might be not most reliable way to find out power level that I have in my memory, can't test braking in reality now. Then again, how much average human can push such pedal, his own weight for sure? That in mind 80kg is good value too, then I would end up to 1676.99 which is right about my previous 1680 which I got by calculating 80% from 2100, so that seems most probably from all the values, I guess. From that maybe 5%? to losses and 1593 might be something that at least sounds quite nice and spot on for me, just guessing however, but I think this guessing is closer than copying value from some other car or by adjusting so that car feels good. 100kph to 0 bit over 40 meters, need to learn how to brake properly to get accurate results, but I would say that is a result?
I think this page has some educational information http://www.stoptech.com/technical-support/technical-white-papers/white-paper---brake-bias-and-performance-why-brake-balance-matters Here are also some coefficient of friction stuff: http://www.remanbrakes.com/coefficientoffriction This one mentions that street pads are in ~0.3-0.4 range and performance pads in ~0.4-0.5 range: http://www.dixcel.co.jp/en/subcontent/literature/literature03.html 0.65 I have seen for race pads mentioned too, there are very wide range of them. edit: found more Source:http://www.powerstop.com/faqs DOT_EDGE_CODES.pdf from here gives letter codes like GG, coefficient of friction 0.45-0.55, there are other codes too and even some temperature effects described: http://www.bimmerfest.com/forums/showthread.php?t=570124
My track day pads are Ferodo 2500, which are FF code, that means that they have CoF between 0.35 to 0.45 and should not fade easily, they are better than standard, but not fadeless. Then I digged bit in garage, found FE pads, Siffert brand (fits Mercedes 200 for example), these have normal temp (300F/149C) CoF between 0.35 and 0.45 but hot temperature (600F/315C) CoF 0.25 to 0.35, so there is clearly fading happening already at that temperature. I also have one old pads from Ford Taunus I believe, really hard to read, but there is barely readable GA, however I don't know about A code and I doubt pad being G rated even at cold. One pad Roadhouse brand, had no letter coding at all. That little samples give at least some range for values to be used and also some idea about point where brake fade should start to occur. Race pads are of course another animal, but they experience fading too, just not so soon as street pads. I wonder if heating could be calculated from friction and force? In Wikipedia there is some formula under "Energy of Friction" but it speaks engineer and I fail at it. https://en.wikipedia.org/wiki/Friction I guess there is some formula for this too, maybe pad CoF, volume of disc and surface area of pad are part of it? BrakeHeating=0.00135 // heat added linearly with brake torque times wheel speed (at max disc thickness)
I'd be willing to bet working it from an energy aspect will get you there. I missed your post stating about the codes. That is awesome information. Off to go look if that sprouts new data *cough P50 cough*
About the Inertia of your Brake Bell: if you have technical drawings or just pictures and the main dimensions of it, i can model it in my 3D-CAD System (Autodesk Inventor)...i can then just read out the exact inertia. wouldn't be much of a hassle for me, and i would be glad to help!
