Let's talk about brakes

Discussion in 'Car Modding' started by jtbo, Aug 2, 2012.

  1. Knut Omdal Tveito

    Knut Omdal Tveito Registered

    Joined:
    Mar 16, 2012
    Messages:
    47
    Likes Received:
    3
    I use MoTeC to analyze the data. To export to Excel you can just right-click the screen in MoTeC and click "export data" and select CSV file format.
     
  2. Domi

    Domi Registered

    Joined:
    Oct 11, 2011
    Messages:
    747
    Likes Received:
    44
    Thanks :)
     
  3. jtbo

    jtbo Registered

    Joined:
    Jan 12, 2012
    Messages:
    1,668
    Likes Received:
    48

    :eek: Wizard stuff :D

    Really big thanks, I had no idea that you could calculate this with few bits needed to be known only, I feel educated again :)

    Regards to surface area and disc volume, do we look only part that wears or disc as whole? My understanding is that disc as whole dissipates heat, also stores heat, but only part that wears will get thinner and if rFactor does calculate wear off from disc, isn't that then less volume but only to part that wears so we can't again use bell area for anything in this?

    However what I understand from formulas disc wear seems not to be affecting to disc heating or cooling, but I may be wrong?

    Sample data I posted gave me discs (using lowest end of FE pad range) that did last around four braking until brakes were cooked and without moving it took around 6 minutes to temperature to decrease from 220C to 200C, minimum cooling was 0.000x
    Meganes have disc with more inertia, so I presume they are bigger and maybe heavier too, don't know if they are steel though, they have minimum cooling of 0.0x range, per unit of velocity is much smaller in those, I wonder what is story behind them?

    With sample data as discs are vented, I figured out there is around 4 disc surfaces that dissipate heat and also edge surface, but I did omit edge surface as fins probably take bit out of surface area, how ever I'm not too confident of what to include to surface area.


    I have played around a bit with drum brakes too, I think that I have good idea of them now. With drum brakes there is leverage from brake shoe also, pivot is at bottom, input is at top and output can be thought to be nearly whole surface, but it has middle point at middle of shoe so if I multiply brakedrum calculation by two I get that additional lever included.

    Rest of calculation I did same way as discs as they have surface, it is just 90 degrees different angle, they have center of forces on that surface, also there is one cylinder that has two pistons at opposite of each other so I took that as two cylinders.

    Of course this needs to be tested with more data and see if it works or not in other situations, but for me it would sound as logical approach. With drum brakes, drum face is rather large area that cools too, also some of the heat is transferred to wheel too, so for those heat dissipating is in reality much larger than just area that wears, even for some disc brakes, but for some, like in my Volvo, heat can escape only to air directly or via wheel bearing to wheel, that is what I have been told and also I have noticed that wheels don't get as hot as with other cars even with brake smoking.

    I wonder now if that heat transferred to wheel should be taken account into with systems that have that capability?


    BTW I found this page that has data from Asbestos brake lining material, CoF 0.35, max continuous temp 1250C, max temp 2600C, I wonder if those really even had any brake fade at all?
    http://www.indiamart.com/asbestoscentre/asbestos-brake-lining.html

    Here are some graphs even, but I can't really tell what product they are for, asbestos or some other?:
    http://www.frictionproducts.com/testing-results.html
    http://www.frictionproducts.com/disc-brake-pad.html

    edit:
    Ditch that what I thought from drum brakes, if I add leverage of brake shoe, at least with my Volvo rear brakes become really lot more stronger than front, even with F70:30R brake distribution that car has.
    I'm getting so different result for rear brakes with these two vehicles that I'm skeptical if same method can be used for drum brakes at all.

    Must experiment more and build better torque calculator sheet, it has bit too much mess right now as there are all my experiments too :p
     
    Last edited by a moderator: Aug 6, 2012
  4. Novis

    Novis Registered

    Joined:
    Oct 7, 2010
    Messages:
    251
    Likes Received:
    4
    Great stuff indeed, thanks!
     
  5. Knut Omdal Tveito

    Knut Omdal Tveito Registered

    Joined:
    Mar 16, 2012
    Messages:
    47
    Likes Received:
    3
    For volume and area I would use the geometry and don't worry about wear (we dont have any control over wear-effect when it comes to heating)

    For the volume I would use the entire volume, except the volume of the front "disc" of the bell that contains the 4 holes for the wheelnuts (this part is in good contact with the rim and will not heat that much)

    As for the Area, I would sum the area that is in contact with the flowing air. So the entire disc except the area that is covered by the brake pads/caliper. i would leave out the front "disc" of the bell again.

    Yup, you are right. It doesn't depend on temperature in rFactor. So you have to choose one "representative" temperature to calculate the wear. Since the wear increases exponentially with temperature, it is the high temperatures that does the most damage. So lets say the disc temperature varies between 100 and 300 degrees during racing. Then I would use one of the higher temperatures, around 280 degrees to calculate the wear rate in rfactor.


    As for calculating brake torque. When you calculated the force from the master cylinder and the resulting force from the caliper pistons you use this calculation:
    - MC piston diameter: 0.02m (20mm),
    - piston set per caliper: 2x (0.025m + 0.03mm + 0.035m) = 0.18m,
    (2x 0.18) / 0.02 = 18

    But you should not use diameter here, you should use Area. (Pressure*Area=Force and pressure is the same for master and caliper) So it should look like this:
    - MC piston area: pi/4*0.02^2
    - piston set per caliper: 2x pi/4* (0.025^2 + 0.03^2 + 0.035^2)
    2*2*(0.025^2 + 0.03^2 + 0.035^2) / 0.02^2 = 27.5

    What does the caliper look like?


    BTW. i found this for the Ferodo DS2500:
    http://www.ferodoracing.com/it/car_racing/ds_2500.htm
    Seems like there is quite small variation in mu. Slight increase up to 200 degrees, but pretty much constant after that. Unknown what happens after 500 degrees, but I guess you will not reach that high anyway.
     
  6. jtbo

    jtbo Registered

    Joined:
    Jan 12, 2012
    Messages:
    1,668
    Likes Received:
    48
    Thanks, that helps a lot to know how to look at disc with rFactor in mind, also might work as good baseline for standard method for modders to do brake surface area calculations.

    With vented discs there are also area of fins and second side opposite of friction surface, if fins are curved they have again more area, I need to explore bit of interwebs to get understanding of how to calculate such areas, especially curved fins looks to be interesting to calculate.

    With my solid discs things are of course easier.

    Area for pistons, doh! I even remember wondering why using diameter when pressure is on area, but failed to really read it as so, silly me :D



    View attachment 3484
    Here are brakes of Volvo as of 2008 and actually state is same even now, I just turned nose to other direction...
    Anyway, you can see in this photo new caliber that has been installed just for test fit, also one can see how I have managed to cook even wheel bearing grease + everything that was contact with brakes, old caliber rubber parts were also destroyed and I did only 5 lap stints including in and out laps, discs were driven some 190 000km or 80 000km if replaced once, which I doubt, discs are under fail limit, but after moving car has just been in garage.

    For Firebird 1967 had 4 piston calibers with 400 ram air model and last year it had those were 1968, I did actually found AMA specs for Firebird brakes, which have slight differences to Camaro brakes.

    Here are original type of brake calibers available for Firebird and naturally there is a picture:
    http://www.firebirdcentral.com/product_p/brc-163.htm

    Ferodo graph is really good, thx :)

    I must fix my calculators and put new ones up then, with enough iterations it will be good ;)

    edit: No, mist is surely in my head, but form should of been "I must fix" instead of "I mist fix", fixed typo
     
  7. Knut Omdal Tveito

    Knut Omdal Tveito Registered

    Joined:
    Mar 16, 2012
    Messages:
    47
    Likes Received:
    3
    Thanks for the picture! Just wondering now what this means:
    2x (0.025m + 0.03mm + 0.035m)
    The caliper has only 1 piston right? what does the three numbers represent?
     
  8. jtbo

    jtbo Registered

    Joined:
    Jan 12, 2012
    Messages:
    1,668
    Likes Received:
    48
    I'm starting to feel confused :D

    Anyway, I think those are numbers from LesiU's example:
    http://isiforums.net/f/showthread.php/6967-Let-s-talk-about-brakes?p=87188&viewfull=1#post87188

    I'm confused from where those came from, had I put them accidently on my calculus?

    As I have 48mm caliber piston, but I have two of them for front loop (both wheels) I should have 2x area of my caliber piston, from my understanding.

    Edit, I have this:

    Brake cylinder diameter 48 mm area of 0.003619115m^2
    Number of cylinders 1
    Master cylinder piston dia. 20.64 area of 0.000669174m^2
    Radius for center of forces 95 mm
    Coeffient of friction 0.35
    Brake pedal lever 3.8 :1
    Force applied to pedal 80 kg
    Vacuum assist 3 :1

    I get these results:

    total power (N) 8946.72
    Pedal * leverage (N) 2982.24
    Pedal (N) 784.8
    Calipers/master cylinder 10.81665765
    brake torque 1 Cof(N) 96773.60736
    brake torque user Cof(N) 33870.76257

    Brake torque for HDV 0% losses 3217.722445
    70% front brake bias torue is 2252.405711

    I would need to invent some losses for that of course, but still it is probably instant locking brakes with that kind of number. Maybe there is still errors, I must go trough all formulas and compare them to posted ones.
     
    Last edited by a moderator: Aug 7, 2012
  9. jtbo

    jtbo Registered

    Joined:
    Jan 12, 2012
    Messages:
    1,668
    Likes Received:
    48
    These give me same result:
    =PII()*(0.02064/2)^2
    =PII()/4*0.02064^2

    First one I understand easily, 2nd one have no logic, but problem is in my head ;)

    I did put area calculation in and made spreadsheet prettier, but now I get very high brake torque of nearly 3000 which will lock wheels instantly, maybe I need to put some losses in.

    I had areas wrong before I believe, double amounts, but that really should not affect much if relation is same, I did verify now with online calculator that I can get right numbers for that:
    0,0018095574m
    0,0003345872m

    View attachment 3490

    I can't spot any more errors, but it does not mean this being error free..
     
  10. Knut Omdal Tveito

    Knut Omdal Tveito Registered

    Joined:
    Mar 16, 2012
    Messages:
    47
    Likes Received:
    3
  11. jtbo

    jtbo Registered

    Joined:
    Jan 12, 2012
    Messages:
    1,668
    Likes Received:
    48
    That is very nice, also I see that I managed to do calculations finally quite close to correct too (at least I end up to similar final number with similar values), which is surprising :D

    Such tool is very valuable for the community, I'm very sure of that as before most have had no way of getting brakes right :eek:


    Now then there is mystery, my original value was 2100 and that resulted instant locking of wheels when I touched brake pedal, now there is values of 3000 to over 4000 depending from pad friction, maybe this needs to be compensated with brake pressure setting for cheap plastic pedals?

    Thanks a lot :)

    edit: I forgot to post this before, it is rather big document about brakes and temperatures, science stuff:
    http://researchbank.rmit.edu.au/eserv/rmit:6207/Stephens.pdf

    There are lot of graphs that are easy to read even for me :D
     
    Last edited by a moderator: Aug 7, 2012
  12. Knut Omdal Tveito

    Knut Omdal Tveito Registered

    Joined:
    Mar 16, 2012
    Messages:
    47
    Likes Received:
    3
    Thanks for that link, very good. BTW I also added the heat conduction from brakedisc to rim. It's mostly important at low speed.
     
  13. jtbo

    jtbo Registered

    Joined:
    Jan 12, 2012
    Messages:
    1,668
    Likes Received:
    48
    As I'm searching now drumbrake calculation methods, I came accross something interesting:
    http://www.engineeringinspiration.co.uk/brakecalcs.html#bt

    Here is pdf that has drum brake torque formula, but it is tad high level for me to get with single reading:
    http://nptel.iitm.ac.in/courses/IIT-MADRAS/Machine_Design_II/pdf/3_2.pdf

    Looks interesting stuff, wizard stuff :D
     
  14. Marek Lesniak

    Marek Lesniak Car Team Staff Member

    Joined:
    Oct 6, 2010
    Messages:
    1,585
    Likes Received:
    101
    Well, that is an overstatement, because what should or should not be used, depends on which way you want to go.

    If one wants to go with pressure then yes, piston area should be used (as there is no other way).
    But if one wants to use force (like I do, for example), then obviously, piston diameter have to be used.
     
  15. Knut Omdal Tveito

    Knut Omdal Tveito Registered

    Joined:
    Mar 16, 2012
    Messages:
    47
    Likes Received:
    3
    OK thats interesting. Could you explain how you can use diameter to calculate the force? What physics is this based on?
     
  16. Marek Lesniak

    Marek Lesniak Car Team Staff Member

    Joined:
    Oct 6, 2010
    Messages:
    1,585
    Likes Received:
    101
    Physics of this planet :)
    I explained that already. Post #8.
     
  17. Knut Omdal Tveito

    Knut Omdal Tveito Registered

    Joined:
    Mar 16, 2012
    Messages:
    47
    Likes Received:
    3
    :) OK, just trying to understand.

    For the pedal we have the same calculation. We take a moment balance around the pivot of the brake pedal so that F_foot*r_foot=F_mc*r_mc (r is the radius from the pivot to point where the force acts). It seems you use the same calculation for a hydraulic system? But there is no moment balance there, it's just simple hydraulics (P*A=F)
     
  18. I_Bellett

    I_Bellett Registered

    Joined:
    Jan 12, 2012
    Messages:
    200
    Likes Received:
    277
    It is very good although a few was used.
    However, since the drawing displayed on left-hand side collapses and is displayed, I would like you to correct also there.
     
  19. Marek Lesniak

    Marek Lesniak Car Team Staff Member

    Joined:
    Oct 6, 2010
    Messages:
    1,585
    Likes Received:
    101
    As I said in the mentioned post, you don't include balance bar in the calculations, as rF already takes care about that.
    Why do you need a "moment balance"? You have a pedal ratio which you use to multiply force from your foot, to get a force acting on MC's rod.
    From that point, you can convert force to pressure and then back to force at calipers or stay with force all the time (like I do). In the end though, both ways should give the same results.
     
  20. jtbo

    jtbo Registered

    Joined:
    Jan 12, 2012
    Messages:
    1,668
    Likes Received:
    48
    If I have more than one caliber pistons and I want to sum them up, I noticed that if I want to use area then I have to calculate each piston's area separately and not to put diameters together and then calculate area.

    For Firebird I have this wheel cylinder bore 2.9375 and there are four pistons in caliber, I get silly high readings if I put that four times in. Disc is ~278mm and one piston would be 74.6125mm so together two would occupy distance of 149.225mm, also from pic I linked earlier one can see that there is space between two and also brake pad is only 137.16mm long, so I'm thinking that maybe in specs that is bore of all fours together, so I would need to divide it by 4 to get 18.653125mm for one, which would be much better fitting photo I saw.

    Just not quite sure how those documents are meant to be read in that part, but for me it looks like that it should be divided by 4, I doubt that brake torque in 1967 car would of been 5 figures even at front.

    http://www.tpocr.com/67-68fbdiscbraksysipc.html

    http://www.tpocr.com/67-68fbbrakepdlmastripc.html

    edit: More science of drum brakes:
    http://eprints.usm.my/8964/1/A_STUDY_ON_THE_EFFECT_OF_OUT.pdf
     
    Last edited by a moderator: Aug 8, 2012

Share This Page