boxsey911

Evening Alex

I chose to add extra cooling, via ducting, first because although not cheap, it was considerably less expensive than fitting big reds. Both in terms of initial outlay and ongoing consumable costs. I also took my lead from 964s that run in the GB Porsche club championship....they have to use standard calipers and rotors as part of the regulatios, hence ducted air is their only option for cooling.

I may well go for bigger brakes later but first I want to see the effect that the new cooling brings. So far, I've generally added one upgrade at a time and run with it to see what it adds before moving on to the next one.

P.s. the other factor missing from your list above is how the brakes are used. Those that can really drive/race have advised me that using them harder and shorter (but still with finesse) allows more cooling time in between. That's a skill I've yet to get good at myself but I guess it would also be made easier with bigger brakes (i.e. I'd have more confidence in going on them harder).

Babalouie

Yes, Boxsey I've been told that too...that being "gradual" with the brakes and coming on and off them slowly means you're on the brakes for a longer amount of time, so even though you might be braking earlier, you've got more heat buildup.

Pro drivers I've ridden with all seem to leave their braking until the last millisecond, then suddenly shove their foot thru the firewall But they all say that coming off the brakes, in a controlled manner as you turn-in is vital though.

Bill Verburg

I don't know what you mean by 'more extreme use'

tires are certainly the limiting factor is applying brake torque in a useful manner, for track use it is expected that the tires will be upgraded to utilize additional torque compared to their street counterparts, it makes little sense to use bbk's on a street car.

the performance oriented driver will always prefer the highest hardest pedal that can be found, the only limiting factor is the ability to apply enough pressure to the pedal to generate the line pressures desired, street cars are biased so that anyone including the proverbial little old lady can operate them, a track oriented car will want a pedal that needs a much more robust touch. Porsche knows is this and the result can be see when comparing 964 pedal ratios to 964RS pedal ratios
964 44.264
964RS 32.786
964RS w/ 993 m/c 37.342
964 w/ 993 m/c 33.161
lower provides a higher harder pedal and more leg to operate, this is modified by the boost ratios

max3.2

I dont see the need for more friction (more pressure, bigger pads...) in a fairly stock 964. The brakes are more than sufficant for the car in terms of braking toque (my opinion..) - as Alex said, put more downforce and bigger tires in the equation, maybe...
And yes, increasing mass will slow down heating and lower overall temp, since theres more time for the heat to leave the brakes (which actually goes a bit slower, since driving temp difference is lower...but thats not so important here...). But - again already said - unsprung mass is really really bad for handling the car. Maybe get the ceramic rotors... ?
For computation of HTC: There are only correlations for different cases - its not that generic. For example, using the HTC for a flat plate, turbulent flow (high Reynolds number), one can find:

Pr=0.7 for air, approx (Prandlt number)
Re=(v*L)/nu
with v: characteristic velocity(e.g. flow velocity, rotor surface speed...) L: char. length (e.g. rotor diameter....) nu: kinematic viscosity
Nu=0.0287 Pr^(3/5)*Re^(4/5) (Nusselt number)

then:

Nu = (h*L)/k

h: HTC (solve for) L: length scale as above k: thermal conductivity of air

then Fourier's law yields

Q° = h*A*(T-T_amb)

and first law of thermodynamics

(dT/dt) = - Q° / (m_rotor*c_rotor)

c_rotor being heat capacity.

Integration yields T(t), will be something like exp(..)

5 minute calc for a first go.
Or do FEA for generated heat and CFD for flow.


And for the ducting: Get as much as possible air in, shrink diameter smoothly to reduce pressure losses (flow speeds up), and let it hit the rotor. If possible, get a stagnation point at the pads (very high HTC because the air jet "explodes"). Thats what I would do..

And again, all of the stuff just typed down at 1AM - if someone finds errors please correct..

alexjc4

So This interesting because the more I read the more it looks like the right way to do the maths is by looking at heat flux (new to me until this thread) instead of plain old kinetic energy, and heat flux is dependent on time on the brakes. So that all links up nicely.

Think I'll redo the spreadsheet using the heat flux method, perelet has kindly pm'd me some reference material...

alexjc4

That sounds like an interesting project. Maybe worth a thread in its own right. Is there a thread or webpage out there that you can point me at the method and meterials involved in DIYing something like this?

alexjc4

Oh my, this thread is exceeding my expectations! Thanks Max

m_rotor is the mass of the rotor right?

I think I follow all of that upto the integration, which I used to be ace at twenty years ago, but it turns out "if you don't use it you lose it" and I don't know where to start with that

Interesting that you can use FEA (assuming thats finite element analysis) for heat calcs, I'd only seen it for load/deformation sort of stuff.

Re ducting; More air is better to a point but a)i guess you start to get diminishing returns, b)you want to maintain some heat in the system for the pads to work effectively. So it would be interesting to game the numbers to see the relative effects of different sized inlets at the bumper at different pressures/speeds. Probably another thing that's derived by experimentation if you're really serious (race teams) but its raining outside and I'm not volunteering the cut holes in the front of my car

Cheeksyboy

Boxsey is taking the methodical and sensible approach and he's got faster and smoother over the past year that I've braved his pax seat, braking is about the last area to master...and one of the hardest to consistently get right (ask me how I know!) As Boxsey will (hopefully) testify I brake late and hard (easing off as the car slows to avoid lock up, or ABS application) and then once the car is settled, turn in.....well this works most of the time (falling off at Lodge a notable recent exception...but you learn!)

Applies all the way up to F1, but with our cars and the rear weight bias it's important to get the car balanced as you come off the brakes and either start, or continue your turn in....I sometimes left foot brake just to get the nose weighted to aid turn in, but this has to be balanced against the rear getting light...

I run the 964 w/ 993 m/c 33.161 setup, braided hoses and big blacks.

I've tried standard pads (good, but do fade on short, brake intensive tracks), red and yellow stuff (alright, but don't give a firm pedal and don't inspire repeated heavy braking confidence, had some fade at Spa), and Hawks (recommended by Boxsey and so far the best pad I've tried for firm, consistent feel and strong retardation). This is with both R888s and Yoko48s

So far my brake ducts cool my oil and aircon, but hopefully you'll note that there's more to it than just maths once you estab the baseline.....as you've got to figure in ambient temps, track grip, driving style, etc, etc.....

If I were you I'd change your pads to Hawks and see how you go from there...

However, as I don't have half your brain power, please keep posting as this is an interesting subject....and I love to learn from the Dr

max3.2

Yes, m_rotor is the rotor mass.
Not too much time right now, so really quick help for the integration: put everything in first law and assume none of the the other values change with temperature but the temperature itself. Keep T_ambience constant. No time functions either. Expanding the bracket yields something like

dT/dt = -c*T - d (with c,d being constants..)

seperating

1/(c*T - d) dT = -dt

right side will be = -t
left side: = ln(c*T - d) / c (please double check someone...)

using exp-function

c*T = exp(-(c*t)) + d

so finally

T(t) = (exp (-(c*t)) + d ) /c


I bet using heating effects too, maybe by implementing step functions for constant friction power, makes this stuff not solvable anymore, at least analytical...
FEA (is finite element..) will solve for whatever you want as long as you use the right inputs. The main problem I'm looking at right now is finding the right software for a quick analysis, I don't want to write my own code, that would take ages...and one is perfect for contact simulations with friction, and the other one for thermal and rotation...damn

Indycam

What I don't see in this discussion is the pad heat range .

alexjc4

Have you got some example figures for ideal operating ranges for various compounds?

Obviously you don't want to over cool, below point that the pad is making decent friction.

Also would be interesting how high temp you can go with readily available pads before you get pad related fade. Obviously you eventually start boiling fluid, even racey 600 f stuff.

Indycam

And then re boiling , there are ways to delay / stop that from happening .
Fluid circulation systems . Water cooling for the caliper .

http://www.performancefriction.com/m...criptions.aspx

https://docs.google.com/viewer?a=v&q...13M4h4gSfKGnAw

alexjc4

Pad compound heat ranges culled from vwvortex:

oem level Performance
# examples: OEM MkIV pads, Mintex Redbox, Hawk Ceramic
# cF=0.30-0.40
# max temp before fade: 600-700°F (Hawk Ceramic is a little higher on cF, redbox is lower on both cF & fade)
# street use only

Performance Enthusiast/Light Autocross
# examples: Hawk HPS, EBC Greenstuff, EBC Redstuff, Porterfield R4-S, Axxis Ultimate
# cF=0.41-0.45 (EBC pads have higher cF, Greenstuff has poor performance on heavier cars
# max temp before fade: 700-900°F (some pads may be rated higher)
# light auto-cross and performance street use

Autocross/Light Track Pad/HPDE
# examples: Hawk HP+, Carbotech Bobcat, Ferodo DS-2000, EBC Yellowstuff
# cF=0.46-0.49
# max temp before fade: 900-1100°F
# light track days, auto-cross, and performance street use

Track Pad/HPDE
# examples: Ferodo DS-2500, Carbotech Panther Plus, Mintex C-tech 1155
# cF=0.50-0.55
# max temp before fade: 1100-1300°F
# track days and high-performance street use

Race Pad
# examples: Ferodo DS-3000, Hawk Blue 9012, Hawk Black, EBC Bluestuff, Porterfield R4
# cF=0.56-0.68 (depends on specific race application)
# max temp before fade: 1300°F+ (depends on specific race application)
# track use only

alexjc4

Thanks to boxsey and cheeksy we have some reference points

Ebc yellow fade with big blacks and 322mm? discs so temps must be 1100deg f + (600deg c) rotor temps

hawk blue fade with stock brakes so must be 1300deg f+ (700deg c) rotor temps

Making the bold assertion that the driving technique (between these two pretty experienced chaps) accounts for a small amount of the difference. We can estimate that under track conditions you might see 100 deg c lower rotors temps with big blacks vs stock brakes under repeated braking.

Edit: i just realised this is a reduction of ~14% and surprisingly close to the 10% difference predicted in my first calculation

alexjc4

Great explanation of the basics in that video. Thanks Indycam.

Did I read somwhere that the 959 had a system for circulating the brake fluid?

Interestingly the rotor temps are line line with those I googled up, its always nice to cross reference these things.