Better brakes

[brownslane]

Nathan, at the risk of a coronary, do you know the ballpark figure for a set of these babies?

(waiting to see if I should cash in my kid's inheritance)

[starkejt]

The AP version are $2500-3000 per caliper IIRC.

[S Lathrop]

The AP calipers have been out for several years.

http://www.apracing.com/calipers/pro...F%3E_2662_2611

[glenn cooper]

Take the first part of the "W' and separate it from the second half. Add in the mark on the top of the second half where the marker rode along the top edge of the box and the word turns into "USED".

I may have taken them out on a wednesday though.

Dan-O, I knew that. Tag - You're it!

Oh yeah, we now take you back to our regularly scheduled programing:
Calipers - how to spend and ass load!

[brownslane]

"Brakes? I don't need no stinking brakes" !!!!

Sorry about the reference to chicano....there was a movie I saw with a guy with an accent.....well, never mind

[Northwind]

Coop,

I don't know if it's a laughing matter but my penmanship is terrible. You know when a doctor writes out a perscription and you try to read it and all you do is think what the heck does that say. Mine is so bad that I have had surgeons look at my notes and ask "What does that say"

Back to da brakes eh!

[glenn cooper]

leaves a lot to be desired, also.

From this it's easy to connect the dots and arrive at the fact that geniuses have crap handwriting!

[nulrich]

Nathan, at the risk of a coronary, do you know the ballpark figure for a set of these babies?

(waiting to see if I should cash in my kid's inheritance)

The Alcon F3 caliper is 1170 British pounds retail, so about $1800 each. I think they've been out for about as long as the AP version, apparently there was a small brake war in F3 at one point.

I don't have a photo, just a CAD model. Here's what it looks like.

Nathan

[RobLav]

Anytime I see the word "optimal" or "optimize", I cringe. Any supposed optimization of a multi-objective function must be based on a human's input for the Measure of Effectiveness or Performance between the variables. And the variables in real life are almost always nonlinear, which makes it even more difficult.

[nulrich]

Part of a brake caliper "drag" is actually aerodynamically created. Aerodynamic drag does not create heat, but it has a significant affect on rotation of the brake assembly.

Tom:

This paragraph made me curious when I first read it. Can you can explain this a bit more? Are you suggesting there is aerodynamically drag created by the interaction between the pads and rotor? Or the rotor itself spinning? Or are you referring to the aerodynamic drag created by cooling flow around the caliper and disk? (That last loss is an important one, for sure, and worth spending the effort to reduce when you are getting down to the last percent of whole car drag.)

One of the tests we've done to measure caliper drag is very simple, and also, I believe, completely accurate. It wasn't my idea originally, so I don't feel like I can disclose it, but I'm sure anyone can figure it out with a little hint: imagine a way of removing all potential sources of caliper drag in less than ten minutes, without unbolting anything from the car.

When we do a back-to-back test, we can't see any difference in acceleration or top speed, either in wheel speed (we measure all four wheel speeds with a sensor wheel that has 20 teeth per revolution) or in the GPS speed. To me, that is definitive proof that our calipers are not producing any drag within our measurement accuracy, which is very high. The test should also show any aero drag associated with the pad/disk interaction, and we don't see any (another hint).

I'm interested to hear your thoughts on this. Thanks.

Nathan

[nulrich]

Anytime I see the word "optimal" or "optimize", I cringe. Any supposed optimization of a multi-objective function must be based on a human's input for the Measure of Effectiveness or Performance between the variables. And the variables in real life are almost always nonlinear, which makes it even more difficult.

Rob:

Please excuse me if I use incorrect terminology in explaining how this "optimization" software works, I just approach it as a design engineer.

The first step is to define the mounting points, the outer envelope of the caliper (constrained by wheel geometry, primarily), and the size and position of the disk, pistons, pads, and abutment plates, all based on the required braking performance.

The desired stiffness of the caliper is then fixed, and the software removes and adds material iteratively to try to achieve minimum mass at that stiffness. When it reaches a "minimum" the caliper design is checked to make sure it can be made, and the model is verified further, then prototyped and tested.

Obviously, the final design is influenced somewhat by the initial choices, but it does seem to be a true optimization around a simple figure of merit (mass). What am I missing?

Nathan

[starkejt]

imagine a way of removing all potential sources of caliper drag in less than ten minutes, without unbolting anything from the car.

Make sure none of the 4 wheels are spinning. Bam, zero drag.

[bill gillespie]

Make sure none of the 4 wheels are spinning. Bam, zero drag.

Any residual parasite/ form drag ?

[RobLav]

Good Morning Nathan,
My comment was not directed at you but more so at the typical marketing hype surrounding optimization. In the case of these brakes, the software is minimizing (optimizing) the mass given a bunch of constraints, design requirements, and assumptions. The objective function is written to minimize mass. It is an entirely reasonable design approach but, in my opinion, when the word optimal is used, literature ought to specify precisely the variable(s) which was (were) optimized.

[Jnovak]

I know for a certainty that there is "other" drag in a brake system in addition to the frictional drag between the pads and the rotor.

I have tested NASCAR race cars at Talladega and Daytona with devices used to keep the pads a large distance from the rotors (approximately .12" clearance). In back to back testing at both tracks with a 2 car team it was worth over .1 seconds/lap. This method was only used in qualifying and has resulted in the pole at both tracks for the team. Of course use of the brakes are not allowed during the 3 lap runs and stopping in the pits requires some extra caution. I am sure that this method is now commonly used in NASCAR land for Daytona & Talladega for qualifying.

I am also certain that this is not due to deflection of the rotors or spindle - bearing assemblies as we had tested for that too. I never understood the entire issue but I do know that it works.

Thanks ... Jay Novak

[nulrich]

Very interesting, Jay. That must be what Tom was talking about earlier. I can imagine some sort of drag associated with the disk boundary layer impacting the pad. I'm guessing someone has done CFD studies on this, and I'm sure someone has measured it on a brake dynamometer. Anyone know?

What kind of speeds are we talking about for those qualifying laps? I would think this kind of "aero" drag would scale with the square of the surface velocity, meaning larger diameter rotors at very high speeds would see much more drag. The top speed in our tests was 140 mph.

In any case, not practical for our cars, we kinda need to use the brakes!

Nathan

[Jnovak]

Lap average speeds approaching 200 mph.

Thanks ... Jay Novak

[R. Pare]

Back in the mid '80's I was involved with the development of the big Alcon Nascar brake system (GTP car calipers on the front, Indy car calipers on the rear), and their 2-stage master cylinders (one version having an anti-knockback check valve).

Spindle flex was a BIG issue - enough so that with the anti-knockback version of the MC on Petty's car brake drag on the right front slowed the car by almost 10 mph at Daytona.

Switching out the MCs gained that back, and pulling the pads all the way back (and gluing them to the pistons to keep them there) yielded another .25 mph.

So yes, turbulent air at the pad/rotor interface is sufficient enough to be measurable.

Nothing new here - this has been known at Indy for 60 years or so.

[Wren]

I think this might be the first time on here that I have seen the word "optimize" used correctly. They(AP/Alcon) actually maximized something within a set of bounds.

Optimized has to be one of the most overused and misunderstood words in marketing. I particularly enjoy when someone offers a frame or bodywork that has been "optimized" and then they release something new the next year that is supposed to be even better.

[starkejt]

Well, if you remember your low-level calculus, maximization is a type of optimization.

[brownslane]

Very interesting, Jay. That must be what Tom was talking about earlier. I can imagine some sort of drag associated with the disk boundary layer impacting the pad. I'm guessing someone has done CFD studies on this, and I'm sure someone has measured it on a brake dynamometer. Anyone know?

What kind of speeds are we talking about for those qualifying laps? I would think this kind of "aero" drag would scale with the square of the surface velocity, meaning larger diameter rotors at very high speeds would see much more drag. The top speed in our tests was 140 mph.

In any case, not practical for our cars, we kinda need to use the brakes!

Nathan

Nathan, you are right on the money. Look at the boundary layer interface with the pads. Add to that the various iterations of brake cooling ducts....there is a significant amount of drag created...or roataional mass influence exerted....Here is my thought; in NASCAR as in other forms of roundy-round racing, there is a measurable increase in average speed in the field after a few laps of racing...in effect, the motion of all those cars rotating acts like a spoon in a bowl of soup. Starts everything spinning....

Now, to use the same analogy of that bowl of soup. Stop the rotation of the spoon, leave it in the bowl, and turn it so its widest part is 90 degrees to the direction of flow.... and what happens.

Picture the same functional rotation of inertia all wrapped around a brake rotor. Add to this the potential radial output from a central source of air (force-fed brake duct)....the rotor is acting like a high-speed air pump. At high velocity, I might add. Caliper shape is more critical than many people may think. Air needs to flow over, through and around the caliper. Then there is the brake pad which is totally blocking (scraping off, if you will) the boundary layer...that is also constantly increasing in heat, separating, allowing new (cooler) atoms to attach, it is literally a swirling maelstrom of highly energized atoms!

There is a heck of a lot going on in those last few thousands of an inch!

Some smart designers in F1 discovered that a lot of this mass/inertial movement around brake rotors can be channeled ....look at the covers over the rotating assemblies. They are not there for good looks or to hide the rotor/caliper design. They (F-1 designers) have engineered the brake caliper/rotor cooling system into a smoothly integrated air pump...the objective of which is to manage temperatures, reduce drag, and actually add thrust by innovative use of air flows.


At least that is my story and I am sticking with it.

Guys, thanks for this thread; this stuff is so cool and I always learn a lot.

BTW, let me wish YOU ALL a Merry Christmas!