I need to order new front brakes now, lol. I think I'm going to stick with the WRX sized setup since they clear my wheels. Probably 2-piece rotors and Hawk HPS pads all around.
I made this post over on the Forester boards to answer some bias questions anyone might have:
**warning** This is a LONG post. A VERY long post. VERY VERY long. I'll CN it at the bottom. If you are a dork, you'll read it all. Data for this thread came from the Subaru Factory Service Manuals and my friend Josh Colombo.
Below is all of the information about brake bias setups. For clarification, all we are doing here is assuming a pedal input pressure and then extrapolating data from that. Because we are only concerned with bias, the actual numbers don't have to have any real world significance, only the ratio really matters. That said, we can guess at a few things and still come up with a reasonably realistic bias guide. There are a few unknowns that we are forced to eliminate. Mainly, the proportioning valve split point. Pressure IS split 50/50, however above a certain threshold the prop valve dramatically limits the pressure flowing to the rear lines. Up until that point, bias will be the same no matter the pedal input. After that point, bias will shift forwards. Since I didn't originally design the system, I have no idea what that point is. Still, these calculations will give us a general idea of how the brakes will feel when paired with different front brake setups. The calculations aren't difficult, just tedious. To cut back on this, we will keep a few things constant. Master cylinder size, brake pad coefficient of friction, and pedal pressure. Keeping these constant will make some bits of the equation constant and save time. When looking at the numbers, there is such a thing as too much rear brake. It's like riding a bicycle. Too much rear brake and the rears lock up and you don't stop nearly as well as you could. Too much front brake and you lose rear wheel traction putting even more stress on the front brakes. The ideal bias is tough to determine and varies for everyone. To get an idea of what you MIGHT like, you could go ride a bicycle and experiment with lever force. Ideally you'll have enough rear bias to not lock the rear tires, but not so much front bias that you dive excessively. This ties a significant amount into suspension setup, tire choice, and alignment settings. All of these things will impact the "right" brake bias for you. Really soft suspension and a front biased setup will make the rear brakes inefficient. Really stiff suspension and not enough front bias, and you'll overload the rear tires and make trail braking difficult. It's kind of a slippery slope and the best thing you can do is look at the stock numbers and then think to yourself how you might like to change it. If you are unhappy with stock, and would really like more rearward bias, then go that way and judge the amount of bias you'd like.
Brakes are a system. You have your pedal input and the associated pressure applied to the master cylinder. The master cylinder applied force across an area giving you some line pressure. The pressure in the lines and the area of the piston give you some force output. The friction between the pad and the rotor with the force applied and the distance from the center of the pad to the center of the hub all compile to produce torque. The actual torque applied to the hubs is not the focus of this discussion. This post is all about ratios. The ratio of front to rear torque, and to some extent, the ratio of one setups brake torque to another. Since we are keeping so much constant, we can assume that if one system produces X brake torque, and another produces 1.25X brake torque, then the latter is proportionally "stronger" by that amount. I wouldn't expect numbers that large. Also, one thing that we will choose to ignore is tire coefficient of friction and its relationship to stopping the vehicle. This gets into mathematical models far beyond the scope of this thread. Since this is a brake comparison and not a tire comparison, the figures to calculate theoretical stopping distances would be constant and directly proportional to brake torque within reason (it's possible that the torque produced at our theoretical pedal pressure is beyond the torque required to stop the vehicle).
So let's begin by laying the ground rules. We need to assume our constants.
Unfortunately, I don't have access to my FSM right now so I don't exactly know the stock MC size. I seem to recall it being 1.00" since I KNOW it's not the 1.0625" STi size and I doubt it's the 0.9375 of the LGT. That's important. Right away, we encounter our first tidbit. Larger MC's mean greater pedal pressure required for the same line pressure. So if you feel like your brakes have to light of pedal pressure, a bigger MC is for you.
We also need to assume a few other things. Mainly piston travel for sliding calipers (.025"), and roll back. These will be constant, since in reality I think they are. We won't get into caliper flex and such, because they aren't necessarily important.
Now we need the following brake data for each of our interests; effective disc diameter, piston area, and number of pistons.
This is where it gets complicated. A fixed caliper has all pads pushing with the same pressure. A fixed caliper has the same pressure for each piston. A sliding caliper, however, applies the pressure from the piston twice (once on each side of the pad). So a 2-piston sliding caliper effectively applies the force from 2 pistons on each side, making it have twice the area. Confused? The pistons in a sliding caliper push the pad into the rotor. When the pad contacts the rotor, the caliper pulls the outboard pads into the rotor. Negating caliper flex, it pulls the outboard pad into the rotor with exactly the same force as it pushes the inboard pad to the rotor. We COULD simply ignore half of the fixed caliper pistons and focus on half of the braking system, or we could multiply the sliding piston calipers by two. Doesn't really matter. Let's do the latter because it makes for more easily handled numbers. This is funny. A sliding caliper actually acts like it has twice its actual piston area. Funny to think, huh? That a fixed caliper and a sliding caliper can have exactly the same effective piston area. No wonder the M3 has never come with monoblock calipers!
Also, outside diameter does not impact brake torque. On the contrary, effective diameter is what matters. The effective diameter is provided in the FSM. It is the average diameter of pad swept area. If you look at the swept area of a rotor, the smaller diameter of the inside and the larger diameter of the outside would produce different torques at each point. The effective diameter can be thought of by the average of these two points, and as such is slightly higher than the center point of the swept area. Funky, I know. But the FSM provides this, so it doesn't REALLY matter how they got it. It is also not rotor specific, but has to do with the pad shape as well. The FSM provides it, so we go with what they give. We will only make one assumption, and that is the effective diameter of the RB 316x18 rear rotors. We will assume that they have the same effective diameter as the STi rotors. It won't change much if they are slightly different.
Another thing to consider is pad compound. There are a bunch of different compounds. Standard pad coefficient of friction ranges from 0.38 to 0.40. Performance street pads range from 0.45 to .050. Race pads range from 0.60 to 0.65. We'll use performance street pads because that's what most of us will run. We'll simply average the values and use 0.475.
The last thing to consider (I promise) is the brake booster. There is some data out there, but we will make some generalizations. Accurate or not, it would give us a good range for where our line pressure should lie. We'll basically ignore this and just make an assumption for the pressure applied to the rear of the MC. We'll call it 100 pounds to try and keep numbers as manageable as possible.
OK, I think we've gone over everything. Now let's do some math!
We'll start with stock FXT brakes. They are easy and give a good base line. I'll only do the front end calculations once and then just post the numbers for the other setups. I'll post the data though, incase anyone wants to double check my math. I'll show the steps as well so that you can play with different MC sizes if you so desire.
The 1" master cylinder has a surface area of 0.785.
So we start by finding line pressure. 100lbs of input force divided by the surface area will give us line pressure.
100pounds/.0785sq.in = 127.3885 psi (so much for keeping simple numbers).
With this, we are armed with our first constant. Line pressure = 127.3885psi. Yay!
Now areas from above. Pressure times area.
Front: 127.3885psi * 8.915sq.in = 1135.69lbs of clamping force on the front rotors
Rear: 127.3885psi * 3.5325sq.in = 450lbs of clamping force on the rear rotors.
Good! Now we can figure out torque, and from that bias information.
What we need to do is this:
Fluid pressure * effective diameter of the disc * coefficient of friction * Effective piston area.
Front: 127.3885psi*9.75in*0.475*8.915sq.in = 5259.664313lb-in of force
Rear: 127.3885psi*9.06in*0.475*3.5325sq.in = 1936.575 lb-in of force
Now, we find the ratios.
5259.664313/(5259.664313+1936.575) = .730891 Front and 1-.730891 rear
That means a stock FXT has 73% of its braking force up front and 27% in the back. Good!
Easy, huh?
So, fluid pressure is the same for all setups as is the coefficient of friction. So we can combine these into one constant term we will call M. 127.3885*0.475 = 60.5095
Now, to find torque we just multiply piston area times effective diameter times M.
Notes: The rear pistons are the same size as the 1-piston rear brakes of the LGT. Effectively, the 4-pots provide LESS clamping force than the 2-piston brakes. This shifts the bias rearward which is why people see braking advantages from the 4-pots. Stiffer calipers and more rearward bias. Ultimate front brake torque is actually less with the 4-pots than with the stock FXT brakes.
Brembo
Front Brembos -
Outer Disc Diameter - 12.83" (326mm)
Effective Disc Diameter - 10.55" (268mm)
Caliper piston diameter - 1.575" x2 and 1.811" x2
Caliper piston area - 9.043743.in
Notes: This is a perfect example of two different outer diameters with the same effective diameters. The STi front caliper is tricky because of the different piston sizes. Still, it works out well and the bias is pretty good.
OK, so that's it for stock calculations. We have:
Stock FXT -
Front = 73% Rear = 27%
Stock LGT -
Front = 72% Rear = 28%
Stock 4/2 pot -
Front = 69% Rear = 31%
Stock Brembo -
Front = 74% Rear = 26%
Now, the calculations for the rear brake torque numbers of the RB big brakes
Notes: Same as LGT rears! In fact, this is the same effective brake torque as the standard 'H6' upgrade. The benefit you have is vented rears over solid rears, allowing for a greater heat holding capacity and better cooling.
Bias is 71.91132567% front and 28.0886743226% rear
So, that about covers it.
We just did calculations for Stock: FXT, LGT, STi, 4/2-pots
We also did the 290x18 with stock: FXT, STi, 4-pots
We also did the 316x18 with stock: FXT, LGT, STi, 4-pots
All of these combinations put you fairly close with an ideal 70/30 split and any of them would work really well.
One thing I would consider when choosing is your style of driving and your suspension setup. If your car is stiffly sprung and you don't mind the tail stepping out a little, then the 316 rears would likely be ideal. More rear bias will take advantage of your suspension and you'll notice far less dive.
I'd say for most people with stock front brakes, the 290x18 kit would work out really well.
For those with 4-pots, it's a tough draw.
Brembo's should definitely get the 316x18 kit.
LGT fronts, I'd go with the 316x18 rear kit.
Also note, RB DOES have a 316mm front brake upgrade for the WRX/FXT which will then turn your numbers into those for the LGT w/ 316 rears.
I hope this has been informative and has helped you make a decision. If you have any questions, please let me know and I'll do my best to answer them.
Also, I'll ask to confirm the effective disc diameter on the RB 316x18mm rotors. I'll correct this post if I find that EDD to be different than the STi rotor.
Ok....read through the mega post, and it's really quite a good post about how to do bias calculations and understand sort of what's going on. I think the one thing that needs to be stressed is that what was posted is a static condition used for comparison, and that static conditions really don't exist in real life.
Not sure if you want to include this, but a small blurb regarding the difference between brake torque and brake bias might be good. I say this because my spreadsheet is done in brake torque only. This was done intentially because of the large variability in master cylinder sizes and brake booster assist.
I had one question already as to why my numbers were different then numbers coming from TCE's Brake bias calculator, and that was the answer, my numbers are brake torque, TCE's are brake bias.
The only exception I had was regarding the calculation of area for fixed vs. sliding calipers.
... We COULD simply ignore half of the fixed caliper pistons and focus on half of the braking system, or we could multiply the sliding piston calipers by two. Doesn't really matter. Let's do the latter because it makes for more easily handled numbers.
In terms of the mathematics, it doesn't matter, but in terms of terminology in the brake industry, it does matter. I had this very discussion with TCE when I was developing my spreadsheet. I argued the same thing, it doesn't matter.
You can mention caliper piston area to "brake" guys and they can probably relate to a certain range of numbers. That range is always using just half of the caliper (all the pistons in a sliding caliper setup, or half the pistons in a fixed caliper setup). So that's how my spreadsheet is setup.
It's up to you whether you want to make that clarification, but I wanted to at least point it out.
Lastly, not sure if you posted it in the forester board, but can you please include a link to my brake math spreadsheet.
BAC5.2 wrote:It's like riding a bicycle. Too much rear brake and the rears lock up and you don't stop nearly as well as you could. Too much front brake and you lose rear wheel traction putting even more stress on the front brakes. The ideal bias is tough to determine and varies for everyone. To get an idea of what you MIGHT like, you could go ride a bicycle and experiment with lever force. Ideally you'll have enough rear bias to not lock the rear tires, but not so much front bias that you dive excessively. This ties a significant amount into suspension setup, tire choice, and alignment settings. All of these things will impact the "right" brake bias for you. Really soft suspension and a front biased setup will make the rear brakes inefficient. Really stiff suspension and not enough front bias, and you'll overload the rear tires and make trail braking difficult.
There are some things in that paragraph that I don't think are quite right that could be re-worded probably.
For one, too much front brake doesn't cause a loss of rear wheel traction. More like a loss of front wheel traction. The fronts do more of the work and lock up first.
You also seem to be implying that suspension stiffness affects the amount of weight transfer. It does not. Stiffness only changes the speed of weight transfer. The amount of weight transfer is mainly determined by G force and CG height.
Jamal brought up some of the points I kind of stayed away from.
Depending on how in-depth you want to explain things, it becomes increasingly more difficult to include everything affecting braking. Braking is truly a very dynamic system.
I tried to keep it as "OK for the masses" as possible.
Dumbing down of some aspects of vehicle dynamics, fluid dynamics, and theoretical vs. practical was required. I could have made comment to the effects of trail braking, suspension motion, and isolated each corner. Not to mention the significant influence of AWD to the braking comparison. I also feel that bias numbers are really important to consider. You can SAY that one brake setup shifts torque 7% to the rear, but to say that the bias is 60/40 helps put it into perspective. Ride a bike and get a feel for the varying bias. I probably should include more about how impractical this practical evaluation is. I might do that later.
It's easy to lose non-engineering types in the maelstrom of calculations and variables.
I still think it's accurate to say that too much front brake reduces rear traction. Ride a motorcycle in a straight line. Apply an even amount of rear brake. Start applying more front brake. Eventually, you'll lock the rear without locking the front. It's more about weight transfer. And on the topic of weight transfer, I am using the term incorrectly. Weight transfer is commonly thought of as suspension motion. Dive under braking. In extreme situations, mega-dive will lighten the rear, decreasing traction while increasing front traction (Less weight on a tire, less ability to make effective use of the contact patch). Again, unrealistic in the real world, but for the purposes of the general public I feel it was a valid explanation.
Tires, not brakes, are the main thing impacting stopping distances. Suspension follows a close second. The thread was about the one thing that impacts stopping distances the least and instead focuses more on balance and steps toes into dynamics of a brake system. In that respect, I feel the post was successful. Still, I see the points mentioned in this thread and I certainly could modify my post to address those who may be more techincal than others. If I have the time, I'll do just that.
BAC5.2 wrote:I still think it's accurate to say that too much front brake reduces rear traction. Ride a motorcycle in a straight line. Apply an even amount of rear brake. Start applying more front brake. Eventually, you'll lock the rear without locking the front. It's more about weight transfer.
Increasing the front brake bias will in no way decrease the rear braking traction or ever cause the rears to lock up. Brake bias and weight transfer are only related to the effect that a properly biased system will be able to generate more braking grip out of all four tires and pull a higher braking g force, which will result in more weight transfer, and therefore more load on the front and less on the rear.
If anything, more front bias will cause the brakes to become less effective, reducing braking force, reducing weight transfer, and increasing the load and traction on the rear tires.
You are trying to tie weight transfer, brake bias, and suspension stiffness all into one using a bike analogy that doesn't really relate to cars, and your reasoning is that explaining things the way they actually are would be too complicated?
I think it's easy enough to explain how things work without having to worry about going into fluid dynamics or theory vs. practice or whatever. You have front brakes, and you have rear brakes. To brake most effectively and stop in the shortest distance possible, you want the bias to be such that all four tires are doing their share of the work to stop the car.
jamal wrote:If anything, more front bias will cause the brakes to become less effective, reducing braking force, reducing weight transfer, and increasing the load and traction on the rear tires.
That doesn't sound correct... More front brake is less effective because it increases the load and traction on the rear tires?
I think what I'm saying is valid, and gets the point across about too much front brake bias. Too much front is bad, and the motorcycle analogy works perfectly to demonstrate that. Keep rear torque static and increase front torque and you decrease load on the rear tires, effectively decreasing traction at the rear wheels. It's a function of suspension setup and brake bias. Weak sauce suspension and mega front bias decreases the ability of the rear brakes to stop the car. Granted in reality the rear brakes will very rarely lock before the fronts with too much front bias, the idea is there. Like doing an endo on a bike.
For the purpose of understanding brake bias, riding a bike is a perfect analogy. You can't vary front and rear brake torque on a car, but on a bike you can. You can feel the difference more front or more rear makes. Yes, the braking dynamics are totally different than in a car, but that's not what I was getting at. When someone says "well, what's the difference between 74/26 front/rear and 72/28 front/rear bias?" Saying "one is slightly more front biased than the other" doesn't tell them anything. But riding a bike, where you can experiment, to some extent, that bias puts it into a real world perspective.
The main goal is to prevent people with stock suspension from dumping Brembo's on the front of the car and thinking they have the mackdaddy brake setup. In reality, a brembo/stock car is probably going to stop far worse than a properly biased setup. More weight transfer, less load on the rear tires and less effective rear brakes. Even if you have REALLY sticky tires that can handle the added load on the front brakes, you still are unsettling the rear end a significant amount.
Only partially related, do you think sending bias rearward will decrease forward weight transfer? As you brake, the rear suspension tries to compress. More rear bias means more compression and less weight transfer, which means more consistent load on the tires and an overall greater braking force, right? You can experiment with the e-brake to see the rear compression under braking.
the only thing here that affects weight transfer is the total stopping force provided by the tires. If two cars are braking at 0.5g and one is doing it with only the front brakes, and the other is doing it with all four, they will both transfer the same amount of weight.
Say, for example, you have a car with no rear brakes. It will not be able to stop as well as a car with four brakes, which means it will transfer less weight, and the rear tires will not be providing any stopping force. Since they're just rolling along it gives them more available traction.
BAC5.2 wrote:For the purpose of understanding brake bias, riding a bike is a perfect analogy. You can't vary front and rear brake torque on a car, but on a bike you can. You can feel the difference more front or more rear makes. Yes, the braking dynamics are totally different than in a car, but that's not what I was getting at. When someone says "well, what's the difference between 74/26 front/rear and 72/28 front/rear bias?" Saying "one is slightly more front biased than the other" doesn't tell them anything. But riding a bike, where you can experiment, to some extent, that bias puts it into a real world perspective.
I don't think the bike example really works well because with a bike you have two completely independent variables: front brake pressure and rear brake pressure.
On a car, you have one variable, master cylinder pressure, your bias split must always equal 100%. To make the same comparison on a bike you'd have to gradually ease up on the rear brake so that the total braking bias/force is equal to 100.
jamal wrote:the only thing here that affects weight transfer is the total stopping force provided by the tires. If two cars are braking at 0.5g and one is doing it with only the front brakes, and the other is doing it with all four, they will both transfer the same amount of weight.
Say, for example, you have a car with no rear brakes. It will not be able to stop as well as a car with four brakes, which means it will transfer less weight, and the rear tires will not be providing any stopping force. Since they're just rolling along it gives them more available traction.
It would seem to me that suspension motion would affect weight transfer as well. Less motion, less weight transfer. More motion, greater dive, greater load transfer to the front wheels.
Either way, I think the point I was trying to make is more about dive. Less dive, greater ability the rears have to maintain traction. Suspension most definitely plays a role in braking distances.
Legacy777 wrote:I don't think the bike example really works well because with a bike you have two completely independent variables: front brake pressure and rear brake pressure.
On a car, you have one variable, master cylinder pressure, your bias split must always equal 100%. To make the same comparison on a bike you'd have to gradually ease up on the rear brake so that the total braking bias/force is equal to 100.
I think you are mistaken. Bias ALWAYS equals 100%. It's a ratio of brake torques.
Whether you modify pressure, or increase diameter you are varying the amount of brake torque. Remember, bias is: Front or rear torque divided by front + rear torque.
So to make the same comparison, you can simply squeeze the front a bit more on a bike to be able to get more front bias while maintaining the same rear brake torque. Very much like putting HUGE front brakes with stock rear brakes.
BAC5.2 wrote:I think you are mistaken. Bias ALWAYS equals 100%. It's a ratio of brake torques.
Whether you modify pressure, or increase diameter you are varying the amount of brake torque. Remember, bias is: Front or rear torque divided by front + rear torque.
So to make the same comparison, you can simply squeeze the front a bit more on a bike to be able to get more front bias while maintaining the same rear brake torque. Very much like putting HUGE front brakes with stock rear brakes.
Honestly....Looking back at the posts, I'm not exactly sure what I was trying to say, because I can't seem to decipher it now. The thought got screwed up going from my brain to the keyboard.
I agree that overall bias has to be 100. I think I was trying to say something about variable brake pressure difference between the bike, and the "fixed" car brake pressure, and how with the bike, you have an on-the-fly adjustable bias, while with the car, you don't really.....but I'm not sure how I was relating that to the previous posts/discussion. Oh well.
BAC5.2 wrote:
It would seem to me that suspension motion would affect weight transfer as well. Less motion, less weight transfer. More motion, greater dive, greater load transfer to the front wheels.
That is completely wrong. Body motion contributes a very very small amount to the total weight transfer, and only due to the slight movement of the center of gravity because of the change in angle of the car.
The force of deceleration is what creates the weight transfer. Softer suspension does move more and creates more dive/pitch/roll, but given the same g force, the weight transfer is the same. It has to be.
I find it hard to believe that suspension motion barely affects weight transfer. The more dive you have under braking the more the center of gravity is shifting towards the front of the car. When the back of the car moves up a couple inches and the front goes down a couple inches it an have a significant effect on where the center of gravity of the car is and where the weight is distributed on a nose heavy car, ala weight transfer. A big reason porsche's always stop sooooo fast is the rear weight bias, and how little weight transfer there is under braking since the motor and transmission are located in the rear of the car.
The opposite can be said for acceleration. RWD drag cars are usually softly sprung to get a lot of weight transfer to the rear of the car to help with traction. A friend with a stiffly sprung 5.0 mustang can't get off the line to save his life, yet another guy here with a stock suspended camaro with smaller tires owns him off the line.
so, say we have the front of the car moving down two or three inches, the front moving up two or three inches... it's not possible for the CG to move even that far. Do you think that would really contribute much to weight transfer?
As for drag cars, the softer sprung cars aren't transferring more weight. They're just doing it more slowly, which loads the tires more gradually, which improves the grip.
The only effect stiffness will really have is how quickly the weight transfers and the chassis reacts. Which will probably result in a slight stopping distance improvement because the car reacts more quickly. However, it does not have an effect on the absolute braking g force the car can create. The tires are responsible for that.
Force of deceleration is only one part of weight transfer. Center of gravity-height also plays as significant of a role, as does wheelbase.
Even if the front and rear of the car move up and down in equal amounts, the effect on the center of gravity height (COGH) is significant on a nose heavy car. COGH height is directly correlated to weight transfer. If a 3000lb car has the COGH drop by 2 inches under braking, maintaining .9g and a 96in wheelbase, that changes the weight transfer by at least 50lb. Since traction is a function of friction AND the normal force at the tires, its kind of a big deal.
Δload=((Hcg*g)/wb)*(Wtot)
Δload= load transfer, lb
Hcg= center of gravity height
g=acceleration in g
wb=wheelbase
Wtot= total weight
So a drag car that raises it nose by only a few inches can mean the difference of many pounds of normal force at the tire, and with a high coefficient of friction at the tire from a sticky tire the effect is even more exasperated. Serious drag cars such as top fuel get around that with tires that have a coefficient of friction much higher than 1.0. If they had too much weight transfer and change in pitch associated with that, they would be impossible to drive.
Thank you Dan, that's what I've been trying to get at. The tires are responsible for stopping the car, but friction is a factor of more than just tire compound. Brake bias and weight transfer are significant factors of brake control, and suspension plays a key role in the latter of the two.
.
