Jump to content


Photo
* * * * * 1 votes

Brake by wire?


  • Please log in to reply
93 replies to this topic

#51 Diablobb81

Diablobb81
  • Member

  • 8,724 posts
  • Joined: August 09

Posted 03 February 2014 - 15:59

What would we do without you? Good to have such technical info.



Advertisement

#52 Guest_4L3X_*

Guest_4L3X_*

Posted 03 February 2014 - 18:11

I still think that even with limited number of variables one team can still write code to allow for ABS or even Traction control like functions.  I do think it's going to be hard to abstract wheel spin from them, they need that variable for a number of things.  But even without using that for brake mapping, one could read the load/resistence the ERS unit is having and read if the wheel/shaft is about to lock (sharply decreasing speed rate) and release it before that, not the hydraulic pressure but the corresponding side on the ERS.  That way, it would go back to the driver if he would feel the release and increase the pedal pressure to THEN lock the wheels on purpose (or go straight to lock in pedal to the metal situations).  If they can read the pedal fluctuations/pressure, they can allow for both ABS like control and panic/spin allowed lock mode, so that's not a problem.



#53 Clatter

Clatter
  • Member

  • 44,648 posts
  • Joined: February 00

Posted 03 February 2014 - 18:28

I still think that even with limited number of variables one team can still write code to allow for ABS or even Traction control like functions.  I do think it's going to be hard to abstract wheel spin from them, they need that variable for a number of things.  But even without using that for brake mapping, one could read the load/resistence the ERS unit is having and read if the wheel/shaft is about to lock (sharply decreasing speed rate) and release it before that, not the hydraulic pressure but the corresponding side on the ERS.  That way, it would go back to the driver if he would feel the release and increase the pedal pressure to THEN lock the wheels on purpose (or go straight to lock in pedal to the metal situations).  If they can read the pedal fluctuations/pressure, they can allow for both ABS like control and panic/spin allowed lock mode, so that's not a problem.

Writing the code wouldn't be that hard, but as everything has to go through the standard ECU it would be a far more difficult task to hide it.



#54 prty

prty
  • Member

  • 8,417 posts
  • Joined: April 05

Posted 03 February 2014 - 18:44

I'm not an expert (i.e. have no clue about the subject), but isn't brake by wire incredibly dangerous? I mean, if it's an extra electronic system between the driver and the brakes that take care of the force applied to them. The more elements into a system, the more sources for failures, and you don't to have one when braking with an F1 car.



#55 RoutariEnjinu

RoutariEnjinu
  • Member

  • 2,442 posts
  • Joined: March 09

Posted 03 February 2014 - 18:46

I still think that even with limited number of variables one team can still write code to allow for ABS or even Traction control like functions.  I do think it's going to be hard to abstract wheel spin from them, they need that variable for a number of things.  But even without using that for brake mapping, one could read the load/resistence the ERS unit is having and read if the wheel/shaft is about to lock (sharply decreasing speed rate) and release it before that, not the hydraulic pressure but the corresponding side on the ERS.  That way, it would go back to the driver if he would feel the release and increase the pedal pressure to THEN lock the wheels on purpose (or go straight to lock in pedal to the metal situations).  If they can read the pedal fluctuations/pressure, they can allow for both ABS like control and panic/spin allowed lock mode, so that's not a problem.

 
Have you ever heard of variable scope? If the area they get to write their own software is a delegate function, then McLaren technologies can limit the variables they get to look at when writing their own code. And that is assuming they actually have that much flexibility and power. It could be locked down further than that to the point they're putting numbers in cells, like in a fuel/ignition map.
 
Just because the ECU has all those inputs available, doesn't mean the area where you write any braking code has access to them.
 
When I was talking about a limited number of variables, I'm talking about variable scope for any custom braking function that will have to share a common interface to work with the supplied ECU. I'm not talking about what the McLaren ECU itself can see, as of course it has an overall picture of the car.
 
If it was so easy to write ABS or even Traction Control like functions, then they'd be there already in the throttle maps. They aren't. Likely because they're only allowed to punch numbers in cells, or at the most, are working with an explicitly defined function delegate with a limited scope of parameters to work with. I'd be surprised if they were allowed THAT much power and flexibility.
 
It's simple and tidy software engineering to only expose what needs to be exposed. It's the only way this standard ECU could have stopped Traction Control to begin with.
 
If McLaren are only providing the hardware, and they get to write it from the ground up, which I doubt, then the SAME process of code review that has stopped Traction Control will stop ABS and Traction Control like functions.
 

one could read the load/resistence the ERS unit is having and read if the wheel/shaft is about to lock (sharply decreasing speed rate) and release it before that

 
One couldn't do that if the closed source binary part of the software didn't expose any wheel/shaft speed, which would make sense. In mapping fuel/ignition the only thing you're exposed to will be engine speeds. If they had access to wheel/shaft speed at that point, they'd be traction control too. But there isn't, because whoever is writing the software for the McLaren CPU isn't that stupid. You're worried about what you don't understand.

I'm not an expert (i.e. have no clue about the subject), but isn't brake by wire incredibly dangerous? I mean, if it's an extra electronic system between the driver and the brakes that take care of the force applied to them. The more elements into a system, the more sources for failures, and you don't to have one when braking with an F1 car.


Fortunately the rules state there must still be a hydraulic link to the rears. The fronts remain as they are. Think of it as a computer having control of the brake bias, rather than it being an extra link in the chain.

Edited by RoutariEnjinu, 03 February 2014 - 18:47.


#56 Victor_RO

Victor_RO
  • RC Forum Host

  • 6,067 posts
  • Joined: March 09

Posted 03 February 2014 - 18:49

I'm not an expert (i.e. have no clue about the subject), but isn't brake by wire incredibly dangerous? I mean, if it's an extra electronic system between the driver and the brakes that take care of the force applied to them. The more elements into a system, the more sources for failures, and you don't to have one when braking with an F1 car.

 

The rule posted earlier (post #56) seems to indicate that if the brake-by-wire electronics fail, there should be a fail-safe which means that the rear brakes will continue to operate on a hydraulic circuit similar to what we've had so far.



#57 saudoso

saudoso
  • Member

  • 6,776 posts
  • Joined: March 04

Posted 03 February 2014 - 19:06

Pirelli better start heating up the PR machine, because there will be a lot of locking and a lot of damaged rubber due to it. And in the end they will take the heat.



#58 Guest_4L3X_*

Guest_4L3X_*

Posted 03 February 2014 - 19:22

 
Have you ever heard of variable scope? If the area they get to write their own software is a delegate function, then McLaren technologies can limit the variables they get to look at when writing their own code. And that is assuming they actually have that much flexibility and power. It could be locked down further than that to the point they're putting numbers in cells, like in a fuel/ignition map.
 
Just because the ECU has all those inputs available, doesn't mean the area where you write any braking code has access to them.
 
When I was talking about a limited number of variables, I'm talking about variable scope for any custom braking function that will have to share a common interface to work with the supplied ECU. I'm not talking about what the McLaren ECU itself can see, as of course it has an overall picture of the car.
 
If it was so easy to write ABS or even Traction Control like functions, then they'd be there already in the throttle maps. They aren't. Likely because they're only allowed to punch numbers in cells, or at the most, are working with an explicitly defined function delegate with a limited scope of parameters to work with. I'd be surprised if they were allowed THAT much power and flexibility.
 
It's simple and tidy software engineering to only expose what needs to be exposed. It's the only way this standard ECU could have stopped Traction Control to begin with.
 
If McLaren are only providing the hardware, and they get to write it from the ground up, which I doubt, then the SAME process of code review that has stopped Traction Control will stop ABS and Traction Control like functions.
 
 
One couldn't do that if the closed source binary part of the software didn't expose any wheel/shaft speed, which would make sense. In mapping fuel/ignition the only thing you're exposed to will be engine speeds. If they had access to wheel/shaft speed at that point, they'd be traction control too. But there isn't, because whoever is writing the software for the McLaren CPU isn't that stupid. You're worried about what you don't understand.


Fortunately the rules state there must still be a hydraulic link to the rears. The fronts remain as they are. Think of it as a computer having control of the brake bias, rather than it being an extra link in the chain.

 

I think you're assuming that all electronics in the car are coded inside the McLaren ECU.  I'm not talking engine mapping, and I'm not assuming the only way to control the brakes would be to control the ECU.

Think about in an non electronic brake control: say one team would develop a rubber where the brake pedal meets the floor, a variable density rubber that would deform and allow different braking for according to how hard the pedal is pressed. Now imagine that this rubber is actually an actuator is in the brake pedal itself, and that it can read enough information from variables, not necessarily connected to the ECU (say wheel speed, car speed, vibration sensors, tire temps, etc.).  So it's not that engine mapping would change,  but the pedal brake itself (in this hypothetical scenario) would be where the engineering creative genius would actuate.

 

Unless you say ALL THE ELECTRONICS IN THE CAR are in the ECU, and that the difference betweeen Ferrari's brakes and Mercedes are ALL in the numbers they enter in the ECU, than okay, your points are valid.  But if the teams have a tire sensor, heat sensor and all sorts of sensors in the brakes, cooling, turbo, ERS and what not, what says they can't use those not to control what the ECU controls, but something like the Brake pedal itself?  



#59 Guest_4L3X_*

Guest_4L3X_*

Posted 03 February 2014 - 19:27

Writing the code wouldn't be that hard, but as everything has to go through the standard ECU it would be a far more difficult task to hide it.

 

Why would they have to go through the standar ECU, if it's not about controlling gears or engine mapping, but pressure on brake lines, pedal pressure, or how to direct hydraulic pressure, or wheatever?  I mean, does the button for the water they drink have to go thorough the ECU too?  If not, loophole.



Advertisement

#60 demet06

demet06
  • Member

  • 126 posts
  • Joined: December 13

Posted 03 February 2014 - 19:28

Have all the teams fitted brake by wire systems or are some still running hydraulic systems? I would imagine some of the smaller teams would still run a hydraulic system as the brake by wire is something else to go wrong with the new complicated drivetrains they have.



#61 redreni

redreni
  • Member

  • 4,709 posts
  • Joined: August 09

Posted 03 February 2014 - 19:31

So in theory this new system should have the same feel and performance of the old system? Will a driver who's exceptional at late breaking with the old system still carry that advantage over to this new one?


Last year the driver had to adjust the brake bias front to rear from corner to corner depending on the amount of KERS harvesting. More KERS harvesting, more bias to the front, as otherwise the rears would tend to lock as KERS harvesting kicked in. This year the driver won't need to do that as ERS harvesting will be automatically compensated for. Presumably there will still be a brake bias lever governing the relative amounts of braking effort front to rear, but it will revert to being more of a setup tool than something a drivrr would use multiple times per lap to compensate for KERS. So if the system works well it will feel to the driver like a pre- KERS F1 braking system, no?

#62 Clatter

Clatter
  • Member

  • 44,648 posts
  • Joined: February 00

Posted 03 February 2014 - 19:34

Why would they have to go through the standar ECU, if it's not about controlling gears or engine mapping, but pressure on brake lines, pedal pressure, or how to direct hydraulic pressure, or wheatever?  I mean, does the button for the water they drink have to go thorough the ECU too?  If not, loophole.

All the controls for the rear brakes has to go via the ECU as per the rules. 



#63 Guest_4L3X_*

Guest_4L3X_*

Posted 03 February 2014 - 19:36

All the controls for the rear brakes has to go via the ECU as per the rules. 

So no more manual levers?  



#64 prty

prty
  • Member

  • 8,417 posts
  • Joined: April 05

Posted 03 February 2014 - 19:39

 
Fortunately the rules state there must still be a hydraulic link to the rears. The fronts remain as they are. Think of it as a computer having control of the brake bias, rather than it being an extra link in the chain.

 

The rule posted earlier (post #56) seems to indicate that if the brake-by-wire electronics fail, there should be a fail-safe which means that the rear brakes will continue to operate on a hydraulic circuit similar to what we've had so far.

 

Thanks for the answers :up:



#65 toofast

toofast
  • Member

  • 512 posts
  • Joined: October 12

Posted 03 February 2014 - 19:51

dms1431ja91.jpg

 

I think the BB-R and BB-F buttons are for changing the brake bias.


Edited by toofast, 03 February 2014 - 19:52.


#66 RoutariEnjinu

RoutariEnjinu
  • Member

  • 2,442 posts
  • Joined: March 09

Posted 03 February 2014 - 19:51

I think you're assuming that all electronics in the car are coded inside the McLaren ECU.  I'm not talking engine mapping, and I'm not assuming the only way to control the brakes would be to control the ECU.


A secondary ECU just for braking? Then why not a piggy-back ECU for traction control from throttle input. One that looks at the wheel speed, and adjusts the drive by wire throttle (they've been that way for years) to lift off at super human speeds when the wheels break loose?

What you're saying isn't unique to the new brake by wire system. There's a reason everything has to go through the McLaren ECU, and such active systems are banned.

Unless McLaren are idiots, the brake by wire system is going to be no more exploitable than the throttle by wire system they have already had and governed perfectly well.

If they're going to have piggy-back electronics for brakes, they may as well have it for throttle too. There's literally nothing special about the brake by wire system in that respect.

It's also imperative for Formula 1 is to stay current or even lead the way. It's all about integrated Power Units and integrated braking systems now. It's the way it's going to go. Brake discs and pads are wasted energy. Hybrid shouldn't be a word only associated with dull and dreary Hondas, and putting smiles on the faces of people that love a nice strong celery soup to go with a glass of fair trade tomato juice.
Racing can get in on it and use it to build faster cars than they otherwise could too. Even without the Energy Store and, the continuous MGU-H to MGU-K feed is a brilliant alternative to a waste-gate. The runway feedback loop has not just been restrained, but fully harnessed.

To call it fake or turning the drivers into operators instead of "real" drivers (not that you did) is to say they should also do away with fuel injection and computerised ignition and injector duty maps, and just have various dials to twist depending on how much throttle they have on and what gear they're in.

Edited by RoutariEnjinu, 03 February 2014 - 19:55.


#67 RoutariEnjinu

RoutariEnjinu
  • Member

  • 2,442 posts
  • Joined: March 09

Posted 03 February 2014 - 19:53

So no more manual levers?


There still has to be a hydraulic link to the rear callipers. I think Clatter means any electronic regulation of its pressure has to be controlled through the McLaren ECU. Just as with throttle input. It's really not any different. It just feels it as traditionally braking systems have always been mechanical. Throttles used to be too. And clutches, in fact.


As long as the throttle pedal actuation directly relates to torque at the driving wheels, with too much causing wheel-spin, and the brake pedal actuation directly relates to retardation of the wheels, with too much causing a lock-up, it doesn't matter if the Power Unit is a combination of ICE, electronic, nuclear or hamsters in wheels all being added together in whatever proportions, or whether the braking is a combination of friction, electricity generation, or a giant shoe coming out and rubbing against the tyre, all being added together in whatever proportions.

As long as those pedal movements directly relate to torque or friction, it doesn't really matter.

Edited by RoutariEnjinu, 03 February 2014 - 19:59.


#68 RoutariEnjinu

RoutariEnjinu
  • Member

  • 2,442 posts
  • Joined: March 09

Posted 03 February 2014 - 20:07

I think the BB-R and BB-F buttons are for changing the brake bias.


Good spot. It does look like they'll make the brake-by-wire system that "falls back" to a hydraulic system nothing more than the baseline brake-bias being adjusted to compensate for MGUK drag. Exactly as they did before, but rather than having to do so manually. If the KERS failed last year they'd be straight on the radio to tell them to put the brake-bias rearward. Or if they were told to adjust the harvesting rate, they were also told so many clicks forward or back.

I suppose with a system as fully integrated as the twice as powerful MGUK, with an energy store 10x as great, it's a bit out of the drivers hands now.

#69 Guest_4L3X_*

Guest_4L3X_*

Posted 03 February 2014 - 20:31

A secondary ECU just for braking? Then why not a piggy-back ECU for traction control from throttle input. One that looks at the wheel speed, and adjusts the drive by wire throttle (they've been that way for years) to lift off at super human speeds when the wheels break loose?

What you're saying isn't unique to the new brake by wire system. There's a reason everything has to go through the McLaren ECU, and such active systems are banned.

Unless McLaren are idiots, the brake by wire system is going to be no more exploitable than the throttle by wire system they have already had and governed perfectly well.

If they're going to have piggy-back electronics for brakes, they may as well have it for throttle too. There's literally nothing special about the brake by wire system in that respect.

It's also imperative for Formula 1 is to stay current or even lead the way. It's all about integrated Power Units and integrated braking systems now. It's the way it's going to go. Brake discs and pads are wasted energy. Hybrid shouldn't be a word only associated with dull and dreary Hondas, and putting smiles on the faces of people that love a nice strong celery soup to go with a glass of fair trade tomato juice.
Racing can get in on it and use it to build faster cars than they otherwise could too. Even without the Energy Store and, the continuous MGU-H to MGU-K feed is a brilliant alternative to a waste-gate. The runway feedback loop has not just been restrained, but fully harnessed.

To call it fake or turning the drivers into operators instead of "real" drivers (not that you did) is to say they should also do away with fuel injection and computerised ignition and injector duty maps, and just have various dials to twist depending on how much throttle they have on and what gear they're in.

 

Well, thanks for taking the time to explain these things.  I do agree it's a good thing in theory ("by wire" control) I just don't like the way it's implemented particularly in F1.

 

I'm also aware that the spirit of the regulations is to avoid having any secondary ECU to control the engine.  What I'm wondering about is a way to circumvent the LETTER of the regulations to gain an advantage - which is cool with me.  One way would be to have a secondary ECU in the pedal brake (or a valve controlling the brake fluid line diameter/flux whatever).   I think it's possible to have a very small Ecu in the pedal brake assembly itself if need be, with no input from the ECU or remote sensors if it needs to be to get around regulations.  Imagine the computer power inside an iPhone (have you seen an open one, without the batteries?  Or Google Glass?  It's nothing.  It could fit one say behind the rubber pads in the brake pedals, at least the brains alone.  You could fit a gyroscope and accelerometer in there, as these devices do and get speed, inclination, temperature and a map of each circuit (I know, totally illegal, but remember, Pat Sysmonds, the Briatore accomplice, is back, so we're not talking about sheeps here).  And some essential variables could be created to control the pedal/valve/distribution of the brake itself, to avoid lock up or tire wear, or whatever they can do with the brakes that feel the need.  In theory.  That's all I'm saying.

It's highly unlikely at this point and in retrospect I regret wasting people's time with speculations.  Still if an F1 engineer is not thinking about getting around the ECU to do stuff, while still looking to avoid breaking the letter of the regulations, is not doing it's job.  Cheers.


Edited by 4L3X, 03 February 2014 - 20:35.


#70 RoutariEnjinu

RoutariEnjinu
  • Member

  • 2,442 posts
  • Joined: March 09

Posted 03 February 2014 - 20:42

Well, thanks for taking the time to explain these things.  I do agree it's a good thing in theory ("by wire" control) I just don't like the way it's implemented particularly in F1.
 
I'm also aware that the spirit of the regulations is to avoid having any secondary ECU to control the engine.  What I'm wondering about is a way to circumvent the LETTER of the regulations to gain an advantage - which is cool with me.  One way would be to have a secondary ECU in the pedal brake (or a valve controlling the brake fluid line diameter/flux whatever).   I think it's possible to have a very small Ecu in the pedal brake assembly itself if need be, with no input from the ECU or remote sensors if it needs to be to get around regulations.  Imagine the computer power inside an iPhone (have you seen an open one, without the batteries?  Or Google Glass?  It's nothing.  It could fit one say behind the rubber pads in the brake pedals, at least the brains alone.  You could fit a gyroscope and accelerometer in there, as these devices do and get speed, inclination, temperature and a map of each circuit (I know, totally illegal, but remember, Pat Sysmonds, the Briatore accomplice, is back, so we're not talking about sheeps here).  And some essential variables could be created to control the pedal/valve/distribution of the brake itself, to avoid lock up or tire wear, or whatever they can do with the brakes that feel the need.  In theory.  That's all I'm saying.
It's highly unlikely at this point and in retrospect I regret wasting people's time with speculations.  Still if an F1 engineer is not thinking about getting around the ECU to do stuff, while still looking to avoid breaking the letter of the regulations, is not doing it's job.  Cheers.


You're not wasting anyone's time at all. This thread is exactly the place to talk about it all. I learned some things from TC3000 too.

As for your miniature ECU, it would be perfectly possible, but as I said, it would be nothing new to this year, and such a system could already have been used in 2009, but hidden in a throttle pedal, to adjust the signal coming out of the potentiometer to give traction control. The trace the ECU would see, would be of exquisite throttle control operating at superhuman speeds, but as far as it would be concerned, it would be coming from the pedal.

It could look at the signal coming from a drivers foot at 100%, work out from its various independent sensors that this is too much, and adjust the voltage down (assuming it's analog) to actually send 85% throttle back to the ECU/Throttle-body actuator.

To put it another way, it's exactly how some cruise control systems work. They literally sit as an intermediary between the throttle potentiometer and the throttle body, manipulating the signal that otherwise comes from the throttle pedal assembly.

By the way, what you described is actually exactly how some of these "chips" you find on eBay work for TDI engines. They're literally a little ECU that piggy-backs the signal coming from one of the sensors, before adding its own value to it, and then sending it to the main untouched ECU, which is none the wiser to it. It might tell it fractionally more air passed through the air mass meter than really did, so that on an otherwise lean running engine, it runs a little richer. On my own car, it bleeds air from a vacuum pipe connected to my turbo with a special injector, to control how the waste gate works, and allow it to run higher pressures, which the standard ECU mapping can account for.

So if you wanted to cheat in F1, you could do it at that level maybe. But you'd way sooner be doing it to the throttle pedal than the brake pedal. Something that has been by-wire since the mandatory FIA regulated ECU's came in.

Edited by RoutariEnjinu, 03 February 2014 - 20:48.


#71 saudoso

saudoso
  • Member

  • 6,776 posts
  • Joined: March 04

Posted 03 February 2014 - 21:19

Just as an general comment in regards to the ABS/TC discussion.

 

Technically there is a difference between preventing something and "controlling" something.

 

You can prevent wheel lockup (for example) by not supplying enough brake pressure (in the extreme case, if you never brake, you will never lock up the wheel)

This doesn't mean, that you have any "control" of the situation, if it does occur, despite your best efforts to prevent it.

 

The main difference is, that an ABS, in this case, can actively do something, if the wheel does lock-up - reducing brake pressure, to get it rolling again.

 

 

I have seen and engaged in lots of ABS/TC discussions during the past years and had never seen someone put it so simply and clearly(while spot on). Attaboy.



#72 SpeedRacer`

SpeedRacer`
  • Member

  • 1,423 posts
  • Joined: October 08

Posted 03 February 2014 - 21:54

Does this mean there will be far less pressure required by the driver on the brake pedal?



#73 Clatter

Clatter
  • Member

  • 44,648 posts
  • Joined: February 00

Posted 03 February 2014 - 21:55

Does this mean there will be far less pressure required by the driver on the brake pedal?

No. 



#74 CoolBreeze

CoolBreeze
  • Member

  • 2,450 posts
  • Joined: January 12

Posted 04 February 2014 - 01:44

And the rain light will also function as a brake light.

 

Source? first time i'm hearing about this actually. 

 

Back in 2002, Ferrari did run brake lights testing at Sepang. It didn't really work, or make any differences. The guys just get used to the braking points. 



#75 scg92

scg92
  • Member

  • 44 posts
  • Joined: October 10

Posted 04 February 2014 - 07:22

You wrote some great posts RE - thank you -  :up: 
But let me "pick" on this one for a moment.
This new system still uses the same brake hardware and all what comes with it. It's nothing magic, you can best think of it as an adjustable proportioning valve.
In the case of an electronic failure, it is required to fail in a way, that all the brake pressure coming from the rear master cylinder passes through the system/valve an goes to the rear calipers, where it will brake/****** the rear wheels, just as any other brake system.

If the MGUK (KERS) now harvests energy from the rear axle, it will reduce the brake pressure which goes to the rear wheel/axle calipers. Ideally in a way, that the net braking torque and thereby the force at the contact patch of the tyre remains the same. In the ideal world the tyre, which is the part that matters in this context, doesn't know where the braking torque of the axle comes from. It doesn't care if it is generated by the brakes or by the MGUK. But if the transition from one to the other causes a change in net torque, the tyre will notice it.

And this is the area where the real challenge for the control and other engineers lays ( the driver is "out of the loop", he will only bitch if it doesn't work as he likes it - just kidding, but that's what drivers normally do - all of them). The easy part is to calculate how much pressure you want/need at any given time to compensate for the MGUK torque (or engine braking in conventional cars, remember that in lower form of motorsport, people can & will lock up rear wheels during down shifting, if the don't match the engine rpm to the velocity of the car correctly - in a nutshell we talk about the same effect, just on a different scale), but one "issue" is that both system use a different "path" to transmit the torque. -  (in control engineering terms you could say, both systems have a different transfer function).
The MGUK has the whole gearbox (inertia effects and ratio change during downshifts), the differential and the drive shafts (torsional windup) between itself and the rear wheels, while the brake system behind this new electronic valve, has to deal with, as you say, compressibility of the brake fluid, expansion of the brake lines & calipers etc. etc..
To make a long story short, both circuits have, very likely, a different dynamic response and there own "delay times" (time constants).
Now, you need to match, this two dynamic responses in a way, that you achieve no change in net braking force at the contact patch - that's no small feast, and a real challenge from a control engineer perspective, because the transfer function of both systems will/can change with things like temperature or other parameters (rotational speed, for inertia for example).

To make things slightly more interesting/challenging, you have secondary effects which have an effect on the contact patch load of the tyre. If you pick up an book on suspension design &/or vehicle dynamics, you will see, that there is a difference between inboard and outboard mounted brakes. So on top of all what was said above, you may deal with the different torque reaction effects which comes with them. Anti-squat/lift settings of the suspension play a role in this too.

I'm not surprised, that the drivers comment on it, but I have a lot of sympathy for the engineers who have to deal with this too, because it's far from easy.
Racing in the rain or on a damp track with changing grip levels, will become a greater challenge - IMHO - as it already is.

 

                                                                                                                         ***************

 

for the forum members, who want to visualize this is simple terms, maybe imagine a "tug of war" game, where two kids, pull on an "solid" object in opposing directions.

as long as both pull with equal force, the object won't move. If one pulls harder then the other, the object moves in the direction of the kid who pulls harder.

Now imagine the same game, but one kid is not pulling with an rope, but with an rope who is made from rubber (like a bungee cord / spring). 

To keep the object centered between the two kids, you still need the same force at the object at the same time, but now the kid with the rubber/bungee cord would move  more, because it needs to account for the "stiffness" of the bungee cord/ the spring, which transfers the force, with a delay ( you need to expand it a certain distance F= k/s

the force [F] is equal to the spring stiffness [k] divided by the distance [s]). So this rope/path has a different "transfer function" then the other rope/path.

The job of the controller would be, to position both kids at the same time so, that the net force on the object remains the same - the object doesn't move.

 

Translated to the F1, this would mean if the object doesn't move, you have "perfect brake balance", if it does move, depending from the direction, you either lock up the rear wheels, or you don't have enough overall retardation/braking, which may let's you overshoot your turn in point, or leads to (more) understeer etc. (whatever comes with a too high entry velocity).

As closer to the overall limit you are, as more serve the consequences. If you only brake at 80% of the max. potential, a bit more brake torque may don't lock up the wheel, just changing the balance a bit. If 3, 4 , 5 (or whatever) higher entry speed, let you still make the corner, no dramas, may just run a bit wide, lose some time, but if you are right at the "edge", the consequences can/will be more serve - interesting times ahead, at least early on, until people get the hang on it.

I'd just like to commend you on a fantastic post. The way you explained all the variables was great, and really put a lot of my own thoughts (as a final year mechanical engineering student) into text. :up:



#76 Clatter

Clatter
  • Member

  • 44,648 posts
  • Joined: February 00

Posted 04 February 2014 - 09:34

 

Quote

 

BBC F1 commentary box producer Tony Dodgins: 

 

"Here in testing, the cars have various warning lights on them. All have a flashing red light on the back of the car. This is a torque-reduction warning light. Basically, in the past, before a driver lifted off, he'd check his mirrors (well, usually) but with the torque distributor deciding where it is going to get the power from (engine, Ers, etc) various fuel-saving strategies could mean a car slowing down, say on the straight, other than simply via a throttle/torque map.

"Therefore, they are experimenting with warning lights to warn following cars whenever there is a torque reduction. 

 

So the light will come on if the driver lifts? That will make for some interesting brake testing arguments.



#77 GreenMachine

GreenMachine
  • Member

  • 2,635 posts
  • Joined: March 04

Posted 04 February 2014 - 11:03

There is a video on another thread that shows the lights in action - I will see if I can find it.



#78 Tenmantaylor

Tenmantaylor
  • Member

  • 18,050 posts
  • Joined: July 01

Posted 04 February 2014 - 12:52

Even if lifting from 100% to 50% throttle? (Sometimes all that is required to "brake" test someone). I thought harvesting would only take place at 0% throttle position? Otherwise the first part of the throttle is becoming a combined accelerator/brake pedal.

 

I'm getting a bit confused now. Before any energy recovery systems were used in F1 engine braking was managed (ie reduced) by supplying an 'over run' of fuel to the engine when off the throttle depending on engine speed (the over run was set by the engine speed as higher revs = more drag). This helped balance and stabilise the car when coasting and braking, getting the amount right was very important, especially in fast corners like Maggots and Becketts where the off throttle balance of the car is key to lap time.

 

I imagine to save fuel the teams are trying to run as little engine braking/over run as possible. Has engine braking/over run been replaced entirely by the electronic systems? If so this would require the car to be consuming electrical energy at 0% throttle rather than harvesting (unless the throttle pedal is indeed turning into a combined throttle/brake).


Edited by Tenmantaylor, 04 February 2014 - 12:56.


#79 thiscocks

thiscocks
  • Member

  • 1,489 posts
  • Joined: October 07

Posted 04 February 2014 - 12:52

If the engineers (all round) have done their job right, the drivers won't notice the difference.

 

I think "tried and tested" is what's made Formula 1 less interesting. I wonder if there were concerns when the throttle control was changed from a cable to a potentiometer. What would happen if it got stuck open?

 

I bet early hydraulic brakes failed. If these electronic ones do, they still have pressure to the front wheels, and the natural drag of the engine at the rear. They won't stop competitively, but they'll still stop hard.

 

In the mean time, pads cook, disks explode, tyres burst and sometimes wheels fall off. I don't think it's worth worrying about. Pirelli last year were more of a danger to braking power than this system will be.

They did (hoses poping off ect..) and the saftey back up was a ditch for the driver to sit in!  As mentioned though if this electronic system fails it will just revert to hydraulic so not an issue in saftey terms.



Advertisement

#80 GreenMachine

GreenMachine
  • Member

  • 2,635 posts
  • Joined: March 04

Posted 06 February 2014 - 01:13

There is a video on another thread that shows the lights in action - I will see if I can find it.




#81 Timstr11

Timstr11
  • Member

  • 11,162 posts
  • Joined: May 02

Posted 06 February 2014 - 10:51

Paddy Lowe explains Brake-by-Wire:

 



#82 RoutariEnjinu

RoutariEnjinu
  • Member

  • 2,442 posts
  • Joined: March 09

Posted 06 February 2014 - 11:33

A very abstracted description of the MGUH and Brake-by-Wire. I know you need to make these things accessible to everyone, but sometimes I wish they went into a litle more detail.

Had this been my introduction to the MGUH, I'd be left wondering just how it extracted energy from a turbo.

Also, from what Paddy was describing, it's as if the rear brakes are isolated from the pedal itself, in how he described the feel. Or I wasn't paying attention.

#83 RoutariEnjinu

RoutariEnjinu
  • Member

  • 2,442 posts
  • Joined: March 09

Posted 06 February 2014 - 13:43

From how Paddy was describing it, there may be more to it than what I originally thought of it just adjusting the brake bias on the fly, as you would do so with the lever/dial in the cockpit.

But then it got me thinking that maybe you couldn't adjust the bias on the fly, and it would need to be set before any pressure is on the system? Which made me think that if you wanted to reduce rear braking effort, you'd need to bleed it back around wherever the bias differential is happening? Which seems to be what you describe with ABS.

Does anyone know how brake bias systems work, and whether, under high effort braking, turning the dials one way and then the other would affect the pressure already in the calipers? Could you even physically turn it?

I really know nothing about the mechanics of a traditional brake bias device.

Edited by RoutariEnjinu, 06 February 2014 - 13:43.


#84 Lee Nicolle

Lee Nicolle
  • Member

  • 11,038 posts
  • Joined: July 08

Posted 06 February 2014 - 22:39

I'm sure drivers won't like this system initially due to the difference in feel and settings but I think it's great that F1 is pioneering this technology. Advancements in this area are as important as the energy output side of the system. I imagine the options available to the engineers these days should ABS and TC become legal again would be mind boggling compared back to when they were perfected in the 90s.

But why are they trialling new technology? Because of some stupid electric gizmo they have to make the brakes work again! And electronics will never get the pedal feel that a good driver needs. Maybe they should just drive by wire! They can sit in the pit bays and drive from there with video screens for forward and rear vision. But they have them, they are called simulators!

#85 Lee Nicolle

Lee Nicolle
  • Member

  • 11,038 posts
  • Joined: July 08

Posted 06 February 2014 - 22:50

There still has to be a hydraulic link to the rear callipers. I think Clatter means any electronic regulation of its pressure has to be controlled through the McLaren ECU. Just as with throttle input. It's really not any different. It just feels it as traditionally braking systems have always been mechanical. Throttles used to be too. And clutches, in fact.


As long as the throttle pedal actuation directly relates to torque at the driving wheels, with too much causing wheel-spin, and the brake pedal actuation directly relates to retardation of the wheels, with too much causing a lock-up, it doesn't matter if the Power Unit is a combination of ICE, electronic, nuclear or hamsters in wheels all being added together in whatever proportions, or whether the braking is a combination of friction, electricity generation, or a giant shoe coming out and rubbing against the tyre, all being added together in whatever proportions.

As long as those pedal movements directly relate to torque or friction, it doesn't really matter.

Most race cvars of the last decades have a form of traction control. Without wheel speed moniters. Ideally the ECU is programmed with them in testing then removed to race. But the ECU limits throttle if the engine 'picks up' too quickly in the lower gears. The one time I drove this type of throttle I did not like it but would get used to it. And it was not my car, I was just doing some testing. This was 20 years ago in a lower tintop category so it would have been very crude to what they would be using these days.

#86 Clatter

Clatter
  • Member

  • 44,648 posts
  • Joined: February 00

Posted 10 February 2014 - 08:59

In the above quote from Massa, he pinpoints the new brake system as one of the main "work in progress" areas. (the full interview is on the AMuS website) - the highlighted part reads loosely translated:

 

The largest area of work, is the new braking system (brake by wire), which isn't just adjusting the brake balance automatically, but is also responsible to recharge the ES/batteries with the energy collected via MGU-K.

 

"It's not easy, to develop 100% confidence in the brakes"

 

In the last part he says, that at the moment the power/torque still develops a bit to quickly/abruptly (at the exit of corners), which leads to many sideways moments  - the rear end stepping out, when the driver applies the throttle (pedal)

 

"We will see this quite often this season" - says Massa

That's sounds like a good thing to me, as a spectator. He isn't suggesting the electronics should be stopping it is he?



#87 Timstr11

Timstr11
  • Member

  • 11,162 posts
  • Joined: May 02

Posted 10 February 2014 - 10:07

That's sounds like a good thing to me, as a spectator. He isn't suggesting the electronics should be stopping it is he?

But you're happy that the electronics are helping to cause the lack of confidence?

 

A driver will always want predictable behavior. How else is he to know when the demand varies between recovery systems and brakes?


Edited by Timstr11, 10 February 2014 - 10:11.


#88 mariner

mariner
  • Member

  • 2,329 posts
  • Joined: January 07

Posted 10 February 2014 - 10:45

Great explanation byt TC3000.

 

 

He makes a very good point that for many, many years drivers had to allow for sharp changes in engine braking as they went down through the gears manually. Lots of fast processing took place in their brains and feet. espeically true pre carbon brakes and in long distance racing. They managed just fine so I think the top drivers wil cope fine with electronic systems which aren't perfect at first.

 

Also all this software isnt new. the Prius and electric trains have been doing this, albeit at slower response rates,for some time.



#89 Clatter

Clatter
  • Member

  • 44,648 posts
  • Joined: February 00

Posted 10 February 2014 - 10:50

But you're happy that the electronics are helping to cause the lack of confidence?

 

A driver will always want predictable behavior. How else is he to know when the demand varies between recovery systems and brakes?

Different issue as far as I'm concerned, but I'm referring to the line about corner exit. I want to see the drivers having to use their skill to control the car and not some form of TC.



#90 Clatter

Clatter
  • Member

  • 44,648 posts
  • Joined: February 00

Posted 10 February 2014 - 13:45

As for the last bit of the quote.

He expresses the the hope, that it becomes better over time, when the engineers have fine tuned the mapping a bit more.

 

He says: "Obviously, the engineers must still work on the engine [power unit] mapping, because at the moment, the torque comes in too abruptly, which leads to many sideways moments" - we will see this quite often this season, he says

 

So he would like, that the engineers, tune this a bit more, so that the torque develops more progressively [more gentle] when going onto the throttle at the exit of the corners - making it easier to drive/put the power down.

 

As for the brakes, I think this indicates, that the response from the system is still not consistent, which is hardly surprising, seeing that this was the first time the cars run in anger.

The drivers would want somethings, which is predictable, and even if not "perfect" can normally deal with a system which, while not perfect, reacts in a consistent manner (let's say a slight time delay for example). If it does the same things all the time, the drivers will take it into account, and just adapt their inputs to cope with it. It becomes a "problem" if the response from the system (any system) is not consistent, when under some conditions it does this, but then next time round it does that.

 

If they can't predict the response they are going to get, they will need to leave themselves a bit more of a margin, in case the reaction from the system is not what they expected. To really "push" a car hard, they need to have confidence in the reaction of the car, to their inputs.

If the reaction of the car, to a given input is inconsistent  (being it brakes, steering, throttle or whatever), they will be hesitate a bit, to really going for it, balancing the car at the edge of adhesion from the tyres.

 

They will give themselves a bit more of a margin, in case the reaction of the car isn't what they wanted/expected. A bit like driving in changeable grip conditions, when you never can be quite sure, how much grip you will have, this time round, before choosing your barking point.

 

For now, it seems, driving these cars, isn't as easy/predictable as they [the drivers] would like, but with more miles under their belt, they will get used to it, as will the engineers - IMHO

Cheers for the translation.

 

I thought that might be the case. There's a lot of work for the teams do, but hopefully it will lead to some exciting times on track until they get it sorted.



#91 vista

vista
  • Member

  • 1,343 posts
  • Joined: April 13

Posted 24 February 2014 - 17:28

Lorenzo de Luca thinks brake by wire can be explored in such a way it could provide benefits similar to traction control and ABS. 

 

http://thejudge13.co...nic-revolution/

 

But how can it be used to emulate TC effects? Surely, it only works in braking phases - and not out of corners where traction is required..

 

Could it be that the system automatically brakes a little bit when drivers are wheel spinning out of corners, thereby controlling the torque so the car is has more driveability?



#92 pingu666

pingu666
  • Member

  • 9,272 posts
  • Joined: October 07

Posted 13 March 2014 - 23:17

i guess the feel will be different, as they did used to have direct hydrolic connection before, and i think kers harvesting feels different to normal brakes too, so retardation might be similer but it might feel different, and if system is changing harvesting alot during braking phase thats going to be noticeable on some level



#93 chipmcdonald

chipmcdonald
  • Member

  • 1,824 posts
  • Joined: November 06

Posted 15 March 2014 - 01:18

I don't think this will ever be sorted.  They're adding extra variables into the braking equation that the software can't possibly anticipate, and in turn the driver can't anticipate. 

 

I think we're going to see a lot of "unexplained" offs in braking zones.  The software would have to either be designed to be very, very compliantly conservative - therefore inefficient in harvesting, or it would have to be dangerously unconservative, risking ruining the battery or motor, in exchange for being consistent in behavior.  I don't imagine engineers taking the later path, or the first one.  

 

You can't have brake feel when the "brakes" also have to incorporate the non-linearities of charging a battery under variable conditions, with variable loading.  It's complicating something that is already non-linear and variable.  You can make non-linear curves dynamically integrated on the fly.   It's also one less skill the driver is having to hand off to "technology".

 

I'm sure there can also be cascade situations from a data point of view, when too many things are happening at once and they probably have the fly by wire system default to some neutral bias when that occurs.  Probably very unfun feeling to the drivers.

 

/ $.10



#94 Lee Nicolle

Lee Nicolle
  • Member

  • 11,038 posts
  • Joined: July 08

Posted 15 March 2014 - 05:03

Great explanation byt TC3000.

 

 

He makes a very good point that for many, many years drivers had to allow for sharp changes in engine braking as they went down through the gears manually. Lots of fast processing took place in their brains and feet. espeically true pre carbon brakes and in long distance racing. They managed just fine so I think the top drivers wil cope fine with electronic systems which aren't perfect at first.

 

Also all this software isnt new. the Prius and electric trains have been doing this, albeit at slower response rates,for some time.

Prius and trains are mechanised transport. AFAIK Prius has standard hydraulic brakes anyway. 

All this fancy unreliable poor to drive stuff is not motorracing. The silly electric engines are bad enough without electric brakes too. All the waffle about 'codes' to make the brakes feel right, work right is just so silly. I think F1 has shot itself in the foot. Cars that move around a lot are good, cars that are unreliable because of pointless electronics, yet alone then become an economy run in a race are boring. Maybe they should ask Greenpeace for sponsorship!