Jump to content


Photo

How does Standard ECU benefit McLaren?


  • Please log in to reply
272 replies to this topic

#251 ATM_Andy

ATM_Andy
  • Member

  • 1,482 posts
  • Joined: October 07

Posted 17 January 2008 - 14:12

Originally posted by rodlamas


Ok, you cannot prevent the wheels from spinning, but you can prevent the driver from excessively pressing the throttle, right?

So you map the throttle according to lots of parameters... And then whatever the driver does to the throttle is gonna be modulated by a certain limit, which the driver can set on his steering wheel. Ferrari and the other teams already have this knob on their 2008 versions.

It's not an engineering problem anymore, it's a legal problem. The engineer says "Ok, the rules says we cannot control the spinning wheels in function of overthrotlleling, but I can minimize the amount of over-throttle. Ok, if there's still some, there will be minimal wheelspin" That's why I expect no wheels spinning at all, or to say the most, just a tiny bit. But we won't see any smoking starts.


Compensating for excessive throttle demand by the driver is NOT allowed i'm afraid.

What you can do is adjust the sensitivity of the throttle pedal, as long as the throttle pedal travel positions correspond to the engine throttle minimum (idle) and maximum (open) positions.

What is not allowed is a throttle pedal which has specific points along the travel that can be identified by the driver to assist him, for example a clicker or stepped pedal.

Advertisement

#252 rodlamas

rodlamas
  • Member

  • 8,010 posts
  • Joined: February 04

Posted 17 January 2008 - 15:35

Originally posted by ATM_Andy


Compensating for excessive throttle demand by the driver is NOT allowed i'm afraid.

What you can do is adjust the sensitivity of the throttle pedal, as long as the throttle pedal travel positions correspond to the engine throttle minimum (idle) and maximum (open) positions.

What is not allowed is a throttle pedal which has specific points along the travel that can be identified by the driver to assist him, for example a clicker or stepped pedal.


A computer isn't 100% and won't react in absolut real time. That's why spins were seen with TC like Kimi's at Monza 2005.

When you say "compensating for excessive throttle demand" that's what a TC device does. That's not allowed as you correctly say.

But, for instance, imagine the throttle as box which has one input and one output. You can't limit its output, as this what a TC is all about. But if you can make the signal (electrical or mechanical) that arrives at the throttle acuation mechanism (i.e., at its input) you're not actually controlle excessive throttle demand. You're controlling the amount of input is given for the throttle. What I mean is that you pre-filter the electrical signal the throttle comands the driveshafts is not illegal, opposing to post-filtering the signal, which is what the TC does.

I'm a control systems engineer and that's how I would do it. Certainly the teams have more knowledge and they know how to bend the regulations as they did with the starting procedings from 2004 to 2007 when LC was officially banned.

#253 undersquare

undersquare
  • Member

  • 18,929 posts
  • Joined: November 07

Posted 17 January 2008 - 15:46

Originally posted by ATM_Andy


What one top team has done is to design and install a driver controled, very simple, adjustable, throttle linkage. An elegant solution this allows the driver to adjust the sensitivity of the throttle pedal at the starts, or even in the very slow corners, to combat wheel spin.


Is this mechanical or electronic adjustment, can you say, Andy?

As an observation, if the min and max are fixed, then it's the slope they're changing? I would have thought drivers would want above all a consistent relationship between pedal position and throttle opening.

#254 ATM_Andy

ATM_Andy
  • Member

  • 1,482 posts
  • Joined: October 07

Posted 17 January 2008 - 16:21

Originally posted by undersquare


Is this mechanical or electronic adjustment, can you say, Andy?

As an observation, if the min and max are fixed, then it's the slope they're changing? I would have thought drivers would want above all a consistent relationship between pedal position and throttle opening.


You’re correct the driver can control the rate of change or 'slope' of the throttle but not change its max/min points. This gives the driver the ability to have more sensitivity in the throttle pedal. With TC systems last year the throttle was more or less a switch, now without it the drives need to control the engine a lot more themselves rather than relying on the ECU.

As far as consistency goes, F1 drivers are pretty adept at changing attributes of the car, they already change the differential settings, revs, break balance etc etc etc.

Originally posted by rodlamas


A computer isn't 100% and won't react in absolut real time. That's why spins were seen with TC like Kimi's at Monza 2005.

When you say "compensating for excessive throttle demand" that's what a TC device does. That's not allowed as you correctly say.

But, for instance, imagine the throttle as box which has one input and one output. You can't limit its output, as this what a TC is all about. But if you can make the signal (electrical or mechanical) that arrives at the throttle acuation mechanism (i.e., at its input) you're not actually controlle excessive throttle demand. You're controlling the amount of input is given for the throttle. What I mean is that you pre-filter the electrical signal the throttle comands the driveshafts is not illegal, opposing to post-filtering the signal, which is what the TC does.

I'm a control systems engineer and that's how I would do it. Certainly the teams have more knowledge and they know how to bend the regulations as they did with the starting procedings from 2004 to 2007 when LC was officially banned.


I’m not really fully understanding you rodlamas, I’m sorry.

Basically the rules are:

All components of the engine can only be controlled by the ECU, all of the control sensors have to be FIA approved and homologated.

The ECU code doesn’t allow for filtering the throttle signal like that, you can change the parameter of the throttle however basically when the pedal is fully depressed the throttle must be fully open and when the pedal is up the throttle must be at engine idle position…

The FIA have tied it up pretty well... But as I say it's the same for everyone...

#255 rodlamas

rodlamas
  • Member

  • 8,010 posts
  • Joined: February 04

Posted 18 January 2008 - 01:16

Originally posted by ATM_Andy


I’m not really fully understanding you rodlamas, I’m sorry.

Basically the rules are:

All components of the engine can only be controlled by the ECU, all of the control sensors have to be FIA approved and homologated.

The ECU code doesn’t allow for filtering the throttle signal like that, you can change the parameter of the throttle however basically when the pedal is fully depressed the throttle must be fully open and when the pedal is up the throttle must be at engine idle position…

The FIA have tied it up pretty well... But as I say it's the same for everyone...


I think we've been talking the whole day and in the end the we share the same ****ing oppinion.

Follow the link bellow, I've made a pic with MATLAB. On the x-axis you would say you have the amount of throttle pressed and on the y-axis you have the amount of throttle open. I would like to have post 12 curves as Ferrari and Mclaren have 12 positions on their knobs for the throttle, but MATLAB offers only 7 colors and it's late on the night and I won't spend anymore time on traces, lines and small dots. :p

Throttle Maps

#256 1&Only1Massa

1&Only1Massa
  • Member

  • 143 posts
  • Joined: January 08

Posted 18 January 2008 - 01:22

Why did you rapidshare a simple jpg lol

also I thought a simple linear relation would be better? edit: oh i get it, i guess it's to take into account the peaky nature of the engines.

#257 rodlamas

rodlamas
  • Member

  • 8,010 posts
  • Joined: February 04

Posted 18 January 2008 - 02:08

Originally posted by 1&Only1Massa
Why did you rapidshare a simple jpg lol

also I thought a simple linear relation would be better? edit: oh i get it, i guess it's to take into account the peaky nature of the engines.


Actually this a throtlle input vs throttle output chart.

It means that if you are on a higher throttle mode, you will have to press the throttle a bit more to get the same amount of power into the wheels than if you were on a lower mode. But when you get closer to full throttle, this is becomes irrelevant.

It's a way of masking TC and minimizing wheelspin.

#258 Lazarus II

Lazarus II
  • Member

  • 4,527 posts
  • Joined: July 04

Posted 18 January 2008 - 03:52

Originally posted by rodlamas


Actually this a throtlle input vs throttle output chart.

It means that if you are on a higher throttle mode, you will have to press the throttle a bit more to get the same amount of power into the wheels than if you were on a lower mode. But when you get closer to full throttle, this is becomes irrelevant.

It's a way of masking TC and minimizing wheelspin.

I've a simple solution to the problem - I call it drive-by-cable throttle.

#259 peroa

peroa
  • Member

  • 8,846 posts
  • Joined: March 05

Posted 18 January 2008 - 10:56

The ECU!

Advertisement

#260 undersquare

undersquare
  • Member

  • 18,929 posts
  • Joined: November 07

Posted 18 January 2008 - 11:36

Originally posted by 1&Only1Massa
Why did you rapidshare a simple jpg lol

also I thought a simple linear relation would be better? edit: oh i get it, i guess it's to take into account the peaky nature of the engines.


Yes, normally a linear relationship is best - 1 on the control gives 1 on the output, but as you say they can't have that in terms of tractive effort because of the torque curve. But whatever the curve is, the system needs to allow the driver to predict the control movement required and make it immediately with good accuracy, not wait for feedback and make an adjustment. These are two quite different types of control action, one fast, one slow. I would be sure, if F1 teams aren't contradicting me, that it would be best for a driver to learn one pedal/throttle relationship very very thoroughly, and stick with it.

So I can see why teams are keen to optimise the input/output curve, but changing it during a lap or during a race does not make sense to me.

#261 Dom77

Dom77
  • Member

  • 264 posts
  • Joined: May 06

Posted 18 January 2008 - 11:36

Ive read in motogp articles that modifying the engine mapping eases the power thru the first few gears. Now no engine mapping on bikes at Barcelona makes you 2 seconds slower :rotfl: If the FIA think of that idea it should put DC in retirement :rotfl:

Active Differential banned this year correct as i see the Differential buttons on the ferrari steering wheel?

dom

#262 rodlamas

rodlamas
  • Member

  • 8,010 posts
  • Joined: February 04

Posted 18 January 2008 - 12:39

Originally posted by Dom77
Ive read in motogp articles that modifying the engine mapping eases the power thru the first few gears. Now no engine mapping on bikes at Barcelona makes you 2 seconds slower :rotfl: If the FIA think of that idea it should put DC in retirement :rotfl:

Active Differential banned this year correct as i see the Differential buttons on the ferrari steering wheel?

dom


Every team has still 3 diff knobs on their steering wheel... I keep wondering why...

#263 Dom77

Dom77
  • Member

  • 264 posts
  • Joined: May 06

Posted 18 January 2008 - 12:55

I know previously it was controlled by computers i dont know if 2008 its manually because of the Standard ECU? Perhaps someone can tell us :)

#264 ATM_Andy

ATM_Andy
  • Member

  • 1,482 posts
  • Joined: October 07

Posted 18 January 2008 - 13:13

Originally posted by Dom77
I know previously it was controlled by computers i dont know if 2008 its manually because of the Standard ECU? Perhaps someone can tell us :)


All components of the engine and gearbox, including clutch, differential and all associated actuators must
be controlled by the Single Electronic Control Unit (SECU).

#265 AFCA

AFCA
  • Member

  • 6,661 posts
  • Joined: February 06

Posted 25 April 2008 - 10:51

There are still problems with the ECU. There are especially complaints about the lack of interoperability with own components. Mentioning a few of the problems of the past three races: At Honda there were problems with the speed control, Glock suffered from sudden bugs in the gearbox whereas Piquet's Renault had an entire system crash which disabled him to do anything.

Michael explains the difficulties in entering parameters: ''We have to check two or three times if the data has really been saved by the system.''

The contentious issue is the fact that the teams don't know the algorithm of the software. When MES comes up with an updated version of the ECU, the engine- and gearbox engineers need to get back to the testbenches to prevent negative effects. Theissen: ''That costs money. McLaren is fine with it. They exactly know their software.''

Marmorini: ''When we change something we not only have to go to the testbench but also test the changes on track. There's still a lot to learn to fully optimise the standard electronics.''

On the costs: ''With the engine freeze and the standard electronics it seems as if no development is necessary any longer at all. But the costs are enormous and the development is comparable with that of a new engine.''

#266 Gecko

Gecko
  • Member

  • 800 posts
  • Joined: April 07

Posted 25 April 2008 - 11:08

Originally posted by AFCA
At Honda there were problems with the speed control, Glock suffered from sudden bugs in the gearbox whereas Piquet's Renault had an entire system crash which disabled him to do anything.

Michael explains the difficulties in entering parameters: ''We have to check two or three times if the data has really been saved by the system.''


All of this sounds more of a typical Microsoft than McLaren issue to me ;)

#267 ATM_Andy

ATM_Andy
  • Member

  • 1,482 posts
  • Joined: October 07

Posted 25 April 2008 - 11:32

Originally posted by Gecko


All of this sounds more of a typical Microsoft than McLaren issue to me ;)


Not really no, McLaren ES write/change/modify the software. From their point of view however it's not an easy job as they have to build in requirements for all the teams...

I dont quite agree with Luca on the cost issue, it is certainly not cheep, however new engine development is significantly more expensive, in my experience anyway.

#268 CaptnMark

CaptnMark
  • Member

  • 1,011 posts
  • Joined: March 99

Posted 25 April 2008 - 12:17

It's surprising that teams agreed to use ECU where they don't have the source code. How stupid of them.

#269 Buttoneer

Buttoneer
  • RC Forum Admin

  • 16,252 posts
  • Joined: May 04

Posted 25 April 2008 - 12:57

Ferrari seems to be doing OK with it.

#270 WOOT

WOOT
  • Member

  • 429 posts
  • Joined: April 08

Posted 25 April 2008 - 13:01

I thought the code was open source for all the teams?

#271 Clatter

Clatter
  • Member

  • 27,211 posts
  • Joined: February 00

Posted 25 April 2008 - 13:35

Originally posted by CaptnMark
It's surprising that teams agreed to use ECU where they don't have the source code. How stupid of them.


They do have the source code.

#272 AFCA

AFCA
  • Member

  • 6,661 posts
  • Joined: February 06

Posted 25 April 2008 - 13:40

Originally posted by Buttoneer
Ferrari seems to be doing OK with it.


I think that after Australia they got things right. At least right enough.

#273 HP

HP
  • Member

  • 14,197 posts
  • Joined: October 99

Posted 25 April 2008 - 16:57

Originally posted by CaptnMark
It's surprising that teams agreed to use ECU where they don't have the source code. How stupid of them.

As already pointed out, they have it.

But having the code isn't that important if the interface is well defined, and the software delivers exactly what it promises.

It would IMO have been better to have an ECU from a company that has no connection with F1 teams and not releasing the source code at all.

With the code, teams might be able to find exploits.