2021 Engine thread

[AJI]

A formula car does not have awd, we should keep that for another 10 years at least.

Why not?

[NL_Fer]

I’d rather see f1 cars behave more like he uncontrolable beasts of the 80’s 90’s, instead of rolling like trains on a track, like they do now.

Laggy engines with limited rear traction, powersliding out of corners.

[Zynerji]

Pat Symonds give an interview in 2009 where he clearly stated that the problem is that they cannot give the driver more engine power than the chassis can handle, like they could in the 90s. He literally stated that the problem was that they could build too good of a chassis to ever bring back that level of driver making the difference.

[OO7]

Pat Symonds give an interview in 2009 where he clearly stated that the problem is that they cannot give the driver more engine power than the chassis can handle, like they could in the 90s. He literally stated that the problem was that they could build too good of a chassis to ever bring back that level of driver making the difference.

I'm not sure I agree with that Zynerji. One of the issues is that the driver is not in control of the engine/PU, the ECU is. The driver demands torque via the pedal and Hal (who resides in the ECU) performs all the voting magic to give the driver exactly what he or she requested. If the driver no longer demands torque and instead the throttle pedal was precisely that (it directly controls the amount of air entering the engine), then I believe things would be different.

[roon]

https://i.imgur.com/RVCklty.jpg

Dont see the need for it to be at the T tray, MGU K can just hang below the front bulk head.

Was thinking at the time that an angled driveshaft, pretty much following the profile of the existing noses, would be less obstructive to the airflow volume under the nose. Then again, the motor can be pretty tiny depending on gearing and construction, so you may be right.

[Zynerji]

Pat Symonds give an interview in 2009 where he clearly stated that the problem is that they cannot give the driver more engine power than the chassis can handle, like they could in the 90s. He literally stated that the problem was that they could build too good of a chassis to ever bring back that level of driver making the difference.

I'm not sure I agree with that Zynerji. One of the issues is that the driver is not in control of the engine/PU, the ECU is. The driver demands torque via the pedal and Hal (who resides in the ECU) performs all the voting magic to give the driver exactly what he or she requested. If the driver no longer demands torque and instead the throttle pedal was precisely that (it directly controls the amount of air entering the engine), then I believe things would be different.

If you gave them 1800HP with turbo lag maybe. Hal is the optimizer, not the authority.

[OO7]

Pat Symonds give an interview in 2009 where he clearly stated that the problem is that they cannot give the driver more engine power than the chassis can handle, like they could in the 90s. He literally stated that the problem was that they could build too good of a chassis to ever bring back that level of driver making the difference.

I'm not sure I agree with that Zynerji. One of the issues is that the driver is not in control of the engine/PU, the ECU is. The driver demands torque via the pedal and Hal (who resides in the ECU) performs all the voting magic to give the driver exactly what he or she requested. If the driver no longer demands torque and instead the throttle pedal was precisely that (it directly controls the amount of air entering the engine), then I believe things would be different.

If you gave them 1800HP with turbo lag maybe. Hal is the optimizer, not the authority.

Hal is both. There are times where a drivers request will not be honoured e.g rev limit and other protection modes. In any event optimisation is hugely influential in the characteristics of the engine. There are also open-loop wheel spin-up protection maps, as well as open loop rear wheel ABS.

[Zynerji]

Hal is the optimizers that delivers the torque that the driver demands. It cannot drive by itself, and the drivers make dozens of adjustments during the race, so it is not all algorithms.

Also, I would state that the last 2 items that you mention are illegal, and doubtful that it is a fact. I agree that MGU strategies can effectively give traction control (through optimization), but individual wheel ABS is specifically banned.

[Tommy Cookers]

the last two items cannot be banned entirely
the genie can't be put back in the bottle - 'they' can control only the size of the genie (ie to 120 kW and 200 Nm equiv)

if you disagree please explain how they can be banned
what steady change response and dynamic response to be specified and enforced - by whom, when, and how ?

banning would need to prevent any inherent dynamic characteristic changing torque with change in load (grip)
even a spec system would have such a characteristic
any stable 120 kW electric motor setup can be called a weakish open loop wheel spinup map - and weakish ABS emulator

[Zynerji]

the last two items cannot be banned entirely
the genie can't be put back in the bottle - 'they' can control only the size of the genie (to 120 kW and 200 Nm equiv)

if you disagree please explain how they can be banned
what steady change response and dynamic response to be specified and enforced - by whom, when, and how ?

banning would need to prevent any inherent dynamic characteristic changing torque with change in load (grip)
even a spec system would have such a characteristic

Yes. I agree with all of that, just like nothing is "fully rigid" even tho specified in that manner.

I guess you could have a diff/MGUK/H/throttle strategy to have individual rear ABS maps and traction control. Watching drivers today losing the rear under braking and Raikkonen throwing away pole under acceleration kind of points to these items being small ripple smoothers more than AI control, however.

[OO7]

My comments weren't even about the MGU side, just the ICE. The open-loop ICE controls mentioned are all legal.

In terms of the ERS perhaps the following may be possible?:
The torque map of the ICE alone is measured and stored. The MGU torque map must then mirror the ICE torque map. A 3% increase in ICE torque would also translate to a 3% increase in MGU torque.

[Tommy Cookers]

the MGU-K torque map or equivalent will be similar (for motoring) to the ICE torque map
the actual map required is a PU map

but these maps define steady state behaviour not the various dynamic MG behaviours under varying load

the Porsche has an induction MG - and a synchronous MG can be designed around asynchronous behaviour
both these and other approaches will passively produce a convenient torque collapse under rapidly varying load
whilst having a steady state map showing no torque collapse

[Zynerji]

the MGU-K torque map or equivalent will be similar (for motoring) to the ICE torque map
the actual map required is a PU map

but these maps define steady state behaviour not the various dynamic MG behaviours under varying load

the Porsche has an induction MG - and a synchronous MG can be designed around asynchronous behaviour
both these and other approaches will passively produce a convenient torque collapse under rapidly varying load

That actually sounds like it would be very Road relevant technology. I wish they would add a front mgu and keep the H on the turbo.

[OO7]

the MGU-K torque map or equivalent will be similar (for motoring) to the ICE torque map
the actual map required is a PU map

but these maps define steady state behaviour not the various dynamic MG behaviours under varying load

the Porsche has an induction MG - and a synchronous MG can be designed around asynchronous behaviour
both these and other approaches will passively produce a convenient torque collapse under rapidly varying load
whilst having a steady state map showing no torque collapse

Can't these behaviours be monitored?

[Zynerji]

the MGU-K torque map or equivalent will be similar (for motoring) to the ICE torque map
the actual map required is a PU map

but these maps define steady state behaviour not the various dynamic MG behaviours under varying load

the Porsche has an induction MG - and a synchronous MG can be designed around asynchronous behaviour
both these and other approaches will passively produce a convenient torque collapse under rapidly varying load
whilst having a steady state map showing no torque collapse

Can't these behaviours be monitored?

Sure, but isn't that limiting operating modes of hardware that passes legal inspection? Are you specifically suggesting that you don't allow the engineers to optimize the machines that they spend hundreds of millions to build?