Honda Power Unit Hardware & Software

[rscsr]

Apologies to resurrect the fuel usage discussion,

Is lift and coast the only way teams seek to conserve fuel?

Is it possible / likely that teams are running at say 90kg/hr for periods of the lap / race rather than at the full 100kg/hr?

If this is true then it would be possible to see Mercedes say running at 85kg/hr, letting their chassis advantage deliver the laptime while other teams are running at 90kg/hr + to maintain their target laptime? Knowing the constraints of the Mclaren chassis and associated drag, maybe this puts Honda into the higher bracket therefore giving a more "thirsty" engine?

If we watch Mercedes onboard from the races they are quite happy to pull from 8krpm or even 6krpm between chicanes rather than dropping it down a further gear, I dont think the power bands are anywhere near as narrow as suggested in this thread. They also tend to be very careful with throttle application in 4th and don't even rush to 100% in 5th.

Using 100kg/h and lift and coast at the end of the straight is generally speaking faster than running at a lower rate and not using lift and coast (so that you use the same amount of fuel for the straight). But those cars are grip limited for a long time. I think Bottas said in 2015 when they use full power they can spin their rear tyres until 4th gear. So it could well be that they are not using full power at those lower gears at all.

Sorry for quoting myself, but I just made a small comparison. I checked if what I said is actually the case, and I created a chart showing what I said.
The top lines show the velocity of a F1 like car, over the distance traveled. They accelerate from 100km/h, with 300-600kW and 702kg. After they used an arbitrary 6000kJ energy they lift and therefore decelerate. The black line is the reference, which uses 600kW.
The bottom lines show how much distance the different energy using schemes lose compared to the reference scheme. So for example, when the reference car traveled 1000m, the car using 300kW is about 200m behind.

What this chart clearly shows is that you don't want to use less power to avoid lift and coast. For any given distance, it is better to lift & coast, than to use less power to avoid it. Additionally it is advantageous to use more power, than to use less.
Another way of saying it is, that the crossover point is always at a distance, where both have to lift & coast. (without a mathematical proof, because acceleration with constant power is not nicely to integrate analytically in a nice form)

This chart obviously ignores draft. Additionally when you lift and coast more, you arrive at the breaking zone with a lower speed and therefore you can break later.

[wuzak]

So then the power units have a useful(efficient) powerband of between 8,700 - 12,500 best case scenario?

Similar more or less to this crude drawing I made using completely fabricated power units so no one thinks I'm quoting official figures.

http://i.imgur.com/6I8kEFb.jpg

So, Ferrari's power peaks at around 9,000rpm and they lose power when they add more fuel?

And the graph shows that Renault have the same or more power in the normally used speed range than the Ferrari.

[wuzak]

Was TC allowed in 85-90 era?

It wasn't disallowed/banned.

I am sceptical that anybody had TC in that period. Certainly in 1993 McLaren had TC, not sure about before then.

[godlameroso]

So then the power units have a useful(efficient) powerband of between 8,700 - 12,500 best case scenario?

Similar more or less to this crude drawing I made using completely fabricated power units so no one thinks I'm quoting official figures.

http://i.imgur.com/6I8kEFb.jpg

So, Ferrari's power peaks at around 9,000rpm and they lose power when they add more fuel?

And the graph shows that Renault have the same or more power in the normally used speed range than the Ferrari.

It's a crude drawing, it's not accurate, nor should it be seen as accurate. It's only meant to illustrate what the power units' powerband resembles. I probably should have shifted all the numbers to the left and added little vertical dashes to mark where I imagine 8 10 and 15k are, and the curves themselves are only a more or less guesstimate.

[gruntguru]

Does anyone think some of the 115mm pedal travel is used for other than power modulation? For example the lowest 20mm could be a region of re-gen braking. The highest 20mm could be a region of additional power with heavy use of ES energy (eg qualy mode)

[godlameroso]

Does anyone think some of the 115mm pedal travel is used for other than power modulation? For example the lowest 20mm could be a region of re-gen braking. The highest 20mm could be a region of additional power with heavy use of ES energy (eg qualy mode)

Possibly, it would be a challenge to do that and comply with the rules but these people are far smarter and more clever than myself.

5.5.3 At any given engine speed the driver torque demand map must be monotonically increasing for an increase in accelerator pedal position.
5.5.4 At any given accelerator pedal position and above 4,000rpm, the driver torque demand map must not have a gradient of less than – (minus) 0.045Nm/rpm.

[Craigy]

Was TC allowed in 85-90 era?

It wasn't disallowed/banned.
I am sceptical that anybody had TC in that period. Certainly in 1993 McLaren had TC, not sure about before then.

Offtopic for this thread, but Williams ran TC on the FW14 in '91. That was the first car with both a semi-auto box and TC in F1, as far as I know.

[Tommy Cookers]

Does anyone think some of the 115mm pedal travel is used for other than power modulation? For example the lowest 20mm could be a region of re-gen braking. The highest 20mm could be a region of additional power with heavy use of ES energy (eg qualy mode)

Possibly, it would be a challenge to do that and comply with the rules but these people are far smarter and more clever than myself.
5.5.3 At any given engine speed the driver torque demand map must be monotonically increasing for an increase in accelerator pedal position.
5.5.4 At any given accelerator pedal position and above 4,000rpm, the driver torque demand map must not have a gradient of less than – (minus) 0.045Nm/rpm.

ok apparently only 2d mapping is allowed, potentially as traction-assisting as in the N-A days

but there is also a driver-selectable wet-weather map ? (and maybe other maps for quali ?)

and the mgu-k will give an effect of continuously modulating its + and - torque with wheel load changes ie helping in incipient slip or skid
(a system that is torque-neutral to dynamic changes in load seems inconceivable - neutral dynamic stability !)
of course turbo characteristics can and fuelling/rpm characteristic will also help the driver and so parallel the N-A mapping

the demanded torque is PU torque not ICE torque ? ie it is the sum of ICE torque and MGU-K (crankshaft-equivalent) torque
driver selection allows the ICE at partial torques to drive the K machine to generation, an efficient means of filling the ERS lap-energy deployment quota
driver selection is presumably involved in other partial torque factors including cylinder cutting


so surely there's plenty of reasons for having such a long accelerator travel ?

[wuzak]

but there is also a driver-selectable wet-weather map ? (and maybe other maps for quali ?)

Nope, just a map for dry running and one for wet running.

[Craigy]

but there is also a driver-selectable wet-weather map ? (and maybe other maps for quali ?)

Nope, just a map for dry running and one for wet running.

Don't they have a start map any more?

[wuzak]

but there is also a driver-selectable wet-weather map ? (and maybe other maps for quali ?)

Nope, just a map for dry running and one for wet running.

Don't they have a start map any more?

That's an engine map, whereas the discussion was about throttle pedal mapping.

There are a number of engine maps which are driver selectable.

[Craigy]

That's an engine map, whereas the discussion was about throttle pedal mapping.
There are a number of engine maps which are driver selectable.

I'm quite surprised to learn they aren't the same thing.

I'd have a start map, where the majority of the throttle travel up to almost max throttle was well under max torque demand, so as to make wheelspin more controllable in the lower gears. Is that banned by regulation?

Can't the teams have maps which do things like regen in the lower gears (which otherwise would equate to wheelspin) in order to map around this sort of thing?

At least in 1st and 2nd, which on most tracks they simply won't use on a normal non-start lap?

[wuzak]

That's an engine map, whereas the discussion was about throttle pedal mapping.
There are a number of engine maps which are driver selectable.

I'm quite surprised to learn they aren't the same thing.

I'd have a start map, where the majority of the throttle travel up to almost max throttle was well under max torque demand, so as to make wheelspin more controllable in the lower gears. Is that banned by regulation?

Can't the teams have maps which do things like regen in the lower gears (which otherwise would equate to wheelspin) in order to map around this sort of thing?

At least in 1st and 2nd, which on most tracks they simply won't use on a normal non-start lap?

The pedal map regulates torque demand, the power unit map regulates the ICE mode, ERS modes, etc.

The reason for defining a pedal map is that with fly-by-wire throttle control the engine can be made to do things which the driver didn't ask it for. Such as the diffuser blowing in the EDB era of the V8s - the engine would open its throttles even though the driver was off the pedal.

Also, several years ago electronic power assisted steering was banned. This was because it was possible through track mapping and electronic control to give the driver some steering assistance. A similar problem existed with the throttles.

[FPV GTHO]

It was common before to also have a non linear power delivery through the pedal map. For a high speed circuit for instance, having the last 25% of pedal travel only controlling the last 10% of the throttle body opening in order to get better throttle control. You could do the same thing for the low-midrange to help wheelspin on the starts. The idea has flowed onto the aftermarket for road cars, with plug in units that can now multiply or dull the throttle inputs

[godlameroso]

That's silly, you can use traction control to much greater effect than altering the throttle response. Good throttle control is what separates great drivers from good ones. Most everyone on the field can nail any apex and get 99% out of the car, the difference is in that small space before and after the apex of each turn. To most people the difference is imperceptible, even to the drivers themselves, the difference is imperceptible only seen in the telemetry traces. There you see, oh you braked a little earlier than your team mate on such and such turn, so you lost a little speed on entry, but you carried more speed through this part, and got a better exit so you net gained .2 seconds.