F1technical.net

Tommy Cookers wrote: ↑

05 Apr 2018, 00:29

then how does it slow eg from 12200 rpm to 10500 rpm ?

there is a lot of energy stored in the PU's rotation at 12200 rpm
where does the energy go ?
the energy difference between the energy at 12200 rpm and the energy at 10500 rpm

are you saying it is all passed through parts that are trying to shift and on to/from Planet Earth ?


we've seen telemetry showing the K generating momentarily with upshifts and motoring momentarily with downshifts

Bit of a late reply:

Part of the energy tends to be dissipated by the clutch (even though the clutch is not operated, it is designed to slip at high torque stabs) and the rest is converted into kinetic and elastic energy and is gradually dissipated by the driveline dampers. Below is a comparison between a conventional and a seamless shift gearbox from the Honda F1 paper. There is no drop in mean input shaft torque.

I do agree that the torque going through the driveline is very high due to inertia sync, but it is designed to cope with that. It is one of the reasons they started introducing compliant shafts between gearbox and crank.

this doesn't show that 'the engine is developing full torque throughout the shift' - your claim on P608

it shows that the torque seen by the input shaft is not less than full engine torque - not the same thing at all

imo the torque seen by the input shaft has been managed to this end by a suitable millisec duration of fuel cutting


anyway the events that you have shown seem to match quite well the response timing estimates I put out

It shows that the input shaft torque is constant. Since presumably the engine is putting out full torque before the upshift and there is no dip in input shaft torque it looks to me as though the engine is making full torque throughout.

in the 100 mS before the shift something is happening or being made to happen
a fall in what was a rising trend of the rpm and a fall in what was a steady trend in the torque

to me that's easing of the torque to enable the shift
the output torque of the PU won't follow exactly a cutoff of energy input (ie a fuel cut reduces not cuts off PU torque)
and the torque in a compliant load system won't follow exactly the rate of reduction of the PU output

there's a lot of strain energy stored in the design for compliance (and a lot of energy stored in PU rotation)
we see here the input shaft torque fall that is slower than the PU torque reduction
we do not see PU torque reduction


IMO if PU behaviour was fixed (not modified for shifts) the telemetry would show a different result

Some pwm to reduce the mgu-k during an upshift?

iirc we have seen telemetry showing the K momentarily going into generation during upshifts
ie making the best contribution it can to slowing the PU rpm and/or reducing PU torque

Tommy - the torque produced by the engine is small compared to the torque required to decelerate (or accelerate in the case of downshifts) the engine inertia. Simply cutting fuel for an instant doesn't even begin to make up for the massive hammer blow that is the gear change. Instead, the driveline has to be designed in such manner as to cope with the load - normally by slipping the clutch.

There is something I found odd about the graphs I've posted though. Before the gearshift, the input shaft shows normal torque fluctuations (looks about 60 Hz which is consistent with the 5th order of a V10 running at 18000 rpm) however something before the gear change excites the driveline (about 160 Hz, most likely its natural frequency) and the response is slowly damped after the shift.

I have a question about the torque of the input shaft, isnt it a reflection of road, tyre etc conditions?

what has changed since the V10 days ......

we need a much longer life from the gearbox
we haven't got free fuel
the ICE response to a DI fuel cut is quicker (than rich mixture PI - which has a float of maybe 3 revsworth of fuel)

ok clutch torque limiting effect is a valuable safeguard (but quickening shifts this way would be unproductive laptimewise)
ok I agree that these shifts taking about 1 rev doesn't give a WOT engine much scope for rpm increase

Mudflap wrote: ↑

15 Apr 2018, 04:42

however something before the gear change excites the driveline (about 160 Hz, most likely its natural frequency) and the response is slowly damped after the shift.

Probably trying to unlock the pawls for gear selection

Is it possible to have the bottom orifice of the TJI to spin with the thrust generated by the jet flames to provide

1) a more complete flame all along the top of the cylinder head

and

2) to create more flow pattern to aid the charge air/fuel mixing

If you want to have it rotating, you'll probably have to design some sort of bearing with lubrication etc. Otherwise you may find it throwing metal particles into the piston, causing excessive wear and what not. Another question is what happens with the (unpowered?) moving head during combustion. It will probably keep rotating, so that's allright, but it needs be able to withstand all the forces working upon it as well.

And the final question, is the current atomization so 'bad', it needs further refining/optimization with a rotating injector nozzle?

iichel wrote: ↑

16 Apr 2018, 09:49

If you want to have it rotating, you'll probably have to design some sort of bearing with lubrication etc. Otherwise you may find it throwing metal particles into the piston, causing excessive wear and what not. Another question is what happens with the (unpowered?) moving head during combustion. It will probably keep rotating, so that's allright, but it needs be able to withstand all the forces working upon it as well.

And the final question, is the current atomization so 'bad', it needs further refining/optimization with a rotating injector nozzle?

Can I ask a question on this which is beyond my knowledge so may be stupid? (I do not know the correct terms either)

Would a similar effect be achieved by directing one of the 'jets' onto a shaped surface? Possibly even a surface retaining more heat and activating or 'vaporising' (wrong word sorry as it is already a vapor, or just an aerosol?) the fuel into a 'cloud'

The spinning orifice could also be designed so as to provide fresh charge air into the prechamber.

Big Tea wrote: ↑

16 Apr 2018, 10:33

iichel wrote: ↑

16 Apr 2018, 09:49

If you want to have it rotating, you'll probably have to design some sort of bearing with lubrication etc. Otherwise you may find it throwing metal particles into the piston, causing excessive wear and what not. Another question is what happens with the (unpowered?) moving head during combustion. It will probably keep rotating, so that's allright, but it needs be able to withstand all the forces working upon it as well.

And the final question, is the current atomization so 'bad', it needs further refining/optimization with a rotating injector nozzle?

Can I ask a question on this which is beyond my knowledge so may be stupid? (I do not know the correct terms either)

Would a similar effect be achieved by directing one of the 'jets' onto a shaped surface? Possibly even a surface retaining more heat and activating or 'vaporising' (wrong word sorry as it is already a vapor, or just an aerosol?) the fuel into a 'cloud'

I'm not sure there is enough time available in the intake/compression stroke to inject fuel in this way. Furthermore you like to keep the fuel off any surface to minimize heat transfer to the head and cylinder walls.