Rain Light. -H or -K harvesting? or -K torque demand?

[supertweet]

Haven't found in the regs or discussions whether the rain light flashes due to reduced torque demand on-throttle from the -K, or when the -H/-K harvesting. Does anyone have the answer?

As I generally see it flashing towards end-of-straight, but before the braking zones, I'd assume it's -H switching from -K transfer to ES storage or -K from propulsion to stasis (either throttle on or off, but before braking), but am unsure.

[Lycoming]

Pretty sure the rain light always flashes. May not always be visible as such from onboards on account of rain and roostertails, but it's always flashing. Nothing to do with ERS.

[supertweet]

Pretty sure the rain light always flashes. May not always be visible as such from onboards on account of rain and roostertails, but it's always flashing. Nothing to do with ERS.

I should have clarified rain light flashing in the dry. My apologies. It's general consensus the rain light flashes in the dry "when lift and coast." Makes sense as we don't see it flashing in acceleration, nor generally while braking.

My specific question is what is harvesting, and can it be occurring whilst still on throttle, but ERS not providing additional power.

[Tommy Cookers]

good question, I asked the same last week

IMO they all mgu-k harvest driven at partial power and when there's disruption by incipient wheelspin (Hamilton) the rain light lights
signalling that momentarily the harvesting has stopped
this seemed to fit Mr Coulthard's explanation (lap 53 IIRC)

this harvesting under power is an open goal that I have been attempting to highlight for ages
(most so-called KERS harvesting was also harvesting under power)

[supertweet]

IMO they all mgu-k harvest driven at partial power and when there's disruption by incipient wheelspin (Hamilton) the rain light lights signalling that momentarily the harvesting has stopped

Interesting theory. However, that would denote a form of "traction or torque control," to me. "Clutching" or modulating -K output under throttle, whether full or partial, dependent upon wheel slip counters FIA traction control ban, to me.

Also, I generally see the rain light flashing near end of straight, before braking, where wheel slip is rare. That's why I'm hypothesizing it's either -K disengagement while under throttle, or under coast w/ -H harvesting turbo energy. Basically, in the space between full ICE/ERS power and the braking zone.

I don't know if any of this makes sense...

[tok-tokkie]

I thought it is as a warning to a following car that the car ahead may not continue accelerating as was before the light comes on. So it signals when power is being harvested by the turbine & -H & being stored in the ES. I picked that up from comments by Brundle during the race. I stress that I am not certain about all this.

[supertweet]

I thought it is as a warning...it signals when power is being harvested by the... -H & being stored in the ES. I picked that up from comments by Brundle during the race.

Anecdotally, I feel the same, that it's the -H harvesting while the ICE is still loaded, or perhaps the -K. The problem is Mr. Brundle and Co. only mention generalities, never specifics. I think if someone knew what was occurring, we'd have a better idea how each PU manufacturer/team is tuning its ECU's (control electronics I believe it's called now?)

A better idea of how the 2 MGU's are operating while on throttle, how they're switching from power to recovery to transfer, might provide clues as to how efficient each turbo is, how much -H assist is needed for spooling, or what relative time -H spends powering -K vs recharging ES. The rain light's a great signal, if we knew what it signified.

[Tommy Cookers]

IMO they all mgu-k harvest driven at partial power and when there's disruption by incipient wheelspin (Hamilton) the rain light lights signalling that momentarily the harvesting has stopped

Interesting theory. However, that would denote a form of "traction or torque control," to me. "Clutching" or modulating -K output under throttle, whether full or partial, dependent upon wheel slip counters FIA traction control ban, to me.
Also, I generally see the rain light flashing near end of straight, before braking, where wheel slip is rare.

the short story ......
(at one time I designed electro-mechanical servo systems for a living)
necessarily for system stability and viability a motor drive will always know and respond to the (mgu) rpm
(and probably its rotational position to within much less than a degree)
since the mgu-k is slaved to the crankshaft
there is an inevitable and automatic (but not unlimited) modulation of the wheel torque that acts to reduce wheelspin
but it's not TC because TC is defined by having wheelspin sensing

the mgu-k can in principle harvest power from the crankshaft any place the driver doesn't want full power
and specially after a late-race safety car has saved fuel
that's what I was looking at on replay and by DC's commentary
agreed it might have been LH momentarily lifting for car control

carry on !

[supertweet]

...a motor drive will always know and respond to the (mgu) rpm...since the mgu-k is slaved to the crankshaft there is an inevitable and automatic (but not unlimited) modulation of the wheel torque that acts to reduce wheelspin but it's not TC because TC is defined by having wheelspin sensing

the mgu-k can in principle harvest power from the crankshaft any place the driver doesn't want full power
and specially after a late-race safety car has saved fuel
that's what I was looking at on replay and by DC's commentary
agreed it might have been LH momentarily lifting for car control

carry on !

If I understand you correctly, you're positing -K will harvest during unwanted "overrun" ICE crankshaft RPM, either through over-rotation or part throttle, which seems backwards to me. My understanding is that -K will harvest solely through byproduct of waste energy via braking friction; the -K influences the crankshaft, but not vice versa, except in full off throttle/brake, in which I agree mechanical drag "sucks" kinetic energy (engine braking). Otherwise, it's in discharge mode to the ICE or perhaps, as I questioned, in a stasis mode during coast. Please correct me if I'm wrong, as you experience and technical literacy surely trumps mine.

I wish I'd seen the BBC commentary as well; despite being in the states, I watch Sky's coverage. Ours is...substandard.

Thanks for taking the time.

[jbolenbaugh]

If I understand you correctly, you're positing -K will harvest during unwanted "overrun" ICE crankshaft RPM, either through over-rotation or part throttle, which seems backwards to me. My understanding is that -K will harvest solely through byproduct of waste energy via braking friction; the -K influences the crankshaft, but not vice versa, except in full off throttle/brake, in which I agree mechanical drag "sucks" kinetic energy (engine braking). Otherwise, it's in discharge mode to the ICE or perhaps, as I questioned, in a stasis mode during coast. Please correct me if I'm wrong, as you experience and technical literacy surely trumps mine.

I wish I'd seen the BBC commentary as well; despite being in the states, I watch Sky's coverage. Ours is...substandard.

Thanks for taking the time.

The MGU - K can certainly and most likely does harvest any extra "unwanted" energy from the ICE. The battery / power unit controls will have a desired battery SOC target that they will attempt to maintain throughout the lap. If the engine is running at WOT it is more efficient, so under torque request of magnitude Z lets say, that doesn't utilize maximum engine torque it may be beneficial to run the engine at WOT producing X amount of torque, and harvesting Y amount of torque, resulting in output torque of magnitude X-Y, Where X-Y = Z, the desired output torque. I wrote Hybrid powertrain controls software for a living for ~3 years. If you have any other questions feel free to message me.

[supertweet]

The MGU - K can certainly and most likely does harvest any extra "unwanted" energy from the ICE. The battery / power unit controls will have a desired battery SOC target that they will attempt to maintain throughout the lap. If the engine is running at WOT it is more efficient, so under torque request of magnitude Z lets say, that doesn't utilize maximum engine torque it may be beneficial to run the engine at WOT producing X amount of torque, and harvesting Y amount of torque, resulting in output torque of magnitude X-Y, Where X-Y = Z, the desired output torque. I wrote Hybrid powertrain controls software for a living for ~3 years. If you have any other questions feel free to message me.

Fascinating, thank you. Mechanically-speaking, how would -K be harvesting the excess-to-speed torque demand (Y)? The engine is still under load there's no unnecessary crankshaft rotation the -K may siphon, and obviously no kinematic braking energy, how's it recharging ES? Are you saying the throttle mapping would allow for increased RPM demand (or more accurately power demand) relative to the car's velocity, which the -K would be simultaneously-siphoning off and converting for ES store, equaling a steady-state velocity?

Hypothetical, as we don't see numerical energy flow diagrams, but wouldn't the lift and coast method be preferable to increased engine demand? In this case, it's the -H harvesting excess turbine speed, w/o the resultant fuel usage of your equation. I realize they're different phases in the car's travel, but just ruminating here.

[jbolenbaugh]

Fascinating, thank you. Mechanically-speaking, how would -K be harvesting the excess-to-speed torque demand (Y)? The engine is still under load there's no unnecessary crankshaft rotation the -K may siphon, and obviously no kinematic braking energy, how's it recharging ES? Are you saying the throttle mapping would allow for increased RPM demand (or more accurately power demand) relative to the car's velocity, which the -K would be simultaneously-siphoning off and converting for ES store, equaling a steady-state velocity?

Hypothetical, as we don't see numerical energy flow diagrams, but wouldn't the lift and coast method be preferable to increased engine demand? In this case, it's the -H harvesting excess turbine speed, w/o the resultant fuel usage of your equation. I realize they're different phases in the car's travel, but just ruminating here.

You seem to have linked torque to RPM which is not necessarily the case. For example...

The engine is operating at 10,000 rpm creating 450 Ft Lbs of torque. The MGU-K is not harvesting at all.

Now lets say the engine ramps up its torque to 550 ft lbs of torque over 1 second. Now lets say the MGU-K ramps from 0 to -100 ft lbs of torque over that same second.

Assuming the torque ramp rates are identical (there will be small error) the engine will be still running at 10,000 rpm at the end, but producing 550 ft lbs of torque, the mgu-k will be harvesting 100 ft-lbs of torque (converting it to electricity), and the output torque will be still 450 ft lbs. The engine rpm stays the same because the engine rpm is a function of the gear ratio, and determined by the vehicle speed, and all other things assumed equal, driving on a straight line on a flat road for example, the vehicle speed remains the same because output torque has not changed. It gets very very complicated when a planetary gearset is used to have the motor be able to effectively change the engine to output speed gear ratio. This is what many hybrids such as the Toyota Prius, Chevy Volt etc. The F1 hybrid powertrains are relatively "simple" (In this sense only!) compared to the planetary gearset stuff.

[Tommy Cookers]

..... but wouldn't the lift and coast method be preferable to increased engine demand? In this case, it's the -H harvesting excess turbine speed, w/o the resultant fuel usage of your equation. I realize they're different phases in the car's travel, but just ruminating here.

do we know whether or not lift and coast includes bleeding down turbo rpm ?
that would always be followed closely by a need for electrical restoration of full turbo rpm

using ICE power to generate for ERS is nearly as efficient as using ICE power any other way
(by increasing ICE power at low power demand it may improve ICE efficiency and be fully as efficient as normal ICE powers)
and that extra bit of stored energy may be priceless at some times and places
at the less fuel-hungry races this may be better than leaving 5 kg of fuel in the pits
what I saw only happened after the safety car, when the cars would have had fuel to spend in any way possible

much of the KERS recovery of previous years has come this way
the 'zero torque' rules allowed full cancellation of engine braking at 0% accelerator
clearly ? this was allowing engine running to power KERS generation
(though such engine running was then of course important for exhaust blowing)

thanks jbolenbaugh !

[supertweet]

The engine is... creating 450 Ft Lbs of torque. The MGU-K is not harvesting at all...

...Now lets say the engine ramps up its torque to 550 ft lbs of torque over 1 second. Now lets say the MGU-K ramps from 0 to -100 ft lbs of torque over that same second.

Assuming the torque ramp rates are identical (there will be small error) the engine will be still running at 10,000 rpm at the end, but producing 550 ft lbs of torque, the mgu-k will be harvesting 100 ft-lbs of torque (converting it to electricity), and the output torque will be still 450 ft lbs...the vehicle speed remains the same because output torque has not changed. It gets very very complicated when a planetary gearset is used to have the motor be able to effectively change the engine to output speed gear ratio. This is what many hybrids such as the Toyota Prius, Chevy Volt etc. The F1 hybrid powertrains are relatively "simple" (In this sense only!) compared to the planetary gearset stuff.

Thanks, I understood cranking force vs/ RPM, but worded it so poorly I can see how it'd be missed (RPM demand ("or more accurately power demand"). So, as I read it, in a perfect world, -K would be extracting torque surplus to car acceleration/velocity demands via the crankshaft and converting it to electrical energy as the motor's under load, correct? This would be via ECU tuning, perhaps if the 4MJ discharge allotment is used or if ES storage drops below a certain %, also correct?

If so, it makes sense. Could you relate this to rain light flashing, that it's seen mostly near end of straight, when I assume the cars at Vmax? They're at max RPM and if geared properly max power, so at a declining torque curve, but I'd still expect minimal "torque pull" from the crank to maintain velocity.

[wuzak]

The rain light comes on when they are harvesting from the MGU-K.

The "lift and coast" strategy isn't so much about saving fuel, as many people have been complainging about, but to generate as much of the 2MJ allowed as they can.