Horsepower of the engines.

[Joseki]

If I got this right it is possible that some teams are reaching the max harvesting from K and H and maybe can have the MGU-K spending the max 120kW all the time they need in every lap of the GP and not only the initially theorised 33s?

Sorry for being a bit slow to respond.

As with everything to do with these PUs I don't think there is a single, one size fits all, answer to this. It varies with circuit, chassis and aerodynamic characteristics and even things like wind direction.

I've tried to work out a method to get a ballpark idea of what it would mean to achieve what you suggest.

First we need a definition of "all the time they need". One definition might be all the time they are not traction limited and running at max ICE power. I'll use that.

Second we need to look at how energy is used around the lap and through the race.

If we take a lap time we can divide the lap into three states :

1 Traction limited - part power running typically the slow medium parts of the course. The ICE is at part throttle or low revs, or both. The MGU-K is not deployed.

2 Not traction limited - full power running, typically the straights and fast corners. The ICE is at full power making as much power as it can. The MGU-K is fully deployed motoring.

3 Braking. The ICE is off. The MGU-K is generating.

There is a fourth state, lift and coast, but trying to deal with that is beyond me.

I think that using the 100 kg and 100kg/hr fuel regs and some estimates for the above it is possible to estimate the output that would be needed from the MGU-H to make it possible to run the MGU-K flat out (120 kW ) while in state 2.

I have worked on three circuits, Barcelona, Spa and Singapore. I'll use the data for Barcelona.

Barcelona runs over 66 laps with a typical lap time around 92 seconds. That makes average fuel useage 1.515 kg a lap.

Brembo give the % braking as 18%. I found a figure of 57% full throttle (52.4 secs) on the f1fanatic website.
This leaves 25 % at part power.

If we assume the full 120 kW is captured during braking the 2 mJ Is captured into the ES every lap.

The time in state 1 Is 23 secs. To calculate the fuel available to run the ICE at full power we need an estimate of the fuel usage in state 1. I assumed an average engine speed of 9000 rpm, giving a Max fuel flow of 83.6 kg/hr. I then assumed an average of 50% of this for part throttle. Arbitrary I know but necessary to get an answer.

This gives a fuel useage in state 1 of 0.275 kg/lap. Leaving 1.24 kg for full throttle state 2.

This converts to 44.6 seconds, with a consequence of needing several seconds coasting Per lap.

If the MGU-K runs at 120 kW for this time it needs 5.4 mJ. 2 come from the ES leaving the MGU-H to supply 3.4.

The MGU-H spends 44.6 seconds generating at its max level and 23 at a lower level which I have assumed is in ratio to the fuel rate, 100 vs 43.

This gives a target MGU-H power of 3420 / (44.6 + 23 X 0.43). 62.76 kW.

I have yet to rerun the figures for a quali lap time but even using these figures and an extra 2 mJ in the ES the power required is 26 only kW.

The figures for Spa are 86 and 58 kW. For Singapore 56 and 22.

I leave everyone to make up their own minds about whether these levels are achievable or not.

In making these calculations I have noticed there is a distinct positive feedback mechanism at play. The faster you go, shorter lap time, the faster you go. This may explain some of the extent of Mercedes advantage. Just a little bit better in a few key areas makes for a big improvement.

Here there are actual data from Magneti Marelli but from Bahrain. 2 MJ recovered from MGU-K are definitely too much.

[gruntguru]

Thanks everyone for the replies these engines are truly formidable. I just had one more thing to ask. While the MGU-H will slow down the turbine to harvest energy and transfer to the MGU-K->crankshaft I guess it also, in special cases like low RPM, will use some energy to accelerate the compressor and compensate the turbo lag, right? Is that energy very little to take into account, or is it being taken into account or am I understanding it poorly?

It would need to be accounted for although it is comparatively small. Total power to run the compressor at max output is only about 80 kW and the turbine is usually able to supply some of this.

80 kW for 0.5 sec = 40 kJ = 1% of ES allowance (4 MJ)

[ringo]

I think some people are wasting their time trying to explain why that MGUK shaft power is limited to 120 kW. Save your energy my friends.

The reason the MGUH to the MGUK is umlimited is to account for the efficiency of the MGUK.

If my MGUK is 50% efficieint it would require 240 kW from MGUH, but it's ouput would still be at the limit of 120 kW.

This is the reason for stating unlimited. There is no loop hole for the MGUK to output 120kW.

It is simpler than that. "Unlimited" means no limit on the ENERGY per lap.

1. From ES to MGUK there is a limit of 4 MJ/lap.
2. From MGUH to MGUK there is no energy limit.

In both 1 and 2 above there is a limit on POWER. The instantaneous sum of 1+2 cannot exceed 120 kW (0.12 MJ/s)

I agree. That's what i've been saying.

[ringo]

If I got this right it is possible that some teams are reaching the max harvesting from K and H and maybe can have the MGU-K spending the max 120kW all the time they need in every lap of the GP and not only the initially theorised 33s?

Sorry for being a bit slow to respond.

As with everything to do with these PUs I don't think there is a single, one size fits all, answer to this. It varies with circuit, chassis and aerodynamic characteristics and even things like wind direction.

I've tried to work out a method to get a ballpark idea of what it would mean to achieve what you suggest.

First we need a definition of "all the time they need". One definition might be all the time they are not traction limited and running at max ICE power. I'll use that.

Second we need to look at how energy is used around the lap and through the race.

If we take a lap time we can divide the lap into three states :

1 Traction limited - part power running typically the slow medium parts of the course. The ICE is at part throttle or low revs, or both. The MGU-K is not deployed.

2 Not traction limited - full power running, typically the straights and fast corners. The ICE is at full power making as much power as it can. The MGU-K is fully deployed motoring.

3 Braking. The ICE is off. The MGU-K is generating.

There is a fourth state, lift and coast, but trying to deal with that is beyond me.

I think that using the 100 kg and 100kg/hr fuel regs and some estimates for the above it is possible to estimate the output that would be needed from the MGU-H to make it possible to run the MGU-K flat out (120 kW ) while in state 2.

I have worked on three circuits, Barcelona, Spa and Singapore. I'll use the data for Barcelona.

Barcelona runs over 66 laps with a typical lap time around 92 seconds. That makes average fuel useage 1.515 kg a lap.

Brembo give the % braking as 18%. I found a figure of 57% full throttle (52.4 secs) on the f1fanatic website.
This leaves 25 % at part power.

If we assume the full 120 kW is captured during braking the 2 mJ Is captured into the ES every lap.

The time in state 1 Is 23 secs. To calculate the fuel available to run the ICE at full power we need an estimate of the fuel usage in state 1. I assumed an average engine speed of 9000 rpm, giving a Max fuel flow of 83.6 kg/hr. I then assumed an average of 50% of this for part throttle. Arbitrary I know but necessary to get an answer.

This gives a fuel useage in state 1 of 0.275 kg/lap. Leaving 1.24 kg for full throttle state 2.

This converts to 44.6 seconds, with a consequence of needing several seconds coasting Per lap.

If the MGU-K runs at 120 kW for this time it needs 5.4 mJ. 2 come from the ES leaving the MGU-H to supply 3.4.

The MGU-H spends 44.6 seconds generating at its max level and 23 at a lower level which I have assumed is in ratio to the fuel rate, 100 vs 43.

This gives a target MGU-H power of 3420 / (44.6 + 23 X 0.43). 62.76 kW.

I have yet to rerun the figures for a quali lap time but even using these figures and an extra 2 mJ in the ES the power required is 26 only kW.

The figures for Spa are 86 and 58 kW. For Singapore 56 and 22.

I leave everyone to make up their own minds about whether these levels are achievable or not.

In making these calculations I have noticed there is a distinct positive feedback mechanism at play. The faster you go, shorter lap time, the faster you go. This may explain some of the extent of Mercedes advantage. Just a little bit better in a few key areas makes for a big improvement.

A waste gate makes all this easier as well; if control electronics are overwhelmed with energy conversion and management.

[ringo]

If the total energy generated by the MGUH plus MGUK per lap is greater than needed to run the K at 120 kW during full throttle operation, the surplus energy can be used to save fuel during part load operation eg 200 kW needed at the wheels could be supplied as 120 from the K plus 80 from the ICE.

What you are saying doesn't seem right. How do you get this 80kW from the ICE related to the MGUH?

If your MGUH is capable of harvesting 200kW you send 80 to the ES and 120kW to the MGUK.

Ideally you should not harvest 200kW if engine performance is need, since it's a great back pressure on the engine.
You may want to harvest this during braking.
Nonetheless can you explain the 80 from the ICE?

[gruntguru]

I was referring to the post before mine in which Henry looks at the total energy budget for a lap. At Barcelona there is 52 seconds at WOT during which the MGUK can use 120 kW x 52 s = 6.24 MJ of energy. If the total energy recovered per lap from the MGUH plus regen from the K is more than 6.24 MJ - how do you best use the rest?

Henry's post also shows that fuel available per lap (1.515 kg at Barcelona) is not sufficient to supply all the WOT and part throttle running. One solution (if the total above is > 6.24 MJ) would be to send some electrical energy to the K during part throttle running eg 80 kW ICE + 120 kW MGUK = 200 kW to the wheels etc.

[henry]

Here there are actual data from Magneti Marelli but from Bahrain. 2 MJ recovered from MGU-K are definitely too much.

Can you point me at these figures? Brembo say Bahrain is heavy on brakes. 20% of a 100 second lap. Which gives an energy recovery potential of 2.4 mJ

[Tommy Cookers]

......If your MGUH is capable of harvesting 200kW you send 80 to the ES and 120kW to the MGUK.

Ideally you should not harvest 200kW if engine performance is need, since it's a great back pressure on the engine.
You may want to harvest this during braking. .......

this seems a good opportunity to say imo ....

there's nothing wrong with 'a great back pressure on the engine'
back pressure means the exhaust gas is denser and so the blowdown process is working against a greater inertial load

the exhaust internal energy (often misleadingly called heat energy) has multiple components
sensible heat is a major component, and is completely useless to an expander device ie the turbine
(in the turbocompound aircraft engines this useless sensible heat was 35% iirc of the total exhaust energy)

back pressure reduces this loss of useful energy as sensible heat
greater exhaust density tends to lower velocities below sonic/supersonic so less pressure energy is diverted into sensible heat energy

back pressure can even exceed induction pressure (without producing underscavenge)
since the momentary exhaust port pressure at valve closure will be lowered by the exhaust system 'tuned length' functionality

1940s work on this seemed to show efficiency benefits in engines with small valve timing overlap but no benefit with larger overlap
ie benefits beyond any benefit from elimination of fuel overscavenge (in these compounded but non-DI engines)

and remember that our N/A road cars run this 'back pressure' way all the time they are partially throttled
as induction pressure will then be sub-atmospheric

[hemichromis]

......If your MGUH is capable of harvesting 200kW you send 80 to the ES and 120kW to the MGUK.

Ideally you should not harvest 200kW if engine performance is need, since it's a great back pressure on the engine.
You may want to harvest this during braking. .......

this seems a good opportunity to say imo ....

there's nothing wrong with 'a great back pressure on the engine'
back pressure means the exhaust gas is denser and so the blowdown process is working against a greater inertial load

the exhaust internal energy (often misleadingly called heat energy) has multiple components
sensible heat is a major component, and is completely useless to an expander device ie the turbine
(in the turbocompound aircraft engines this useless sensible heat was 35% iirc of the total exhaust energy)

back pressure reduces this loss of useful energy as sensible heat
greater exhaust density tends to lower velocities below sonic/supersonic so less pressure energy is diverted into sensible heat energy

back pressure can even exceed induction pressure (without producing underscavenge)
since the momentary exhaust port pressure at valve closure will be lowered by the exhaust system 'tuned length' functionality

1940s work on this seemed to show efficiency benefits in engines with small valve timing overlap but no benefit with larger overlap
ie benefits beyond any benefit from elimination of fuel overscavenge (in these compounded but non-DI engines)

and remember that our N/A road cars run this 'back pressure' way all the time they are partially throttled
as induction pressure will then be sub-atmospheric

What are the practical implications of this?

That charging the MGU-H while at full throttle and shunting the power to the MGU-K may be more efficient than reducing the back pressure and relying on the ICE to produce the power?

[gruntguru]

That is always going to be more "efficient". For the fixed fuel input permitted, the highest work output (crank + MGUH) will always be at a high back pressure. If there were no MGUH and the engine was optimised for crankshaft power only, a BTE in the low 40% region would be the limit. With the MGUH I believe Mercedes is now in the high 40's.

[Joseki]

Here there are actual data from Magneti Marelli but from Bahrain. 2 MJ recovered from MGU-K are definitely too much.

Can you point me at these figures? Brembo say Bahrain is heavy on brakes. 20% of a 100 second lap. Which gives an energy recovery potential of 2.4 mJ

Sorry I forgot to put the link, here it is:

http://it.motorsport.com/f1/news/f1-sim ... smissione/

"ERS La MGU-H e la MGU-K comportano un guadagno di 2”5 per giro, corrispondenti a +21 km/h di velocità massima. In Bahrein, la MGU-K può recuperare fino a 1.085 kJ al giro in frenata, mentre la MGU-H può arrivare fino a 2.815 KJ al giro per un totale di 3.900 KJ per giro."

MGU-K can recover 1.085 kJ in a lap, the MGU-H 2.815 kJ.

[gruntguru]

the exhaust internal energy (often misleadingly called heat energy) has multiple components
sensible heat is a major component, and is completely useless to an expander device ie the turbine
(in the turbocompound aircraft engines this useless sensible heat was 35% iirc of the total exhaust energy)

"Internal energy" is heat energy.

Internal energy is the largest contributor of energy input to an exhaust gas turbine. Sure its pressure x volume that drives the turbine but it is internal energy (heat) that drives and prolongs the expansion.

For Isentropic expansion of a 0.5 kg/s flow of ideal gas at 1000*C and 3 atm to atmospheric pressure you get the following:
Outlet temperature = 657*C (Temp change = 343*C)
Turbine power = 172 kW
ie 1 kW for each 2*C of temperature change through the turbine.

If we take the same situation except the gas starts at 25*C you get:
Outlet temperature = -55*C (Temp change = 80*C)
Turbine power = 40 kW
ie 1 kW for each 2*C of temperature change through the turbine.

[livinglikethathuh]

Do we have a formula to relate exhaust back pressure to ICE power? It would make it easier to see the tradeoff between ICE and MGU-H power.

[wuzak]

the exhaust internal energy (often misleadingly called heat energy) has multiple components
sensible heat is a major component, and is completely useless to an expander device ie the turbine
(in the turbocompound aircraft engines this useless sensible heat was 35% iirc of the total exhaust energy)

"Internal energy" is heat energy.

Internal energy is the largest contributor of energy input to an exhaust gas turbine. Sure its pressure x volume that drives the turbine but it is internal energy (heat) that drives and prolongs the expansion.

For Isentropic expansion of a 0.5 kg/s flow of ideal gas at 1000*C and 3 atm to atmospheric pressure you get the following:
Outlet temperature = 657*C (Temp change = 343*C)
Turbine power = 172 kW
ie 1 kW for each 2*C of temperature change through the turbine.

If we take the same situation except the gas starts at 25*C you get:
Outlet temperature = -25*C (Temp change = 80*C)
Turbine power = 40 kW
ie 1 kW for each 2*C of temperature change through the turbine.

In the second example the delta T should be 50C?

[gruntguru]

the exhaust internal energy (often misleadingly called heat energy) has multiple components
sensible heat is a major component, and is completely useless to an expander device ie the turbine
(in the turbocompound aircraft engines this useless sensible heat was 35% iirc of the total exhaust energy)

"Internal energy" is heat energy.

Internal energy is the largest contributor of energy input to an exhaust gas turbine. Sure its pressure x volume that drives the turbine but it is internal energy (heat) that drives and prolongs the expansion.

For Isentropic expansion of a 0.5 kg/s flow of ideal gas at 1000*C and 3 atm to atmospheric pressure you get the following:
Outlet temperature = 657*C (Temp change = 343*C)
Turbine power = 172 kW
ie 1 kW for each 2*C of temperature change through the turbine.

If we take the same situation except the gas starts at 25*C you get:
Outlet temperature = -55*C (Temp change = 80*C)
Turbine power = 40 kW
ie 1 kW for each 2*C of temperature change through the turbine.

In the second example the delta T should be 50C?

Whoops - typo. Fixed now - thanks Wuzak.