Renault Power Unit Hardware & Software

[gruntguru]

2, it is likely that the compressor will not be able to provide sufficient air at engine speeds of 10500rpm as there is no head room in left in turbo speed.

Look at the compressor map provided by Edis. Providing a pressure ratio of 3.5 at airflows from 50 - 80 lb/min requires rotational speeds from 105,000 to 120,000. Rotational speed affects boost. Flow is largely independent of shaft speed in the efficient operating range.

[stevesingo]

No, you don't want to reach 125,000 rpm as soon as possible as that would force the compressor into surge as a centrifugal compressor can't supply pressure unless there is adequate flow and the flow depends on how much air the engine can use (engine speed, throttle position and so on). The turbocharger speed needs to be controlled depending on what boost level you want to reach, where a higher boost pressure in general means a higher turbocharger speed. So, if for instance the boost pressure is lowered at engine speeds above 10,000 rpm due to the fuel flow limit it is possible that you need to reduce the turbocharger speed above that point.

From a purely turbo point of view I would agree. But the MGU-H is driven by the turbine also.

So we have two loads applied to the turbine Compressor and MGU-H

Power generation from the MGU-H is unlimited in the regulations. Boost is unlimited in the regulations, but is effectively limited by the fuel flow at about 2.75 Bar and mass air flow <1500kg/hr. As power is a function of force over distance and in the case of a generator torque applied over rotational speed, we can either apply high torque to the turbine at a lower speed or low torque at a higher speed. There will be a compromise of where in the scale of MGU-H tq/rpm gives the least backpressure to MGU-H benefit.

My thinking is, the work done by the turbine should favour the unlimited output-MGU-H. The turbine can have a finite amount of load applied before losses from increased back pressure exceed the gains from MGU-H.

Could the Turbine be sized where it makes up to 125k rpm at 7krpm engine speed or even at part throttle. In normal circumstances the compressor will surge, but if we bleed off excess mass air flow we could prevent surge. Yes there will still be load applied to the turbine by this work done, but there may be a balance to be had where total load applied to the turbine is less than the gains from MGU-H generation.

I am just theorising here and the 125k rpm may not be attainable, but I believe there will be a balance to be had, which will be forever changing with load and engine speed, where MGU-H power might take priority over compressor function and in the case of too much boost, then that could be bled off in order to prioritise MGU-H function?

[toraabe]

There is not a significant pressure loss anywhere in the inlet manifold and runner length does not have a significant effect on pressure loss. Perhaps you are confusing "loss" with pressure wave tuning and the relationship between runner length, rpm and cylinder filling.

No.. What i meant was because of pipe length, intercooler, bends and so on and if this in total longer than it competitor you will have Answer is yes. And another thing the spring effect on longer pipes....


As a general rule, a cooler engine is less efficient in terms of "thermal efficiency" (the one that counts underthe current formula). You may be confusing volumetric efficiency, which does increase in a cooler engine but is less important under the current formula.

What I tried to say, if the engine is running cooler higher vomuletric efficiency it is benificial in terms of power and reliability.. Same --- anyway. As long as you understood what I ment

[Vortex37]

No, you don't want to reach 125,000 rpm as soon as possible as that would force the compressor into surge as a centrifugal compressor can't supply pressure unless there is adequate flow and the flow depends on how much air the engine can use (engine speed, throttle position and so on). The turbocharger speed needs to be controlled depending on what boost level you want to reach, where a higher boost pressure in general means a higher turbocharger speed. So, if for instance the boost pressure is lowered at engine speeds above 10,000 rpm due to the fuel flow limit it is possible that you need to reduce the turbocharger speed above that point.

From a purely turbo point of view I would agree. But the MGU-H is driven by the turbine also.

So we have two loads applied to the turbine Compressor and MGU-H

Power generation from the MGU-H is unlimited in the regulations. Boost is unlimited in the regulations, but is effectively limited by the fuel flow at about 2.75 Bar and mass air flow <1500kg/hr. As power is a function of force over distance and in the case of a generator torque applied over rotational speed, we can either apply high torque to the turbine at a lower speed or low torque at a higher speed. There will be a compromise of where in the scale of MGU-H tq/rpm gives the least backpressure to MGU-H benefit.

My thinking is, the work done by the turbine should favour the unlimited output-MGU-H. The turbine can have a finite amount of load applied before losses from increased back pressure exceed the gains from MGU-H.

Could the Turbine be sized where it makes up to 125k rpm at 7krpm engine speed or even at part throttle. In normal circumstances the compressor will surge, but if we bleed off excess mass air flow we could prevent surge. Yes there will still be load applied to the turbine by this work done, but there may be a balance to be had where total load applied to the turbine is less than the gains from MGU-H generation.

I am just theorising here and the 125k rpm may not be attainable, but I believe there will be a balance to be had, which will be forever changing with load and engine speed, where MGU-H power might take priority over compressor function and in the case of too much boost, then that could be bled off in order to prioritise MGU-H function?

I agree with you. We seem to be in a minority here. You and I are thinking in terms of a system, rather than individual elements. This link to Garrett is useful. It's not a case of just dumping 'all' exhaust pressure or compressor outlet, but a small reduction by modulating the wastegate and BOV. This is all happening in a short time period. The MGU provides a controlling load, so that runaway will not happen. Not forgetting that fuel/ignition is also cut/modulated.

[gruntguru]

What I tried to say, if the engine is running cooler higher vomuletric efficiency it is benificial in terms of power . .

Not trying to be picky here but my point is that increased volumetric efficiency due to a cooler engine has no power payback under a fuel-flow-limited formula. These engines have the potential to flow vastly more air than they can burn with the limited fuel available.

[gruntguru]

No, you don't want to reach 125,000 rpm as soon as possible as that would force the compressor into surge as a centrifugal compressor can't supply pressure unless there is adequate flow and the flow depends on how much air the engine can use (engine speed, throttle position and so on). The turbocharger speed needs to be controlled depending on what boost level you want to reach, where a higher boost pressure in general means a higher turbocharger speed. So, if for instance the boost pressure is lowered at engine speeds above 10,000 rpm due to the fuel flow limit it is possible that you need to reduce the turbocharger speed above that point.

From a purely turbo point of view I would agree. But the MGU-H is driven by the turbine also.

So we have two loads applied to the turbine Compressor and MGU-H

Power generation from the MGU-H is unlimited in the regulations. Boost is unlimited in the regulations, but is effectively limited by the fuel flow at about 2.75 Bar and mass air flow <1500kg/hr. As power is a function of force over distance and in the case of a generator torque applied over rotational speed, we can either apply high torque to the turbine at a lower speed or low torque at a higher speed. There will be a compromise of where in the scale of MGU-H tq/rpm gives the least backpressure to MGU-H benefit.

My thinking is, the work done by the turbine should favour the unlimited output-MGU-H. The turbine can have a finite amount of load applied before losses from increased back pressure exceed the gains from MGU-H.

Could the Turbine be sized where it makes up to 125k rpm at 7krpm engine speed or even at part throttle. In normal circumstances the compressor will surge, but if we bleed off excess mass air flow we could prevent surge. Yes there will still be load applied to the turbine by this work done, but there may be a balance to be had where total load applied to the turbine is less than the gains from MGU-H generation.

I am just theorising here and the 125k rpm may not be attainable, but I believe there will be a balance to be had, which will be forever changing with load and engine speed, where MGU-H power might take priority over compressor function and in the case of too much boost, then that could be bled off in order to prioritise MGU-H function?

I agree with you. We seem to be in a minority here. You and I are thinking in terms of a system, rather than individual elements. This link to Garrett is useful. It's not a case of just dumping 'all' exhaust pressure or compressor outlet, but a small reduction by modulating the wastegate and BOV. This is all happening in a short time period. The MGU provides a controlling load, so that runaway will not happen. Not forgetting that fuel/ignition is also cut/modulated.

In general, bleeding boost is a very inefficient way of controlling boost - have a look at the compressor map for starters - the boost doesn't come down until you are bleeding a lot of air mass and the compressor efficiency is falling dramatically (more heat added to the air). Also - what the map doesn't show - as you move the airflow to the right at constant compressor speed, the compressor power requirement goes up, leaving less power for the MGUH to harvest.

[stevesingo]

In general, bleeding boost is a very inefficient way of controlling boost - have a look at the compressor map for starters - the boost doesn't come down until you are bleeding a lot of air mass and the compressor efficiency is falling dramatically (more heat added to the air). Also - what the map doesn't show - as you move the airflow to the right at constant compressor speed, the compressor power requirement goes up, leaving less power for the MGUH to harvest.

The compressor map shown I would say is larger than the requirment for the 1600cc turbos used in F1. 100kg/hr fuel at 15:1 AFR means 1500kg/hr or 55bl/hr. I work out the boost pressure to be 2.75bar, so the compressor shown is not in it's most efficient area of the map, nor is it turning near the maximum allowed, where I would imagine it should be for best MGU-H function. I would suspect it is not as focussed in it's design as something used on a modern F1 ICE.

I will take your point that bleeding air is not efficient, but the trade off of power lost through bleeding air vs power gained from MGU-K must have been explored, the feasability of which may change with a turbine/compressor, MGU-H system optimised for such a purpose.

[PlatinumZealot]

Yeah PZ.. very funny, very droll..

Hmmm..
..isn't it something about the greater the temp/density gradient in the working fluid, the higher the energy transfer?

The ICE component certainly works harder on 'ice cold' induction air.. &,
How ever do those gas-turbine powered planes manage at altitude in sub-zero temperatures?

The low temperature is beneficial before any form of compression. However that is definitely not in the turbine section of the engine! Two contrasting ends of the engine. You need the inlet to the turbine section as hot as your engine can handle it... (without melt-down) and the exit to be as cool as you can get it.
It is the continuous goal of gas turbine designers to make hotter running turbine engines. By hotter I mean combustion temperature. The limits are material properties, blade cooling and flame propagation. Special coatings and cooling channels are applied to the turbine section to keep the components from failing (creep).

The body of the gas has different levels of momentum the bulk momentum of the gas, and the internal momentum of the individual particles (and then even lower level is particle vibration and rotations). The internal energy. The momentum of the particles can be mathematically equated to the internal energy, however I don't remember any of that stuff right now.. been too long out of school. The point is the hotter the gas the more energy you can extract from it. The gas can give up its internal energy by expanding.

The turbine is indeed a heat engine.

[PlatinumZealot]

2, it is likely that the compressor will not be able to provide sufficient air at engine speeds of 10500rpm as there is no head room in left in turbo speed.

Look at the compressor map provided by Edis. Providing a pressure ratio of 3.5 at airflows from 50 - 80 lb/min requires rotational speeds from 105,000 to 120,000. Rotational speed affects boost. Flow is largely independent of shaft speed in the efficient operating range.

Looking on the actual graph posted that is a brave theory.

OK. example..

Compressor spinning at 70k rpms. I decide to increase boost pressure across the entire range of the engine using the MGUH. The load for the engine kept the same.

To increase boost, the mguh/compressor rpm will need to increase (assuming throttle is open and engine accelerating). It goes to 90k rpms - still in the efficient range. The result of this rpm increase, the boost will increase and the air flow will increase as well. Because the load is kept the same the engine rpm will increase faster and allow more flow to pass through the engine.


Soo.. your interpretation of the graph I do not totally disagree, if you can pick any point on the map to "jump" to but in real life the operating point will follow some progression on a path across the map.

Well this is my interpretation.

[gruntguru]

I will take your point that bleeding air is not efficient, but the trade off of power lost through bleeding air vs power gained from MGU-K must have been explored, the feasability of which may change with a turbine/compressor, MGU-H system optimised for such a purpose.

I agree. It would have been explored - and quickly rejected IMO. If you wanted to bleed air from the compressor it would make most sense to send the bled air to the exhaust manifold where it will be added to the mass flow through the turbine.

I considered this possibility some time ago as a means of recovering some waste heat from the exhaust post turbine by heating some bypass air from the compressor to help drive the turbine. My back of the envelope calculation showed it only gained about 5 kW so was pretty marginal for the extra complexity and weight required. The concept would require a heat exchanger, additional plumbing, a control valve, larger compressor, larger turbine etc etc.

[gruntguru]

Looking on the actual graph posted that is a brave theory.

Compressor spinning at 70k rpms. I decide to increase boost pressure across the entire range of the engine using the MGUH. The load for the engine kept the same.

To increase boost, the mguh/compressor rpm will need to increase (assuming throttle is open and engine accelerating). It goes to 90k rpms - still in the efficient range. The result of this rpm increase, the boost will increase and the air flow will increase as well. Because the load is kept the same the engine rpm will increase faster and allow more flow to pass through the engine.

Soo.. your interpretation of the graph I do not totally disagree, if you can pick any point on the map to "jump" to but in real life the operating point will follow some progression on a path across the map.

At W.O.T. and constant boost (height on the map) mass flow (horizontal position on the map) at the operating point is determined entirely by engine speed.

The normal control strategy would be to control MAP via compressor speed - which is controlled via wastegate or MGUH. Engine speed will determine where the flow sits.

[PlatinumZealot]

Looking on the actual graph posted that is a brave theory.

Compressor spinning at 70k rpms. I decide to increase boost pressure across the entire range of the engine using the MGUH. The load for the engine kept the same.

To increase boost, the mguh/compressor rpm will need to increase (assuming throttle is open and engine accelerating). It goes to 90k rpms - still in the efficient range. The result of this rpm increase, the boost will increase and the air flow will increase as well. Because the load is kept the same the engine rpm will increase faster and allow more flow to pass through the engine.

Soo.. your interpretation of the graph I do not totally disagree, if you can pick any point on the map to "jump" to but in real life the operating point will follow some progression on a path across the map.

At W.O.T. and constant boost (height on the map) mass flow (horizontal position on the map) at the operating point is determined entirely by engine speed.

The normal control strategy would be to control MAP via compressor speed - which is controlled via wastegate or MGUH. Engine speed will determine where the flow sits.

Ok, yes. let me examine the scenario you presented (on full boost which is held constant)... (Correct me if i am wrong here).

So the driver demands and Engine rpm increases; this reduces the restriction the compressor "sees," more flow passes through the engine as a result. (I agree that engine speed will determine flow very straight forward).

Back on the compressor graph the operating point will move to the right on a horizontal line in response to mass flow increasing and holding boost constant.

If you are on the left side of the best efficiency point: Compressor speed only increases slightly..
If you are on the right side of the best efficiency point: The compressor speed increases even more.

So yes it is determined by engine speed, but the flow through the compressor and turbine speed are still related based on the curvature of the graph.

For those who need an explanation: If you examine the general slope of the rpm plots on the graph they start out horizontal, then they get steeper to near vertical at the far right as you go past the best efficiency point. This says that at lower air flows the compressor speed is almost constant (the most part independent) of the air flow...but at higher air flow the compressor speed is no longer so, the line curves away.. This makes sense because there is a choke point where the air flow will not increase any more even when you increase the compressor speed. but anyway the point is It depends on if you are at full boost already (the scenario you Edis set) and if you are operating on the left side of the best efficiency point (which is advised!)..

[ozzimark]

Not trying to be picky here but my point is that increased volumetric efficiency due to a cooler engine has no power payback under a fuel-flow-limited formula. These engines have the potential to flow vastly more air than they can burn with the limited fuel available.

That's not entirely true. With a higher natural engine VE, the compressor will have to do less work to pump air into the engine to achieve the required Air/Fuel ratio, since it will need to produce a lower amount of boost at a given load point. This will result in more MGU-H energy harvesting, and an overall more powerful power unit.

[Abarth]

What I tried to say, if the engine is running cooler higher vomuletric efficiency it is benificial in terms of power . .

Not trying to be picky here but my point is that increased volumetric efficiency due to a cooler engine has no power payback under a fuel-flow-limited formula. These engines have the potential to flow vastly more air than they can burn with the limited fuel available.

Two questions remain:
Do they encounter more engine knock problems if they do not cool down the charge air sufficiently? Or in other words, they can increase CR with cooler charge air.

How much more air flow would be useful? Methinks that at these high engine speeds, leaning the mixture over a certain limit will not work due to slow burning of lean mixtures. Fuel preheating would help though.

[PlatinumZealot]

Not just the inlet air temperature is important, the temperature just before compression is most important.
You can have very cool inlet air and there is some remaining hot hair left back from the exhaust in the cylinder undoing your cooling efforts. Designers try their best to evacuate the cylinder of exhaust air to prevent this. You can tune the intake and exhaust valve timing, the speed of the air as it flows past the valves, the exhaust back pressure, all in the aim of blowing out the hot exhaust gas before compression.