69 replies to this topic

[Wuzak]

I think we touched on this in another thread, but I want to explore it a little further.

In 2014 it is rumoured that turbocompounding will be allowed (why it won't for 2013 is beyond me).

It is likely that the compounding turbine will take the place of the second turbo they will be required to run in 2013.

WW2 era engines that had turbocompounding used mechanical supercharging, but this may be as much about control and reliability as suitability (turbochargers in WW2 were mainly used to compensate for altitude, and provided sea level air pressure to the main engine, with its integral supercharger).

I see the benefit of using a supercharger with a turbocompound turbine is that there is simplified plumbing and if the supercharger is driven through a variable speed drive it can provide good low end boost for torque. A fixed ratio drive would be simpler and lighter. I presume a centrifugal compressor would be used in a supercharged application.

The downside is that there are mechanical losses in the supercharger drive, as well as the drive from the turbine to the crank.

Would these losses be comparable to the losses in the turbocharger/power recovery turbine? I am thinking there would be some pressure losses in the piping and bypass systems.

What sort of engine speeds would the turbine power exceed the supercharger power?

For 2013 there is no question that the supercharger would be less efficient than the turbocharger, but when compounding is in the mix will that remain the case?

I suppose that this is all moot, as the turbocharging solution is likely to be fixed anyway.

[WhiteBlue]

The split of the new engine project into the core engine for 2013 and the advanced turbo solutions in 2014 is easily explained. There is a variety of solutions which are basically all untried for an F1 racing engine. To split the project is simply an efficient way of cost control and a clever way to stretch innovation over several years. I see at least five basic ways to do turbo compounding:

• The way it was done in the Wright engine in the Super Constellations with a supercharger and turbo compounder.
• The way Scania did it with a separate sequential turbo downstream from the turbocharger.
• A single turbocharger with attached high speed MGU.
• A single turbocharger with attached IVT connected to the crank shaft or the gear box.
• Separate turbine with attached high speed generator and separate compressor with attached electric motor.

It will obviously be a costly process to have the various solutions converge towards the best F1 solution. It makes sense IMO to get into that rat race only once the engines are sorted out.

[gruntguru]

A high efficiency positive displacement supercharger (eg screw or scroll) would offer zero lag throttle response without a large efficiency penalty. (Zero lag may have limited benefit as KERS could be used to fill holes in the engine response.)

There is a significant additional benefit if boost pressure can be maintained higher than exhaust back pressure under all conditions (not possible under transient conditions with a turbo.) The main benefit is knock resistance (allowing higher compression ratios) due to scavenging of hot residual combustion products and exhaust valve cooling.

Exhaust energy would be recovered with a turbine which could be better matched and more efficient since it would be dedicated to that task ie no need to match compressor speed.

[desmo]

gruntguru, on Jan 21 2011, 23:08, said:

A high efficiency positive displacement supercharger (eg screw or scroll) would offer zero lag throttle response without a large efficiency penalty. (Zero lag may have limited benefit as KERS could be used to fill holes in the engine response.)

How does this last bit not become traction control? Not that I care if it does.

[saudoso]

desmo, on Jan 22 2011, 18:13, said:

How does this last bit not become traction control? Not that I care if it does.

A lot of hi tech fantasies will be shatered when the 2013 rules hit reality head on...

[gruntguru]

desmo, on Jan 23 2011, 06:13, said:

How does this last bit not become traction control? Not that I care if it does.

"Filling-in" the power curve or the response curve (lag) with KERS is purely an powerplant function. Wheelspeed (wheelspin) is not sensed. To expand the meaning of those functions:

1. "Filling-in" the power curve. Topping up engine power to produce the desired power (torque) curve. This may remove "dips" or simply extend performance to the off-boost (lower) rev range.

2."Filling-in" the response curve. Adding power during rapid throttle opening to give "steady state boost equivalent power" during the period when boost is still catching up to the throttle setting (lag).

[WhiteBlue]

gruntguru, on Jan 22 2011, 06:08, said:

A high efficiency positive displacement supercharger (eg screw or scroll) would offer zero lag throttle response without a large efficiency penalty. (Zero lag may have limited benefit as KERS could be used to fill holes in the engine response.)

This could all be much easier be done by an asymmetric design of the turbine and the compressor plus the addition of a fast revving electric servo MGU. The turbine would be bigger than the compressor to extract additional energy for electricity generation at full throttle. In start up and overrun the MGU would spool up the compressor to supply optimum boost from the electrically stored energy. I'm pretty sure such an electric turbo compounder would still make a positive contribution to the balance of electric energy. The surplus can be used in the KERS boost.

[Wuzak]

WhiteBlue, on Jan 23 2011, 11:13, said:

This could all be much easier be done by an asymmetric design of the turbine and the compressor plus the addition of a fast revving electric servo MGU. The turbine would be bigger than the compressor to extract additional energy for electricity generation at full throttle. In start up and overrun the MGU would spool up the compressor to supply optimum boost from the electrically stored energy. I'm pretty sure such an electric turbo compounder would still make a positive contribution to the balance of electric energy. The surplus can be used in the KERS boost.

Won't your way be heavier and less efficient?

Edited by Wuzak, 23 January 2011 - 11:05.

[WhiteBlue]

Wuzak, on Jan 23 2011, 12:04, said:

Won't your way be heavier and less efficient?

It depends of what you compare it to. Another option is to replace the MGU by a Torotrak IVT. It would do absolutely the same. Take power from the engine to support the boost or take power from the turbine to the engine to prevent over boosting. I think both systems would have an efficiency of above 80% for the full energy circle of boost support and energy generation. So it would depend of the weight which of the two systems is better. I'm pretty convinced both systems would be more efficient than using separate units for compressing with an electric motor and expanding with a generator on separate shafts. By use of a common MGU you avoid one more shaft, inverter and MGU.

[Wuzak]

WhiteBlue, on Jan 23 2011, 14:39, said:

It depends of what you compare it to. Another option is to replace the MGU by a Torotrak IVT. It would do absolutely the same. Take power from the engine to support the boost or take power from the turbine to the engine to prevent over boosting. I think both systems would have an efficiency of above 80% for the full energy circle of boost support and energy generation. So it would depend of the weight which of the two systems is better. I'm pretty convinced both systems would be more efficient than using separate units for compressing with an electric motor and expanding with a generator on separate shafts. By use of a common MGU you avoid one more shaft, inverter and MGU.

Grunt, I'm sure, was talking of a supercharger directly geared to the crank, with no variable speed, and similarly for the power recovery turbine. The supercharegr and turbine would be optomised around the normal operating conditions of the engine.

[WhiteBlue]

Wuzak, on Jan 23 2011, 13:10, said:

Grunt, I'm sure, was talking of a supercharger directly geared to the crank, with no variable speed, and similarly for the power recovery turbine. The supercharegr and turbine would be optomised around the normal operating conditions of the engine.

Yes, I do understand this but I'm not sure that such a fixed geared solution would have the highest efficiency. The systems which I described can be very effectively controlled by digital algorithms and multi dimensional maps. I think that is the way to go. Relatively simple mechanical systems with high software sophistication. If you look at aircraft the current crop of high performance fighters use unstable aerodynamic configurations that can only be made to work by fly by wire computer control. I think similar trends will apply to racing engines. Throttle less engine management and cascaded integrated control of the ICE and the exhaust turbine should become the standard.

[MatsNorway]

WhiteBlue, on Jan 23 2011, 14:52, said:

Yes, I do understand this but I'm not sure that such a fixed geared solution would have the highest efficiency.

It will be the highest at a spesific point. Overall more efficient in the relevant powerband is more arguable. Personally i don`t believe in electric bizniz to push the charger. How much power does the charger on a turbo use? does anyone know?

Got some other figures then? like normal liters of air pr min for instance.

Personally i would think a centrifugal charger would be the most efficient over a screw or roots. At least the industry variants is that way.

Variable vanes and so on could pull of a trick or two as well.

[Magoo]

http://www.enginehis...ht/TC Facts.pdf

[Tony Matthews]

Very interesting, thanks.

[gruntguru]

Very interesting indeed - a good explanation of blowdown energy etc.

When reading this article, don't be confused by the frequent mention of exhaust turbochargers NOT utilising blowdown energy. Almost all race turbo applications these days are designed to utilise bowdown energy. This will definitely be the case for all the new crop of engines since blowdown energy is (mostly) "free" and will improve engine efficiency and thus power.

More interesting facts from the article. Additional power recovered from the blowdown turbines - 13% to 18%. Of course current turbo race engines would already be recovering some of that but expect it to improve when axial turbines and compounding arrive.

There will be some arguments on this one but - I predict the new engines will be running mixtures around 1.1 (10% excess air) at full power!

[Wuzak]

gruntguru, on Jan 24 2011, 03:55, said:

Very interesting indeed - a good explanation of blowdown energy etc.

Yes, it was a very good explanation, and an interesting article.

gruntguru, on Jan 24 2011, 03:55, said:

When reading this article, don't be confused by the frequent mention of exhaust turbochargers NOT utilising blowdown energy. Almost all race turbo applications these days are designed to utilise bowdown energy. This will definitely be the case for all the new crop of engines since blowdown energy is (mostly) "free" and will improve engine efficiency and thus power.

So little effect on back pressure?

If you set blowdoen turbines into the exhausts of the current engines, would they need significant modifications? The Wright article suggests that the blow down turbine solution has only a minor effcet on the cylinder operating conditions.

gruntguru, on Jan 24 2011, 03:55, said:

More interesting facts from the article. Additional power recovered from the blowdown turbines - 13% to 18%. Of course current turbo race engines would already be recovering some of that but expect it to improve when axial turbines and compounding arrive.

If turbine development is allowed. And I doubt that is what they want.

Previously you doubted the value of axial flow turbines.

gruntguru, on Dec 14 2010, 07:37, said:

An axial turbine would need to be multi-stage to achieve the required pressure ratio. The compressor would remain centrifugal since similar efficiencies are achievable in a compact, single stage unit. You neglected to mention the main drawback of the axial turbine in an F1 application - bulk.

For the uninitiated, pressure ratio in turbo machines, is largely governed by contact length - the length of the gasflow path which is in contact with the runner (rotating part). Clearly a radial-flow wheel has an inherently longer contact length than axial-flow.

From What will be in the 2013 F1 engines?

The Wright turbines were impulse turbines, which I didn't know before. In the article it suggests that pressure turbines (reaction turbines?) could be placed downstream of the impulse turbines for additional power recovery without any more influence on the cylinder conditions. Would this be worth persuing?

[gruntguru]

Wuzak, on Jan 24 2011, 10:31, said:

So little effect on back pressure?

Very little if you only take the blowdown energy

Wuzak, on Jan 24 2011, 10:31, said:

If you set blowdown turbines into the exhausts of the current engines, would they need significant modifications? The Wright article suggests that the blow down turbine solution has only a minor effect on the cylinder operating conditions.

The Wright experience (little modification required) would apply to most engines.

Wuzak, on Jan 24 2011, 10:31, said:

If turbine development is allowed. And I doubt that is what they want.

Only turbines in the sense of "exhaust" turbines.

Wuzak, on Jan 24 2011, 10:31, said:

Previously you doubted the value of axial flow turbines.

The value of axial flow is higher efficiency. I know radial flow turbines have improved but don't know how far behind they remain. It will be interesting to see what happens. Axial flow is a possibility in 2012 and probable in 2013 when exhaust energy extraction becomes more important.

Wuzak, on Jan 24 2011, 10:31, said:

The Wright turbines were impulse turbines, which I didn't know before. In the article it suggests that pressure turbines (reaction turbines?) could be placed downstream of the impulse turbines for additional power recovery without any more influence on the cylinder conditions. Would this be worth persuing?

Downstream reaction turbines would effect the backpressure on the engine. They would also reduce the blowdown energy available in the impulse turbines.

[Wuzak]

gruntguru, on Jan 24 2011, 07:05, said:

Only turbines in the sense of "exhaust" turbines.

That's what I meant.

For 2013 it is likely that the turbos will be a common unit, or at least heavily restricted. So, I'm not sure how keen they will be on the development of turbochargers and power recovery turbines.

gruntguru, on Jan 24 2011, 07:05, said:

The value of axial flow is higher efficiency. I know radial flow turbines have improved but don't know how far behind they remain. It will be interesting to see what happens. Axial flow is a possibility in 2012 and probable in 2013 when exhaust energy extraction becomes more important.

Downstream reaction turbines would effect the backpressure on the engine. They would also reduce the blowdown energy available in the impulse turbines.

The question then becomes how much energy can you extract from the exhaust, and will that be more than the penalty due to back presure.

[gruntguru]

Wuzak, on Jan 24 2011, 13:54, said:

That's what I meant.
For 2013 it is likely that the turbos will be a common unit, or at least heavily restricted. So, I'm not sure how keen they will be on the development of turbochargers and power recovery turbines.

Yes although I can't see a problem with using an axial flow turbine on the turbo. Advantage would be reduced backpressure so increased efficiency (lower pumping losses - possibly even negative)

Wuzak, on Jan 24 2011, 13:54, said:

The question then becomes how much energy can you extract from the exhaust, and will that be more than the penalty due to back presure.

That will be interesting. For one thng, the higher the boost and the lower the compression (expansion) ratio, the more blowdown energy there is in the exhaust.

[Wuzak]

gruntguru, on Jan 24 2011, 08:44, said:

That will be interesting. For one thng, the higher the boost and the lower the compression (expansion) ratio, the more blowdown energy there is in the exhaust.

I think boost is likely to be limited. So lowish boost levels and high CRs.

[PJGD]

Most turbocharged automotive engines try to make use of the pulse energy in the exhaust, but I think it is easier and thus more efficient with 3-cylinder engines than with 4-cylinders because there is no overlap (in a normal engine) between pulses. That is why 6-cylinder engines use twin-scroll turbine housings.

To complement Magoo's link to the Wright Turbo-Compound information, Click here is another approach. The Wright TC approach applied pulse energy capture to an otherwise conventional piston engine and extracted a modest benefit. This link is a brief review of a paper by Sir Harry Ricardo in which he proposes a more fundamental concept that was subsequently taken up by Napier as the Nomad. In the Nomad, the compressor was a 12 stage axial unit giving a pressure ratio of 8.25:1 in single-shaft format with the 3 stage turbine. That might be OK for a near fixed speed engine, but two-shaft solutions are obviously better for automotive applications. I am not promoting this for F1, but there is a certain logic to Ricardo's argument if energy efficiency is important.

PJGD

[Magoo]

gruntguru, on Jan 23 2011, 18:55, said:

Very interesting indeed - a good explanation of blowdown energy etc.

When reading this article, don't be confused by the frequent mention of exhaust turbochargers NOT utilising blowdown energy. Almost all race turbo applications these days are designed to utilise bowdown energy.

Sure, and then some.

gruntguru, on Jan 23 2011, 18:55, said:

More interesting facts from the article. Additional power recovered from the blowdown turbines - 13% to 18%.

At high constant load.

[gruntguru]

Magoo, on Jan 26 2011, 12:00, said:

At high constant load.

18% at takeoff rating. 14% at cruise (50% power)

Another interesting fact. With compounding the cruise ignition timing is 5 deg less advanced. Clearly the small loss in BMEP is compensated by an increase in blowdown energy (higher cylinder pressure at EVO) and additional recovery in the turbines.

Edited by gruntguru, 26 January 2011 - 06:26.

[Wuzak]

PJGD, on Jan 25 2011, 05:15, said:

To complement Magoo's link to the Wright Turbo-Compound information, Click here is another approach. The Wright TC approach applied pulse energy capture to an otherwise conventional piston engine and extracted a modest benefit. This link is a brief review of a paper by Sir Harry Ricardo in which he proposes a more fundamental concept that was subsequently taken up by Napier as the Nomad. In the Nomad, the compressor was a 12 stage axial unit giving a pressure ratio of 8.25:1 in single-shaft format with the 3 stage turbine. That might be OK for a near fixed speed engine, but two-shaft solutions are obviously better for automotive applications. I am not promoting this for F1, but there is a certain logic to Ricardo's argument if energy efficiency is important.

Ricardo's proposal is the piston engine becomes a gas generator for the turbine. The piston engine turnes the blower/compressor and the fuel is burned in its combustion chambers, but the turbine supplies most of the output power, if not all.

Some of the late WW2 proposals for the Crecy also revolved around it being a gas generator for a turbine.

In the Nomad MkI the Diesel engine drove a propellor, while the turbine drove a second, coaxial propellor. For teh MkII the turbine fed power back to the crank of the Diesel engine. I guess the MkI is close to teh desciption Ricardo put forward.

I don't know why a single shaft compunding turbine is not suitable for automotive use. It is different for turboshafts - the power turbine is separate from the gas generation as it would for a turbocompound engine. Adding an extra shaft just adds to complications.

[Wuzak]

Here's some specs for the Allison V-1710-127(-E27) from Dan Whitney, Vees for Victory - The Story of the Allison V-1710 Aircraft Engine 1929-1948

Model				   V1710-117 (E27)			
Takeoff power				2825 hp		2108 kW
@ rpm						3000 rpm		
Takeoff MAP				   100 inHgA	 3375 kPa

Military rating			  2430 hp		1813 kW
@ rpm						3200 rpm		
@ altitude				  17000 ft		5182 m
MAP							85 inHgA		2869 kPa

War Emergency rating		 2980 hp		2223 kW
@ rpm						3200 rpm		
@ altitude				  11000 ft		3353 m
MAP						   100 inHgA	 3375 kPa

Fuel Grade				115/145			

Compression Ratio		  6.00:1			
Supercharger OD			   9.5 in	   241.3 mm
Supercharger gear ratio	8.10:1			
Auxiliary S/C OD		  12.1875 in	   309.6 mm
Auxiliary S/C gear ratio   7.64:1			
Propellor Gear ratio	  2.478:1			

Dry weight				   1983 lb		 901 kg

Had they proceeded Allison was willing to guarantee the following performance from the turbocompound V-1710.

Normal rating				1740 hp		1298 kW
@ rpm						3000 rpm		
@ altitude				  33000 ft	   10058 m
MAP						  52.5 inHgA	 1772 kPa
				
Military rating			  2320 hp		1731 kW
@ rpm						3200 rpm		
@ altitude				  28000 ft		8534 m
MAP							85 inHgA	 2869 kPa
				
War Emergency rating		 3090 hp		2305 kW
@ rpm						3200 rpm		
@ altitude				  28000 ft		8534 m
MAP						   100 inHgA	 3375 kPa

Maximum cruise			   1340 hp		1000 kW
@ rpm						2700 rpm		
@ altitude				  26000 ft		7925 m
Specific Fuel Consumption   0.365 lb/hp/hr 0.222 kg/kW/hr

The turbine used for the prototype was a modified version from the C-series GE turbocharger - usually used for engines like the R-2800.

Exhaust temps at WER power were well above the safe rating for the turbine, and had it proceeded Allison wer going to develop air cooled blades.

The engine on which the -127 was based had a WER of 2200hp, and, IIRC, military power of 1600hp-1800hp. Unfortunatly I do not have my book with me to check.

Thought that this would make an interesting comparison to the Wright turbocompound. One difference is that the V-1710 is about half the capacity, but being water cooled was able to take much higher boost ratings. Also, it was a military engine.

[gruntguru]

Wuzak, on Jan 26 2011, 17:39, said:

Ricardo's proposal is the piston engine becomes a gas generator for the turbine. The piston engine turnes the blower/compressor and the fuel is burned in its combustion chambers, but the turbine supplies most of the output power, if not all.

I don't think I would put it that way. The piston engine produces about 40% of the shaft power and the turbine about 60%. How the outputs are connected and which machine drives the blower is not really relevant.

[Grumbles]

Wuzak, on Jan 26 2011, 08:39, said:

Ricardo's proposal is the piston engine becomes a gas generator for the turbine. The piston engine turnes the blower/compressor and the fuel is burned in its combustion chambers, but the turbine supplies most of the output power, if not all.

I remember someone (Duckworth?) proposed the use of a two-stroke engine purely as a gas generator many years ago. Or how about a Wankel as a light and compact gas generator? On second thoughts it wouldn't work; the turbine would jam up with bits of apex seal...

[Wuzak]

gruntguru, on Jan 27 2011, 01:55, said:

I don't think I would put it that way. The piston engine produces about 40% of the shaft power and the turbine about 60%. How the outputs are connected and which machine drives the blower is not really relevant.

From Ricardo Proposal for Turbo-Compounding (1946):

Ricardo said

the combined plant will consist of a blower, preferably of the axial flow type with, perhaps, a centrifugal for the last stage, a very small and simple piston engine in place of the usual combustion chamber, and a gas turbine. Of these three components the piston engine will deliver, and the blower will consume, about the same power, while the turbine will deliver about 50 per cent more than either.

That says to me that the piston engine is used to power the compressor.

[Wuzak]

Grumbles, on Jan 27 2011, 03:27, said:

I remember someone (Duckworth?) proposed the use of a two-stroke engine purely as a gas generator many years ago. Or how about a Wankel as a light and compact gas generator? On second thoughts it wouldn't work; the turbine would jam up with bits of apex seal...

I believe that Duckworth began working on a turbocompound engine when Ford asked him to design a turbo engine in the early '80s. I understand that he actually had a single cylinder development engine running.

I know that during WW2 some German companies were working with the idea. I seem to recall a small X-16 being used as a gas generator for the turbine, and I definitely remember a swing piston design which was to be used as a gas generator.

Campini came up with a dufferent solution.

The piston motor drove a compressor, which fed a combustion chamber, the exhaust being used as a jet.

[gruntguru]

Wuzak, on Jan 27 2011, 10:43, said:

That says to me that the piston engine is used to power the compressor.

Not really he's just comparing the power produced/consumed by each component. It doesn't really matter what drives what. Ideally the power produced by each component would be summed, and the power required to drive the compressor would be subtracted from the total. That would allow the percentages to be flexible to suit varying operating conditions. The ultimate flexibility would probably be the use of MGU's as suggested by Whiteblue. This would allow each component to operate at the optimum speed and torque for each operating point.

[gruntguru]

Wuzak, on Jan 27 2011, 10:56, said:

I believe that Duckworth began working on a turbocompound engine when Ford asked him to design a turbo engine in the early '80s. I understand that he actually had a single cylinder development engine running.

I know that during WW2 some German companies were working with the idea. I seem to recall a small X-16 being used as a gas generator for the turbine, and I definitely remember a swing piston design which was to be used as a gas generator.

If the piston engine is to be purely a gas generator you would use a "free piston engine". Wikipedia

[Wuzak]

gruntguru, on Jan 27 2011, 04:37, said:

Not really he's just comparing the power produced/consumed by each component. It doesn't really matter what drives what. Ideally the power produced by each component would be summed, and the power required to drive the compressor would be subtracted from the total.

So, the engine makes X hp, the compressor consumes X hp, and the turbine produces X/2 hp (according to Ricardo).

So, X-X+X/2 = X/2.

Ricardo's proposal is for a piston engine to act as a gas generator to drive a turbine.

gruntguru, on Jan 27 2011, 04:37, said:

The ultimate flexibility would probably be the use of MGU's as suggested by Whiteblue. This would allow each component to operate at the optimum speed and torque for each operating point.

Not sure that MGU's were available to Ricardo, or that it was desirable in weight terms for aircraft.

That would provide the flexibility and optimisation, but one wonders how much heavier they would be for an F1 engine, considering that the normal operating range is quite narrow comapred to road cars.

[Wuzak]

gruntguru, on Jan 27 2011, 04:43, said:

If the piston engine is to be purely a gas generator you would use a "free piston engine". Wikipedia

Gas Generators

Quote

After the success of the free-piston air compressor, a number of industrial research groups started the development of free-piston gas generators. In these engines there is no load device coupled to the engine itself, but the power is extracted from an exhaust turbine. (The only load for the engine is the supercharging of the inlet air.)

[gruntguru]

Wuzak, on Jan 27 2011, 11:45, said:

So, the engine makes X hp, the compressor consumes X hp, and the turbine produces X/2 hp (according to Ricardo).

The Turbine produces 1.5X actually.

[Wuzak]

gruntguru, on Jan 27 2011, 04:55, said:

The Turbine produces 1.5X actually.

Right..."50% more"

[Wuzak]

gruntguru, on Jan 22 2011, 09:08, said:

A high efficiency positive displacement supercharger (eg screw or scroll) would offer zero lag throttle response without a large efficiency penalty. (Zero lag may have limited benefit as KERS could be used to fill holes in the engine response.)

There is a significant additional benefit if boost pressure can be maintained higher than exhaust back pressure under all conditions (not possible under transient conditions with a turbo.) The main benefit is knock resistance (allowing higher compression ratios) due to scavenging of hot residual combustion products and exhaust valve cooling.

Exhaust energy would be recovered with a turbine which could be better matched and more efficient since it would be dedicated to that task ie no need to match compressor speed.

gruntguru, on Jan 27 2011, 05:37, said:

The ultimate flexibility would probably be the use of MGU's as suggested by Whiteblue. This would allow each component to operate at the optimum speed and torque for each operating point.

What about using an axial flow compressor driven by an electric motor? Or even a cvt drive to the compressor?

The compressor efficiency should be better than a centrifugal compressor, or the positive displacement types, and therefore would need less power to drive and the temperature rise should be less.

Any chance that the temperature rise in the compressor will be low enough to negate the need for an intercooler?

[MatsNorway]

Wuzak, on Jan 27 2011, 07:47, said:

What about using an axial flow compressor driven by an electric motor? Or even a cvt drive to the compressor?

The compressor efficiency should be better than a centrifugal compressor, or the positive displacement types, and therefore would need less power to drive and the temperature rise should be less.

Any chance that the temperature rise in the compressor will be low enough to negate the need for an intercooler?

Got a source that states that a axial compressors got better efficiency up to 2-4bars? how many stages would it need? two?
Got a source about axial compressors at all. other than airplane motors.
Just want some easy reading.

I just see the centrifuge to be a easy and simple and efficient compressor in that pressure range.

[Villes Gilleneuve]

Wuzak, on Jan 27 2011, 06:47, said:

What about using an axial flow compressor driven by an electric motor?

With electric from KERS?

[Wuzak]

Villes Gilleneuve, on Jan 27 2011, 23:39, said:

With electric from KERS?

I think you'd be better off with fixed gearing to the supercharger than rob energy from KERS. KERS will become more and more significant from 2013, both in energy available to use and power permitted.

If you used an electric motor for the supercharger drive then you'd probably need some sort of generator on the power recovery turbine.

WB suggests just having a turbo with an oversize turbine, with an MGU between them. That way the disadvantage of having a larger turbine is offset by the MGU being able to spin it up to keep the responsiveness. When there the MGU is generating there will need to be some stored for use later, and some fed back to teh motor - presumably through the KERS motor. Which would then become complicated to police.

I think that there is more benefit to be had if you can control the turbine and compressor separately. To use electric drive you would need the motor for the compressor, a generator on the power recovery turbine, and a storage device. For the turbine I can see you do it two ways - have the turbine drive the generator, whose power is then fed to the storage device or the KERS motor to increase the power of the engine, or have the turbine directly geared back to the engine with a generator on the drive shaft between them. I suppose you could use the KERS motor to generate when it wasn't being used for braking energy recovery or deploying KERS power.

Alternatively you could run a CVT between the engine and compressor with, say, a variation of 2:1. At 6000rpm you could potentially pump as much air in as at 12000rpm. The turbine would be directly geared to the engine, and would use variable vane/geometry turbines.

[gruntguru]

Wuzak, on Jan 27 2011, 16:47, said:

What about using an axial flow compressor driven by an electric motor? Or even a cvt drive to the compressor?

Then you need a generator connected to the turbine. Easier to connect the turbine and compressor mechanically, use a single MGU to import or export energy as required, plus VNT to juggle the operating points of compressor vs turbine (since speed is a shared parameter)

Wuzak, on Jan 27 2011, 16:47, said:

The compressor efficiency should be better than a centrifugal compressor, or the positive displacement types, and therefore would need less power to drive and the temperature rise should be less.

Not much difference in efficiency. For high pressure ratios the axial loses on cost and bulk.

Wuzak, on Jan 27 2011, 16:47, said:

Any chance that the temperature rise in the compressor will be low enough to negate the need for an intercooler?

No. Even a 100% efficient (isentropic) compressor has a significant temperature increase. One advantage of the axial though - its multi stage nature means intercooling between stages is possible. This will reduce the compressor work requirement but I doubt the gains would justify the complexity, cost, mass, bulk etc

Edited by gruntguru, 28 January 2011 - 00:05.

[Wuzak]

gruntguru, on Jan 28 2011, 03:05, said:

Not much difference in efficiency. For high pressure ratios the axial loses on cost and bulk.

Bulk?



A 4 stage 550 CFM Axial Flow compressor.


I forgot that from 2013 F1 is becoming cheapF1.

[Wuzak]

gruntguru, on Jan 28 2011, 03:05, said:

Then you need a generator connected to the turbine. Easier to connect the turbine and compressor mechanically, use a single MGU to import or export energy as required, plus VNT to juggle the operating points of compressor vs turbine (since speed is a shared parameter)

With the aim of effciency, would it be better to control the speeds of the supercharger and power recovery turbine separately?

gruntguru, on Jan 28 2011, 03:05, said:

No. Even a 100% efficient (isentropic) compressor has a significant temperature increase. One advantage of the axial though - its multi stage nature means intercooling between stages is possible. This will reduce the compressor work requirement but I doubt the gains would justify the complexity, cost, mass, bulk etc

What sort of inlet temperature could the engine cope with?

[gruntguru]

Wuzak, on Jan 28 2011, 10:24, said:

With the aim of effciency, would it be better to control the speeds of the supercharger and power recovery turbine separately?

Probably, but not much advantage when the two machines are correctly matched (they have almost identical mass flow) and the VNT allows a fair bit of balancing.

Wuzak, on Jan 28 2011, 10:24, said:

What sort of inlet temperature could the engine cope with?

Its more a question of increasing air density which increases mass-flow through the engine. Increasing detonation tendency is another consequence of raising inlet air temp.

[Wuzak]

Another alternative is to have something like this:

Antanov 2 speed supercharger

That way the boost can be achieved at low rpm as well as high rpm.

[gruntguru]

Wuzak, on Jan 28 2011, 10:18, said:

Bulk? A 4 stage 550 CFM Axial Flow compressor.

Do you know the max pressure ratio for that compressor? A 4 stage unit is unlikely to produce the pressures required.

Wuzak, on Jan 28 2011, 10:18, said:

I forgot that from 2013 F1 is becoming cheapF1.

Price is always a consideration. If two solutions are comparable in every other respect you select on price.

[Wuzak]

gruntguru, on Jan 28 2011, 04:32, said:

Do you know the max pressure ratio for that compressor? A 4 stage unit is unlikely to produce the pressures required.

Nope. Unfortunately they don't seem to show that information.

[Tony Matthews]

gruntguru, on Jan 28 2011, 01:32, said:

Price is always a consideration. If two solutions are comparable in every other respect you select on price.

In F1 this has tended to be the higher price!

[MatsNorway]

gruntguru, on Jan 28 2011, 02:32, said:

Do you know the max pressure ratio for that compressor? A 4 stage unit is unlikely to produce the pressures required.

Well designed and constructed Axial compressors can have adiabatic efficiencies of up to 80% if the path and the blades are well designed. However they can only develop pressure ratios of around 1.1 to 1.4 per stage so that's why this device has four stages.. Given this isn't General Electric lets say its 1.2 PR then the max pressure it can develop is twice ambient pressure (1.2^4) at ideal speeds. Now this device is an axial flow compressor which, because of the drive shaft turns the flow to radial at the end of the unit. That's going to increase pressure and reduce flow at the final stage but reduce the over all PR.

Centrifugal compressors can get around 2.8 pressure ratios at 85% which is why they are used so often in engine boost applications. Per volume of air and similar PR's a centrifugal compressor will take less power to turn than an equivalent multistage axial.

Edited by MatsNorway, 28 January 2011 - 11:25.

[Wuzak]

MatsNorway, on Jan 28 2011, 15:25, said:

Well designed and constructed Axial compressors can have adiabatic efficiencies of up to 80% if the path and the blades are well designed. However they can only develop pressure ratios of around 1.1 to 1.4 per stage so that's why this device has four stages.. Given this isn't General Electric lets say its 1.2 PR then the max pressure it can develop is twice ambient pressure (1.2^4) at ideal speeds. Now this device is an axial flow compressor which, because of the drive shaft turns the flow to radial at the end of the unit. That's going to increase pressure and reduce flow at the final stage but reduce the over all PR.

Centrifugal compressors can get around 2.8 pressure ratios at 85% which is why they are used so often in engine boost applications. Per volume of air and similar PR's a centrifugal compressor will take less power to turn than an equivalent multistage axial.

Interesting that I couldn't find a comressor better than 80% efficiency in the Garrett catalogue. Actually the best I could see at a quick glance was 78 or 79%.

A pressure ratio of 4 would be more efficient using a 2 stage centrifugal supercharger as well.

I though that the main reason why centrifugal compressors are used is that they are cheaper and easier to make, and they can get good pressure ratios in one stage.

The reference to the "drive shaft turns the flow to radial" I guess means that it is a vortex, liek a cyclone, and not radial, which is what the centrifugal compressor does. In most gas turbines that is taken care of by guid vanes which straighten the flow, usually between each stage (thereby maximising each stage's efficiency). In smaller GTs the axial flow compressors are followed by a centrifugal stage. The twisting effect can also be lessened by having each alternate compressor stage counter-rotate, but that is more complicated, but takes awa the need for guide vanes and leads to better efficiencies.

Edited by Wuzak, 28 January 2011 - 13:04.

[MatsNorway]

Wuzak, on Jan 28 2011, 13:53, said:

Interesting that I couldn't find a comressor better than 80% efficiency in the Garrett catalogue. Actually the best I could see at a quick glance was 78 or 79%.

A pressure ratio of 4 would be more efficient using a 2 stage centrifugal supercharger as well.

Cool. The post is not a good source.
Was more to check the responses. Besides i don`t believe in anything else than centrifuge before i get some good info to read.
Possibly a screw type or something at low engine rpms might be efficienct but i don`t know so much about engines and you guys mostly ignore my posts so..

How big does the garrets go? They guy i quoted might be talking about bigger units.

Edited by MatsNorway, 28 January 2011 - 13:05.