Jump to content


Photo
- - - - -

Mercedes-AMG achieves 50% thermal efficiency on 2016 F1 engine


  • Please log in to reply
111 replies to this topic

#1 TDIMeister

TDIMeister
  • Member

  • 286 posts
  • Joined: January 06

Posted 26 February 2016 - 17:48

http://www.enginetec...hp?NewsID=77881

 

http://viewer.zmags....3d#/c55fc83d/34

 

Discuss!



Advertisement

#2 gruntguru

gruntguru
  • Member

  • 6,733 posts
  • Joined: January 09

Posted 27 February 2016 - 08:02

The wording in both articles is ambiguous.

"the ICE was running at 45% efficiency, which was boosted to 50% when the energy recovery system was factored in."

"a total output of 750ps"

 

Overall it sounds to me like a thermal efficiency of 45% which is consistent with the 750ps figure. To an engineer the word "efficiency" as applied to a heat engine should only refer to a steady state "work-out/heat-in" relationship. Heat-in should be the 1240 kW (1650 hp) available from the fuel. Work-out should only include power available at the crankshaft plus the MGUH (turbine minus compressor power). To claim "50% TE" (those words never appear in either article) the compound engine should produce a sustained 620 kW (825 hp) on the dyno.

 

Either way, it is a remarkable achievement. Toyota claimed 42% for a development Prius engine (turbo but not compounded) so 45% is totally believable, 50% - incredible if true.


Edited by gruntguru, 27 February 2016 - 08:20.


#3 Wuzak

Wuzak
  • Member

  • 6,894 posts
  • Joined: September 00

Posted 28 February 2016 - 06:13

The wording in both articles is ambiguous.

"the ICE was running at 45% efficiency, which was boosted to 50% when the energy recovery system was factored in."

"a total output of 750ps"

 

Overall it sounds to me like a thermal efficiency of 45% which is consistent with the 750ps figure. To an engineer the word "efficiency" as applied to a heat engine should only refer to a steady state "work-out/heat-in" relationship. Heat-in should be the 1240 kW (1650 hp) available from the fuel. Work-out should only include power available at the crankshaft plus the MGUH (turbine minus compressor power). To claim "50% TE" (those words never appear in either article) the compound engine should produce a sustained 620 kW (825 hp) on the dyno.

 

Either way, it is a remarkable achievement. Toyota claimed 42% for a development Prius engine (turbo but not compounded) so 45% is totally believable, 50% - incredible if true.

 

 

These numbers fit with what Cowell has said comparing them to the V8, V8 + KERS and V10.

 

45% TE gives 747hp @ 45% efficiency. About the power of the V8s.

 

50% TE gives 831hp, roughly the same as the V8 plus KERS. Note that the ERS is giving just over 80hp, which presumably could be had continuously if you were running the engine a the peak for long periods.

 

That leaves another 70-odd hp which can be used from the ES to bring the total over 900hp.

 

If 45% for the combustion engine without compounding is believable, then surely 50% with compounding is too?

 

If talked with some other people about this and they do not believe the numbers. They suggest non-compounded power to be closer to 600hp than 700hp. Their argument is that the efficiency numbers are equivalent to those of large, slow revving marine Diesels. 



#4 saudoso

saudoso
  • Member

  • 6,742 posts
  • Joined: March 04

Posted 28 February 2016 - 12:36

The art of the spin doctor. Make you believe he said something when he actually didn't. Make you believ in something that really isn't exactly like that.

 

Cowell and Symonds should quit F1,  team up and start a public relations business.


Edited by saudoso, 28 February 2016 - 12:43.


#5 gruntguru

gruntguru
  • Member

  • 6,733 posts
  • Joined: January 09

Posted 28 February 2016 - 23:49

These numbers fit with what Cowell has said comparing them to the V8, V8 + KERS and V10.

 

45% TE gives 747hp @ 45% efficiency. About the power of the V8s.

 

50% TE gives 831hp, roughly the same as the V8 plus KERS. Note that the ERS is giving just over 80hp, which presumably could be had continuously if you were running the engine a the peak for long periods.

 

That leaves another 70-odd hp which can be used from the ES to bring the total over 900hp.

 

If 45% for the combustion engine without compounding is believable, then surely 50% with compounding is too?

 

If talked with some other people about this and they do not believe the numbers. They suggest non-compounded power to be closer to 600hp than 700hp. Their argument is that the efficiency numbers are equivalent to those of large, slow revving marine Diesels. 

 

So what do you believe the numbers are for:

 - Crankshaft only in self sustaining mode

 - Compound (crank + MGUH) in self sustaining mode.

 - Self sustaining + ES (high efficiency max output)

 - Qualifying mode. (Low efficiency max output)

 

Your friends comparisons to marine Diesels are actually irrelevant. There is no question that F1 is setting new standards for SI efficiency.



#6 Wuzak

Wuzak
  • Member

  • 6,894 posts
  • Joined: September 00

Posted 29 February 2016 - 01:12

So what do you believe the numbers are for:

 - Crankshaft only in self sustaining mode 750hp

 - Compound (crank + MGUH) in self sustaining mode.830hp

 - Self sustaining + ES (high efficiency max output) 910hp

 - Qualifying mode. (Low efficiency max output) Maybe 920-930hp

 

Your friends comparisons to marine Diesels are actually irrelevant. There is no question that F1 is setting new standards for SI efficiency.

 

The first two equate to 45% and 50% efficiency.

 

Those are the numbers for last year's PU. This year's has even higher efficiencies.



#7 Kelpiecross

Kelpiecross
  • Member

  • 1,456 posts
  • Joined: October 10

Posted 29 February 2016 - 04:32

No question etc. F1 SI efficiency?    What would be the TE of a "bare" current F1 engine without the turbo-compounding etc.?     About the same as any modern  road-going engine I would think.   Why would the "bare"  F1 engine have a better TE than an "ordinary" engine?   The F1 engine presumably doesn't have a very high CR?  


Edited by Kelpiecross, 29 February 2016 - 04:33.


#8 gruntguru

gruntguru
  • Member

  • 6,733 posts
  • Joined: January 09

Posted 29 February 2016 - 08:12

The first two equate to 45% and 50% efficiency.

 

Those are the numbers for last year's PU. This year's has even higher efficiencies.

Surely Cowell is only claiming 50% for the 2016 PU?



#9 gruntguru

gruntguru
  • Member

  • 6,733 posts
  • Joined: January 09

Posted 29 February 2016 - 08:21

No question etc. F1 SI efficiency?

 

 

What would be the TE of a "bare" current F1 engine without the turbo-compounding etc.? Wuzak says 45% and I don't disagree.

 

About the same as any modern  road-going engine I would think. No, I would expect better than any road - going engine.

 

Why would the "bare"  F1 engine have a better TE than an "ordinary" engine?  What do these things cost? How long do they have to live? What operating band is required? (Hint. Road-going engines spend most of their time at light loads). What is the over-riding design goal? (Answer = TE). Do F1 engines need to meet emissions regulations? Toyota are testing engines designed for mass production (ie non exotic) with 42% TE - no compounding.

 

The F1 engine presumably doesn't have a very high CR? I would guess around 12:1. Is that "very high"?


Edited by gruntguru, 05 March 2016 - 02:33.


#10 Wuzak

Wuzak
  • Member

  • 6,894 posts
  • Joined: September 00

Posted 01 March 2016 - 00:12

The F1 engine presumably doesn't have a very high CR? I would guess around 12:1. Is that "very high"?

 

12:1 CR with around 3.5bar MAP. I think that would be considered quite high compression.



#11 Wuzak

Wuzak
  • Member

  • 6,894 posts
  • Joined: September 00

Posted 01 March 2016 - 00:25

Surely Cowell is only claiming 50% for the 2016 PU?

 

The (approximately speaking) 750hp/830hp/910hp was stated last year by Cowell, maybe before the end of the season. That equates to 45% crankshaft efficiency and 50% in compounding mode.

 

I have seen it suggested that the 2016 PU could be going up to ~47% efficiency for 2016. That would be 583kW/782hp crankshaft power, ~860-870hp compounding mode and ~940hp with full ERS deployment.



#12 gruntguru

gruntguru
  • Member

  • 6,733 posts
  • Joined: January 09

Posted 01 March 2016 - 01:32

I am a bit suspicious that they would quote a TE for crankshaft only. It is pretty meaningless when you have significant recovery of useful energy from the exhaust/MGUH ie the 830 hp is the real output of this engine.

 

12:1 compression. Yep that's high.



#13 Kelpiecross

Kelpiecross
  • Member

  • 1,456 posts
  • Joined: October 10

Posted 01 March 2016 - 03:17

GG - why did you suggest a 12:1  ratio?   As far as I can discover the fuel is essentially petrol - 12:1 would seem to be very.   I would have expected something less than 10:1.

 

 I think it is a bit simplistic to suggest that a 45% is achievable by spending a lot of money.  

 

Going back to basics - what is the origin of the 45% TE claim?    There is no way I can see to really determine the accuracy of this claim.

 

To jump from the accepted 30%  TE (or thereabouts on a very good day)   to 45% TE  is a. enormous leap  and,  if true,  would require some sort of truly fundamental change  in the way we understand how engines work  - yet the basic F1 engine is essentially conventional in design and layout. 

 

 I think it is generally accepted that turbo engines are a little more thermally efficient than NA engine - but not 15% better.

 

 Possibly this Cowell bloke  (who appears to be the sole source for this  45% claim)   is mistakenly including some of the recovered exhaust energy.

 

  I don't know for certain - but I doubt they would have tested the engine all that thoroughly (if at all)  in non-turbo NA form. The 45% is probably estimated.

 

  Of course it may all be true - spending a lot of money on a engine and not running it in low-speed traffic may be enough to defeat the laws of thermodynamics.     



#14 TDIMeister

TDIMeister
  • Member

  • 286 posts
  • Joined: January 06

Posted 01 March 2016 - 03:22

With the likely valve timings used for 15,000 RPM, the intake closure will be somewhere north of 60 deg. after BDC so the effective compression ratio will be something more sensible as seen in engines running 98 RON. Audi is already employing a compression ratio of 11.65:1 with early intake valve closure Miller cycle, as reported in the same issue of Engine Technology International linked above.

 

F1 doesn't have the luxury of breaking the laws of physics but it can certainly push the envelopes, which is what makes this exciting! I remember posting here years ago my wish for a fuel-limited formula but was pooh-poohed by many as making the racing boring.  :rolleyes:  :cool:


Edited by TDIMeister, 01 March 2016 - 03:28.


#15 Wuzak

Wuzak
  • Member

  • 6,894 posts
  • Joined: September 00

Posted 01 March 2016 - 04:03

Possibly this Cowell bloke  (who appears to be the sole source for this  45% claim)   is mistakenly including some of the recovered exhaust energy.  

 

Andy Cowell made the following claims last year:

 

The V6T without ERS is equivalent in power to the V8s. ie ~750hp.

The V6T with self sustaining ERS is equivalent in power to the V8s + KERS (80hp). ie ~830hp.

The V6T with full ERS power (ie MGUK is fed by MGUH + ES, or just the ES) is 10% more powerful than the V8 + KERS. ie. ~910hp.



#16 Wuzak

Wuzak
  • Member

  • 6,894 posts
  • Joined: September 00

Posted 01 March 2016 - 04:07

I don't know for certain - but I doubt they would have tested the engine all that thoroughly (if at all)  in non-turbo NA form. The 45% is probably estimated.

 

No, they wouldn't have tested the ICE without the turbo. However, there are modes where the MGUH does not feed power back to the MGUK, by either sending it to the ES or using the wastegates.

 

I doubt that 45% is estimated by MPHE. They know LHV/HHV of the fuel - which was stated as 1240kW for 100kg/h fuel flow rate last year.



#17 Wuzak

Wuzak
  • Member

  • 6,894 posts
  • Joined: September 00

Posted 01 March 2016 - 04:09

F1 doesn't have the luxury of breaking the laws of physics but it can certainly push the envelopes, which is what makes this exciting! I remember posting here years ago my wish for a fuel-limited formula but was pooh-poohed by many as making the racing boring.  :rolleyes:  :cool:

 

F1 at the moment has a fuel flow rate limit and a race fuel limit. The latter has more effect on the racing due to need for saving fuel.

 

There are some that still think that the fuel flow rate is responsible for fuel saving in the race, so blame it for boring racing.



#18 Kelpiecross

Kelpiecross
  • Member

  • 1,456 posts
  • Joined: October 10

Posted 01 March 2016 - 04:37

Woozy and GG - just for the record - do you believe the 45% TE figure to be true - without any qualms or slight doubts?

 

 Again for the record -  no I don't believe the 45% TE figure.    



#19 gruntguru

gruntguru
  • Member

  • 6,733 posts
  • Joined: January 09

Posted 01 March 2016 - 04:57

GG - why did you suggest a 12:1  ratio?   As far as I can discover the fuel is essentially petrol - 12:1 would seem to be very.   I would have expected something less than 10:1.

1. The fuel is highly developed for energy content and anti-knock properties.

2. Although the MAP is around 3.5 bar, much of it is excess air that doesn't get burned so doesn't contribute fully to peak cylinder pressure.

3. Ultra lean mixture reduces combustion temperature => reduces knock

4. Stratified charge reduces end-gas temperature and flame travel distance  => reduces knock

 

To jump from the accepted 30%  TE (or thereabouts on a very good day)   to 45% TE  is a. enormous leap  and,  if true,  would require some sort of truly fundamental change  in the way we understand how engines work  - yet the basic F1 engine is essentially conventional in design and layout.

Honda RA168e (1986) had 32% TE from a race engine running PFI and lambda 0.98. Last years Prius had 37%. Test bench Prius engines - up to 42%. 45% is not far from there.

 

 I think it is a bit simplistic to suggest that a 45% is achievable by spending a lot of money.

Don't forget - these guys started with a (relatively) clean sheet - to design an engine with the highest possible TE and last a few thousand km. The money allows the use of exotic materials, fuels and lubricants, complex designs, expensive manufacturing techniques etc etc

 

 

Woozy and GG - just for the record - do you believe the 45% TE figure to be true - without any qualms or slight doubts?

Yes. No doubts. . . .

Even the 50% claim is possible.


Edited by gruntguru, 01 March 2016 - 05:00.


Advertisement

#20 Wuzak

Wuzak
  • Member

  • 6,894 posts
  • Joined: September 00

Posted 01 March 2016 - 05:59

Yes, I would believe the TE numbers.



#21 Wuzak

Wuzak
  • Member

  • 6,894 posts
  • Joined: September 00

Posted 01 March 2016 - 06:04

Don't forget - these guys started with a (relatively) clean sheet - to design an engine with the highest possible TE and last a few thousand km. The money allows the use of exotic materials, fuels and lubricants, complex designs, expensive manufacturing techniques etc etc

 

And running in a small operating window.



#22 RogerGraham

RogerGraham
  • Member

  • 183 posts
  • Joined: October 12

Posted 01 March 2016 - 11:25

The V6T without ERS is equivalent in power to the V8s. ie ~750hp.

The V6T with self sustaining ERS is equivalent in power to the V8s + KERS (80hp). ie ~830hp.

 

Question from a dummy: is that ~80hp increase the amount of energy that can be continuously pumped into the energy store by the MGUH?  

 

If so, then surely that value should always be included in the definition of the overall TE (unless you're specifically rating the TE of just the ICE for whatever reason)?

(Given unlimited fuel flow, presumably you'd get a lot more than an additional ~80hp from traditional turbocharging.)



#23 TDIMeister

TDIMeister
  • Member

  • 286 posts
  • Joined: January 06

Posted 01 March 2016 - 15:35

1. The fuel is highly developed for energy content and anti-knock properties.

2. Although the MAP is around 3.5 bar, much of it is excess air that doesn't get burned so doesn't contribute fully to peak cylinder pressure.

3. Ultra lean mixture reduces combustion temperature => reduces knock

4. Stratified charge reduces end-gas temperature and flame travel distance  => reduces knock

 

Interesting theory about being stratified lean - do you have some inside knowledge to know this is a fact, or is it your speculation? It is certainly plausible, but let's straighten out a few things. Lean mixtures would actually promote knocking by reducing the mixture laminar flame speed and thus leaving increased residence time in the end-gas before being consumed by the flame front, thus causing it to spontaneously combust, leading to knock. I admit that it makes for interesting theory, but I have read nothing about knocking at 15000 RPM. I did schlieren imaging on the mixture formation and combustion process in an optically accessible engine at what I believed from the published literature to be unprecedented at 5000 RPM when most were done at 1200, 1500 and 2400 RPM.

 

Charge stratification is certainly one way to address the above. When I have more time I'll do some back of napkin calculations to figure out what the air mass flow of a 15000 RPM 2 litre engine with at 3.5 bar MAP and typical volumetric efficiency of an F1 engine is - we could use the last naturally aspirated engine figures as a rough guide. Then we can know much more accurately the global lambdas these engines are running.

 

I'm extremely hungry for more info like this - whether exotic fuel injection tricks, charge motion and stratification (subjects of my PhD dissertation) or advanced combustion modes. If anyone has articles or papers on the subject, I would be grateful if you would share it and I'll do the same.



#24 Ross Stonefeld

Ross Stonefeld
  • Member

  • 62,658 posts
  • Joined: August 99

Posted 01 March 2016 - 20:48

Are you using 15000rpm as potential as part of your engine calculations? Because the engines aren't actually getting anywhere near 15000 on the track. Aren't they stuck at around 12 because of the fuel flow limits?

#25 TDIMeister

TDIMeister
  • Member

  • 286 posts
  • Joined: January 06

Posted 01 March 2016 - 21:50

Are you using 15000rpm as potential as part of your engine calculations? Because the engines aren't actually getting anywhere near 15000 on the track. Aren't they stuck at around 12 because of the fuel flow limits?

Anyone?

 

As for the calculations, does anyone have any firm clues of the thermal efficiency of the last naturally aspirated V8s? And I'll have to assume that they were running close to stoichiometric?



#26 J. Edlund

J. Edlund
  • Member

  • 1,323 posts
  • Joined: September 03

Posted 01 March 2016 - 22:10

Yes, they typically don't run to 15 000 rpm, peak power is achieved earlier, maybe around 12 000 rpm or so.

 

I also doubt that a compression ratio of 12:1 is enough for these engines. Le Mans engines like AER's P90 ran with more than 12:1 a few years ago, with a reasonable amounts of boost. With P60 I think the compression ratio was raised further and with MAP as high as 4 bar.

 

I seem to recall an article about Renault Sport and that they were installing some large industrial type compressor there some years ago, so I suspect they can actually run their engines without turbos too, with boost pressure supplied by an external compressor. This is not really new, Cosworth did some of their early turbo engine development this way back in the eighties.

 

Some F1 engine chief (Rob White I think) was asked about stratified combustion, he sort of avoided to answer by going on about that they aren't necessarily using the same solutions as road cars. But I suspect that to manage knock and getting a reasonably low combustion duration with lean mixtures are very important factors for the power output of the current F1 engines no matter how you achieve it.



#27 J. Edlund

J. Edlund
  • Member

  • 1,323 posts
  • Joined: September 03

Posted 01 March 2016 - 22:34

Anyone?

 

As for the calculations, does anyone have any firm clues of the thermal efficiency of the last naturally aspirated V8s? And I'll have to assume that they were running close to stoichiometric?

 

Cosworth have published 275 g/kWh in power mode and 262 g/kWh in lean mode for the CA. Lambda values for these modes have been published too I think.



#28 gruntguru

gruntguru
  • Member

  • 6,733 posts
  • Joined: January 09

Posted 02 March 2016 - 00:17

I doubt the V8s were close to stoichiometric. Perhaps 0.95 in economy mode, 0.9 in power mode? (WOT of course)

 

AFR calc.

Mass Airflow (assume intercooling to ambient) = rho x Displacement/2 x RPM/60 x VE x PR = 1.2 x 1.6E-3 x 1/2 x 11,000/60 x 1.1 x 3.5 = 0.68 kg/s

Mass Fuel flow = 100/3600 = 0.0278 kg/s

AFR = 0.68/0.0278 = 24.4 = lambda 1.66

Higher CAT, lower MAP or lower trapped mass will make this richer but I don't think they are running less than 1.4. (I doubt there will be a lot of scavenge - they want to run the exhaust pressure as high as possible for turbine power)

 

Stratified charge is a no-brainer:

1. They are very lean - unlikely to be homogeneous.

2. No fuel near the walls where quenching kills combustion.

3. Shorter flame travel.

4. Cool end-gas.



#29 OO7

OO7
  • Member

  • 19,170 posts
  • Joined: November 04

Posted 02 March 2016 - 21:20

Paddy Lowe recently stated during a Qualcomm presentation, that the compounded mode is providing an efficiency of over 45%: http://www.youtube.c...bi77NY&t=44m45s



#30 OO7

OO7
  • Member

  • 19,170 posts
  • Joined: November 04

Posted 02 March 2016 - 21:29

Woozy and GG - just for the record - do you believe the 45% TE figure to be true - without any qualms or slight doubts?

 

 Again for the record -  no I don't believe the 45% TE figure.    

Ignoring what people such as Andy Cowell have openly admitted, there were hints about 45% plus efficiencies back in 2014.  During a press conference with the power unit directors present, Andy Cowell was a little coy and mentioned efficiencies in the 40% range.  Rob White (I think from Renault) almost half chuckled and said that the efficiency figures were a bit higher than that, obviously knowing the numbers for the Renault unit and witnessing the significant advantage Mercedes had.



#31 MatsNorway

MatsNorway
  • Member

  • 2,730 posts
  • Joined: December 09

Posted 03 March 2016 - 19:14

They are aiming at peak power as soon as peak fuel flow kicks in. That is at 10 500rpm. To put it differently, thats where they want the stars to align.

 

Edit: GG correction added.


Edited by MatsNorway, 03 March 2016 - 21:42.


#32 gruntguru

gruntguru
  • Member

  • 6,733 posts
  • Joined: January 09

Posted 03 March 2016 - 20:41

Peak fuel starts at 10,500 actually. Your point makes for an interesting topic though.

 

I had always assumed that the engines would be optimised a little above 10,500 say 11,000 to give a good spread of power from say 10 - 12k. Of course if the peak was at 10,500 it would be slightly higher due to the lower friction but that would force the use of lower revs - say 9,500 on upshifts and perhaps a lower "average power" during operation.

 

It would be interesting to see some projected power curves to judge the best choice of peak-power rpm. Clearly it is a compromise between rising friction above 10,500 and reducing fuel flow below 10,500.



#33 MatsNorway

MatsNorway
  • Member

  • 2,730 posts
  • Joined: December 09

Posted 03 March 2016 - 21:45

I see what you mean. The average hp could be higher across the used rpm range if you optimise a area a bit further up because the effect spreads both ways.

 

optimising could at 10,5k make it possibly a bit more narrow in the powerband.

 

You can allways speculate and share your thoughts by manipulating this spreadsheet we made a few years back.

https://dl.dropboxus...l delivery.xlsx

 

Edit: Second thought. By removing a gear or two you could then by your reasoning force them to optimise at a higher rpm as the drop of below 10,5k is bigger..

Personally i strongly hope they make the fuel flow limit more flat. More hp and turbo spool becomes a bigger challenge then.


Edited by MatsNorway, 03 March 2016 - 21:56.


#34 gruntguru

gruntguru
  • Member

  • 6,733 posts
  • Joined: January 09

Posted 03 March 2016 - 22:39

More flat?



#35 Kelpiecross

Kelpiecross
  • Member

  • 1,456 posts
  • Joined: October 10

Posted 04 March 2016 - 06:45

It may seem like an obvious question to some people - but - does a typical turbo engine have some degree of "natural"  power recovery from the exhaust gases? - meaning without turbo-compounding etc.  The pressurised air from the turbo pushing down on the pistons on the induction stroke must add some power to the engine.  I remember reading a few years ago that of the power needed  to drive a "mechanical"  supercharger about 1/3 was recovered by the extra pressure on the pistons on the induction stroke    If this is true the same must also be true of turbo superchargers - but mostly using "free" energy from the exhaust - leading to an overall net gain in power. 

  I don't see any reference to this effect on the internet or elsewhere.   Maybe if this effect does occur possibly M-B have been able to develop it to a much greater degree than in a normal turbo engine?   

I don't see how a bit (or a lot) of fiddling around with better lubrication, stratified charge, exotic materials could add more than a few percent at best to the TE.          


Edited by Kelpiecross, 04 March 2016 - 06:46.


#36 gruntguru

gruntguru
  • Member

  • 6,733 posts
  • Joined: January 09

Posted 04 March 2016 - 06:57

Yes - turbo engines have potentially higher efficiency for the reasons you give. This is especially true if exhaust blowdown energy is utilized and no wastegating is happening. (Those two factors will reduce exhaust backpressure => reduce negative pumping work while gaining the positive pumping work during the intake stroke.)

 

MB do not need to better-utilize this effect because the MGUH can do the recovery. The effect you speak of is only a phenomenon which converts recovered exhaust energy to crankshaft work.

 

Correct. Toyota have test engines operating at 42%. Mercedes have added a few percent by "fiddling around with better lubrication, stratified charge, exotic materials"  - taking them to say 45%, then added compounding, taking them to say 49.9%.



#37 Kelpiecross

Kelpiecross
  • Member

  • 1,456 posts
  • Joined: October 10

Posted 04 March 2016 - 11:52

The MGUH may better utilize recovering energy from the exhaust - but - the engine is highly supercharged so the inherent self-recovery of exhaust energy by a turbo engine is still there.   This inherent exhaust energy recovery would be in addition to the MGUH recovery.

 

 This 42% TE Toyota engine you have mentioned several times  - presumably a turbo engine?   Or it uses a very high CR/ER Atkinson Cycle effect?  

 

  What would be the best possible TE of an NA engine of conventional CR and not using any Atkinson/Miller or other "tricks"?  -  but using best lubrication/ stratified charge and exotic materials etc.  - I would guess maybe no better than about 32% or so.   These engines that are around the 42% to 45% range (uncompounded) if not Atkinson etc.  must be utilizing some turbo energy recovery effect.   

 

 I saw a lot of testing at the local uni engineering department of some kind of twin-cam Toyota engine on a dyno.   The engine was conventional - but every possible trick was being used to achieve the best TE  (this was the whole point of the testing)  - they never did better than 27%  and concluded that the often-quoted  30% or slightly better  was just not possible.    

 

 Anyhow all interesting stuff.         



#38 Canuck

Canuck
  • Member

  • 2,072 posts
  • Joined: March 05

Posted 04 March 2016 - 21:19

I never cease to be amazed by the arrogance displayed in those conclusions. "We've tried everything we know or can think of and were unsuccessful, therfore it can't be done".

#39 TDIMeister

TDIMeister
  • Member

  • 286 posts
  • Joined: January 06

Posted 04 March 2016 - 22:03

Funny, I achieved 42.4% ITE in a 2-stroke H2DI SI outboard engine...

http://gasturbinespo...ticleid=1700312



Advertisement

#40 gruntguru

gruntguru
  • Member

  • 6,733 posts
  • Joined: January 09

Posted 04 March 2016 - 22:04

The MGUH may better utilize recovering energy from the exhaust - but - the engine is highly supercharged so the inherent self-recovery of exhaust energy by a turbo engine is still there.   This inherent exhaust energy recovery would be in addition to the MGUH recovery.         

For each 1 bar increase in boost with no exhaust back pressure, the maximum recovery via pumping work is about 20hp @11,000 rpm. Each 1 bar increase in exhaust backpressure costs 20 hp - lost as pumping work in the exhaust stroke.

 

OTOH 1 bar increase in exhaust backpressure generates more than 50 hp in additional turbine recovery. (The 50hp figure is very general. It diminishes rapidly with increasing BP and is highly sensitive to massflow which is sensitive to boost pressure)

 

 

This 42% TE Toyota engine you have mentioned several times  - presumably a turbo engine?   Or it uses a very high CR/ER Atkinson Cycle effect?

 

There were two development engines in the article (which I can no longer find). One NA at 41-41% TE, one turbo at 42-43% TE. Here are the efficiency for production Prius engines.

 

"The next generation Toyota Prius engine achieves 40% thermal efficiency, says a report in Nikkei Technology. Toyota made the announcement at last week’s Spring Conference of the Society of Automotive Engineers in Yokohama, Japan. The engine in the current generation of Prius automobiles is rated at 38.5% efficiency while that in earlier models was rated at 37%, according to Green Car Reports."

 

http://gas2.org/2015...mal-efficiency/

 

they never did better than 27%  and concluded that the often-quoted  30% or slightly better  was just not possible.

Yes Canuk - amazing arrogance. Scientific conclusions?


Edited by gruntguru, 04 March 2016 - 22:06.


#41 Greg Locock

Greg Locock
  • Member

  • 5,656 posts
  • Joined: March 03

Posted 04 March 2016 - 22:13

The original Prius engine was not turbocharged and they claimed 37%. That was slightly 'overexpanded' to use Heywood's more accurate description than 'Atkinson'. I'd like to see some more best BSFC numbers for naturally aspirated engines.


Edited by Greg Locock, 04 March 2016 - 22:18.


#42 gruntguru

gruntguru
  • Member

  • 6,733 posts
  • Joined: January 09

Posted 04 March 2016 - 22:22

Funny, I achieved 42.4% ITE in a 2-stroke H2DI SI outboard engine...

http://gasturbinespo...ticleid=1700312

 

I don't suppose you would like a job at an Aussie university?  :cat:



#43 Greg Locock

Greg Locock
  • Member

  • 5,656 posts
  • Joined: March 03

Posted 04 March 2016 - 22:55

Um, not a great idea to confuse indicated and brake, is it?



#44 gruntguru

gruntguru
  • Member

  • 6,733 posts
  • Joined: January 09

Posted 05 March 2016 - 02:18

All the same - the 42% ITE result was probably better than 27% BTE.

 

EDIT. According to the abstract the BTE was 32.3% at rated power. I assume the peak BTE was at some other point so maybe >32.3%.

 

KC. I have edited post #9 to include the elephant in the room. Production engines have to meet emissions standards. In most cases this precludes the use of lean-burn technology at WOT and best-efficiency point due to NOx/3 way catalysts.


Edited by gruntguru, 05 March 2016 - 02:32.


#45 TDIMeister

TDIMeister
  • Member

  • 286 posts
  • Joined: January 06

Posted 05 March 2016 - 03:21

I don't suppose you would like a job at an Aussie university?  :cat:

I visited Oz in October 2014 to visit relatives and friends - loved it there. I'd be quite open to the idea. :)

 

Um, not a great idea to confuse indicated and brake, is it?

I'm well aware of the distinction between indicated and brake efficiencies - I didn't intend to conflate them, I apologise. I have no reason to doubt Cowell's figures of 45% BTE for the base engine and 50% once heat recuperation is taken into account as Kelpiecross does. It would mean ITE in the order of 50% if we assume 90% mechanical efficiency, which is altogether plausible. Getting 50% ITE on an SI engine would almost certainly mean geometric expansion ratios north of 13:1, application of Miller cycle, and extremely fast heat release while minimizing wall heat transfer and combustion exergy losses. All really neat stuff as this plays right into my current research work.

 

All the same - the 42% ITE result was probably better than 27% BTE.

 

EDIT. According to the abstract the BTE was 32.3% at rated power. I assume the peak BTE was at some other point so maybe >32.3%.

 

KC. I have edited post #9 to include the elephant in the room. Production engines have to meet emissions standards. In most cases this precludes the use of lean-burn technology at WOT and best-efficiency point due to NOx/3 way catalysts.

 

Best point BTE was indeed better than 32.3%, but that was not a focus of my investigations (another way of saying we didn't actually record the measurement in the experiment that achieved 42.4% ITE) because as an outboard engine there's also a transmission and churning losses in the water tank, so mechanical efficiency is very low compared to an automotive engine measured at the flywheel. The ITE that corresponded with the recorded 32.3% BTE was 39.7%, yielding a mechanical efficiency of 81.4%; if I were to apply the same figure on ITE=42.4% - a conservative approach as the mechanical efficiency at best point is better than at rated power - I'd arrive at 34.5% BTE.

 

The salient point I wanted to make is that this was achieved on an otherwise unremarkable production 246cc 2-cylinder 2-stroke engine with very few efficiency "tricks" as it were except for being modified to run on direct-injected hydrogen with a geometric compression ratio 14.5:1.  



#46 Kelpiecross

Kelpiecross
  • Member

  • 1,456 posts
  • Joined: October 10

Posted 05 March 2016 - 05:18

To be fair to the local uni  -  the were working under a severe handicap  -  they were being totally honest.   The results they got  - they reported as they found them.    The point of the testing was to remain totally road-legal - so there was not a huge amount they could do. 

 

  Maybe they should have "massaged" the results a little and announced an improvement to 35% or so and then announced their amazing result to the world.   Who would know the difference?

 

  Of course it is a well-known fact that no big car company or research institution  has ever been known to even slightly tell any form of "porkie".    

 

 I have to admit though that maybe modern TEs are a little higher than I thought. 

 

  On the subject of surprising TEs but in the opposite direction - in the modern model live-steam  scene  (some of these locos weigh quite a few hundred kilos)  they have TE contests.  Most of the contenders  achieve a best TE of less than 1%  - the winner is usually under 2%.     


Edited by Kelpiecross, 05 March 2016 - 05:20.


#47 Greg Locock

Greg Locock
  • Member

  • 5,656 posts
  • Joined: March 03

Posted 05 March 2016 - 09:46

There are very good reasons why a small scale steam plant is inefficient, not least of which is that large scale mobile steam plants are pretty bad as well. For instance, for a 1900s era steamship, the boiler was maybe 88% efficient, and a triple expansion (better technology than a railway engine)  engine would be 11% or so. Multiply those together, you get 10%. Turbines are heaps better than piston engines, when run at the right speed/geometry, for steam. Figure on 17% instead of 11%, back then, and a great deal better than that now.

 

But if you are engineering minded, there are few more interesting/nostalgic things to play with than a decent sized steam piston engine. Casey Jones out.



#48 Kelpiecross

Kelpiecross
  • Member

  • 1,456 posts
  • Joined: October 10

Posted 05 March 2016 - 11:32

Power station turbines can be up towards 50% efficient - using every possible method of waste heat recovery.

 

 There has been much debate in model engineering circles as to why the model loco engines are so spectacularly inefficient   when they are pretty much exact scale replicas of a full size engine that may be approaching 10% efficient.     

 

 I have dabbled a little in experimental small steam engines  - mostly in the steam inlet region - using something like  a  petrol injector to meter the steam.  But without any great success.   Also - I am a little wary of home made boilers  - the Mythbusters episodes on exploding household water heaters (at about 350psi from memory)  are something of a warning.       



#49 Greg Locock

Greg Locock
  • Member

  • 5,656 posts
  • Joined: March 03

Posted 06 March 2016 - 01:02

 

 

 There has been much debate in model engineering circles as to why the model loco engines are so spectacularly inefficient   when they are pretty much exact scale replicas of a full size engine that may be approaching 10% efficient.     

 

 

 

I'd have thought the answer was obvious, why is that wrong? Things don't scale directly. An exact 1/6 scale (ie 10 ton) Apollo wouldn't get into orbit even with no payload.(I think it would burst on ignition in fact) A properly designed rocket of that size could. Model aircraft suffer similar problems,



#50 Kelpiecross

Kelpiecross
  • Member

  • 1,456 posts
  • Joined: October 10

Posted 06 March 2016 - 07:02

I imagine it does have something to do with scaling factors - but just exactly what scaling factors seems to be somewhat debateable. 

 

 I don't know why it is - maybe you could explain more fully?

 

  When it comes to 4-stroke petrol engines I think the TE for a 24 litre engine  is much the same (or can be made to be) as for a 24cc engine - scaling factors don't seem to matter much.    Thermodynamics  for both IC and steam engines would seem to be the same  - expanding hot gases are expanding hot gases  no matter what scale they are on.  Maybe the heat loss through the cylinder walls in a model engine is a much greater in proportion  to a full-size steam engine?  - but then this effect doesn't seem to happen with IC engines  of greatly different capacities.   

 

 I don't know the reason for the great difference in TE -  and from what I read in model engineering magazines - neither do a lot of other people.