Tuning for BSFC

So lets say one were not worried about maximizing power out of a given engine, but rather trying to maximize BSFC across the rev range at what the driver considered 'full throttle'. Would it not help to keep the engine out of the zones that need to be run fully 'rich'?  Could this be achieved by restricting inlet size or max throttle opening percentage? How far does it have to get choked down to safely start running closer to stoich? Does it get down to the point where the pumping losses overtake the efficiency gains? Would limiting the rev range be more or less effective than limiting intake flow? Or would the two methods be best performed simultaneously? What else don't I understand about this, besides pretty much everything.

I put this in your other thread but I suspect this is one of the main reasons throttle by wire exists.

And I think it's not about limiting the ultimate size of the opening, but more about the transient states. 

At WOT and a steady rpm, you can get to a good bfsc. But it's when the throttle changes rapidly that you need to make fueling "guesses".

Let's say I want the best fuel economy possible while pumping out 200hp from one of two versions of the same basic engine with factory type tunes. Should I be looking at:

A) Run a 200hp engine at WOT and max power RPM.

B) Run a 250hp engine at 200hp throttle and max power RPM.

C) Run a 250hp engine at WOT and 200hp RPM.

D) Run a 250hp engine at less-severely reduced both throttle and RPM.

I know that generally you want to reduce pumping losses by removing restrictions, but can being able to safely make 200hp at 13.5:1 instead of 12.5:1 make an overall improvement? From what I recall, max BSFC generally happens at large (but not quite wide-open) throttle opening near the torque peak.

What else could be done within the tuning itself to sacrifice peak power for efficiency? What else could be done for the 250hp engine to maintain the 200hp engine flat-out efficiency at any amount >200hp? Assume DBW is an option.

Well, it probably depends more on heat management than throttle. 

Running rich keeps the chamber cool, and running high power output for sustained periods does the opposite.

So if you have the cooling capacity, then you can use less fuel. Regardless of engine potential or throttle opening. 

Keep in mind I am just an amateur with limited understanding.

The answer to your questions will vary heavily by engine, because there is so much variation in power cylinder geometry, compression ratios, specific output, cam design, VE, etc.

Yes, steady state is typically more efficient than transient operation. Most engines I've worked with have some combination of preemptive ignition retard and acceleration enrichment that hurt you during rapid transients.

Drive by wire can be used for this, but I've only seen it done in some motorcycle application where the "base" model had limited throttle and the higher end model allowed the throttle to open fully. Not for economy though, just for horsepower differentiation between models.

I've only seen cars limit throttle openings to control cylinder pressure in specific situations (low RPM lugging, basically only limited near idle speed), never explicitly for BSFC. I don't think it would help for BSFC because you're incurring throttling losses, which are inherently inefficient.

Most of the questions in your original post (limit RPM, limit airflow through restrictors/throttle control/etc) will not help BSFC necessarily, but would reduce fuel consumption. Reducing BSFC means doing the same work with less, as opposed to using less but also doing less. However, they may put you in a part of the engine operating map that happens to have higher BSFC, again depending on how the engine is designed. For example, engine might make the most power around redline, but if peak BSFC is at a lower engine speed, a lower rev limit might keep the driver in that high BSFC range.

I would guess (because I've never tried it) that most modern engines are tuned well for BSFC within a particular rev range until you get to the last 10-15% of throttle opening or whenever they come out of closed loop. At the expensve of power, in that open loop region you could probably back off timing and work your way up towards stoich to see how close you can get without knock at WOT, or without melting things. If temperatures go crazy, then use electronic throttle to limit airflow/cylinder pressure to where the engine can handle it.

Of course, generally speaking anything you can do to reduce intake and exhaust restrictions will also have some positive impact on BSFC.

And to your example multiple choice, I would guess that the answer is somewhere between A and C. Option B and D both add throttling losses. Option C allows you to reduce RPM without incurring throttle losses which is good, as long as the area you're now limited to is still an area where volumetric efficiency is high. If it isn't, then you're probably better off with the engine designed for your power output and let it eat.

My comments are mostly guesstimations from various stuff I've tuned for whatever it is worth.

In reply to Robbie (Forum Supporter) :

Good point on the cooling capacity.

In reply to gearheadE30 :

Thanks for sharing your thoughts. So it sounds like basically trying to keep a standard tune in the closed-loop range probably isn't likely to be a very effective overall strategy, incurring more fueling losses than gains for a given output. Am I wrong in thinking that a strategy that also increases waste heat (retarding timing) at a given power level would have to be less fuel efficient?

What about altering the valve timing to run more of a 'mild atkinson cycle' at WOT?

Are you talking about production engined or F1 engines?   As they are designed to make power very differently.  

There's no production engine where BSFC matters at peak power/ peak torque, and all but HD truck motors, WOT BSFC isn't considered.  

Where it is designed for is in the 1500-2000 rpm range, for sort of high loads.

That being said, pretty much every modern engine out there is capable at running WOT at peak power speed and peak power torque at stoich for a really long time.  Some older ones will have valve recession issues, but most modern ones are capable of a lot.

IF (and that's a capitol IF) an engine isn't knock limited, the difference between 12.5:1 and 14.6:1 on gas is only about 5%.  But since engines are designed to use their compression at low speeds, every engine is knock limited- so the big gain in going rich is gaining some MBT spark back.  Still, combustion efficiency kind of sucks because of many reasons- rich plus spark retard to keep things alive.

BTW, I doubt many engine limit throttle at WOT other than at speeds lower than 1300rpm.  Delivering safe power is pretty important if the driver is asking for it.

If this is an F1 motor, where getting the most out of ~100l/hr fuel at WOT- those are very different beasts.

alfadriver said:

If this is an F1 motor, where getting the most out of ~100l/hr fuel at WOT- those are very different beasts.

This is exactly the type of thing I'm asking about, but as applied to common production engines. Turning my original question around, if I want to reliably make the most power I can, over a prolonged period, while averaging 10 gal/hr on a closed course... What would be the attributes to look for and strategies to use for maximizing this? I'm largely thinking within the different variants of a single engine family, for example 4-valve port-injected V6's with a range (3.0/200, 3.0/240, 3.2/260, 3.5/260, 3.5/285) of displacement/hp, but secondarily what about comparing against a similar tech but higher strung 2.0L-2.5L 4cyl still capable of 200+hp? My gut (and anecdotally) also seems that boost would likely be worse for this than N/A, but I may very well be wrong on that too.

In reply to Driven5 :

Not entirely sure how current F1 motors really work- I do know they have pre-chambers for ignition, where I'm betting the fuel mixture is reasonably rich, or at least stoich, whereas the rest of the mixture is as lean as possible to support combustion.  Like way lean of stoich.   Searching in google doesn't answer it- but one person thinks it's a lambda of 1.3-1.5.  

Their goals would be 1) to use as little fuel as possible, and 2) burn every single little bit of it.  Doing that on an engine that's not designed around igniting a 20:1 mixture at WOT would be really difficult.

I've not seen my friends who were part of the Ford GT racing program- I'd wager they run as lean as they can, too.

In reply to alfadriver :

Ok, maybe not *exactly* the type of thing I'm asking about... Think less LeMans tech and more LeMons tech.

In reply to Driven5 :

Even that would be pretty tough- from what I can think of, the real key is to burn the leanest mixture you possibly can- one that is so lean that knock isn't a problem- so you can boost the compression.

Then again, BOP means that all of the racers are not really running the top of the top's engine capability.

But I will have to see if I can remember my friends who worked on the racing GT program, and see if they know. 

alfadriver said:

In reply to Driven5 :

Not entirely sure how current F1 motors really work- I do know they have pre-chambers for ignition, where I'm betting the fuel mixture is reasonably rich, or at least stoich, whereas the rest of the mixture is as lean as possible to support combustion.  Like way lean of stoich.   Searching in google doesn't answer it- but one person thinks it's a lambda of 1.3-1.5.  

 

That sounds like Honda's old CVCC system which used "Stratified Charge" 

 >Then Honda took a look at combustion pre-chambers, a design employed in diesel engines. The CVCC engine uses the pre-chamber area to start combustion with an isolated region where the fuel mixture is 14.7:1. When it lights, a jet of flame shoots from the pre-chamber to decisively ignite the lean fuel mixture in the main combustion chamber. The resulting emissions from the CVCC engine were clean enough to meet the requirements of the U.S. Clean Air Act for new cars in 1975 without needing an expensive catalytic converter with its precious metal active ingredients.

IIRC, the main cylinder ran at something like 20:1.  All of these engines were carbed, using a "2 & 1/2" barrel carb. (The "half barrel" was for the rich mixture for the tiny prechambers, and the other two barrels were a main and a vacuum secondary I think. Last car I had like that was an '85 Accord.  I thought this was a bit of a technological dead end, so it's suprising to hear that something that sounds broadly similar is used in F1. 

In reply to flat4_5spd :

It is, but with digital control- including an air injector instead of just a small intake valve.  But beyond that, I don't know what the mixture is.  

Oddly enough, that pre-chamber is on the research docket for many OEM's these days.  It will be interesting to see if it ends up being worth it.

Driven5 said:

alfadriver said:

If this is an F1 motor, where getting the most out of ~100l/hr fuel at WOT- those are very different beasts.

This is exactly the type of thing I'm asking about, but as applied to common production engines. Turning my original question around, if I want to reliably make the most power I can, over a prolonged period, while averaging 10 gal/hr on a closed course... What would be the attributes to look for and strategies to use for maximizing this? I'm largely thinking within the different variants of a single engine family, for example 4-valve port-injected V6's with a range (3.0/200, 3.0/240, 3.2/260, 3.5/285) of displacement/hp, but secondarily what about comparing against a similar tech but higher strung 2.0L-2.5L 4cyl still capable of 200+hp? My gut (and anecdotally) also seems that boost would likely be worse for this than N/A, but I may very well be wrong on that too.

I'm glad you're doing this research. It's intensely fascinating but I would have never even thought to run down this line of thinking. I'm definitely a "drop whichever engine in we find and see what happens" kind of guy vs the "think about which engine and which restriction method is most effective". 

Back to the thread...

Well, if you can change no other efficiencies about the car, I still think that your cooling (and maybe knock sensing and adjustment strategies) will be your biggest friend here. Maybe you can run more coolant and keep it regulated at a lower temperature so that you can run lean during power cycles and let the coolant heat up but then shed more heat. Like maybe run a couple laps lean then a lap normal to cool back down. If you have good knock sensing, maybe there is a tad to gain here. 

In actual practice though, I would think there are bigger fish to fry in the efficiency game. For example, any braking is gas that is thrown away. Can you brake less? Can you cut fuel entirely during engine braking?

And the big picture as well is always "how much time can be saved by reducing the pit stops by one?" Maybe you can just run a larger fuel tank, or use hydramat or similar to run the tank further down before you have starvation problems?

In reply to Robbie (Forum Supporter) :

Yes, we'll be considering other efficiencies as well, but one of the big ticket items right now is picking which engine to install... And for me personally, if I can have (and maximize) alternative fueling pit stop strategy options for minimal additional cost, I'd rather have them and end up not needing them than needing them and not having them.

Driven5 said:

alfadriver said:

If this is an F1 motor, where getting the most out of ~100l/hr fuel at WOT- those are very different beasts.

This is exactly the type of thing I'm asking about, but as applied to common production engines. Turning my original question around, if I want to reliably make the most power I can, over a prolonged period, while averaging 10 gal/hr on a closed course... What would be the attributes to look for and strategies to use for maximizing this? I'm largely thinking within the different variants of a single engine family, for example 4-valve port-injected V6's with a range (3.0/200, 3.0/240, 3.2/260, 3.5/260, 3.5/285) of displacement/hp, but secondarily what about comparing against a similar tech but higher strung 2.0L-2.5L 4cyl still capable of 200+hp? My gut (and anecdotally) also seems that boost would likely be worse for this than N/A, but I may very well be wrong on that too.

Ideally you'll size the engine (power wise) as close as you can to your need.  This is a proven approach in aircraft, marine and stationary engines.  Ideally you'll have the throttle close to wide open while the load holds your engine speed at peak torque.  Since you're presumably changing speeds you'll want lots of gears to keep the engine in its happy place. 

There may be an edge to naturally aspirated but in the stationary engine world where engines are designed from the ground up to maximize efficiency at a given power output turbos are the norm. 

In reply to Driven5 :

Since you are at that point, plan on running premium fuel- which should help reduce knock issues if you lean the mixture out.  BTW, running more aggressive spark reduces exhaust temps- quite a lot.   As well as any waste heat to the cooling system.  Lots of knock on effects by advancing the spark to as close to MBT as possible.

Can you pick an engine that has a really high BSFC to begin with? I've heard honda k series is up there but also ford ecoboost. Here's an article with lots of cool info but I havent read through the whole thing:

https://www.epa.gov/sites/default/files/2018-10/documents/sae-paper-2018-01-0319.pdf

Which includes gems like this, which may indicate that with a turbocharged engine you may want to keep load high but limit rpm as the bsfc seems to slide as RPM increases. 

So for the 3.5l V6 engine taken to LeMans and Daytona- turns out that there are screw in pre-chamber ignition systems.  But they require controls for the air and fuel injector, and a dyno to fully tune it.  But that's probably how engines can run way lean of stoich and still survive- and make good power.  With the BOP adjustments, the power lost vs. the economy gained is straightforward to deal with.

edit- just did a search for pre-chamber spark plugs- one of the picture came from rusefi....  So a poster here seems to be working on it for his DIY EFI system.

PSS- the one reason you don't really see them on production vehicles is that lean burn combustion is really hard to keep clean.  Just like a diesel.