Understanding BSFC maps

[yoshimitsuspeed]

After growing intimately familiar with compressor maps, turbine maps, ECU and other types of maps and graphs I figured BSFC maps would be pretty easy to wrap my head around. I thought I had a pretty good idea of how it all tied together but recently someone made a comment about BSFC that didn't sound right and I realized I didn't understand things well enough to know for sure one way or another.
Part of the problem is that there are so many different types of maps and none of them are clearly defined for someone who doesn't have a complete grasp on reading them.
I have spent probably the last 45 min searching online for answers but from my experiences in learning things like turbo maps I have come to expect that about 90% of what you find online will be wrong and much of it does contradict it's self or my expectations.
The topic that got me off on this tangent had to do with part throttle BSFC and I am having a hard time figuring out which maps have enough data to solve for varied throttle and which are tested or only show for a single throttle position.
Someone claimed that lower RPM and more throttle would always net a better BSFC. Now I know this can't be entirely true because most carbs and ECUs will tend to start adding a richer mixture with enough throttle. With EFI that starts at the point when the ECU trips into open loop. Obviously it seems like you would want to maintain closed loop for best BSFC. This got me wondering how OEM BSFC is generally charted. Is it usually at a fixed AFR or does it include the OEM ECU fuel mapping?


Now on to a chart that I believe has enough data to quantify part throttle BSFC.

From the reading I have done it looks as though the curved lines represent the BSFC path for a given power level right?

BMEP should basically be a combination of throttle position, volumetric efficiency, and combustion efficiency right? Really it is telling us how much throttle you need to give it in that range of efficiency if I understand it right.

So let's say we are cruising using 10 KW in fourth gear which puts you at 1500 RPM it will require say 2.8 bar BMEP with a BSFC of about 300 g/KWh.

Drop to third gear and 2500 RPM and now it only requires a BMEP of 1.8 bar (very likely less throttle) but now we are at about 450 g/KWh so our gas mileage has dropped significantly.

So this would mean cruising at a speed requiring less than 20 KW the lower RPM the better?
Does this account for high load enrichment?
This goes against much of what I have learned about keeping a motor in it's sweet spot more in the middle of it's rev range. Many people on sites like eco modder seem to go with the lower RPM more throttle philosophy but others say that you will bet better BSFC near peak torque even at partial throttle.

So now if I'm reading it right that would mean that if you are using 40 KW your lowest BSFC would be around 2250 RPM right?
But that's a lot of power. Why tune a vehicle for peak BSFC at a power level that would put a small passenger car around 100 MPH cruising on flat ground?
Wouldn't it be better to put the peak closer to 20-30 KW?


And finally, most graphs I have seen look like this.

This graph doesn't provide enough information to calculate for various load conditions does it?
Does this graph just represent one throttle position like wide open?

 

[GregLocock]

" This got me wondering how OEM BSFC is generally charted. Is it usually at a fixed AFR or does it include the OEM ECU fuel mapping? "

It includes it.

In the first graph the curved paths descending left to right are lines of constant power. that's easy to work out bmep in Pa*capacity/2000*n/60*2*pi, so they often aren't plotted. they are identified on the RHS.

The heavy dashed line is road load in bmep for one particular gear, often you'll see the road loads for all or many of the gears plotted, handy for developing transmission schedules.

The contours are of constant bsfc. Best bsfc for that engine is at 8 bar 2200rpm, something like 240 g/kWh

"So this would mean cruising at a speed requiring less than 20 KW the lower RPM the better?"


Yes, you'd be better off at low speed, more throttle, the standard result.

"Does this account for high load enrichment?"

No, high load enrichment is used at high load, not 25% power.

"This goes against much of what I have learned about keeping a motor in it's sweet spot more in the middle of it's rev range."

You learned wrong
"
Many people on sites like eco modder seem to go with the lower RPM more throttle philosophy but others say that you will bet better BSFC near peak torque even at partial throttle. "

Depends on gearing. That's why you want a CVT.

"Why tune a vehicle for peak BSFC at a power level that would put a small passenger car around 100 MPH cruising on flat ground?"

Because the shape of that bsfc map is pretty much fixed. If you want a certain max power there's a lot of things pushing a typical SI towards min bsfc at 1/3 to half redline 80% throttle and it is very hard to move it very far, usefully. The general shape of the contours is also fairly uniform and untunable in a good way.

The second graph can be turned into the first graph by adding some simple maths and a road-load curve, it contains all the tricky info.

Check out Heywood: Internal Combustion Engine Fundamentals for enough info to undertand the reasons for the shape of this graph.







Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[yoshimitsuspeed]

Thanks I will check out that book.

"No, high load enrichment is used at high load, not 25% power."

If we look at that first map I would assume the left hand side of the constant power lines represent full throttle?

The engine management I am most familiar with is L-jetronic EFI like you find in 80s Toyotas. All that I have seen flip into open loop even at very low RPM if given enough throttle. I would expect most EFI especially anywhere from the 70s to early 2000s to behave similar. As RPM and load increase it takes less throttle to trip into open loop and the quicker it increases the AFRs as RPM and load increases however it's fully possible to say flip into open loop with very rich AFRs at 3/4 throttle bogging in 3rd gear at 2000 RPM when in second at 1/2 throttle it maintains closed loop. It seems the latter should net a lower BSFC. It also seems like that transition to open loop should be very obvious on on a BSFC map as load increases.

Or is the whole map charted below the open loop trip point?
Not that I am sure this motor behaves in the same way as I am used to seeing but all maps I have seen look very similar and don't seem to have any dips that I would recognize as a transition into open loop AFRs.


"Because the shape of that bsfc map is pretty much fixed. If you want a certain max power there's a lot of things pushing a typical SI towards min bsfc at 1/3 to half redline 80% throttle and it is very hard to move it very far, usefully. The general shape of the contours is also fairly uniform and untunable in a good way."

So basically the intake, exhaust, cams etc are primarily designed around getting the desired power curve and at that point there isn't much you can do to change BSFC without negatively impacting your power curve right?

What are your thoughts on the fact that the top graph only goes to 4500 but the BSFC island is slightly higher in the rev range than the second graph that goes to 6000? Can we assume both of these map to the engines redline? If so what characteristics might cause one island to be in a much lower percentage of the rev range?

 

[Nereth1]

Keeping the motor at the peak of its torque curve is only an easily-false rule of thumb, and even then only really *seems* to apply (as far as I can reason at least) under full load where the peak of the torque curve the manufacturer publishes is the actual torque curve you're riding on. At part throttle the peak moves downwards. Otherwise this would be an argument for driving around at peaks as high as 4k RPM in 3-4L motors at 40kph. At some point you have to admit you are cracking the throttle more at 4k RPM to keep the engine at 4k RPM than you were at 2k RPM to keep the car rolling. Lower really is better in my limited experience, as long as the ECU doesn't start stepping in/enriching, and as long as you don't get too silly (I believe a few books describe thermal losses through cylinder walls etc at very low RPM becoming dominant).

I am actually a little surprised about ECU effects being in the BSFC map as well - but it makes sense, that's the vehicle being driven so the BSFC map without the ECU effects would lead to some pretty poor gearing decisions and so on when matching a transmission to the rest of the vehicle.

 

[GregLocock]

cams don't affect part throttle efficiency very much because there is enough time for the air to get in and exhaust to get out anyway.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[Tmoose]

The first BSFC curve says 300 SD. Isn't that a turbocharged Mercedes Deeezul?

In any case, the uppermost dark curve is the "full" throttle and max power curve, but really the torque curve, as at any one rpm the only way to make more HP is to make more torqueage.

All intermediate throttle positions at any rpm are included, with the BSFC to be interpreted from interpolating between the islands.

2190 rpm, and BMEP as necessary to make 16 HP, or whatever it is you need, or are using, and there's your BSFC.

With a gasser at 2000 rpm the full power/max torque might be obtained with the throttle partially open, but the manifold pressure would be zero. Opening the throttle further would not make more power. Hence carburetors with vacuum controlled secondaries, etc

 

[pickler]

I have found with modern MPFI/DGI vehicles it's best to just drive smoothly (normally) and not to worry about anything such as BSFC. This is if fuel economy is important to you. With modern CVT/6-8 speed Auto transmissions most engines are at low enough RPM to keep the manifold pressure high enough to eliminate pumping losses.

However if this is for older carburated engines or EFI systems from 80s/90s I would accelerate at 1/3 throttle. This usually completely loads out the engine at low RPMs and helps the transmission to shift near peak torque (highest volumetric efficiency) and just before open loop fuel enrichment.

Every engine is different, which care are you exactly studying?

The short answer is yes cruising induces pumping losses, pulse and glide technique fixes this. But also most modern engines are designed to eliminate pumping losses.

 

[pickler]

also have a look at this thread

http://www.eng-tips.com/viewthread.cfm?qid=339585

 

[BrianPetersen]

Driving around with a scangauge connected can be a useful exercise, too. You can monitor MAP (which gives an indication of the throttling losses) and instantaneous fuel consumption.

I haven't found a circumstance yet in which the best fuel consumption is not in the highest gear that the engine will take up the load.

It's also pretty apparent that the stock programming attempts to run the engine at as high MAP as it can (in an attempt to minimize pumping losses) and it plays with the valve timing to modulate power output in a way that keeps the MAP high. At part load it intentionally sets the valve timing in a way that reduces volumetric efficiency so that it can keep the MAP high while only delivering the torque that the driver requests.

 

[PedroCG]

The shape of the map can be understood simply,

At high revs the BSFC is high (low efficiency) because the mechanical losses are high.

Under low load the BSFC is high because the mechanical efficiency is low (mechanical losses high relative to work produced) (this is worst for gasoline because of pumping losses)

At low revs the BSFC is high because the heat losses are high compared to work produced.

At high load, diesel engines will not have minimum BSFC because combustion efficiency will be lower then at partial load (more fuel to burn and same air mass). Gasoline engines without catalytic converters will not have minimum BSFC here due to mixture enrichment. Engines with catalytic converters will because you have lambda=1 (otherwise the catalytic converter doesn't work).

Some special cases for gasoline engines: High load @ low revs it's an exceptionally bad place for efficiency because the ignition needs to be retarded to avoid knock and so efficiency drops. High load @ medium-high revs mixture will be enriched regardless of the catalytic converter.

As for my driving if the engine can take another gear give it to him, when accelerating on open road 2/3 throttle on an rpm range around max torque.