People often talk about horsepower per liter when they're comparing two engines. For example, you might say that the JDM B18C engine in the Integra Type-R produced 210hp from 1.8 or an impressive ~ 117 hp per liter. You may then be tempted to look at an engine like the 2JZ-GE found in project Lexus and scoff at it's 215hp from 3.0L, which is a mere ~ 72 hp per liter. Many IS300 owners are previous Honda guys, so, naturally, this is a very common 'scoff' :).
However, there is a better method to compare two engines (and also figure out how 'developed' an engine already is) which while also with its faults will give you a much better idea of what you're dealing with.
BMEP is one of many figures that you can use to compare engines. It stands for Brake Mean Effective Pressure and it refers to the amount of average pressure generated on the piston during combustion. BMEP gives us a much better idea of how 'developed' an engine is, how much stress/pressure is on the engine internals, and to some degree, how effective that engine is at converting gasoline to power. While this article mostly deals with Naturally Aspirated engines, it can also be used on forced induction engines - however, note that the BMEP values for these engines will be much much higher.
So let's show you the formula and then do a few examples:
(kW x 1200) / (Liters * RPM)
kW = power output in kW (you can convert using google, type 200 horsepower in kW - make sure you use correct capitalization or google won't recognize it). A simple conversion is HP * 0.735 = kW.
Liters = Displacement, so 1.8L or 3.0L for example.
RPM = Peak power output rpm (usually given with the specs of the engine, for example, 215 hp @ 5800 rpm).
The result is the pressure in the pressure unit, Bar.
Now, let's look at a few "classic" Honda engines first, note that I rounded in several places so your numbers you may not exactly match mine.
The Integra LS, GS-R, and Integra Type R engines (USDM spec):
The Integra LS engine, B18B1, has a displacement of 1.834L and produces 106kW (142hp) at 6300 rpm. It's BMEP is therefore: (106*1200) / (1.834*6300) = ~ 11.0 bar. Not bad.
The GS-R engine, the B18C1, has a displacement of 1.797L and produces 127kW (170hp) at 7600 rpm. It's BMEP is (127*1200)/(1.797*7600) = ~ 11.16 bar. A little better, as we might expect.
The Integra Type-R engine, the B18C5 (USDM) has a displacement of 1.797L as well and a power output of 147kW (197hp) @ 7800 rpm. The BMEP is (147*1200)/(1.797*7600) = ~ 12.59 bar.... which is mightily impressive, better than a Ferrari Enzo engine as we will see later.
Compare that B18C5 to the modern 2.0L K series Type-R engine, which has a BMEP of only 11.9 bar.
Does that mean the old Type-R engine is better than the new one? Not necessarily, as there are other factors such as gearing and actual torque curve produced. What the BMEP ultimately tells us is how much engineering work has been done to the engine to get to the output we now have. Better put, how much power they've squeezed out of every revolution and cubic inch of cylinder volume.
So, naturally, the modern Type-R engine having more displacement doesn't score quite as well since it has basically the same peak power output as the old Type-R engine at a much smaller engine size while both have a similar peak rpm.
The Type-R always gets compared to the H22 engine, so it's nice to know that the H22's BMEP is 12.01.
What this actually tells us more than anything, is that the H22 has a little more room for improvement than the B18C5 since there IS a limit on how high the BMEP can go. Especially on a street-able motor without boost. I'll give you a hint: both of these engines are quickly approaching that limit. NA F1 engines are somewhere around 13.8 bar to give you an idea and perhaps one of the most sophisticated engines ever designed - the F series engine in the original JDM S2000 has a BMEP of 13.32 (the USDM version has a BMEP of 12.8 while having the same hp peak and displacement - due primarily to a compression increase in the JDM version).
So let's look at a completely different engine to gain more perspective, the 2JZ-GE found in our project Lexus IS300. The BMEP for this particular engine is right at 11.04, so it's getting close to being as "efficient" as the GS-R engine even though the GS-R engine has closer to 95hp per liter.
These are two very different engines, one is a high revving 4 cylinder, one is a fairly low revving 6 cylinder. The BMEP is so close however because while the GS-R engine produces an impressive amount of power for its size, the 2JZ-GE also produces an impressive amount of power for its size...at such a low RPM. That's the key of the BMEP, it takes out some of the bias that power/liter has due to its ignorance of the RPM that the power is produced at.
So maybe we should also look at a fairly lame engine, let's say a 98 Ford Mustang V6. This engine produced a pathetic 150 hp @ 4000 rpm using a 3.802 L engine. The BMEP for this engine is ~ 8.84. This is a very primitive engine.
Domestics always get a bad rap for bad engines, so let's look at a new and seemingly impressive one, such as the new engine for the 2011 Ford Mustang GT. This engine produces ~ [email protected] with 4.951L of displacement. That's a BMEP of 11.45 - better than the old GS-R engine, despite it's seemingly low HP/Liter value of ~ 82.4hp/liter. So before we say an engine is inefficient, we do need to look at slightly better indicators such as BMEP.
How about some exotics, just for fun and reference? A Ferrari 360's engine has a BMEP of 11.81, and a Ferrari Enzo? Well, it has a BMEP of 12.4 - and you KNOW that's an almost no-limits engine.
BMEPs are not without fault. They do not, as previously hinted at, take into account the actual torque curve, gearing, and the other several dozen factors that can make an engine slightly better than another. However, because the BMEP takes account of the HP and at what RPM, it does give us a pretty dang good idea of how efficient one engine is versus another at cramming its cylinders full of oxygen and fuel and converting that into power.
It is also a decent indicator of the amount of stress being put on the engine by combustion as well. You'll find that higher compression, higher strung engines have higher BMEP values. Add boost into the equation and BMEPs can be extremely high!
Finally, let's think about some of the things that the BMEP equation tells us.
The top part, the power multiplied by a constant (1200), will of course get larger or smaller with the power figure. If you have a bigger number on top, you get a higher BMEP. If you have a smaller number on top, you'll get a lower BMEP. In other words, more power = higher BMEP all else being equal.
That's pretty straight forward.
However, the bottom gets a little more interesting. If you increase either the displacement or the peak horsepower RPM, you'll get a DECREASE in BMEP. This means that if you increase displacement of a given engine, you'd decrease the stress needed on the pistons and rods to produce a specific power output. You would also not need a super sophisticated cylinder head and intake manifold in this case. On the contrary, if you decrease displacement, you'll need to put more stress on the internals by say, increasing compression or increasing cylinder head flow. Also, for those of you wanting to rev your engine to the moon, not only does stress increase with RPM (which we can see more accurately with other calculations), but your BMEP value also goes down assuming you don't make any more power at that higher RPM.
I will write a lot more about equations such as this one in the future. This is more of a 'for fun' equation with some interesting applications as well. However, in the future, I will cover how to calculate theoretical volumetric efficiency which will allow you to determine whether power goals are achievable or not. I will also be talking about why PEAK horsepower isn't really important and why most folks have it all wrong about making a car fast
Extremely Limited Time Offer - 3 Days Only
Get a MASTERS DEGREE in Performance Tuning from the BEST in the Field for ONE low package price
For 3 days only, get ALL of our PREMIUM (not available anywhere else) Tuner University Courses featuring some of the greatest minds in motorsports:
- Top 10 Performance Myths Class (MP3 and Edited Transcript) - $29.95 value
- Performance on a Sip of Fuel Class (MP3 and class manual) - $69.95 value
- Header Design Secrets with John Grudynski (MP3 and transcript) - $69.95 value
- Dirty Secrets of Oil with Ryan Stark (MP3 and transcript) - $69.95 value
- Engine Tuning Secrets with Ben Strader of EFI University (MP3 and Transcript) - $69.95 value
Get all of the above courses (many of which not currently available anywhere else) for one single price of just
$309.75 $69.95. A HUGE package savings.
You can also buy any of these individually through our resource center, but I don't know why you would as this is basically the whole store for the price of just 1 course. You'll get the MP3 recordings of each class as well as a transcript or companion manual with each course and you'll be able to download them all INSTANTLY.