Break Mean Effective Pressure

[Howard Gibson]

We keep discussing engine performance. I am researching this in WWII aircraft and I ought to share stuff.

I won't go into the math, but the work done by a cylinder doing a single stroke is the volume swept by the piston, multiplied by the presumably constant pressure. Power is among other things, the rate at which you do work. In the case of a piston, power is the amount of work you do per unit time. The pressure in a spark ignition engine cylinder, not surprisingly, is not constant. When you test an engine, you ought to curious about mean effective pressure. "Mean" is the mathematician's word for "average". Mean effective pressure is average pressure, not pressure that hates children and puppies.

Brake mean effective pressure is pressure calculated from brake horsepower. If you think about it, it is a weird concept. If you mount an engine on a dynamometer with a brake, you can measure torque and speed, and work out power in horsepower or Watts, and brake mean effective pressure (BMEP). Indicated mean effective pressure (IMEP) is measured with the pressure gauge you don't have.

Most of the following data is sucked down off of Wikipedia. Note the cell calculating BMEP. I have a sheet in this spreadsheet that provides unit conversions, mostly to SI.

BMEP gives you a good idea of what is happening inside the engine cylinder. BMEP should give you a good idea of what octane fuel they were using, plus the presence of water and/or methanol injection.

Feel free to speculate wildly about all this.

 

[Howard Gibson]

I graphed something else interesting.

From the table in the previous post, you should be able to see where your favourite engine ranks in power to displacement and mass.

 

[Engineman]

Hi,
Some of the dates and ratings are incorrect. This is probably due to poor data. Unfortunately, that can give very misleading information. As for speculating wildly about this, I don't feel that that is any problem for many people. As always, the devil is in the detail.

Eng

 

[Howard Gibson]

Hi, Some of the dates and ratings are incorrect. This is probably due to poor data. Unfortunately, that can give very misleading information. As for speculating wildly about this, I don't feel that that is any problem for many people. As always, the devil is in the detail. Eng

Some of the dates are extremely arbitrary.

The other numbers should be good for at least some versions of the engines. Some of these engines were produced in many, many versions.

 

[Engineman]

Some of the dates are extremely arbitrary.

The other numbers should be good for at least some versions of the engines. Some of these engines were produced in many, many versions.

Hi Howard,

OK. I think that I understand you are encouraging consideration of BMEP in engine comparisons, and that is certainly a big factor. However, I am pointing out that some of that data is misleading and accurate data is important for meaningful comparison.

Eng

 

[Daggerr]

BMEP gives you a good idea of what is happening inside the engine cylinder.

Not really.
IMEP might give some idea.

BMEP and IMEP are ficticious pressures that can only be calculated, not measured.

For a piston engine alone, one can calculate an IMEP.
For the total combination of piston engine plus supercharger, oil pump, magnetos, and friction losses, one can calculate a BMEP

BMEP should give you a good idea of what octane fuel they were using, plus the presence of water and/or methanol injection.

Maybe in combination with a crystal ball.

BMEP also depends on whether a supercharger or a turbocharger is used, whether an intercooler (aka aftercooler) is used.

Moreover engine power depends on altitude.

Note also that British HP is not the same as German HP (PS).

The engine BHP per pound (or kW/kg) is in practice more important than BMEP, but for watercooled engines you should then include the weight of the cooling fluid, piping, radiators, intercooler (if present), et cetera, when comparing with aircooled engines.
If a turbocharger is included that weight plus ducting plus intercooler should also be included for aircooled as well as watercooled engines (P-47 is aircooled with turbocharger, P-38 is watercooled with turbocharger).
If a system for injection of methanol/water or N2O is included that also adds to the weight.
If you like you can make a lot of different comparisons, including this, or excluding that, but be careful not to drive yourself crazy.

 

[Howard Gibson]

Not really.
IMEP might give some idea.

BMEP and IMEP are ficticious pressures that can only be calculated, not measured.

For a piston engine alone, one can calculate an IMEP.
For the total combination of piston engine plus supercharger, oil pump, magnetos, and friction losses, one can calculate a BMEP


Maybe in combination with a crystal ball.

BMEP also depends on whether a supercharger or a turbocharger is used, whether an intercooler (aka aftercooler) is used.

Moreover engine power depends on altitude.

Note also that British HP is not the same as German HP (PS).

The engine BHP per pound (or kW/kg) is in practice more important than BMEP, but for watercooled engines you should then include the weight of the cooling fluid, piping, radiators, intercooler (if present), et cetera, when comparing with aircooled engines.
If a turbocharger is included that weight plus ducting plus intercooler should also be included for aircooled as well as watercooled engines (P-47 is aircooled with turbocharger, P-38 is watercooled with turbocharger).
If a system for injection of methanol/water or N2O is included that also adds to the weight.
If you like you can make a lot of different comparisons, including this, or excluding that, but be careful not to drive yourself crazy.

IMEP is fairly real if you have a pressure gauge attached to your engine. BMEP is significantly lower than IMEP for reasons that should be obvious.

Actually, BMEP is independent of supercharging. Your actual problem is your fuel/air mixture immediately prior to ignition. When you increase the pressure of a gas, its temperature goes up. Cylinder pressure immediately prior to ignition is a function of atmospheric pressure, supercharge boost, and compression ratio. When the temperature exceeds the ignition temperature of your fuel, it ignites. WWII pilots watched their manifold pressure gauges, unless they were flying Fw190s.

Your second limitation is rich fuel mixture. Once there is enough fuel to burn all the oxygen in your fuel air mixture, there is no point in adding any more.

Higher compressions, resulting in higher BMEP, are made possible by high octane fuels, which allow more compression, by intercoolers and water methanol injection, which cool the air and allow more compression, and by nitrous oxide injection, which provides more oxygen, allowing for a richer fuel air mixture.

 

[Shortround6]

Once there is enough fuel to burn all the oxygen in your fuel air mixture, there is no point in adding any more.

The extra fuel acts as a coolant, both for the mixture before ignition and for the cylinder (and parts) in general after ignition. The unburned fuel carries heat out the exhaust valve and out of the engine.

Actually, BMEP is independent of supercharging.

BMEP it quite dependent on supercharging.

Higher compressions, resulting in higher BMEP, are made possible by high octane fuels, which allow more compression, by intercoolers and water methanol injection, which cool the air and allow more compression,

I think we are arguing about what compression really is.
High octane fuels allow higher density of fuel/air mix right before ignition. Now does the the high density fuel/air mix come from supercharging or from the compression ratio of the cylinder. ?

for car applications

Now please note that for comparing to WW II aircraft applications, this chart is using boost pressure over the standard 14.7/15psi much like the British.
Also note that is very simplistic and does not take into account valve overlap/duration and intake (or exhaust) flow.

Now when figuring the BEMP you are taking into account burning the fuel and getting the energy from the burning fuel (around 30 something percent) so our 6.0:1 compression engine with 15lbs of boost is burning twice the fuel per power stroke that a 12.0:1 compression non-supercharged engine will even if they have the same "compression" ratio.

We can see how this works with the Merlin engine which stayed at 6.0:1 compression for it's whole life. The increased power came from the higher boost/burning more fuel which created the the higher BMEP/IMEP.

Please note that an un-supercharged high compression engine has a very short high pressure peak and the pressure falls quickly as the piston descends. The supercharged engine, with the greater mass of fuel/air and the greater energy content will maintain the pressure for a longer period of time (badly stated, it doesn't fall off as quickly as the piston descends)

High supercharger boost is inefficient, it takes power to turn the supercharger and some (a lot) of pressure/temperature energy goes out the exhaust but for a given size/weight engine it offers the most gross or net power if you don't care about the actual amount of fuel used.
And even liquid cooled engines were using excess fuel as supplemental cooling of the engine itself, not just the mixture.

 

[pbehn]

Back in the day when I got interested in bikes in 1978 Mike Hailwood made his comeback on a 900cc 4 stroke Ducati desmo V twin, Honda made their comeback to GP racing with a 500cc V8 that had oval pistons and 8 valves per cylinder. The ins and outs of getting engines to produce power at high RPM and/or high piston speeds, , it made my head ache reading it, all sorts of stuff about droplet sizes, flame fronts, thermal loads, piston speeds and scavenging efficiency that produce the pressure on the piston which leads to the BMEP figure. much of it applies to supercharged aero engines, it is a dark nether world of satanic magic.

 

[Shortround6]

, it is a dark nether world of satanic magic.

It is pretty straight forward compared to propeller design

I will note that diesel engines with compression ratios of 14-17 to 1 had BMEPs lower than the Lycoming O-235 flat 4 running on 73 octane fuel.

 

[Daggerr]

IMEP is fairly real if you have a pressure gauge attached to your engine.

You can't measure IMEP with a pressure gauge.

IMEP is calculated from IHP (indicated horsepower) in the same way as BMEP is calculated from BHP (brake horsepower), and therefor just as fictitious as BMEP. Only of interest to academics.

Difference between IHP and BHP (and consequently between IMEP and BMEP) is the power that is consumed by supercharger, oil pump, magnetos and friction losses, which reduce the available power from IHP down to BHP.

 

[Deleted member 68059]

Its very difficult to usefully compare aero engine across time and countries, BMEP is certainly a good start and shows you the overall progression - but (for example) its less useful to actually compare similar era engines from different nations in a better/worse way. For example, one might have very low BMEP but have needed to use very low octane fuel because of the year it was built or some other constraint.

Gas turbine people use what they call a "Technology Factor" to describe compressor performance. Which is a sort of ratio of what the engine puts out vs the theoretical maximum.
Probably the best thing is something like: "Hp/kg/sq-m frontal-area", but again that doesn't take into consideration constraints from different nations, there is no easy answer to this.

 

[Howard Gibson]

You can't measure IMEP with a pressure gauge.

IMEP is calculated from IHP (indicated horsepower) in the same way as BMEP is calculated from BHP (brake horsepower), and therefor just as fictitious as BMEP. Only of interest to academics.

Difference between IHP and BHP (and consequently between IMEP and BMEP) is the power that is consumed by supercharger, oil pump, magnetos and friction losses, which reduce the available power from IHP down to BHP.

Pressure gauge on an engine. This one is 250 cubit inch GM six cylinder running at 2000rpm. The work done per stroke is the area of the big curve minus the area of the small curve. This was all done in college about forty years ago. I understand they still have the engine. This is a polaroid of an oscilloscope output. We saw all sorts of weird and wonderful curves on the oscilloscope. We judged this one to be typical. Don't think you are seeing any sort of precision here.

For the record, IHP, calculated from the pressure gauge output was 115psi. IHP was 72.6. BHP was 52, and BMEP, calculated from rpms, displacement and BHP was 82psi. I have no idea of what condition the engine was in. Given it was in the basement of what was then a polytechnical college, the exhaust system may have been very interesting.,

 

[Deleted member 68059]

View attachment 723622

Pressure gauge on an engine. This one is 250 cubit inch GM six cylinder running at 2000rpm. The work done per stroke is the area of the big curve minus the area of the small curve. This was all done in college about forty years ago. I understand they still have the engine. This is a polaroid of an oscilloscope output. We saw all sorts of weird and wonderful curves on the oscilloscope. We judged this one to be typical. Don't think you are seeing any sort of precision here.

For the record, IHP, calculated from the pressure gauge output was 115psi. IHP was 72.6. BHP was 52, and BMEP, calculated from rpms, displacement and BHP was 82psi. I have no idea of what condition the engine was in. Given it was in the basement of what was then a polytechnical college, the exhaust system may have been very interesting.,

I think the point being made was that you cannot DIRECTLY measure BMEP / IMEP from a gauge.

You have a lot of work to do here to translate that PV diagram into IMEP (sorry for pun).

You`ll need to integrate the area of the curve then decide if you`ll subract the suction area or not (which
contributes if its forced induction or subtracts if its naturally aspirated)>

https://ntrs.nasa.gov/api/citations/19840008393/downloads/19840008393.pdf

It says they used to do this by cutting out the shape on the indicator card then weighing it !

Which if you`ve got some precision card and you remembered to weigh the sheet before you cut it up, is as good a way as any I suppose !

 

[special ed]

It is this kind of explanation why I could never be an engineer. It makes my brain hurt.

 

[pbehn]

It is this kind of explanation why I could never be an engineer. It makes my brain hurt.

That is the simple stuff to draw the foolhardy and unwary into their trap. No one has mentioned flame front velocities, rates of expansion of the chamber and the burning fuel or even energy losses. Unplug your computer, immerse it in bleach and walk away, then you will have been saved.

 

[ThomasP]

re "It says they used to do this by cutting out the shape on the indicator card then weighing it !" for figuring the area under the curve.

This was (and to a degree still is) a common way of calculating the area under the curve. For those of you interested, this is where the term 'weight' came from in calculus when used to refer to a unit of area or 'weight' of the area under the curve. The method was used from the (late?) 1700s til today when the number crunching was/is beyond the people's skill, or beyond the equipment and/or time constraints.

The inconsistency of the density of the paper sheet/card will of course influence the precision to a degree - but this can be overcome by using larger area sheet/card stock, and/or graphing the curve on multiple sheet cards and averaging the results.

I do not know if it is still done but beginning in the late-1800s there were precision paper and vellum material production runs - commissioned by some universities, scientific research institutions, and businesses - used just for the purpose.

I last saw it used as a 'standard' method in the early-1990s, in a materials research lab at Hewlett-Packard. They used calibrated ink-jet printers and plotters to generate the graphs, then cut off the area outside the curve with scissors. They sometimes used specially selected or vellum paper runs for the higher precision requirements, and regular paper of various sizes for the lower precision calculations. It was kind of funny in a way, watching a lab full of PhDs sitting around cutting out paper shapes.

Incidentally, once NC (Numerical Control) or CNC (Computer Numerical Control) milling machines became available they were sometimes used to graph the curve on large pieces of sheet metal, and then have the area outside the curve cut off. The remainder was then weighed to get the resulting area.

I used the method on occasion back before capable computers and software became so prolific. When the mathematical calculations are as complex as integrals can be - when not using a computer generated plot and calculation, I aimed for values within 1-2% precision. The paper/card method is quite capable of this if you are careful and use large enough sheet/card and your method of weighing the paper before and after is precise enough.

 

[emu27]

I think the point being made was that you cannot DIRECTLY measure BMEP / IMEP from a gauge

Radials often used a BMEP gauge for setting power and for adjusting mixture settings, used on the R-1820 in the S-2 Tracker and R-2800, R-3350, R-4360 installations.

BHP = BMEP psi*displacement cubic inches*RPM/792,000

Transpose as necessary to calculate BMEP

 

[Shortround6]

I could be very wrong but I think what they had were torque meters?
Or I am confusing things?

For some reason the radials got torque meters and the V12 engines did not. Something to do with the reduction gears. With Torque and RPM you can figure out the actual power and you can work backwards from torque to figure out the BMEP.
With the appropriate conversions (or 2nd input) done inside the gauge your needle can run over the outside dial giving you HP or ft/lbs of torque or BMEP?

 

[Howard Gibson]

I think the point being made was that you cannot DIRECTLY measure BMEP / IMEP from a gauge.

You have a lot of work to do here to translate that PV diagram into IMEP (sorry for pun).

You`ll need to integrate the area of the curve then decide if you`ll subract the suction area or not (which
contributes if its forced induction or subtracts if its naturally aspirated)>

https://ntrs.nasa.gov/api/citations/19840008393/downloads/19840008393.pdf

It says they used to do this by cutting out the shape on the indicator card then weighing it !

Which if you`ve got some precision card and you remembered to weigh the sheet before you cut it up, is as good a way as any I suppose !

Weighing a cut out piece of paper would be more accurate than selecting the most typical polaroid photo of the batch.

We measured the area with a planimeter. Did I mention the grizzly bears?