BMEP and other technical terms
Discussion
Ok on Gazboys request on another post I'm going to cover my understanding of the term BMEP.
I'm actually going to shameless rip-off/quote some of my previous posts and copy and paste here.
BMEP stands for Break Mean Effective Pressure.
Think of BMEP it as specific torque- or torque per litre- that's all.
It allows you to compare how well an engine produces torque for the capacity, so you can compare submarine engines (say) with Fiat Cinqecentos!
Worked example:E36 3.2 litre M3 engine makes 350 Nm of torque at 3250 rpm .
The engine capacity is 3.201 litres.
So 350/3.201=109.34
109.34 X 0.1257= 13.744 bar BMEP.
And finally, to put things into perspective, I've seen the ignorant comparing pushrod Chevy V8 engines with twin cam 4 valve per cylinder VCT engines- thinking that GM haven't been trying- which is as pointless as comparing a Turbo charged engine with a naturally aspirated engine.
To be a bit fairer let's NOT compare boosted engines with naturally aspirated
The next thing should be to seperate home modifed cars from production cars by the manufacturers, the manufacturers are far more constrained (with emissions and durability etc etc), and there's alot of unscrupilous "pub talk"- with people claiming to get 100/Bhp / litre from naturally aspirated 2 valvers all the time!
This post will focus on the most difficult of all: Production naturally aspirated cars design to be sporty road units- NOT race engines.
The most difficult task is to obtain high BHP/litre from a Naturally aspirated engine (like BMW and Honda) and THEN achieve good BMEP (or specific torque at low speeds) Like BMW have with the E36 3.2 litre S50 M3 engine. Infact I would rate the E36 M3 3.2 litre way above in terms of achievement over the honda S2000, no other car can achieve close to 14 bar BMEP at 3250 rpm and over 100 Bhp/litre at 7400 rpm-such incredible range. (The new CSL is even more impressive as it achieves 14.3 bar BMEP and 360 Bhp from a 3.2 litre)
I’d say that in a production type of engine it is very very difficult to get 100 Bhp per litre from a 4 Valve. It’s also pretty impressive going getting 105 Nm/litre (77lb ft/litre)(13.2 bar BMEP) out of a 4 valve unit and quite hard to achieve in practice.
The very best two valve I’ve seen approach perhaps 80/Bhp/litre from the 911/993 Porsche RS but this was a twin plug affair. A more realistic and achievable figures are like the Chevy Vette C5 engine and the BMW M20 325i E30 unit-for two valves they produce around 68-70 Bhp/litre. Obviously you can compromise low speed and mid range to up this figure- but getting above 80/litre is good going! The best specific torque you’re likely to see is 93.9Nm/litre (11.8 bar BMEP) or 69 lb ft per litre- that’s good going!'
I'm actually going to shameless rip-off/quote some of my previous posts and copy and paste here.
BMEP stands for Break Mean Effective Pressure.
Think of BMEP it as specific torque- or torque per litre- that's all.
It allows you to compare how well an engine produces torque for the capacity, so you can compare submarine engines (say) with Fiat Cinqecentos!
wheeljack888 said:
Textbook equation for BMEP is:
BMEP (Pa) = n * P / D * N
or because P = T * 2 * pi * N
BMEP = n * T * 2 * pi / D
where
n = no. of revs per power stroke (2 for 4 stroke)
P = Power (watts)
D = displacement (m^3)
N = revs / sec
T = torque (Nm)
BMEP (Pa) = n * P / D * N
or because P = T * 2 * pi * N
BMEP = n * T * 2 * pi / D
where
n = no. of revs per power stroke (2 for 4 stroke)
P = Power (watts)
D = displacement (m^3)
N = revs / sec
T = torque (Nm)
Worked example:E36 3.2 litre M3 engine makes 350 Nm of torque at 3250 rpm .
The engine capacity is 3.201 litres.
So 350/3.201=109.34
109.34 X 0.1257= 13.744 bar BMEP.
And finally, to put things into perspective, I've seen the ignorant comparing pushrod Chevy V8 engines with twin cam 4 valve per cylinder VCT engines- thinking that GM haven't been trying- which is as pointless as comparing a Turbo charged engine with a naturally aspirated engine.
To be a bit fairer let's NOT compare boosted engines with naturally aspirated
The next thing should be to seperate home modifed cars from production cars by the manufacturers, the manufacturers are far more constrained (with emissions and durability etc etc), and there's alot of unscrupilous "pub talk"- with people claiming to get 100/Bhp / litre from naturally aspirated 2 valvers all the time!
This post will focus on the most difficult of all: Production naturally aspirated cars design to be sporty road units- NOT race engines.
The most difficult task is to obtain high BHP/litre from a Naturally aspirated engine (like BMW and Honda) and THEN achieve good BMEP (or specific torque at low speeds) Like BMW have with the E36 3.2 litre S50 M3 engine. Infact I would rate the E36 M3 3.2 litre way above in terms of achievement over the honda S2000, no other car can achieve close to 14 bar BMEP at 3250 rpm and over 100 Bhp/litre at 7400 rpm-such incredible range. (The new CSL is even more impressive as it achieves 14.3 bar BMEP and 360 Bhp from a 3.2 litre)
I’d say that in a production type of engine it is very very difficult to get 100 Bhp per litre from a 4 Valve. It’s also pretty impressive going getting 105 Nm/litre (77lb ft/litre)(13.2 bar BMEP) out of a 4 valve unit and quite hard to achieve in practice.
The very best two valve I’ve seen approach perhaps 80/Bhp/litre from the 911/993 Porsche RS but this was a twin plug affair. A more realistic and achievable figures are like the Chevy Vette C5 engine and the BMW M20 325i E30 unit-for two valves they produce around 68-70 Bhp/litre. Obviously you can compromise low speed and mid range to up this figure- but getting above 80/litre is good going! The best specific torque you’re likely to see is 93.9Nm/litre (11.8 bar BMEP) or 69 lb ft per litre- that’s good going!'
Very nice.
Any chance of a more complete equation though
Quite odd getting metric Pa BMEP values and then your referring to Bar BMEP values wrt to the M3 etc.
Just to save people getting confused anyway
Also, what does BMEP suggest then? A more complete picture of the engine's specific output apart from bhp/litre?
Are revs irrelevant then? Surely it's pretty flawed in itself if like lots of modern engine's they are "de-tuned" to give flat torque outputs, then the peak torque may not be indicative of the engine's actual ability. 3.2 VR VAG lump for example... Phaeton, R32, A3 3.2 and 3.2 TT all have varying outputs for example...
Just curious why people use BMEP.
BMEP = n * T * 2 * pi / D
All constants except T and D, and thats basically BMEP is proportional to T/D = Nm/litre?
Dave
Any chance of a more complete equation though
Quite odd getting metric Pa BMEP values and then your referring to Bar BMEP values wrt to the M3 etc.
Just to save people getting confused anyway
Also, what does BMEP suggest then? A more complete picture of the engine's specific output apart from bhp/litre?
Are revs irrelevant then? Surely it's pretty flawed in itself if like lots of modern engine's they are "de-tuned" to give flat torque outputs, then the peak torque may not be indicative of the engine's actual ability. 3.2 VR VAG lump for example... Phaeton, R32, A3 3.2 and 3.2 TT all have varying outputs for example...
Just curious why people use BMEP.
BMEP = n * T * 2 * pi / D
All constants except T and D, and thats basically BMEP is proportional to T/D = Nm/litre?
Dave
Mr Whippy said:
Very nice.
Any chance of a more complete equation though
Quite odd getting metric Pa BMEP values and then your referring to Bar BMEP values wrt to the M3 etc.
Just to save people getting confused anyway
Also, what does BMEP suggest then? A more complete picture of the engine's specific output apart from bhp/litre?
Are revs irrelevant then? Surely it's pretty flawed in itself if like lots of modern engine's they are "de-tuned" to give flat torque outputs, then the peak torque may not be indicative of the engine's actual ability. 3.2 VR VAG lump for example... Phaeton, R32, A3 3.2 and 3.2 TT all have varying outputs for example...
Just curious why people use BMEP.
BMEP = n * T * 2 * pi / D
All constants except T and D, and thats basically BMEP is proportional to T/D = Nm/litre?
Dave
Any chance of a more complete equation though
Quite odd getting metric Pa BMEP values and then your referring to Bar BMEP values wrt to the M3 etc.
Just to save people getting confused anyway
Also, what does BMEP suggest then? A more complete picture of the engine's specific output apart from bhp/litre?
Are revs irrelevant then? Surely it's pretty flawed in itself if like lots of modern engine's they are "de-tuned" to give flat torque outputs, then the peak torque may not be indicative of the engine's actual ability. 3.2 VR VAG lump for example... Phaeton, R32, A3 3.2 and 3.2 TT all have varying outputs for example...
Just curious why people use BMEP.
BMEP = n * T * 2 * pi / D
All constants except T and D, and thats basically BMEP is proportional to T/D = Nm/litre?
Dave
As indicated in the post it's an indication of Specific torque for comparison purposes, it's all written up there.
Bar BMEP is a common convention that's well known-it's not desined for laymen- more for technical folk, who usually know what other comparable engines in the field produce- but it can be an eye opener when enthusiasts use it to on their favourite sports cars. No more different using bar over Pa than using Bhp rather than KW, and BHP/litre. It's just a convention that people are familiar with over here- Americans often use PSi, nothing to get excited about or split hairs about, just do the conversion
Who's informed you that the 3.2 litre VW lump was tuned for low speed torque? VW themselves?
Nothing flawed about using BMEP as a measure as long as it is used in context, i.e comparing what BMEP other competitor engines make at 1000 rpm, then at 2000 rpm then at 2500 rpm etc. Quite simple really.
Marquis_Rex said:
Mr Whippy said:
Very nice.
Any chance of a more complete equation though
Quite odd getting metric Pa BMEP values and then your referring to Bar BMEP values wrt to the M3 etc.
Just to save people getting confused anyway
Also, what does BMEP suggest then? A more complete picture of the engine's specific output apart from bhp/litre?
Are revs irrelevant then? Surely it's pretty flawed in itself if like lots of modern engine's they are "de-tuned" to give flat torque outputs, then the peak torque may not be indicative of the engine's actual ability. 3.2 VR VAG lump for example... Phaeton, R32, A3 3.2 and 3.2 TT all have varying outputs for example...
Just curious why people use BMEP.
BMEP = n * T * 2 * pi / D
All constants except T and D, and thats basically BMEP is proportional to T/D = Nm/litre?
Dave
Any chance of a more complete equation though
Quite odd getting metric Pa BMEP values and then your referring to Bar BMEP values wrt to the M3 etc.
Just to save people getting confused anyway
Also, what does BMEP suggest then? A more complete picture of the engine's specific output apart from bhp/litre?
Are revs irrelevant then? Surely it's pretty flawed in itself if like lots of modern engine's they are "de-tuned" to give flat torque outputs, then the peak torque may not be indicative of the engine's actual ability. 3.2 VR VAG lump for example... Phaeton, R32, A3 3.2 and 3.2 TT all have varying outputs for example...
Just curious why people use BMEP.
BMEP = n * T * 2 * pi / D
All constants except T and D, and thats basically BMEP is proportional to T/D = Nm/litre?
Dave
As indicated in the post it's an indication of Specific torque for comparison purposes, it's all written up there.
Bar BMEP is a common convention that's well known-it's not desined for laymen- more for technical folk, who usually know what other comparable engines in the field produce- but it can be an eye opener when enthusiasts use it to on their favourite sports cars. No more different using bar over Pa than using Bhp rather than KW, and BHP/litre. It's just a convention that people are familiar with over here- Americans often use PSi, nothing to get excited about or split hairs about, just do the conversion
Who's informed you that the 3.2 litre VW lump was tuned for low speed torque? VW themselves?
Nothing flawed about using BMEP as a measure as long as it is used in context, i.e comparing what BMEP other competitor engines make at 1000 rpm, then at 2000 rpm then at 2500 rpm etc. Quite simple really.
Yeah I get the idea of BMEP purely being for comparison.
I understand Bar BMEP is nicer than Pa values which are huge in comparison, and obviously not imperial
I just thought since you used metric Si values for the inputs in your equations you'd use them in the rest of your post thats all. The 3.2 VAG lump wasn't tuned for low speed torque, it's output may not well be the limits of the engine thats all. The 3.2 TT for example appears more peaky in it's delivery (by numbers) than the R32 Golf. I'm sure all the engines are identical inside, but the ECU controls the delivery.
Perhaps I'm confused here, but why use BMEP when torque/litre has the same relative value to the BMEP figure?
BMEP = n * T * 2 * pi / D
All you really do is halve the effectiveness of the two stroke engines relative to the four strokes, with 2 * pi being just a constant.
Surely just saying 350Nm/3.2 litres is good? 109.375 Nm/litre @ 4900rpm, 87.5Nm/litre at 3000rpm.
Or using BMEP
4*350*2*pi/0.0032 = 2,748,891 Pa BMEP @ 4900rpm (peak torque)
4*280*2*pi/0.0032 = 2,199,113 Pa BMEP @ 3000rpm (80% torque at 3000rpm according to BMW figures and dyno's)
So comparing we get 2,748,891 Pa BMEP / 109.375 Nm/litre = 25132.7 Pa BMEP / Nm/litre (for any four stroke engine)
Ie, 87.5Nm/litre @ 3000rpm ... 87.5 * 25132.7 = 2199113 Pa BMEP...
Not having a go, just I can't see any huge significance in BMEP when Nm/litre or lbft/cu.in is simple mental arithmetic in comparison and has the same relative values to BMEP as shown above.
Just curious thats all
Dave
Just a quickie.
You need to divide your no.s by a factor of two (otherwise you've got it as a powerstroke every 4 revolutions). Also to get bar from Pa or nm^-2 simply divide by 100000.
BMEP is simply just the prefered convention for normalising output wrt capacity. You can further break up bmep into factors like imep, fmep, pmep (indicated, friction, pumping mean effective pressure).
BMEP = IMEP - FMEP - PMEP
If your really clever you can predict some of these factors using mathematical models, benchmarking data, etc and then working backwards predict the flywheel output of an on-paper concept engine.
You need to divide your no.s by a factor of two (otherwise you've got it as a powerstroke every 4 revolutions). Also to get bar from Pa or nm^-2 simply divide by 100000.
BMEP is simply just the prefered convention for normalising output wrt capacity. You can further break up bmep into factors like imep, fmep, pmep (indicated, friction, pumping mean effective pressure).
BMEP = IMEP - FMEP - PMEP
If your really clever you can predict some of these factors using mathematical models, benchmarking data, etc and then working backwards predict the flywheel output of an on-paper concept engine.
wheeljack888 said:
You need to divide your no.s by a factor of two (otherwise you've got it as a powerstroke every 4 revolutions). Also to get bar from Pa or nm^-2 simply divide by 100000.
Ah ok, that makes more sense now
So two strokes just get a 50% reduction in BMEP for the same torque and capacity
What is a wankel rotary engine classed as wrt to power strokes per rotation?
Dave
Marquis_Rex said:
350matt said:
Hang on a sec chap there's a couple of 4.5 cerbys making about 90Bhp / litre now and thats only 2 valve
Addmittedly slightly streching the term 'production' engine
Matt
Addmittedly slightly streching the term 'production' engine
Matt
Phil you can handle this one seeing as you're more the TVR expert
Is that Melling Pony Powers per litre on a friendly dyno? I jest, I think it's a stonking engine but dare I say there has to be some slight exaggeration?
The best I've seen quoted otherwise is the twin-plug 2-Valve Ducati SS is 992cc 95bhp @ 8000 (10.8 Bar) and 96 Nm @ 6000 (12 Bar).
Mr Whippy said:
What is a wankel rotary engine classed as wrt to power strokes per rotation?
Dave
Dave
Isn't it equivalent of a two-stroke? Three power strokes per rotor rev but shaft output is three times rotor speed, therefore equivalent of one power stroke per output shaft rev.
If so the RX8 is 1.3 litres and the peak torque is ~220Nm that gives a peak BMEP of ~10.6 bar, sounds about right to me!
Sorry chaps
but with a few mods ( short induction hoses and a re-map) a 4.5 cerby will generally make 400Bhp this has been verifed on about a fair few rolling roads ( inc surrey, nobles, emerald, Trackcar) by now and there are quite a few cars out there at least 25+ all pulling this sort of power the best so far is at 425Bhp (without touching internals)
If your're talking from the factory power however then most make 360ish and once re-mapped typically 385Bhp ( yes the std mappping is cr'p)
Matt
but with a few mods ( short induction hoses and a re-map) a 4.5 cerby will generally make 400Bhp this has been verifed on about a fair few rolling roads ( inc surrey, nobles, emerald, Trackcar) by now and there are quite a few cars out there at least 25+ all pulling this sort of power the best so far is at 425Bhp (without touching internals)
If your're talking from the factory power however then most make 360ish and once re-mapped typically 385Bhp ( yes the std mappping is cr'p)
Matt
What kind of engine dyno was that measured on?
Was the dyno calibrated to DIN72000? (even if the engine dyno was calibrated to DIN72000 there is always dyno to dyno variation- if the dyno is rated at a much higher output than the engine being tested- it's generally accepted than there will be less accuracy.
Was that measured with EEC correction factors, SAE or DIN?
Was the mapping checked for MBT ignition/DBL and mixture for LBT/or catalyst protection?
Was the dyno calibrated to DIN72000? (even if the engine dyno was calibrated to DIN72000 there is always dyno to dyno variation- if the dyno is rated at a much higher output than the engine being tested- it's generally accepted than there will be less accuracy.
Was that measured with EEC correction factors, SAE or DIN?
Was the mapping checked for MBT ignition/DBL and mixture for LBT/or catalyst protection?
well oddly enough none of them have been removed from the cars and fitted to an appropiate dyno, as I mentioned they've all been measured on reputable rolling roads. Which while aren't as accurate or repeatable as a proper engine dyno can still give workable results and some might argue are more relavent to everyday driving being as they measure power at the wheels so measuring what the installed engine is producing in its working enviroment.
I dare say the dyno's were calibrated when installed and mabye once a year after that, however being as there are a couple of cars which have been tested on 3 of these dynos and produced repeatable results to one another ( as I recall a variation of 8 Bhp across all three was reported) they are probably not too far from the truth.
I'd imagine everybody is using the DIN std correction factors (SAE? not this side of the pond) and I know on at least 2 instances the cars were mapped on those rollers so yes MBT spark timing and best fuelling for power as the TVR catylsts aren't exactly looked after from the factory.
The coastdown factor which as you know is critical for determining flywheel horsepower when measuring power at the wheels has been set so when a std factory car is tested is records the manufacturers quoted horsepower.
This does mean that some rollers actually under read from what the engine is actually producing and some overread depending on how honest that particular manufacturer has been.
What's with the agressive tone? I generally enjoy reading your posts as they are informative and well written, I'm not out to challenge your knowledge rather looking to share what I know, so we can all learn from one another.
Matt
I dare say the dyno's were calibrated when installed and mabye once a year after that, however being as there are a couple of cars which have been tested on 3 of these dynos and produced repeatable results to one another ( as I recall a variation of 8 Bhp across all three was reported) they are probably not too far from the truth.
I'd imagine everybody is using the DIN std correction factors (SAE? not this side of the pond) and I know on at least 2 instances the cars were mapped on those rollers so yes MBT spark timing and best fuelling for power as the TVR catylsts aren't exactly looked after from the factory.
The coastdown factor which as you know is critical for determining flywheel horsepower when measuring power at the wheels has been set so when a std factory car is tested is records the manufacturers quoted horsepower.
This does mean that some rollers actually under read from what the engine is actually producing and some overread depending on how honest that particular manufacturer has been.
What's with the agressive tone? I generally enjoy reading your posts as they are informative and well written, I'm not out to challenge your knowledge rather looking to share what I know, so we can all learn from one another.
Matt
Edited by 350matt on Friday 25th August 22:45
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff