what is the maximum HP that an tuned NA M20 can reach

[Ray Smoodiver]

The M20 is the only BMW engine never built for racing (*Wes Hill) and was really built for
the gas crisis of the late '70's. They never built an S20- and there WAS even an S42.

Funfact: the S42 oil reservoir for the dry sump was in the gearbox bellhousing.

S42 cams won't fit in the M42 sadly.

I have an S42 parts pdf from my time at BMW, I'll have a hunt for it.

[dirtysix]

Link to thread?

Here's one http://performanceforums.com/forums/...th-TK-goodness

Even if the dyno is a tad optimistic, still impressive.

[morerevsm3]

http://www.youtube.com/watch?v=LFWsGb_jo5Q

[morerevsm3]

quality of vid is a bit average, video file was corrupt, and after going through repair program, I could watch video fine, but failed with 3 attempts to upload to youtube, so I filmed video playing on my monitor and uploaded, car performs ok for 2 valve SOHC 2.8L engine on triple carbs and 1100+kg race weight....

https://www.youtube.com/watch?v=HiSMd0_i7RA

[Q5Quint]

If you rule of thumb yourself at 100hp/liter for a near perfect/amazingly volumentric efficiency 2.5 liter engine.... you get 250 hp.

An older example is a tweaked out s14 e30 m3 engine: about 200hp out out a 2.5 liter 4 cyl for the road engine, and tweaked into race mode almost 250hp out of a 2.5 liters. The much newer m3 s54 is 3.2 liters with 333hp from the factory. The internet is saying the honda s2000 hits 120hp/liter from the factory 240hp/2liter, with the m3 right behind it at 105/hp/liter. The m3 and the honda are cheating... and I will tell you why.

You wont get much over 100hp/liter NA without getting really funky, redesigning intake and exhaust manifolds and combustion chamber swirl stuff. That is why you are boring and stroking to get more liters out of a m20... it is about your only option without serious intake exhaust head flow work to be done.

350hp out of 2.8 liters? Maybe possible on 93 octane! Ohhhh.... I see you are running ethanol... you cheaty bastard ;-)

125hp/liter, slightly better than our stock s2000 or m3. Nice work. The s2000 and m3 also use cheaty 4 valves per cylinder on twin cams that can vary timing (vtec/vanos). I wonder what it would they could do on e85 tuned up?

This vid says 375, but thats only 117hp/liter for the m3. http://www.youtube.com/watch?v=3qtW7dtkVM8

Look on s2000 or m3 websites for quick power upgrades~ there literally are none because the engine is flowing near perfectly from the factory already.

Internet is saying your volumetric efficiency on a NA engine can get to 110-115% due to things like exhaust gas scavenging, velocity stacks etc. once you get the intake, manifold, head, and exhaust flowing perfectly. Sure you can make any engine do that at any size if you want- things like vanos/vtec/ and variable intake/exhaust geometry are how little 1.6 liter cars today have 150hp and still somehow get 37mpg. God help us when they start turbocharging them~ 200hp 1.2 liter turbo engines will probably be the future.

The question for you is do you want to spend all the $$ to get the less efficient m20 up to snuff with the newer better flowing engines, or do you just want to use one of them to start with? Neither is right or wrong, just a $$ and style question really. Dropping in a chevy 350 might get you more 'power' cheaply but not many do it for their own reasons.

[LJ851]

A yamaha R6 was the first production NA engine to break 200 hp/liter, with fixed cam timing and a warranty too.

Obviously a M20 will not reach these lofty heights.

[digger]

If you rule of thumb yourself at 100hp/liter for a near perfect/amazingly volumentric efficiency 2.5 liter engine.... you get 250 hp.

hp/L has nothing top do with volumetric efficiency (VE) really.
VE is basically in proportion to hp/L/rpm

or for the ones who understand the hp, tq and rpm relationship

it is the same as TQ/L

TQ/L is also esentially proportional BMEP

best naturally aspirated engines make about 110lb-ft/L in Prostock an S54 does about 85lbft/L at peak torque in stock form and at the time it was the best of any production engine. The R6 above is 81lb-ft/L at peak hp rpm but would be slightly higher at peak torque rpm obviously (by definition)

[digger]

A yamaha R6 was the first production NA engine to break 200 hp/liter, with fixed cam timing and a warranty too.

Obviously a M20 will not reach these lofty heights.

2L engine with a lot of development will make more hp/L than a 2.8L but probably top out at the same approx peak hp number. the biggest issue will be making the engine live at over 10,000rpm. imagine a 2L m20 showing the s14 guys what high revs and hp really mean......

[morerevsm3]

the shift light is set on 8500rpm, but engine has seen 9300rpm several times, no ecu means no rev limiter...

[nando]

these days a lot of guys are getting ~330whp and 285ft/lbs on the S54 with stock internals and a tune. I assume they spend a lot of time on VANOS tweaks and possibly the knock sensor settings, which appear to be a bit too conservative. that's around 360bhp.

Do people still remember Dave Length's stroker build? Back then there were no guys like E21jps, digger, morerevsm3 to follow making 250-300+hp M20s. It was an example of everything NOT to do - questionable headwork, parts picked essentially at random (272 cam on a 3.1? no tune? stock exhaust? stock intake manifold/AFM?). I don't even think he broke 200whp which should have been no sweat.

I think that stupid motor set the bar for a long time which is why everyone swaps in lame M50s that make like 10whp more than a healthy stock M20. Because one guy tries to run a 3.1 on a stock chip, gets poor output and everyone thinks that's the most you can get out of it. Too bad the car was wrecked before he could finish sorting it out, although with headwork that looked like somebody took an angle grinder to the ports I don't think it had much more potential.

[digger]

these days a lot of guys are getting ~330whp and 285ft/lbs on the S54 with stock internals and a tune. I assume they spend a lot of time on VANOS tweaks and possibly the knock sensor settings, which appear to be a bit too conservative. that's around 360bhp.

Do people still remember Dave Length's stroker build? Back then there were no guys like E21jps, digger, morerevsm3 to follow making 250-300+hp M20s. It was an example of everything NOT to do - questionable headwork, parts picked essentially at random (272 cam on a 3.1? no tune? stock exhaust? stock intake manifold/AFM?). I don't even think he broke 200whp which should have been no sweat.

I think that stupid motor set the bar for a long time which is why everyone swaps in lame M50s that make like 10whp more than a healthy stock M20. Because one guy tries to run a 3.1 on a stock chip, gets poor output and everyone thinks that's the most you can get out of it. Too bad the car was wrecked before he could finish sorting it out, although with headwork that looked like somebody took an angle grinder to the ports I don't think it had much more potential.

cant see anything but a heavily worked 3.5L S54 making 285wtq which is the pic of the bunch for BMW 6 cyl engines. the S54 always made more torque per cubic inch than the rest and even latest NA engines. Some of the uber $$ Porsches for example GT3 etc have eclipsed this now though.

certainly simply throwing parts together doesn't make a good engine. and the power certainly comes from the induction, head and exhaust. here is an old plot of my first dyno with the 3L bottom end and what simply playing with exhaust, cam and induction can do with tuning. This didn't even include an ITB setup which would have added even more particularly to the topend and useable rpm.

All these runs are with the same bottom end you can’t just use a big bottom end and leave it at that.

[SoundByte]

All these runs are with the same bottom end you can’t just use a big bottom end and leave it at that.

So, are you basically saying that it's not worth building a motor with a 2.7 block unless you're prepared to tear into the block?

I only ask because I plan to take the path of least resistance when I first assemble my 2.7i motor (stock b27 bottom end, stock b25 top end, slightly reworked exhaust) simply because money is an issue, and I want to kinda take some baby steps into building myself a performance motor.

I realize that I'm not gonna be breaking 200 whp with that build (I'm not exactly building a racecar here; It'll be DD'd), but I figured that there's always room to further modify it, as long as I'm willing to tear it apart and put it back together again.

[digger]

So, are you basically saying that it's not worth building a motor with a 2.7 block unless you're prepared to tear into the block?

I only ask because I plan to take the path of least resistance when I first assemble my 2.7i motor (stock b27 bottom end, stock b25 top end, slightly reworked exhaust) simply because money is an issue, and I want to kinda take some baby steps into building myself a performance motor.

I realize that I'm not gonna be breaking 200 whp with that build (I'm not exactly building a racecar here; It'll be DD'd), but I figured that there's always room to further modify it, as long as I'm willing to tear it apart and put it back together again.

i wouldn't be doing a built a bottom end that is 2.7L unless you are abiding by some rules or regulations for a racing class. it just limits the performance for a street driven vehicle if staying NA. if you dont have money for a new bottom end just leave it stock or freshen it.

going for extra capacity will give you extra bottom end and midrange but you wont inherently make good power, to make power (and torque) its more about improving the inlet, head and exhaust.

all these budget 2.7L dont have the bases covered to make more power than a stock B25 hence they never do

just look at what alpina, hartge, etc did......mild port job and cleanup, 270ish cam, nice exhaust, proper tune. do the same on a healthy stock bottom end and you will make more power than what a fresh B25 does. these days there are more options with ECU etc for better tuning and less restrictive inlet systems

[SoundByte]

The part I'm still fuzzy on is how I know how to do the fuel map/tune for whatever build I end up doing.

Any suggestions on improving the b27 block? I've read that using a b25 crank and pistons is the optimal cheap solution, even though it enlarges the combustion chamber by a tiny bit.

Also, I don't think paying a shop for a port job is gonna be feasible for getting the car drivable in as short a time as I'm planning. I've got a mated b25 intake/head/exhaust that will give me adequate gains for a first build. I'm just trying to get comfortable with the process while building a reasonably economical DD motor to use while I slowly piece together a truly nice M20.

Should I expect to have problems with DEQ on a 270 cam?

[Q5Quint]

Not if you can tune the air/fuel properly to pass emissions.

We don't butt sniff anything older than 96 in my state, so it isnt an issue.

I tried to do some cam-emissions research and came up with mostly camaro stuff:

http://ls1tech.com/forums/generation-iii-internal-engine/453311-cam-lope-emissions-compliance.html

Some cams are carb certified and apparently with the overlap correct should still pass emissions.... but it will be a 'tune to pass emissions' job I suspect.

The key to passing the sniffer and having lope is by selecting a cam that has -7 degrees of overlap at .050", then having the idle fairly high (900rpm) when you need to go for inspection and 700 rpm for the lope you want. Even the stock cam will lope very hard with a 450 rpm idle. You can tailor your lope tastes by raising and lowering your idle speed, it's as simple as that.