3.0 litre stroker

[jaybrand]

anybody done one of these and what parts are involved .done a quick search but could only find a link which is not active

[StuBeeDoo]

Ireland Engineering in America do 2.9, 3.0 and 3.1 M20 kits.
I'm not sure what cranks they use, but IIRC they bore the block +1mm and use custom pistons (and possibly rods).

Big money as well, I'd have thought.

Do a Google on Ireland Engineering.

[StuBeeDoo]

I've done some calculations, but they're based on assumption so they may well not be 100% accurate.

2.9 = 2.8 (M52??) crank + 1mm overbore.
3.0 = 3.0 (early E36 M3 89.6mm) crank + no overbore
3.1 = 3.0 as above with 1mm overbore.

If there are any serious errors in my thinking I'm sure someone will be along to correct me.

[BadDave]

There could be some info in this thread

http://www.e30zone.net/modules.php?name ... c&t=150844


AND
It may be worth taking a look at this one

http://www.gevans.info/m20/m20.php

[StuBeeDoo]

Some interesting info in that second link, Dave. But I dispute a couple of items on his stroker info.

[HairyScreech]

^^ does that last site belong to mr march 109?

[eta]

M54 crank with the same modifications as required for the M52 crank to fit the M52 block and there was some discussion in HairyScreech's thread about the possibility of using 130mm con rods and modified 2.5i piston or unmodified alpina pistons. An 885 head is a must.

[AlpineAde]

Check out these blokes, too:

http://www.metricmechanic.com/

[BadDave]

^^ does that last site belong to mr march 109?

It would indeed

[HairyScreech]

anyone seen him recently?

the 3.0 crank has an 89.6mm stroke, so i was working on the theory that it has 7.3mm extra throw over a 75mm 325.

which left me thinking:

320 rod =5mm shorter
alpina piston =2mm shorter
knocking 0.3mm from the squish clearance leaves over 1.4mm to the head.

cant see any reason for this not working besides the pistons clearing the crank, which may be bear tight.

im considering looking into getting a batch of the b28 pistons done, which would be 3.5mm ish shorter (i forget off the top of my head) and if i did then i would alter them to have a slipper style skirt which would clear all of the stroker cranks.

just knocked this up, should explain a few things.

[eta]

According to that the M20B28 does not work as the stack hieght too much at 214mm!

Also with the M54 crank the stack height is 216.8 mm. Knocking 2mm of with the Alpina pistons would not solve that according to that sheet. Although your earlier logic is sound as previously discussed.

I have to say I wonder how a 3ltr stroker with 130mm conrods would be in reality. As for the B28 pistons would they not remove the squish band that you like so much?

[HairyScreech]

an m20b28 will work with 130mm rods, any longer rods and you need shorter pistons.

the works true/false is only if the combination is under 207mm stack height with out modification.

if you wanted to be strict the m20b26 with 323 crank would also be an issue as its sat 4mm down the bore.

i think i have made a formatting error with the bottom of the sheet, will check correct and update as required now

edit - simple formatting error, fixed.

[Andyboy]

The connecting rod length must never be less than the crank stroke x 1.5. The 3.0 M54 crank stroke is 89.6mm so that times 1.5 is 134.4mm. Therefore you cannot use the 320i rods because the side thrust would be horrendous - massive friction, power loss and rapid bore/ring wear. It would just be rubbish. The 135mm 325i rods are also very, very borderline and not really acceptable. You need rods as long as possible to reduce piston side thrust to a minimum, which is why race engines use long rods. With the 81mm Eta crank it's 121.5 but Alpina used the longer 135mm rods and special pistons. When I built a 2.8 around 7-8 years ago I was a bit worried about using 130mm rods with the 84mm crank - 126mm which was a bit close really. It did quite good power but it wasn't really a 'revver'.

Therefore, for a 3.0 you need the longer M52/M54 rods (iirc these were about 140mm....?) and custom pistons. Or even better, just start with an M50 2.5 and do it properly. 3.0 M20's are a waste of time because you will never get enough air past one inlet valve per cylinder to make the expense worthwhile. I'm sure the American tuners will quote all kinds of dreamland power figures - well, whatever!

Put it this way - armed with a shagged 3.0 M54 with head gasket woes and a decent complete M50 2.5, you can screw together a very respectable 230 bhp engine for well under a grand including new shells, rings, rods bolts and gaskets. You already have a modern ECU and a MAF free of charge.
Or you can p1ss about with some ancient 12v boat anchor and spend three times that and get 20 bhp less. Hmmm.........

[Ziggy]

The connecting rod length must never be more than the crank stroke x 1.5.

Less than? All news to me, but it would seem to follow the rest of your post rather better!

[GeoffBob]

The connecting rod length must never be more than the crank stroke x 1.5.

In general agreement with all your comments Andy, but I think you meant to say that the connecting rod length should never be less (not more) than the crank stroke x 1.5.

M20's are a waste of time because you will never get enough air past one inlet valve per cylinder to make the expense worthwhile.

The fact that mean piston velocity is directly proportional to the stroke is the major contributing factor here, given that the faster the mean piston velocity the higher the required velocity of the inducted air in order to fill the cylinders. And once the valves are choked (meaning the air flow passed the valves is approaching sonic velocity) you can stroke that engine till you're blue in the face, you still won’t get any more air in those cylinders!

Sorry folks, not meaning to start a fight, but I do believe that Andy has raised a number of good points.

[phelix]

I believe M52B28, M52TUB28 and M54B30 rods are all 135mm.

Metric Mechanic sells 2 different engine kits for the 89.6mm crank. The lower spec one uses custom pistons with 135mm rods from an M3; the higher spec one uses custom pistons with custom 138mm rods.

[eta]

Those points had occured to me as well but I really had not worked anything out so I really did not want to coment until know.

I have a spreadsheet which computes cylinder wall forces against crank angle based on a thermodynamic model of the basic otto cycle. While this makes a few simplication on what really takes place in an engine, the peak cylinder wall forces for a M20B30 are 4% greater than and M20B27. This is hardly earth shattering and does assume I have done the sums right. I have calculated this using 130mm con rods and a CR of 9.34:1 for the 2.7i and a Cr of 10.3:1 for the 3.0i. The CR of the 3.0i is probably wrong but I don't know what the volume of bowl of the modified 325i piston.

So while the friction forces are greater whether they would cause a noticable wear increase I don't know.

As for the piston speed the average piston speed for the 3.0i is 16.7 m/s and for 2.7i it is 14.85 m/s at 5500 rpm. Peak piston speed for the 3.0i will be 26.03 m/s and for the 2.7i peak piston speed will be 25.95 m/s. Whether these differences are enough to choke the air flow into the engine I again don't know but when I plot piston speed against crank angle there is not a whole lot of difference for a 130mm con rod length in both engines.

In conclusion, while the arguments above are sound in theory I am not sure if the maths backs them up.

[GeoffBob]

Hmmmm, interesting results Eta. I assume that you have multiplied the combustion pressure (Force per unit area) by the area of the piston face to determine the downward force as a function of crank angle? Then (by trigonometric method along the line of the connecting-rod) determined the lateral force against the cylinder wall?

If so then your results dispute what Andy said, as well as my general feelings on the matter (but then I didn't do any calculations, and you did) so, unless something is missing from your calculation, I am inclined to take your word on this.

As for the piston speed, those velocities are indeed sufficient to choke the inlet valve. As a rough guide, imagine the valve aperture were a perfect round hole of exactly one tenth the area of the piston (for example). The air velocity through the aperture would thus be required to be ten times the velocity of the piston in order to fill the cylinder. Thus, as the piston approached a peak velocity of 30m/s, the aperture would begin to choke.

The situation is obviously quite different with a real engine since the valve lift (and thus aperture size) is a function of crank angle, and the aerodynamics of the valve lip and seat come into play. It is, however, quite possible to determine an effective aperture area of a valve (provided one can determine/estimate the valve discharge coefficient) as a function of crank angle. The result is (and this is well documented) that engine speed (on any engine) is prohibited beyond a certain value due to the airflow passed the valves progressively becoming choked, and progressively increasing internal engine friction at higher rpm. This is the inherent reason why torque graphs fall down the way they do at high rpm. The modern ”atuned”a intake manifold usually only exacerbates this problem since they are OEM tuned to operate at ”asensible”a rpm and not intended to provide any sort of VE gain at high engine speed. So in short, stroking a motor can only worsen this situation.

To my mind this dictates a motor that would be short-changed on the rpm front, exactly as Andy described, and thus absolute power gains (due to limited engine speed) may be marginal. There is no doubt in my mind, however, that torque gains are there to be had at lower rpm!

[Mikey_Boy]

M20's are a waste of time because you will never get enough air past one inlet valve per cylinder to make the expense worthwhile.

The fact that mean piston velocity is directly proportional to the stroke is the major contributing factor here, given that the faster the mean piston velocity the higher the required velocity of the inducted air in order to fill the cylinders. And once the valves are choked (meaning the air flow passed the valves is approaching sonic velocity) you can stroke that engine till you're blue in the face, you still won’t get any more air in those cylinders!

Sorry folks, not meaning to start a fight, but I do believe that Andy has raised a number of good points.[/quote]

Spot on Geoff - if you look at NASCAR, they achieve some pretty impressive power numbers on 2 valves per cylinder and limited regulations - there however, the power comes from huge intake valves, good porting, optimised components (really optimised components - NASCAR pistons and valve train is something to behold) and revs revs revs...

Are M20's a waste of time? Certainly not - it is very robust and simple design for it's application - a road car designed back in the 70s. The BMW guys did a fantastic job and in the road car that it was, it ticked every single box and then some. Some would argue that BMW got it right again as the S14 exists - high specific output (for the time), tuneable, handles revs etc etc etc. Clearly there was a motorsport aspiration and vision with that application, same as the Sierra Cosworth etc etc. Of course S14s are rare, expensive to buy and rebuild and once tuned, pretty compromised.

Clearly the M20 is always going to be limited when tuning N/A as we don't have the luxury of limitless funds so I think we all have to accept that whatever is done it isn't going to be optimal. The fun of course is squeezing a touch more where we thought it couldn't be done.

If all we were worried about is power and more power then we would all be turboing the arse off M20's, swapping for S5x, turboing them or installing S85 and then turboing them.

The great thing here is that we have options - BMW have done a great job with the progession of time and being German and realising what isn't broke doesn't need fixing, the basic architecture stays similar which means we can play with various engine options.

Like Geoff, I don't wish to start a fight - I will be quite happy with my 2.8 once build and 'only' 220 hp or so, satisfied that it revs nicely, sounds good and has been well built - for my track car application, that I hope will be enough... For now...!



Cheers,
Mike

[eta]

Geoff that is exactly the method I used used to determine wall loading. It took a while to get the spreadsheet to throw out sensible numbers as I kept on entering the formula's incorrectly; incorrectly placed brackets.

Your comments about the valves size make perfect sense and I did think about trying to work out the mass flow rate through the valves but the valve discharge coefficent and something else that I forget are unknown to me unless I take an educated stab in the dark.

I do in a sense agree with Andyboy in that if you want a 3.0i it is proably cheaper and easier to buy an M54B30 and an M50B25 and get to work.

[Andyboy]

Are M20's a waste of time? Certainly not - it is very robust and simple design for it's application - a road car designed back in the 70s.
Clearly the M20 is always going to be limited when tuning N/A as we don't have the luxury of limitless funds so I think we all have to accept that whatever is done it isn't going to be optimal. The fun of course is squeezing a touch more where we thought it couldn't be done.

If all we were worried about is power and more power then we would all be turboing the arse off M20's, swapping for S5x, turboing them or installing S85 and then turboing them.

That's just it. As good as it is (and I like the old M20 a lot) it's a seventies design. There exists a law known as the one of diminishing returns. As soon as you get past around 210 bhp, it all starts to cost a lot of money and you will never, ever get as much power as you will from a 24v M50 engine. Consider that the Alpina 3.4 engine in the Z Roadster - still an iron block unit with single Vanos - develops 300 bhp. It also does 30 mpg on a run, is turn key reliable and has emissions more than capable of passing an MOT.

The optimal M20 is one with 2.7 or 2.8 litres. A 2.8 won't really give a lot more power than a good 2,7, but the cranks are easier to find and with the extra 3mm stroke you're not machining chunks off the block or having to mess around with vernier pulleys. This optimal engine also uses Motronic 1.3 and an air flow meter, nice and simple like the engine and well suited to it. The problem these days is the lack of decent tuners who can accurately map a Motronic ECU - not some halfarsed generic chip, but a proper job. To fit a MAF, mess about with Unichip/wahetever and set it up on a dyno costs about 3 times what a decent running M50 engine does so I don't see much point in that either. Yes, the results are good but at what cost? At the end of the day, a solid 200 bhp with 200lb.ft is enough to make an E30 with the right gearing plenty quick enough.


And yes, I meant to say less, not more! The longer the rods the better - less side thrust, less friction, less wear, more power. That's why BMW lengthened the rods on the 325i - the 323i had a longer stroke and the 130mm rods and they were buggers for piston slap.

[eta]

I really wish there was someone local who properly tune motroic 1.3 but there isn't. Apart from A-tech who actually does it?

[Sebastian35]

anybody done one of these and what parts are involved .done a quick search but could only find a link which is not active

I am in the middle of one now and its in the reassembly process.
I Used M5 crank and rods, custom JE pistons +1mm 11-2-1 compression ratio, also you need an adaptor for crank machined. head is 885 +1 valves with tapered shank for better flow, new double valve springs guides and a little work on ports, Hot cam form topend USA with titanium retainers. As yet I am not sure about the rockers new standard or something better. Also have a war chip and MFA to fit when run in. I am told i could have used the M52 2.8 crank rods and pistons with the pistons having 3-4mm off but them but they are flat top and you then need to use the 731 head like the 323i which had a good cam better than 325i sport. I used the 135 mm rods as the throw ratio is better: this helps keep the piston wear down. if you need help message me. If you read up on metric mechanic you can get a lot of info there or email steve at topend performance this guy is so helpful and knows his stuff.

Sebastian

[skipunda]

anybody done one of these and what parts are involved .done a quick search but could only find a link which is not active

I am in the middle of one now and its in the reassembly process.
I Used M5 crank and rods, custom JE pistons +1mm 11-2-1 compression ratio, also you need an adaptor for crank machined. head is 885 +1 valves with tapered shank for better flow, new double valve springs guides and a little work on ports, Hot cam form topend USA with titanium retainers. As yet I am not sure about the rockers new standard or something better. Also have a war chip and MFA to fit when run in. I am told i could have used the M52 2.8 crank rods and pistons with the pistons having 3-4mm off but them but they are flat top and you then need to use the 731 head like the 323i which had a good cam better than 325i sport. I used the 135 mm rods as the throw ratio is better: this helps keep the piston wear down. if you need help message me. If you read up on metric mechanic you can get a lot of info there or email steve at topend performance this guy is so helpful and knows his stuff.

Sebastian

Taking note of the above posts definitely a photo writeup needed here. I can't remember seeing many 3litre m20's on here.

Nice one!

[Andyboy]

How did you fit an M5 crank into an M20 block then?

[StuBeeDoo]

I am told i could have used the M52 2.8 crank rods and pistons with the pistons having 3-4mm off but them but they are flat top and you then need to use the 731 head

I've often wondered about doing this. Going by the piston specs on the Mahle website, I reckon you'd get away with taking 2-3mm off the pistons. It shouldn't be a problem because the valve cut-outs are that, anyway. The only think that might be a problem is if there wouldn't be enough metal left in the M52 piston crowns to allow the cut-outs for the M20 valves.

If I ever have to do another engine rebuild, I'll look into the above very closely.

[reggid]

As far as rod ratio goes there are standard engines from BMW that use 89.6mm crank with 135mm rods namely the S52B32. The S54 isn’t great either IIRC so there is no reason why it can not deviate a bit away from this given this was offered as a factory setup albeit designed that way. It is by no means ideal. An m20 such as this is not going to be a screamer though.

As for the head I use the same stoke crank as M54B30 and make than 25% more midrange and top end on the same dyno though I have a 1.5mm bigger bore. The M54 makes more low end due to vanos and other gizmos, plus the inherent low end compromises on a cammed SOHC engine so you usually can’t have everything. Obviously the M54 built to the same level would be better but a 24V non M is not mind blowingly better than a M20 IMO. I certainly wouldn’t be putting anything short of an S50B32 in mine.

I am at point with mine with around 260bhp give or take that I can’t make any real serious topend improvements without drastic changes to intake manifold (runner size and length, plenum etc) and/or exhaust tuning. This costs lots of $$$ because it involves R&D testing to see what works and what doesn’t.

Having been through the process for cost of a 3L stroker the following parts are what performance ones I believe is necessary to make good power and torque. Then there are the normal rebuilt items that are needed e.g. seals, gaskets, oil pump etc.

89.6mm crank
Custom pistons 10.5+ CR
Schrick / Catcams camshaft
Ported 885 cylinder head (similar to Alpina)
Full exhaust + 6branch e.g. BTB2, RD etc
Miller MAF/WAR

Cost of which depends where you are located and specifics of what you get.

[huliits]

Could anyone tell me, will this type of engine work. I'm planning to put m52b28 crank with m54b30 rods and pistons in m20b25 block, what problems i'm facing? is it actually possible? any suggestions?
thank you.

[phelix]

the "standard" build is an M52B28 crank with 130mm rods - using the 135mm rods you would need to machine the tops of the pistons considerably (5mm?) to keep them from contacting the head. I don't think it can be done - there wouldn't be enough material left to hold the first compression ring.

[reggid]

the "standard" build is an M52B28 crank with 130mm rods - using the 135mm rods you would need to machine the tops of the pistons considerably (5mm?) to keep them from contacting the head. I don't think it can be done - there wouldn't be enough material left to hold the first compression ring.

what is the stack height of the 84mm crank, M54 rods and pistons?

[phelix]

Dunno but here's a picture of an M54B30 piston - if you take 5 mm off the top I would have thought you'll take the ring with it before too long.

[reggid]

Dunno but here's a picture of an M54B30 piston - if you take 5 mm off the top I would have thought you'll take the ring with it before too long.

then how did you come to the conclusion that 5mm needs to be taken off?

[DanThe]

M54B30 is one of the shortest pistons out there, you dont want to be machining them at all realistically

[phelix]

From eyeballing the wrist pin centre to crown height between an M25B25 piston and an M54B30 piston...