535 shortblock tech.

[Brian D]

Using this:
http://www.mye28.com/viewtopic.php?t=55 ... c&&start=0
As a jumping off point, let's discuss some of the more esoteric options regarding BMW 6 cylinder shortblocks. Shortblock meaning the block itself, crank, connecting rods, and pistons. Seems simple, doesn't it? But as alluded to in the above thread, there's power to be made or lost with decisions made during shortblock assembly.

One of the questions that popped up in my head is "what about longer than stock rods"? This is a rather hotly debated subject among racers and engine builders, and there is quite a bit of info to be googled. Specific to this forum, has anybody experimented using longer rods in a BMW 6? When I was thinking about building an M20, I found a few good options for longer rods than the stock 135mm. Now that I've switched allegiances to the M30, I've found far fewer choices.

I just spent some time going over the "Lucifer's Hammer" build threads, and saw that Ken went with Carrillo rods, but I couldn't find any discussion of rod length. To my mind, there are a few advantages to a longer rod, not the least of which is reduced piston compression height, resulting in reduced piston weight.

One other question, I believe I read somewhere that all M30 cranks are forged. Is that accurate?

[paul burke]

Brian, the reason there is no info on Ken Hs shortblock "tune up" is because rod length (rod ratio) is one of the most important tuning factors there is. This info was held back and is one of the many "keys" to making the kind of power it does at such low boost/rpm. Rod length is directly related to intake/exhaust tract volume and flow capability as it relates to piston speed/rpm and dwell time at TDC. All of the M30 cranks are forged , the later 86mm stuff weights less due to reduced counterweight mass. As far as rod choices I have many , 5.580,5.640, and 5.670 are a few of the more popular choices as well as custom (I hate using that word) lengths. Stock rod is 5.315.

Paul

[paul burke]

Brian, you may find that some of the more technical stuff stalls on this forum. In the future that may change (I hope). Keep posting though, I know that I enjoy It.

Paul

[Ken H.]

Brian, I did cover some of the more fundamental short block considerations in the LH series; Part 7 IIRC has some of the formulas.

For openers, let's look at redline. The standard formula is Redline = 600,000/ stroke length in millimeters. Example. Your M30 build is going to use a 90 mm E34 M5 crank..
So 600,000/90 = 6666.66. Having a modicum of common sense , you round down to at least the nearest 100, so 6600 rpm. Since I tend to think in terms of endurance motors and am more than willing to give up the last 5-10% in output in exchange for 6-figure lifespans, I'd cut it back to 6500 rpm.

The rpm limit is also influenced by piston speed. Canuck's inclusion of some of the formulas on intake charge velocities point towards fairly high feet-per-minute levels for piston travel. My thinking has tended to keeping piston speeds limited to 4000 ft/min. This is from the perspective of engine longevity and reduced wear on the rings. Perforce, this means somewhat reduced upper limits on RPM. If engine longevity isn't a major concern, then proceed as you wish.

Formula: rpm = piston speed in fpm x (6/stroke length in inches).
So using our 90 mm stroke example from above, we get 6500 = allowable piston speed x (6/ 3.543 inches) [90 mm/25.4 = 3.543 inches]
6500 = speed x 1.693.
6500/ 1.693 = 3839 fpm. So we are comfortably under the 4000 fpm "ceiling."

Con rod length is a topic subject to a great deal of discussion. Paul has accurately pointed out that determining the rod length is interactive with intake and exhaust tract volumes, piston speeds and acceleration, anticipated Brake Mean Effective Pressures to name just a few.

Another consideration is piston design, as short rod lengths will tend to place a lot of lateral loadings on the piston skirt and the cylinder walls at the lower end where they join the crankcase. This area tends to be relatively thin, and with increased dynamic cylinder pressures, is subject to a fair amount of distortion. The result is piston slap, scuffing on the skirt, less-than-ideal loadings on the rings, and the cylinder walls themselves going out-of-round.

Generally speaking, one will see rod-to-stroke ratios in the 1.75 to 1.50 range. Shorter rods tend to accelerate the piston more quickly away from TDC and BDC which aids in cylinder fill with the fresh charge and starting the spent charge out the exhaust. How much of this is a Good Thing relates closely to the volume of the incoming charge. If you're running intake pressures above atmospheric, i.e., forced induction, the incoming charge will more readily fill the rapidly increasing cylinder volume.

Higher RPM tend to make this balance more critical.
The general rule I've seen is to keep the rod ratio around 1.67, so on a 90 mm stroke, we are looking towards a center-to-center rod length of about 150.3 mm, but the number is not hard and fast. Piston design will affect this, with pin height, placement of the rings, ring land widths and crown thicknesses all playing significant roles.

Along with the the c-to-c length of the rod, one needs to have the deck height numbers as the compression height of the piston enters into the design parameters as well.

I hate to keep referencing back to Bell's Forced Induction Performance Tuning, but he does a really good job of discussing the stroke length subject matter.

This ought to do as a start. I hope someone else can pick up on the considerations in crank design, including the whys, wherefores and benefits of knife-edging and some of the preparation work done on the parts themselves.

[Brian D]

Paul, that's very interesting. Most discussions and tech I've seen tends to (unfairly, I think) minimize the importance of rod length, although it's been a few years since I've done much research. One thing that has frustrated me is the lack of BMW info on the 'net to "brainstorm" different combo's, compared to sbc's. I will definitely be leaning on your knowledge when it comes time.

Ken, your thinking parallels mine as regards to engine longevity-I drive 200 miles a day, so obviously I want the engine to last a while! I've never been comfortable really winding an engine into banshee territory, which is one of the factors that led me to plan for an M30 to begin with. 6500 is a limit I feel very comfortable with.

I know you're just spitballing, but based on what I found with the M20, a 150mm rod is going to be very hard to actually fit in the engine. I haven't found the deck height on M30's (another piece of info that is almost laughably easy to find for a Chevy), to compare it to the M20. I had a set of 145 rods for the M20, but I couldn't use a crank larger than 75mm without running out of room. I need to do more homework on the M30.

My amateur thoughts on crankshafts? I want the lightest, strongest one available! For cost effectiveness, I'll be sticking with the stock B35 crank (86mm stroke, right?). I haven't looked much at knife-edging, one more thing to research.

Brian D

[paul burke]

Brian, deck height on an M30 is 8.560. I have see some a little taller, but this is where most of them land. Don't get to carried away with rod length, affects rival camshaft duration. I would stick with the 86mm stroke BMW crankshaft for your project. Knife edging can be done on a lathe.The larger effect It has, Is the reduction of counterweight mass. Inline 6cylinder engines do not require opposing counterweight mass to offset piston/rod weights.


Paul

[Rich Euro M5]

Paul,

What about destroking to 84mm (S38b35) and running way oversquare, like the old M90 3.5 Liter engine in the E12 M535i. Except bump the compression ratio up to about 10.3:1 or whatever modern pump gas will support. Then utilize a tricked out B35 head with a modern camshaft. What do you think we would get from something like this ?

Rich

[Ken H.]

Rich, I doodled your idea into the Not2Fast model and came up with 275 fwhp @6800 rpm. As per usual, I'm being conservative with redlines here, tho' I think this layout could probably stand 7K.
N2F will permit inputting values with zero boost, so it can model an NA motor.
My inputs assumed 84 mm S, 94 mm B, .5 psi pressure drop to the intake (air filter restiction), 95% Ve @ 6800, A/F ratio @ 13.3 and .49 BSFC.

I think the 275 fwhp number can probably be beaten with sufficient air flow. Your comments about a using a well-goosed head plus one of Paul's uber-B35 grinds is definitely going in the right direction.

Anyone want to try a roller rocker arm layout ?

I tend to think getting into the 300 fwhp realm is very doable, but I suspect it will need airflows on the order of 220 cfm or better to get there. That and 47 mm intake valves along with exhausts around 38-39 mm.

Set up like you describe, the thing ought to flat-ass scream

[Rich Euro M5]

Ken,

Thanks for running the numbers, I thought the short stroke might be an interesting design.

As we've been discussing,getting the head to flow is the primary challenge. I'm going to throw another idea out that most here haven't heard of. I think we can use the exhaust to increase cylinder filling. instead of a conventional header design, build up a set of snakes that takes advantage of exhaust scavanging occurring in one cylinder to pull the intake charge into another cylinder. I would also advocate using a antireversion cone built into the primary pipe to minimize charge dilution. If the exhaust is properly designed I think one could achieve nearly 100% VE.The obvious challenge is sizing the primaries to encourage high velocity exhaust flow through a fairly wide RPM range and at the same time we need to minimize reversion. I don't know if it's possible to do this with the BMW firing pattern. But I know it was done with certain Chevy straight 6 engines back in the early 60's.

Rich

[paul burke]

Rich, your 84mm stroke idea has its place. You could say It would make a good "Clutch Car " motor as it would require a little deeper gear, offer a little narrower power band (3800 to 7200) to acheive 275 hp. It would be a fun little track motor at 10.5 compression, B35 head with a nice little camshaft.

Paul

[paul burke]

The scavenging affect that a properly tuned header can give a motor can be a huge. This type of tuning cannot be found in an iron exhaust manifold and is not even comparable. One can however over scavenge to the point of drawing raw fuel out of the exhaust on overlap and killing power. There have been many improvements to header design in recent years including merge collectors, stepped primarys and back to a tri-y type header. All in an effort to try to balance exhaust/intake tracts throughout a broader rpm range.


Paul

[Ken H.]

There have been many improvements to header design in recent years including merge collectors, stepped primarys and back to a tri-y type header. All in an effort to try to balance exhaust/intake tracts throughout a broader rpm range.

MmmmmwhuuAhahahahahaha

[Canuck YYC]

I suspect that a 300hp NA M30 street motor is going to be very peaky and not a lot of fun to live with as a daily driver. If you're keeping the piston speed under 4000 ft/min (M3 territory), and maxing out at 94mm bore, you need to generate BMEP in the 170psi range and as Kenh noted, a fair bit of air. There's no way to keep the port flow around .6 mach at 6500rpm with only two valves (unless you can stuff a 50mm valve in there). If you can't maintain effective port velocity, your VE is going to go out the window taking the BMEP with it.

I don't think it's impossible to build it, just that it would tend to be a high-strung thing that becomes arduous to drive anywhere but the track (or the back roads if that's your thing). However...I love being proved wrong - means I learned something new. I'd like to see this done to see what the results are.

[paul burke]

I've been pondering a little about Rich's 84 stroke deal. You might be able to get it to come in at around 6800 rpm with a shortblock adjustment. A rod length of around 5.150/5.200 should get the piston speed up there enough to move the air required. I would build it if I had some of the parts lying around. Ken you want to donate that 84 stroke arm you have holding down you garage floor. Remember compression motors start pulling air earlier and with a stronger signal do to less combustion area. With some camshaft adj. could get away with 11to 1 compression on pump gas.

Paul

[paul burke]

Andrew, a 300hp NA M30 is not a problem with a 90mm arm. If you were making reference to doing it with a 84mm stroke crank that would make it a little tuff, and drivabilty would suffer a little.

[Ken H.]

Canuck, I think you're absolutely correct in your observations about the 84/94 being the automotive equivalent of an ADHD victim. I never said it was going to be the ideal power plant for Buffy to get little Samantha to soccer practice

All this much TIC. But I do think it is a feasible undertaking. Practicality is another topic altogether. This has been another one of my personal hobbyhorses: we spend an awful lot of effort, time and $$ trying to get maximum outputs from our engines. Nothing wrong with that, but I look at just how much time I actually spend between the topof the torque curve and redline, and the answer is damn little.
So over time I've come to look at what kind of responsiveness and power is available from, say. 2500 up to 4500 rpm--where I spend my time in the real world.

That being said, as much as I enjoy building engines, I'd much rather spend my time with my family and my money heliskiing. 8) I use That Thing for long distance autobahn-type runs, along with the shots up the passes. This as opposed to drags, DEs and autocross. These choices bias me to longevity at the expense of seeking maximum output. This isn't dissing the competitive event guys; it's simply a matter of personal choice.

With the goal of improving what we've got, I'm still going to try to provide as much knowledge as I can to the boardlings . . .I just hate to see people reinventing the wheel or pursuing blind alleys which experience has shown don't offer much except a quick way to deplete one's savings account.

[Ken H.]

I would build it if I had some of the parts lying around. Ken you want to donate that 84 stroke arm you have holding down you garage floor.

Absolutely. I won't give the crank away, but I will make a very good deal on it if I know it's going into a solid project.

[Canuck YYC]

Canuck, I think you're absolutely correct in your observations about the 84/94 being the automotive equivalent of an ADHD victim. I never said it was going to be the ideal power plant for Buffy to get little Samantha to soccer practice

Ahhhhh...I misunderstood the comment then. The huge advantage to FI that I see is 300hp is readily do-able and won't make the car behave like a bi-polar supermodel. It's not particularly elegant necessarily and doesn't require the fine touch of experience and knowledge, but it's very cost-effective.

I'd like to build up an M73 one of these days. There's nothing quite like a V12 in full song.

[Canuck YYC]

Andrew, a 300hp NA M30 is not a problem with a 90mm arm. If you were making reference to doing it with a 84mm stroke crank that would make it a little tuff, and drivabilty would suffer a little.

I suppose I did see the post about a 250cfm head. Interesting stuff indeed. I need to finish the project I have before embarking on another just yet.

Out of curiosity, how far can you bore the smaller blocks? Can you drop a 94mm piston into a formerly 86mm 2.8 block?

[Rich Euro M5]

Ken, Paul, and Andrew,

My thoughts about using an 84mm crank vs the more typical 86mm part is get some conversation going about what it takes to make power. As most know, torque is king and you don't improve this parameter by destroking. During this thread we've sort of loosely bound ourselves to an upper RPM limit of 6.8K. This forces the discussion towards how do we make power when we're torque limited. As Paul mentioned, it involves using the right combination of bits and pieces, but it is possible. And for the DE junkies it would be a great engine, especially if it was matched to the correct transmission / diff combination.

To Ken's point for a pure street engine that needs to make torque in a more sane RPM range. Then stroking in combination with the correct rod / piston combination is were we need to concentrate our efforts.

About boring a 2.8 liter block to 94mm. I don't know if there's enough material to do it. I would be looking for a clean M90 block as my starting point. This brings up another thought, can the M90 block be bored enough to gives a little more wiggle room for a larger intake valve? Would a 50mm valve might be possible?

Rich

[Canuck YYC]

Torque and stroke are not related, at least in the fashion people think. IE - bigger stroke does not equal bigger torque. The crank is not a typical lever in this scenario and the longer stroke (at the same RPM) travels farther generating more friction.

Mathematically we know that (torque x rpm)/5252 = hp. Re-arranged to find torque we have (hp*5252)/rpm. Another hp equation, one that I use most of the time is (Piston Area x BMEP x Piston speed)/132,000 (talking 4-stroke engines using imperial measurements). You'll notice there's nothing about stroke or displacement there.

Years ago I simulated two same-displacement engines with different bore/stroke combinations and the results surprised me. I did this after being told that a shorter-stroke engine could produce the same or more torque at the same engine speeds as a longer stroke version.

Short vs. Long

[Rich Euro M5]

Torque and stroke are not related, at least in the fashion people think. IE - bigger stroke does not equal bigger torque. The crank is not a typical lever in this scenario and the longer stroke (at the same RPM) travels farther generating more friction.

Mathematically we know that (torque x rpm)/5252 = hp. Re-arranged to find torque we have (hp*5252)/rpm. Another hp equation, one that I use most of the time is (Piston Area x BMEP x Piston speed)/132,000 (talking 4-stroke engines using imperial measurements). You'll notice there's nothing about stroke or displacement there.

Years ago I simulated two same-displacement engines with different bore/stroke combinations and the results surprised me. I did this after being told that a shorter-stroke engine could produce the same or more torque at the same engine speeds as a longer stroke version.

Short vs. Long

I agree this can be the case if the parts aren't carefully chosen. An example would be to try to utilize the same piston and then pushing the crank stroke too far. This results in an excessively short rod, high piston side loads, parasitic losses, etc. But if the crank is correctly matched to the correct rod/piston combination, then torque output should go up.

Rich

[Canuck YYC]

I'm not sold on that however a longer stroke will increase the piston speed speed for given RPM, which would increase output. But then valve size is limited by the bore...

What are the parameters that we're working with. I've been going around thinking same displacement, different b/s ratios. If however we stick with a given bore, say 94mm, we can say our ROT max valve size will be 50mm with careful prep. I have no idea if you can physically get that into a B34 or B35 head without significant re-work . I'm still getting ahead of myself. The concept is intriguing to me (300 useable na hp M30), so let's lay out the targets and constraints.

In no particular order, 100% open to input:
Peak torque 4000-4500rpm with a fairly flat curve (ha) from 3500-5500.
Pump gas
94mm bore
Stock castings (as opposed to a scratch-built head, not an untouched casting)
Attainable - If it's going to be a $15,000 OTC build it's going to be a pretty tough sell - an economic dead end akin to Ken's technological version.
Others?

[Ken H.]

My thoughts about using an 84mm crank vs the more typical 86mm part is get some conversation going about what it takes to make power.

Excellent! This is exactly what Paul and I have talked about off-line, namely getting some intelligent discussion going which helps us all move ahead in figuring out how we can get more out of these lumps.

Item for Mike W: Looking around somewhere I found a mention of some of power delivery from the M30 that ///Motorsport was getting for engines used in some of the sedan series in Europe in the early 80s . . .in the region of 400+, using mechanical Kugelfischer injection, ITBs and so on.
So will you please bring me another plate of crow to go along with my comment about M30s not being a hi-po engine. More

As most know, torque is king and you don't improve this parameter by destroking. During this thread we've sort of loosely bound ourselves to an upper RPM limit of 6.8K. This forces the discussion towards how do we make power when we're torque limited. As Paul mentioned, it involves using the right combination of bits and pieces, but it is possible. And for the DE junkies it would be a great engine, especially if it was matched to the correct transmission / diff combination.

Rich, the 6.8-7K limit tends to be reinforced to some degree by the block's design as well as what the reciprocating parts can tolerate on a protracted basis. If you're looking at replacing or rebuilding the motor after say, 20-30 hours of use (half a season), then by all means, wind 'er up. Keep in mind the M30-based blocks have a c-to-c bore spacing of 100 mm and a deck height of around 217 mm. These dimensions frame what we can do with the moving bits. Additionally, the blocks don't have a vast amount of cylinder wall thickness, especially at the bottom of the bores, where lateral loadings can permit wall distortion and resultant loss of dynamic pressure.
When BMW was doing the major FI program in Formula 1 in the early 80s, they used a basic M10 block, but deleted a lot of the water passages to the head and sleeved the bores down significantly. The sleeving, besides bringing displacement down to the 1.5 liter maximum, resulted in much thicker cylinder walls where the bores siamesed--1-2, 2-3, 3-4.
These walls were as much as 20+ mm thick which made the block substantially stiffer. Which made sense, seeing that the motors were running 60 + psi boost and were seeing better than 10K revs. Running some kind of carcinogen, these 1.5 liter churns yielded over 1350 fwhp. 1350 is used as the power output, as that was the design limit on BMW's engine dyno and these motors regularly exceeded that. These numbers are from the qualifying motors, which had a life expectancy of about 20 minutes.
Anyway, where all this goes is I don't have any kind of problem with big rev numbers; I'm just hard-pressed to see where they are usable other than in a highly-specialized application.

To Ken's point for a pure street engine that needs to make torque in a more sane RPM range. Then stroking in combination with the correct rod / piston combination is were we need to concentrate our efforts.

About boring a 2.8 liter block to 94mm. I don't know if there's enough material to do it.

If we're looking at an M30 casdting, it's possible, but I'd like to go back to the point about the c-to-c spacing on the bores. The distance is 100 mm. A 94 mm bore give you 6 mm of material at the siameseing. This isn't much, and leaves the cylinder skirts pretty thin. Begin to increase cylinder dynamic pressures--high compression NA or FI and distortion becomes a real issue. Some people have suggested pushing things out to the 95 mm level--which leaves 4 mm (!) of interbore thickness. BMW did this on the S38B38s and experienced less-than-happy results with block failures.

Unless one wants to step up for opening up the blocks big-time and using Darton cylinder sleeves, http://www.dartonsleeves.com my tendency is to think staying very close to stock bores. No one is going to give a flying rip if your engine is 3542 cc or 3668 cc. OK, yah, bench racing and bragging rights, but I come back to my basics of simple is better, meticulous workmanship, and build for longevity, and the power will take care of itself.
Incidentally, Paul and I talked about sleeving LH in the interests of the ultimate build, but this got into the realm of being stupid expensive.

I would be looking for a clean M90 block as my starting point. This brings up another thought, can the M90 block be bored enough to gives a little more wiggle room for a larger intake valve? Would a 50mm valve might be possible?

Someone else is going to have to step in here, as I simply am not familiar enough with the M90 to comment.

[Brian D]

Is the idea behind using the smaller stroke solely to increase the RPM limitation, or is it also to improve engine responsiveness? Using lighter components should improve responsiveness, so will a 86 (or even 90)mm stroke engine have the same "revability" as an 84mm engine if it is built using lighter components? I know the longer stroke engine will still be comparitively rev limited. I hate to give up any displacement, even in my lighter car.

[turbodan]

All this science makes a bolt in turbo kit that'll do 400rwhp look pretty sweet.

Just sayin.

[Brian D]

Just sayin.

Over, and over, and over....

Honestly Dan, every time someone suggests doing something other than a turbo, you make some comment. Let it go already.

[turbodan]

Dont get upset. Like I said, I'm just saying. This thread lays out some realistic options for potential NA power production and what it could cost to pursue them. There seems to be a generally held misconception that NA tuning our engines is cheaper, simpler and as effective as turbocharging. Its good to clear that up.

[Jeremy]

It's also good to note that everything discussed thus far is directly applicable to a forced induction build as well. This isn't just for the air suckers out there.

Jeremy

[Rich Euro M5]

Someone else is going to have to step in here, as I simply am not familiar enough with the M90 to comment.

Ken,

I'm not intimately familiar with the M90, but it appears the M90 block established the basis for the later M88/1, M88/3, and S38b35 engines.

The M90 in stock form has a 93.4mm bore x 84mm = 3453cc, it's the same as a S38b35 in this regard. Since there's a little more bore from the factory, there might be a slight difference in how the water passages were cast into the block. Zain, the guy that owns the badass'd E12 Turbo could probably fill us in with all the details about the differences between the M90 block and the standard M30b34 block.

Rich