535 shortblock tech.

[paul burke]

Rich, the M90 has a cast in water manifold to get cooler water to the middle of the block. As we have seen with the STD B34 and B35 block, the area around the #4 and #5 cylinder is the common point of head gasket failure. Many base head gasket failure in this area on tuning issues, when in fact it is the result of not being able to control heat generated in this part of the block. There are cures for this and I incorporate them into my turbo builds.The M90 style block also enjoys siamese cylinder walls and use steam holes like a 400cuin Chevrolet.

Paul

[Rich Euro M5]

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.

The 84mm stroke 3.5 liter is the original 3.5 liter BMW designed for use in the E12 / E24 cars in the late '70's. This engine has a unique personality and was also the basis for the M88 series and S38b35 engines. It basically had the same output as the later M30b34 dirty engine.

Rich

[paul burke]

As far as valve sizes and bore clearance, this really isn't an issue. The M30 has a 26+ degree valve angle and clears the block with ample room. The problem lies with valve to valve clearance on overlap.

Paul

[turbodan]

The M90 style block also enjoys siamese cylinder walls and use steam holes like a 400cuin Chevrolet.
Paul

And the 2.5/2.7 M20 blocks.

[paul burke]

Hey, Dan did you get familiar with the E46 M3 valve train?

Paul

[Ken H.]

Dan, in rereading the thread, I can't find anything that quantifies what the specifics are going to cost in terms of dollars spent. If someone wants to compile a list of the individual tasks to be done in prepping a short block, then price that work out, fine. But those labor rates are going to differ, depending on your location, if you have some kind of an arrangement with a shop, or if you have access to the tooling gratis.

Jeremy's right that everything that has been laid out is the engineering footing for making a build work better. This stuff applies, regardless of whether it's an atmo engine or not. Given the added stresses from Forced Induction, it just adds that much more reliability. Hopefully if this thread and the "535 cams" companion continue, we'll be able to get into more of the specific details which, cumulatively, make for a stronger build.

This kind of attention to detail tends to be time consuming, but in the process, it reduces or eliminates areas which are subject to failure, accelerate wear, or augment parasitic power losses. If they didn't make sense and show tangible results, they wouldn't be used by builders whose nuts are on the anvil every time one of their products goes out on the track.
Do you really think Hendrick Racing or Turner Motorsports just throws their engines together without doing literally man-years of work to get everything they can out of them?

These discussions have deliberately stayed away from the FI vs. NA debate. Rich raised the question of what kind of power might be attainable NA out of the M30. To that end, a number of technical areas need to be investigated. Further, if someone is interested in pursuing the question, block tuning holds a lot of promise in reaching that goal. I don't think anyone is saying that for a given power delivery NA is quicker/better/cheaper/more good-looking/politically correct than FI. It's a different avenue and may or may not point to the exact same end result as an FI approach.

From a pure cost-per-hp, bolt-on FI such as you have done may be less expensive. It will be so if you can source P-n-P blocks. If disposable engines are what blows up your dress, fine. But a lot of people want a motor that is assembled with "best practices" techniques and is representative of the best that BMW can offer. This can be either NA or FI. Obviously, I chose to go FI. From time to time I wonder if I did the right thing.

Dan, this isn't dissing your approach to finding power; I'm just skeptical about how long that power delivery will last. I'm also certain that if proper block tuning techniques are used, you'd find that there's more in there than you might otherwise find.
Which is the fundamental point of putting the subject out there for discussion.

[Rich Euro M5]

Rich, the M90 has a cast in water manifold to get cooler water to the middle of the block. As we have seen with the STD B34 and B35 block, the area around the #4 and #5 cylinder is the common point of head gasket failure. Many base head gasket failure in this area on tuning issues, when in fact it is the result of not being able to control heat generated in this part of the block. There are cures for this and I incorporate them into my turbo builds.The M90 style block also enjoys siamese cylinder walls and use steam holes like a 400cuin Chevrolet.

Paul

Paul,

Thanks for the info about the M90. About that HG failure point on the b34 engine, both of my HG failures occured here with my present engine. The first failure was between the #5 cyl and the water jacket. Lots of steam and some very cool boil over action going on in the coolant reservoir. The HG failure I just experienced was in the same general area, but between a coolant passage and the oil return galley. No drama at all kind of boring actually.

Back to M90 discussion. Would this be a better block for our usage? As we know, BMW doesn't change a design without reason. Do you believe the changes are the result of a inherent weakness with the M90 design, or a cost savings measure?

Rich

[paul burke]

Brian, It is true that a shorter stroke motor can pickup the throttle quicker, but it requires a shorter gear to take advantage of it. Stick with your plan, it will suit your current needs.

Paul

[paul burke]

Rich, siamese blocks have a problem cracking across steam holes. I suspect this is one reason among others BMW dropped some of the big bore stuff. The water can be redirected in the B34/B35 blocks with a little work. Or one could add external plumbing as I do on my race stuff.

Paul

[Canuck YYC]

Dan, this isn't dissing your approach to finding power; I'm just skeptical about how long that power delivery will last.

I have a problem with this statement. In fact, I think when talking equal horsepower from equal displacement, I believe that is backwards. This 300hp NA build is a perfect example, and your reliability statement was what started me doing homework on the concept the other evening.

In order for a NA engine to develop that power, it's necessary to take great care building the bottom end to withstand the piston speeds, a great deal of work in preparing the cylinder heads to flow and in the end you find yourself with a high-strung tempermental engine. On the other hand, generating 300 hp with a turbo will only increase charge heat and cylinder pressures. Reasonable prep with the M30 seems to be a new headgasket and cylinder studs (assuming an otherwise healthy engine). With efficient charge cooling, you could probably get away with just 8 to 10 lbs of boost. It would be a very tame build, easy to drive and probably work fine with a plain old auto box and torque converter.

I'm not taking sides in the NA/FI debate (if there is one), but the allusion (or my interpretation of your post) is that FI = hard on parts and lb for lb, hp for hp I think that's a misnomer, especially with the contraints of 50%+hp M30 based engines. Indeed, if you applied that level of prep to an FI engine, I think you're correct - you'd find more than you would otherwise.

Even in the confines of this thread the topic seems to be all over the place. Is this about a 300hp NA M30 bottom end, and 300hp M30 bottom end, bottom end prep in general, high-output NA engines...? I'm sorry OP, what are we supposed to be talking about here anyway?

If you want high specific outputs, look no further than your nearest "bic bike" at the local import store. The 08 GSXR 600 sits at a gob-smacking 205hp/litre with the 750 version at 200 and the 1000 at 185. That crushes the M3's 104 and utterly demolishes the Z06's paltry 72hp/l. Incidentely, the 'Vette, for all the praise it's received just has a ton of piston area and some big piston speeds. It's BMEP is only 148psi compared to the M5's 168.

[paul burke]

Hey Andrew, things do tend to stray a bit. I am way more into NA stuff than FI stuff, keeps the brain a little more active. Just stay on topic and it will come back around. I would still like to have a separate area to post on, maybe things wouldn't get so convoluted.


Paul

[Canuck YYC]

I was only half kidding when I asked what the topic was supposed to be focused on It's nice to find a collection of interested parties and knowledgeable gearheads - I expect I'll learn more than just a thing or two and with any luck I can help someone else out in the process.

[Ken H.]

Canuck, I think we started with Brian asking about what goes into a short block tune. I pointed out some of the more common rules of thumb about establishing functional limits.
Rich wanted to know what would happen with a short stroke layout. I ran some numbers througfh an engine model I use. The answer was around 275. The next question to come up was, "is 300 fwhp possible NA?" The answer was/is "yes, but . . ."
Dan rightly suggests that FI would put you there with probably a lot less hassle and expense, all things considered.

I think where some confusion arises is my biases towards preparation is at some variance with Dan. If I have read Dan's posts correctly, he seems to go in the direction of "use whatcha got, and if it breaks, get another one." Using components with questionable or unknown histories makes me uncomfortable, and my ingrained paranoia tells me the thing will give up at the most inopportune time, hence my remark.

Back to the 300 hp topic. I agree with you about needing to do the bottom end work, but hp is hp, and not doing this on a FI motor doesn't make the need go away. The need to maintain signal strength on the NA motor is going to raise difficulties and perforce we have a more touchy motor. But I'm not suggesting such a build is your everyday grocery-getter, simply that it is doable.

As far as direction on these tech threads. I really don't want to see these to degenerate into pissing contests. Unless and until we have a substantial library of posts on a range of subjects, I think hard and fast, narrowly focused threads are going to be hard to find or maintain. In a very real sense, we are finding our way here. There are a lot of players with a great deal to contribute, and even more who may start to look at these threads as source material for proceeding with their own projects.
That being said, I'm not so unsufferably arrogant that I can't stand a difference of opinion or an opposing viewpoint supported by facts. It's only where differing viewpoints intersect that true knowledge begins.

[Canuck YYC]

The last thing I want to do is come off like I'm picking a fight - not my intention. I'm a great big "red" and tend to trample around pissing people off without intending to. I'm here to learn and share and make some new aquaintances while I'm at it.

Hp is hp - agreed, but how it's produced matters as it relates to engine stresses. I don't know how much cylinder pressure you could reasonably build before the crown or a rod gives in but, and I'm basing this on the little that I've read on this forum, the M30 will take a fair whack of boost on the stock internals. Mahle is, in my opinion, the only piston manufacturer. I've used spec. ordered stuff from J&E, Wiseco and OEM aftermarket parts of unknown manufacturer and hands-down the Mahle parts were always the nicest. Anyway, with a forged crank and Mahle pistons stock, it's going to take a good deal of boost to start bending rods. The increase in power with the FI comes primarily from the massive, easy increase in BMEP.

On the other hand, to get the power out of the NA engine, you're going to have to find as much BMEP as you can, and then increase the piston speed by some 30%. That's where your stress is. We still have a solid crank and pistons (theoretically) but now we have a significant increase in friction (rings), much higher piston forces (4100lbs vs. 2940 at TDC, based on 700 gram of recip. weight).

My instinct is that the higher piston speeds and forces are more detrimental than the higher cylinder pressures, but I can't back that up with anything significant

We're both on the same page personally however - prep good, run-whatcha-brung...okay, but... In all fairness though, not everyone has access to the tools and knowledge, or the economic resources to "do it right", and I believe you can learn all kinds of things just doing it. I'm not a fan of half-assed anything, but I'd rather see someone doing it than not because they can't afford to build an engine that would make a BMW engineer smile.

If it were up to me (and not my client's finances), all of my builds would have gone out with AlBeMet rockers and rods, MMC liners, Ti valves, single valve springs, coatings out the wazoo and an unbelievable stunner of an engine. As it was, people always left happy and came back happier, even without the really exotic stuff. Some folks were willing to pay for whatever I recommended and others had a more modest budget for us to work with. It's all good.

And yes - anything that fails will do it immediately after closing time on a long weekend, 4 hours from the nearest tow truck, in the blazing heat or the freezing cold.

[paul burke]

Andrew, Instead of using simulation software to predict peak and dynamic running pressures in an engine. I setup shaft encoders and strain gauges on my dyno (driver vs driven) to measure crankshaft pulsing ( turned it into phase shift) In doing so I was able to see pressure (running) Imbalances between cylinders. Not only did the pressures change between rotations, the numbers were all over the place. I can tell you this, the perfect multiple cylinder, common plenum engine is not a mathmaticly perfect one.

Paul

[derrith]

I wanted to pop in and say that this is an immensely intriguing topic that I'm enjoying reading. Thanks guys!

[Canuck YYC]

You were measuring main to rod or down the length of the crank on the centre axis? I'm not surprised that the complex effects of harmonics, 1st/2nd/3rd order vibrations, accoustic ineraction in the plenum etc etc aren't modelled particularly well. Were you tracking in-cylinder pressures against the strain gauges or calculating variances based off the strain data? I've not had the opportunity to insturment an engine like that. before.

[turbodan]

Hey, Dan did you get familiar with the E46 M3 valve train?

Paul

I did. I also checked out the S65 in the new M3, and oddly enough BMW got rid of the rockers again.

[Rich Euro M5]

Something that often times is overlooked by the FI proponents is not everyone can run FI, even if they want it. I know there are a number of guys out in Cali that would be running FI if it wasn't a problem getting past the Enviro - whacko regulations they have.

Here in Houston I could run FI, but with the tough emissions testing every year I would have to spend too much time on tuning and testing. Cars have to be 25 YO for exemption from testing, this means owners of the '83 - '84 models years can go wild.

My next build engine build will be a replacement lump for my DD, the '86 535i/A. The info that's been presented in this thread makes me believe the M30b35 is what I should start with. Don't know about what stroke I'll use, might stick with 86mm, might search around for something a bit more interesting. Whatever I decide, it will have a base compression ratio of about 9:1 for a future upgrade. The upgrade is FI, but maybe we'll try to do something different. 8)

Rich

[Ken H.]

Good catch on the environazi/emissions issue, Rich.
This snag alone will tend to push lotsa people into the NA column, regardless of the cost to performance benefit of FI.

Political aside:
Don't be falling offn yer chair, youse guyz if BO's EPA doesn't impose the Cali CARB regulations nationwide. BO has stated he wants action by the US on Global Warming and arbitrarily tightening emissions on older vehicles puts the $$ load on individuals and not the government. The EPA has been pushing for years to tighten standards, with specific interest in banning anything not running OBD 2 or newer. Remember the gearhead community represents very few votes in the scheme of things, so if a couple hundred thousand classic or restored cars are forced to the crusher, so what?

Point here is very soon we all may very well have to submit to a visual as part of the emissions check, and if it ain't stock . . . .

End of rant. Now back to our regularly scheduled programming.

[turbodan]

I'm glad so many chose to vote for change.

[Ken H.]

Andrew, I looked at the "275 hp" M30 case in the N2F model, and came up with a BMEP of 149.5. Using the same assumptions (94 mm bore, 47 mm intake) in your 2 vs. 4 formulas, I'm getting a mean piston speed of 3875 fpm with Z = .6.

You've pointed out the immediate benefit of higher BMEP from intake pressures above atmo (FI), and the need to increase piston speeds. I think that one item which needs consideration is the fact that in a FI motor, not only is BMEP higher, but it is maintained through significantly more of the piston's stroke. Thus the power impulse is maintained over more degrees of crank rotation and imparts that much more acceleration to the rotating mass.

Since nothing's free, the higher BMEP translates into higher dynamic running pressures throughout the piston's cycle. This means the bottom end of the cylinder walls see significantly greater radial loading and heat transfer. This is an area in most blocks where the wall thickness is quite thin . . .possibly as little as .080" So distortion and the potential for blowby (insufficient ring sealing) definitely come to the party.
If we add to this a short skirt configuration on the piston and a short rod ratio, increasing the liklihood of the piston cocking in the bore, things can get frustrating.
In the scheme of things, I tend to agree that higher piston speeds probably create more problems than the higher running pressures, but Darton Sleeves has a pretty good business giving the 1/4 mile guys a solution to the pressures situation.

This situation isn't helped when piston speeds start going up towards 5000-5200 fpm and we begin to crowd things on the ring's ability to provide a consistent seal, both immediately and over the engine's life expectancy between teardowns. Use of a Dykes configuration on the #1 oil scraper ring can help a lot to maintain seal and reduce ring flutter, but once again, lower fpm on piston speed is probably better from a wear standpoint if we can achieve it.

Playing with the 2 vs.4 formula, it also points to getting as much intake valve area (and by inference, valve curtain area) as possible, this reducing Z, all other things being equal.

[Rich Euro M5]

Good catch on the environazi/emissions issue, Rich.
This snag alone will tend to push lotsa people into the NA column, regardless of the cost to performance benefit of FI.

Political aside:
Don't be falling offn yer chair, youse guyz if BO's EPA doesn't impose the Cali CARB regulations nationwide. BO has stated he wants action by the US on Global Warming and arbitrarily tightening emissions on older vehicles puts the $$ load on individuals and not the government. The EPA has been pushing for years to tighten standards, with specific interest in banning anything not running OBD 2 or newer. Remember the gearhead community represents very few votes in the scheme of things, so if a couple hundred thousand classic or restored cars are forced to the crusher, so what?

Point here is very soon we all may very well have to submit to a visual as part of the emissions check, and if it ain't stock . . . .

End of rant. Now back to our regularly scheduled programming.

Ken,

I don't like to hear about the Cali CARB stuff being implimented nation wide. I'm hoping his advisors tell him we can't go down the CO2 cap/tax use road until the economy is headed back to good health. I've been reading some of the green countries in the EU are trying to squirm out of their committments they so willingly allowed themselves to be duped into. They are now realizing the CO2 caps they've sign on to are going to tank their economies.

Now back to our regular programming .... how to squeeze the most HP from an ancient BMW lump.

Rich

[Ken H.]

Sorry Rich, but that's the way it looks from downstairs in the 2-story privy. It's a "Change You Can Believe In." Stay Tuned . . .

[paul burke]

Ken, the M30 block can handle a considerable amount of load. Cylinders are quite thick on the mid '80s blocks. I have not had (or heard of) a wall failure in a long time.Sonic testing can reveal wall thickness. BMWs casting methods are pretty consistent regarding core shift and such.


Paul

[Ken H.]

Ken, the M30 block can handle a considerable amount of load. Cylinders are quite thick on the mid '90s blocks. I have not had (or heard of) a wall failure in a long time.Sonic testing can reveal wall thickness. BMWs casting methods are pretty consistent regarding core shift and such.

Good to know, boss. I wasn't trying to infer that the M30 (or S38) blocks had problems in this area. But this area is known to be weak in other makes. IIRC, the GM Performance blocks--SBC to name one--make specific reference to the fact that they used undersized cores the better to add material in stressed areas.

The only drawback I'm aware of with sonic testing is it doesn't always differentiate between solid cast metal and scale buildup. In newly cast blocks, it's not unheard of for chunks of casting sand to remain in the galleries; poor mold prep can result in a thin area which will show up as OK when sonic tested. Not often, but it does happen.

[Canuck YYC]

Andrew, I looked at the "275 hp" M30 case in the N2F model, and came up with a BMEP of 149.5. Using the same assumptions (94 mm bore, 47 mm intake) in your 2 vs. 4 formulas, I'm getting a mean piston speed of 3875 fpm with Z = .6.

You've pointed out the immediate benefit of higher BMEP from intake pressures above atmo (FI), and the need to increase piston speeds. I think that one item which needs consideration is the fact that in a FI motor, not only is BMEP higher, but it is maintained through significantly more of the piston's stroke. Thus the power impulse is maintained over more degrees of crank rotation and imparts that much more acceleration to the rotating mass.

Since nothing's free, the higher BMEP translates into higher dynamic running pressures throughout the piston's cycle. This means the bottom end of the cylinder walls see significantly greater radial loading and heat transfer. This is an area in most blocks where the wall thickness is quite thin . . .possibly as little as .080" So distortion and the potential for blowby (insufficient ring sealing) definitely come to the party.
If we add to this a short skirt configuration on the piston and a short rod ratio, increasing the liklihood of the piston cocking in the bore, things can get frustrating.
In the scheme of things, I tend to agree that higher piston speeds probably create more problems than the higher running pressures, but Darton Sleeves has a pretty good business giving the 1/4 mile guys a solution to the pressures situation.

This situation isn't helped when piston speeds start going up towards 5000-5200 fpm and we begin to crowd things on the ring's ability to provide a consistent seal, both immediately and over the engine's life expectancy between teardowns. Use of a Dykes configuration on the #1 oil scraper ring can help a lot to maintain seal and reduce ring flutter, but once again, lower fpm on piston speed is probably better from a wear standpoint if we can achieve it.

Playing with the 2 vs.4 formula, it also points to getting as much intake valve area (and by inference, valve curtain area) as possible, this reducing Z, all other things being equal.

Ring seal...that's a book unto itself isn't it? Cylinder finish plays such a significant role in ring design and they both tie into our precious BMEP rather significantly.

Something's fishy...47mm valve, 94mm bore. That gives us 6940^2mm Ap, and 1735^2mm Ai. Based on the 2v4 bit, that gives me Z=.7 at 19.5m/s (3840ft/min). To achieve Z=.6, we need a 51.1mm valve. In all fairness I'm a bit groggy - maybe I'm out to lunch. There was more to this post but I have to go away and re-think things first - too tired, too many 4am starts.

[Ken H.]

Hmmmm. Andrew, I think I may have used Ci = .35 in the formula.
94 mm bore = Ap =6940 ^2 mm
47 mm intake valve = 1735^2 mm
Ci = .35 (assumed; closer to the real world?)
a = 1200 f/s; 366 m/s
s = Z= .6

6940/1735 = 4.000
.35 x 366 = 128.1
Using the formula, we get 4.000/128.1 = .0312.
with Z =.6, we have .6/.0312 = 19.23 m/s
1 meter = 3.2808 feet. 19.23 x 3.2808 = 63.09 f/s 63.09 f/s x 60 sec = 3785 feet/min.

Good points about the ring seal topic, along with looking into bore finish . . .I had these itemized as part of the schedule I put aside as things needing exploration as part of "Short Block Tuning."
So it looks like I also got dyslexic and transposed digits in my post. My bad. Sorry.
As to valve diameter. Paul reminded us that in the M30 going beyond 47 mm--the size of the Hartge intake valve--you run into valve interference problems, so the head layout itself imposes a constraint and indirectly limits improvements in valve curtain area as well.

For sure I don't make any claims as to knowing what approach, technique or components are "best" in engine building. Over the years I've relied on discussing the subject with people who have a lot of hands on experience and fairly good results with SBCs, Porsches, and BMWs. Done a couple of those along the way. That and getting a pretty good introduction to Internal Combustion Engine theory courtesy of a friend of mine who is an professor emeritus of Engineerring at Colo School of Mines. ("Bill, if I provide you good single-malt and my bride's cooking, will you tutor me in IC theory.")
Reading everything I can get my hands on hasn't hurt and at least lets me ask some reasonably intelligent questions and avoid repeating the more egregious mistakes.

Paul noted that he might get into a discussion on cam lobe design. I hope other players who have some in-depth understanding of what goes into optimizing IC engine layout will speak up.
To you and the gruppe: If you know something about a particular subject say, intercooler design and layout, or squish areas, Shaddup, Jeremy ! let's get it on the agenda.

[paul burke]

Ken, the M30 will accept larger valves but there are limitations, a 50mm plus intake valve is out of the question without moving things around a little.

Paul

[Rich Euro M5]

Ken, the M30 will accept larger valves but there are limitations, a 50mm plus intake valve is out of the question without moving things around a little.

Paul

Paul,

I was wondering about that when you repllied to Jeremy's question about changing rocker arm ratios via roller rockers. If we are willing to move rocker shafts, it wouldn't be too much of a stretch to think about changing the valve angles. Would doing this allow larger valves without interference ?

Rich