MyE28.com

As you may have guessed from the title I intend to supercharge my M535i. I have been thinking about doing a supercharged BMW project for years, initially with my E30 325i and now with the M535i. The actual details are still in the conceptual phase right now. But I am about to get things rolling with initial parts procurement and getting my spare M30 installed under the house to use for testing and building up parts on. So it’s time I put this out there and get some ideas floating around.

Here are some thoughts that I’ve had so far, in no particular order.

Type of supercharger

The supercharged BMW projects that I’ve heard of have all used centrifugal compressors. These are good for lots of top end boost and power, and would be great for a track car. However I’m intending to maintain my car as a daily driver and as such I’m more interested in low-mid range torque and power. This leads me towards a fixed displacement type compressor.

Ideally I’d like a twin screw type because of their high efficiency and compact dimensions – something like the Autorotor 3150 would be great. But they’re not readily available second hand and a new one would be beyond my means, so that’s ruled out. So I’m thinking of an Eaton M90. There are plenty of them around since they’re used on the supercharged GM V6.

Power output goals

Some preliminary back-of-the-envelope calculations suggest that with good intercooling 300 bhp is quite realistic. I’m not aiming for stratospheric power figures here because it’s got to be driven daily and I’m not planning on a full rebuild of the base engine so I don’t want to over stress the bottom end.

Intercooling

This will be required to make the desired power output. At the moment my thoughts centre around what type to use and where to put it. Air-to-air would be possible but the question is then where to put it? With a front-mount setup there is a lot of piping to be done which equals increased airflow resistance. There is also the consideration that with a fixed displacement compressor the throttle is conventionally placed at the compressor inlet, meaning that a front mounted air-to-air intercooler setup would have a large throttled volume and therefore may have unacceptably poor throttle response.

So if I were to go the air-to-air route I’d want to fit the intercooler on top of the engine as close to the compressor outlet and inlet plenum as possible. This would require a bonnet scoop along the lines of a Subaru WRX. I think that this could be made workable from a functional viewpoint – but how would this affect the aesthetics? On an E28 it would either look very cool or very stupid and right now I’m not sure which!

An air-to-water setup would seem to be the better way to go. The intercooler can be mounted with the minimum possible air piping and the heat exchanger(s) mounted remotely. Several heat exchangers could be mounted at various locations in the front end such as in the standard brake ducts and immediately behind the centre grill. I could also remove the fog lamps and maybe combine with the brake ducts to allow decent sized units to be fitted here.

The obvious downside with air-to-water is the increased weight over air-to-air. This would be especially bad if I were to mount the heat exchangers up front – don’t want any more front end weight. Are there options for mounting in other locations, possibly with forced air flow to assist cooling? Hmmm.

Compressor Drive

There is not a great deal of room at the front of the engine in standard form and some re-engineering will be required here. If I were to leave the existing drive pulleys and belts in place and fit the compressor drive pulley at the very front this would necessitate removal of the standard cooling fan to give the compressor drive belt clearance. I guess that’s not a major thing. I’d then replace the standard electric fan with one or more high-volume fans. The water pump could be left in place without the viscous coupling and fan.

Throttle Placement

Conventional wisdom says that with a fixed displacement compressor you have the throttle at the compressor inlet and use a vacuum actuated bypass valve to control unwanted boost at low engine loads. The problem I have with this is that you have a much larger throttled volume (compressor, intercooler, plenum plus whatever interconnecting pipe work there is). You wouldn’t know until testing time whether the throttle response were acceptable or not, by which stage it’d be a major job to re-engineer things.

Right now I have an idea to leave the throttle in the stock location and blow into it. For compressor bypass I’d build a solid linkage system that would set the bypass valve fully open at idle and gradually close it as the throttle were increased. To avoid having boost under cruise conditions it could be made so that the mechanical linkage doesn’t engage and start closing the bypass until say 1/4 or 1/3 throttle or something. I’d work that out in testing.

I’ve done zero research on this aspect, it’s just as idea at this stage. Thoughts? Is this workable?

Electrical Load

With the extra electrical load (a couple of fans for engine cooling and maybe a couple for heat exchanger cooling) will the standard alternator be up to the job? And if I fit a larger alternator will I drive it of the existing pulley or together with the compressor? Hmmm.

Fuel System and Engine Management

I’m sure I can keep the fuel rail and maybe even the stock fuel pump and injectors – testing will tell. An adjustable FPR would probably be a good idea though. As far as the electronics are concerned I’ve pretty much discounted the idea of even trying to make this work with the Motronic system. At the moment I’m thinking of a Haltech E8. It seems to do everything I’d need it to do and I’ve had a play with the software, which looks very user friendly compared to others that I’ve seen

I would ditch the AFM and use a 2 bar MAP sensor. Standard pulse pickups and ignition would remain.

Base Engine

The engine is currently in standard tune with 250k under the belt. The bottom end is original and the head was rebuilt ~10k ago with a mild port’n’polish job. Inspection has shown that the engine is very tight and generally in great shape so I don’t intend to do a rebuild straight up. I would check the oil pressure and fit a new pump if this were in any way less than optimum, otherwise the engine will be left alone.

Driveline and Brakes

I haven’t given much thought yet what to do with the driveline, if anything. It’s got the 265 (C/R) box and standard clutch. I’d probably keep this initially and when the clutch goes I’d replace with an overdrive box and M5 clutch. I think that with the increased torque an overdrive would be the way to go for relaxed cruising. What else in the driveline do a need to look at? Possibly diff ratio, and I know the LSD is on the way out so that’d get rebuilt at some point.

I already have rebuilt E32 brakes on the front. When the rears need pads and rotors I’d do the E32/E34 conversion. I think that this should be more than adequate for a street car.

Summary

So here’s where I’m at now. I’ve just moved into a place where I have heaps of room under the house to fill up with cars and tools and junk. I’m getting the spare car (85 535i auto) brought over tomorrow. The plan is that I’ll rip out the engine and transmission and mount the engine downstairs. I can then use this as a base for fabricating parts and fitting them up, and I’d still have the M535i in one piece to check engine-to-body clearances and the like. I plan to acquire a compressor asap and will be getting straight into getting it mounted up on the spare engine. Everything else will follow from there.

So that’s it guys. What do you think? Am I crazy? Who dares me to build this thing?!

when you supercharge, do you generally have to lower the compression like when you turbo?

Depends where you start off. WIth a high-comp Euro motor you probably would. My car, being the Australian version, is 8:1 comp standard. My research so far suggests that with good intercooling this will be fine.

You wouldn't really know til it were on the dyno though so careful tuning would be required to get the boost right without overdoing it.

wouldnt having a boost controller help that?
im not TOO familiar with turbo cause im all about N/A motors, but thats the system where you can control how much boost leaves the "wastegate" correct?
I know there are systems that regulate unwanted boost, just dont know what parts are called
Also if thats a 535.....the USA one is 8:1 correct?


-----
forgot to add, its obvious you can calculate everything that goes into a motor and the weight of everything etc....but thats a ton of work in itself

With a supercharger the bypass valve controls the boost. What it does is allow an amount of air to circulate from the compressor discharge back to the inlet and by varying the amount of bypass you can control boost.

In a conventional setup (throttle before the compressor) the bypass valve is vacuum operated. A vacuum line is taken from the downstream side of the throttle and configured so that under high vacuum it opens the bypass and effectively eliminates boost at low loads. Then as the throttle is opened and vacuum reduces (absolute pressure rises) the vacuum dashpot closes the bypass valve and allows boost to build.

I should have mentioned another thing I've been toying with and that's the idea of having a servo-actuated bypass valve. This could be hooked up to an output on the ECU to give full electronic boost control. My only reservation with this is that it's not fail safe and the consequence of having the bypass valve fail closed is that something will break due to pressure build up with a fixed displacement compressor.

For now I favour the idea of a downstream throttle but with hard-linked bypass valve control.

Yes a project like this will be a ton of work, I don't underestimate that. But I'm in no rush here - things will happen as time and $$$$ permit. My goal is to have it running within 12 months.

AFAIK, the bypass on the GM Eaton blowers is vacuum controlled -- you shouldn't have to re-engineer that.

Good luck -- I really like hot-rod projects!

I TRIPLE DOG DARE YOU!!!!!

J3

Yes the M90 uses a vacuum controlled bypass valve in the applications that I know of. In fact I found one at a wrecker over the weekend and had a good look at it as I was considering buying it for my project. It was from a VT Commodore and came complete with bypass valve. It looked like it had been sitting out in the rain for a few months and the lobes had some surface rust so I didn’t buy it.

The need to re-engineer the bypass valve stems from my desire to keep the stock throttle placement. For a vacuum controlled bypass valve the throttle is located at the compressor inlet – it needs to be so that the bypass valve gets the right vacuum signal to operate. As stated I want to avoid this because the throttled volume is so much larger and this will blunt the throttle response. I propose to place the throttle after the intercooler (in the stock location) but this won’t support a vacuum controlled bypass valve. I can still use the standard M90 bypass valve and keep it in the same location but I need to engineer an alternative way of actuating it.

The diagram below shows the general idea that I have in mind. Please excuse the quality – it’s just a 30-second sketch that shows the conceptual layout and rough location of the primary components. The idle control valve isn’t shown but would remain stock. See the linkage directly coupling the throttle and bypass valve? You will notice that they are set up to be rotationally opposite. By this I mean that with the throttle closed the bypass valve is fully open. As the throttle is opened the bypass valve progressively closes and reaches the fully closed position when the throttle is wide open. The blue arrows show the movement as the throttle is opened.

Such an arrangement should provide a nice smooth progression of boost from zero at idle to maximum boost at full throttle. In addition to being progressive it effectively provides a rough sort of feed-forward boost control. Whereas a vacuum controlled bypass valve responds to a higher manifold pressure and starts to close the bypass valve to build boost, a direct linked arrangement closes the bypass in unison with an increasing engine load (throttle opening). This should further sharpen throttle response.

I’ve never seen this kind of arrangement. The only mention I’ve seen of post-compressor throttle placement was in Corky Bell’s book “Supercharged” where he says something along the lines that a method exists for blowing into the throttle with a fixed displacement supercharger but it’s not available in the aftermarket. He doesn’t say exactly what the method is. This idea seems pretty simple and potentially quite effective so I’d like to know if there’s some fundamental engine management or thermodynamic reason why it shouldn’t be used.

One reason that I can think of is that there would be no means of controlling maximum boost. It’s possible that the chosen drive ratio may lead to over-boosting and detonation, hence the subsequent need to limit maximum boost. A simple way of doing this would be to provide a mechanical relief valve in parallel with the bypass valve.

Thoughts anyone?

Not to down play the supercharger, but TCD has a intercooler intake on a 745i he sold last year. I think it was water cooled and would keep the more stock look that you are looking for.
Excuse my ignorance, I am not a turbo or supercharger guy, just throwing something out there that might be usefull to you.

I thought about doing this on my 535i even talked to my buddy that has worked for Porsche, Panoz and was one of the lead designers for Downing's Z3 SC project. The hardest part he said would be new manifold. Let me know how it goes you might change my mind.
Kevin

Making a new manifold won't be necessary for this project, I plan to use the stock inlet plenum and runners. The only manifold mod I see at this stage would be tapping a hole for the MAP sensor.

What's the Z3 project that you refer to? Is that the one featured in CB's book?

http://www.downingatlanta.com/index.htm
Actually CB might have mentioned it, I bought that book awhile back. Great info.
Kevin

so in the actual "compressor" is there something that compresses all this or does a turbine just draw in all the air.
When i think of compression, i think about something being smushed like a piston smushing air in the cylinder :p

Thanks for that link. That sounds very similar to the design/build case study used in CB's book. Not sure if it was an Eaton though - I seem to recall that it was a twin screw type, an Autorotor or something?

Supercharged is a good read, although as an engineer there are a couple of major aspects that bug me. It is obviously aimed at non-technical people and whilst the writing style employed makes it an interesting and easier read there are times where the use of overly colloquial langauge and sentence structure mask the technical clarity. My second beef is the insistence on non-SI units . Any calculations involving thermodynamics are so much easier with SI units and given that he’s trying to simplify the topic I don’t understand why he didn’t go SI.

Overall a good book though![/i]

Mr Diesel – basically the answer to your question is "yes and no" - although the question's a bit vague . Read on...

Superchargers fall into two categories: fixed displacement and non-fixed displacement.

The centrifugal supercharger is non-fixed displacement and is basically like half a turbo. It has a centrifugal air pump (like a turbocharger) to give compression but instead of having an exhaust gas driven turbine it is driven directly from the engine crankshaft. The centrifugal has internal compression, but both the air volume moved per revolution and the compression ratio are very much dependent on rotational speed. Like a turbo they provide virtually no boost at low RPM.

In the fixed-displacement camp you have the Twin Screw and the Rootes. They are called fixed displacement because they move a fixed volume of air per revolution (approximately fixed - I won't get complicated here) so they have similar characteristics to a reciprocating compressor in this regard. A twin screw has internal compression, which means that the act of pumping the air actually compresses it too. This leads to greater thermal efficiency. The Rootes basically just pumps the air and discharges at the same pressure – the compression happens downstream as the supercharger (call it an air pump if you like) squishes the air into a confined volume, such as the intake system.

The Eaton superchargers are all of the Rootes type – so I think I have answered your question?

Anyway I don't have time to go into a detailed technical description of supercharging right now. I'm sure there are plenty of good sites you can read up on it and loads of good books too. The book I mentioned by Corky Bell is worth reading even if you’re not planning on doing an SC project. A lot of the thermodynamic and engine management theory is very useful for anyone into modifying engines.

I think if you have the knowledge and money then go for it! Sounds like you got it all planned out! Keep us up to date!

Um,
First, if your going to use the stock maniflold, then SC placement is very precarious. I like the intgrated manifold/SC setup better like the M52 twin screw setup from http://www.activeautowerke.com.

You glossing over the fuel enrichment too lightly. The 059 needs help, alot of help for forced induction, especially for a low rpm twin rotor setup. Go aftermarket, it'll be cheaper in the end. Haltech would be good.

The engine is good for 400hp stock. No problems there.

With a SC, you can get away with A/A easier than W/A. The plumbing is the bitch but not insurmountable.

Check my website for my abandoned SC project in my profile.

RussC

RussC

Thanks for the comments Russ. I’m aware that getting the SC to fit with the stock manifold will be tricky. I’ll find out just how tricky when I haul my “test bed” engine from the spare car this weekend. Then I can get a really good look at where everything might go.

I’ve seen some pics of that AAW twin screw setup – very nicely done!

As mentioned in my initial post I’m not intending to try to adapt the Motronic setup – the stock engine management will be totally ditched. I currently favour the Haltech E8 with a 2 bar MAP sensor replacing the AFM. I’m figuring on keeping the stock injectors but bumping up the fuel pressure using an adjustable (but fixed pressure) FPR. The E8 (or whatever I end up using) will look after maintaining the correct A/F ratio. That will ultimately be more accurate and tunable than using a rising rate FPR.

As an aside I have just revisited my calculations and I think I may have stuffed up somewhere. It doesn’t seem like 300 bhp is anywhere near possible with the M90. The more likely scenario looks like ~275 bhp using the M112. That’s a bugger ‘cause M112’s are harder to come by than M90’s.

I dunno how I got so far off as I can’t find my original calcs right now. Anyway I’ll post my recent calcs here and hopefully someone can have a look and tell me whether they’re half-baked or not!

OK here are my calcs for the M112. These are just quick estimations as I've had to do a rough extrapolation on the performance curves for my proposed operating conditions. Also I've assumed that the stock max power is 186 bhp at 6500 rpm, which I know is not exactly correct. I don't have the stock power curve for my engine with me right now to do an accurate calculation.

Air flow (normally aspirated) = (6500 rpm x 3.43 litres) / 2 = 11148 litres/min

For 2:1 drive ratio, read flow for 13000 rpm from curves:
Boosted air flow = 730 cfm x 28.317 litres/cubic foot = 20671 litres/min

Pressure ratio = 20671 / 11148 = 1.85

Boosted available power = stock power x pressure ratio = 186 bhp x 1.85 = 344 bhp

Actual boost = (pressure ratio – 1) x 14.7 psi = (1.85 – 1) x 14.7 = 12.5 psi

Blower drive power (est. from curve) = 70 bhp (@13000 rpm / 12 psi)

Boosted flywheel power = boosted available power – blower drive power = 344 – 70 = 274 bhp

So it would seem that using the M112 the maximum possible power will be 274 bhp @ 6500 rpm / 12.5 psi. Of course this assumes no temperature rise and doesn’t allow for belt losses. I’m going to guess therefore that actual power may be around 260 – 265 bhp with really good intercooling.

Of course we all know that maximum power only tells part of the story and I’m more intesrested in low and mid range torque than the top end. So I’ll have to do some calcs across the band to see what I'll get down low.

I think I may have previously neglected to subtract the blower drive power

If your using Haltech, then your ahead of the game. Most people don't do to $$$ reasons.

Your calculations seem OK.

The install will be your problem. There have been a couple TS installs on E28s, none have looked any good and were ghetto(hack) jobs.

RussC

I'm doing two projects simultaneously, just for the heck of it. Both very different.

I have an e34 535i 5 spd with a bone stock motor except for headers and resonnator delete. That's going to get an Eaton M-90 out of a T-bird supercoupe with an A/A intercooler, probably a SAAB unit I have laying around. The big problem for me is placement of the blower. Luckily the Eaton units can be instaled in any orientation. The e34 has the oil filter facing upwards (very convenient, but very in the way of where I want the blower to go so I'll probably go with a remote mount filter or a standard down-facing one like the 28's and e24's use). There used to be a guy on the Roadfly board named Usic who installed an M-90 with an A/A intercooler on an e24 635i. He made lots of power (well into the threes) and he said he used some scavenged motor mounts for the blower and an autozone generic belt tensioner. He sent me 2 pics of his install, but they were not illuminating. He apparently did nice work but liked the low-key look so everything was painted black. He mounted the M-90 on the driver's side, on its side next to the motor, just above the steering box, so it was hard to see how he did it.

If you work out the M-90 mount geometry, I'd certainly apreciate peeking at your notes, if that wound not be an iposition. I can weld myself so all I'd be looking for is your sketches, again, if that's not an imposition.

For my other project, I'm using a Procharger blowing through a nice big carb. I really like the simplicity of a properly jetted carburator and not having to go through a series of electromechanical patches when the air flowing through a system goes beyond its original design limitations.

Both projects are moving along at an unfortunately slow pace- reality intrudes on daydreams.

Anyway, if/when you get the Eaton mounts designed, would you mind sharing with us? If I get around to it before you do, I'll commit to doing the same.

Thanks...Steve5and6.

Steve5and6 wrote:I'm doing two projects simultaneously, just for the heck of it. Both very different.

I have an e34 535i 5 spd with a bone stock motor except for headers and resonnator delete. That's going to get an Eaton M-90 out of a T-bird supercoupe with an A/A intercooler, probably a SAAB unit I have laying around. The big problem for me is placement of the blower. Luckily the Eaton units can be instaled in any orientation. The e34 has the oil filter facing upwards (very convenient, but very in the way of where I want the blower to go so I'll probably go with a remote mount filter or a standard down-facing one like the 28's and e24's use). There used to be a guy on the Roadfly board named Usic who installed an M-90 with an A/A intercooler on an e24 635i. He made lots of power (well into the threes) and he said he used some scavenged motor mounts for the blower and an autozone generic belt tensioner. He sent me 2 pics of his install, but they were not illuminating. He apparently did nice work but liked the low-key look so everything was painted black. He mounted the M-90 on the driver's side, on its side next to the motor, just above the steering box, so it was hard to see how he did it.

If you work out the M-90 mount geometry, I'd certainly apreciate peeking at your notes, if that wound not be an iposition. I can weld myself so all I'd be looking for is your sketches, again, if that's not an imposition.

For my other project, I'm using a Procharger blowing through a nice big carb. I really like the simplicity of a properly jetted carburator and not having to go through a series of electromechanical patches when the air flowing through a system goes beyond its original design limitations.

Both projects are moving along at an unfortunately slow pace- reality intrudes on daydreams.

Anyway, if/when you get the Eaton mounts designed, would you mind sharing with us? If I get around to it before you do, I'll commit to doing the same.

Thanks...Steve5and6.

Steve,

Sounds like fun. I fixed an Eaton 90 equipped e24 this year. At 6psi was pretty cool. Made full boost quickly but it was finished at 5000rpm. Not sure if it was the drag or if the supercharger was out of air of what?

It was stupid loud as well.

Not my cup of tea.

Todd

[quote="50 Hz Sine Wave"]Thanks for the comments Russ. I’m aware that getting the SC to fit with the stock manifold will be tricky. I’ll find out just how tricky when I haul my “test bed” engine from the spare car this weekend. Then I can get a really good look at where everything might go.

I’ve seen some pics of that AAW twin screw setup – very nicely done!

quote]

You should be motivated by the fact that the AA marketed unit was design by a person who decided he wanted a twin screw for his 325i. He did everything himself, except his tuning relationships fell though hence the relationship with AA.

He did, however, spend a trunk full of money getting it done.

50 Hz Sine Wave wrote:See the linkage directly coupling the throttle and bypass valve? You will notice that they are set up to be rotationally opposite. By this I mean that with the throttle closed the bypass valve is fully open. As the throttle is opened the bypass valve progressively closes and reaches the fully closed position when the throttle is wide open. The blue arrows show the movement as the throttle is opened.

If you set this up using a bell crank arrangement, it would let you time the bypass opening for the best combo of throttle response and boost. You could have the bypass close and open faster than the throttle, which would let boost come on quicker but would still avoid a big reversion as you close the throttle.

That's my thought!

hate to bring an old topic up from the dead, but I have dug out an M90 Eaton unit from the depths of my shop and started toying around with the idea of supercharging as opposed to turbocharging. the unit is out of a 95 t bird. has anyone made any serious headway with the M90 unit in a M30 motor? thanks guys!

change of plan. damn diversions... TURBO ALL THE WAY!!!

It's interesting that there aren't more supercharged e28s. I suppose that as simple as a SC system seems, in practice it's so much easier to turbo the car. Packaging the SC requires compromises such as no ac or a custom one off intake manifold. The level of skill, effort the fact that a turbo will likely outperform a SC setup seems to make turbo the choice of most.

Also...If you've ever driven a turbo vs. an S/C car. There is quite a difference. The S/C has power immediately where as the turbo takes a bit to to spool. The way a turbo builds boost is just intoxicating, way more than that of an S/C IMO.

I remember an article in European Car magazine with a supercharged E34 with an M30 motor. I also remember the name Pavoncelli being associated with that car. It was a 535ia if my memory is correct.

DaRedRocket wrote:Also...If you've ever driven a turbo vs. an S/C car. There is quite a difference. The S/C has power immediately where as the turbo takes a bit to to spool. The way a turbo builds boost is just intoxicating, way more than that of an S/C IMO.

Not all of them. Different kinds of superchargers vary in efficiency and performance. Its always a trade off too. A supercharger sized to make boost off idle isn't going to be as efficient at high rpm. Some behave a lot like turbos in that the engine must be over a certain RPM to make boost.

No matter which kind of supercharger you use and no matter how you do it, you can be sure that the same system with a turbocharger will be more efficient and make more power.