84 533i turbo build thread

[tschultz]

For the record and likely my later reference, the turbo has the following specs

Journal bearing/oil cooled
Exhaust:
IN:88.2
EX:77.5

Compressor:
IN: 69
EX:98.1

Which seems to put the ratios at the following, (there is a part of me that hopes the turbo never truly spools)
Exhaust A/R 1.05
Intake A/R 0.66

On reference to a catalog from precision turbo, these dimensions are almost identical to one of their larger models. I chose this turbo as once I realized the supercharger was going to be a real thing, I felt that I needed an engineering justification for the blower, and a large laggy turbo felt like it would legitimize the design. (I suppose by this same legalization logic, the fact I no longer have an oil dipstick is going to justify an external oil tank and belt driven pump) The driving experience I want to achieve is something "raw & intense" which I recognize doesn't equate to "fast". Further riding on bags doesn't exactly rhyme with "fast".

I openly admit that this project has become an art piece first and a "performance vehicle" as a lesser priority. I am really just perusing an experience at this point, and perusing this vision in my head. I have a good friend who couldn't relate to the goals and thought the combination of 4" throttle body, manual steering rack, and un-sprung 6-puck clutch combined with the weird motor would be really unpleasant. Which I hope this is not what I create, but regardless I have painted myself into a corner here and need to continue building a representation of my self-perception.


I really liked the clean unbranded look as the smooth lines visually harmonize with the valve cover shape.

Thanks for sharing those specs... that is a big turbo! But you will be pushing more air through the engine with the supercharger.

Did you already share specs of the supercharger?
It seems like to me you'd want the SC to be able to push 30% more airflow than the standard M30 via the supercharger boost level to be able to really take advantage and feed the turbo. With that different intake manifold, theoretically you could be flowing a fair amount more than the stock engine.

Curious how much you have worked through these numbers at a few rpm ranges. Or are you really just going to see how it runs and then change the SC pulley and boost level?

[Tiit]

Thanks for the update.
45mm ITB and 4" throttle body size comparison is not correct. Intake runners are not flowing continuously. Intake runner is fed from plenum. Throttle body needs to be big enough to feed the plenum.
You're over thinking the plenum design. Those fancy plenums are not proven by science. Well at least I haven't seen any any proof, that these fancy plenums are superior compared to standard versions. I've seen plenty of examples where a standard plenum works just fine.
Obviously, if you're doing it for artistic reasons, more power to you. There is no right or wrong in art.
Here is an example of "fancy" turbo plenum dyno testing.
https://youtu.be/3cMCi8L3qDE

[r.martin]

Thank you guys for engaging!

I knew I was sort of flying ignorant on so many things here, I have never really built a car before and am exposed to so much trying to learn how to weld, machine, and to understand the additional complexities of forced induction and EFI.

Thanks for sharing those specs... that is a big turbo! But you will be pushing more air through the engine with the supercharger.

Did you already share specs of the supercharger?
It seems like to me you'd want the SC to be able to push 30% more airflow than the standard M30 via the supercharger boost level to be able to really take advantage and feed the turbo. With that different intake manifold, theoretically you could be flowing a fair amount more than the stock engine.

Curious how much you have worked through these numbers at a few rpm ranges. Or are you really just going to see how it runs and then change the SC pulley and boost level?

The supercharger is an Eaton M90, eaton suggests that this unit is appropriate for 3-5L engines. And it was commonly found on 3.8L Ford and GM V6s.

here is a excerpt i was able to find from a writeup on the M90:

"The Eaton M90 supercharger is a positive-displacement, roots-type blower that displaces 90 cubic inches per minute of air per stroke. It has an rpm limit of 12,000, where it produces about 550 CFM of air at 10 psi of boost pressure while increasing the temperature. On the 3800 it was fitted with a 3.8“ diameter pulley and produced about 8 psi of boost from the factory."

When I was dreaming of twin-charging I was told by a well respected builder in the area that if I were to do this that the best results would be had if I used a roots/positive displacement compressor as it would not really care about the PSI of the incoming air and it would just further compress a given volume of air each revolution. Because of this the compressor is going to be a "stage 2" that is fed by the turbo. Which roughly made sense to me.

My plan has been to roughly ballpark and try to over-build everything as budget and life permit. long ago I had fantasies as everyone does about raising RPM limit and weird head stuff, but have decided to just keep stock 6500rpm limit and have only ported the intake side of the head to match my flange.

I also have deliberately made the plenum large for 2.5 reasons. 1 is that I read that the factory plenum runs out of capacity at a higher RPM and wanted to eliminate this concern. 2 I theorized that a larger plenum would take longer to pressurize giving me some degree of buffer in respect to boost spikes. 2.5 I felt that the factory manifold would lose its style if I modified it and thought I might be able to replicate the design language and visual interest while adding functionality.

Now that I am deep in this and my self-respect depends on me building this, I feel I need to revisit my conrods, as well has my cam and rocker arms. I think this because even napkin math suggests the very real possibility of crazy boost spike, and I do not know how strong the factory rods are.

I have got a price on rockers, and a set is expensive enough that my belt-driven oil pump is going to have to wait.

Also I do not know what kind of cam profile is appropriate for forced induction? I see several options and suspect that some grinds are for NA to promote scavenging and this would be counter productive on a turbo car?

294°/294° - 243°/243° - 9.45mm/9.45mm - 2.00mm/2.00mm - sport
297°/297° - 256°/256° - 10.55mm/10.55mm - 3.45mm/3.45mm - hot street - dirt track
294°/273° - 245°/229° - 10.35mm/8.45mm - 2.30mm/0.90mm - turbo conversion
326°/326° - 269°/269° - 11.15mm/11.15mm - 4.00mm/4.00mm - tarmac rally - race
315°/315° - 276°/276° - 11.85mm/11.85mm - 4.85mm/4.85mm - full race
320°/320° - 280°/280° - 12.00mm/12.00mm - 5.05mm/5.05mm - full race
301°/294° - 260°/246° - 11.80mm/10.65mm - 3.80mm/1.65mm - turbo conversion

[r.martin]

Thanks for the update.
45mm ITB and 4" throttle body size comparison is not correct. Intake runners are not flowing continuously. Intake runner is fed from plenum. Throttle body needs to be big enough to feed the plenum.
You're over thinking the plenum design. Those fancy plenums are not proven by science. Well at least I haven't seen any any proof, that these fancy plenums are superior compared to standard versions. I've seen plenty of examples where a standard plenum works just fine.
Obviously, if you're doing it for artistic reasons, more power to you. There is no right or wrong in art.
Here is an example of "fancy" turbo plenum dyno testing.

Ha! I have probably watched this exact video 4-5 times and many of his others. When I saw him cut apart his "ram's horn" style intake, I thought "Oh Jesus, am I getting myself into a bad position here?"

And I will absolutely give up the ghost to the statement that I am over thinking this plenum. I concluded a few months ago that I was spending too much time thinking and no enough time doing. I realized that so long as I did not make any tragically bad decisions the flow would be "good enough" and that I just needed to design with art first so long as the decisions could be "legalized" by some kind of engineering explanation. Even if all parties know that my engineering case for some design choice is weak, so long as I can say with a wink and a nod that "I built it this way for fashion and in an effort to equalize boost to the cyls"

I think so long as I do not mislabel this project as some ultra-engineered design that is backed by science, and more of art with a nod to engineering, the audience will not bring charges against me.

In the motorcycle world, I have a particular fondness of MV, their machines are not the fastest, but I feel their art could not be more perfect. The friend who I share this shop with loves his BMW S1000, and I acknowledge the engineering is supreme. But I seem to relate far more with the smaller F3-800. No where close to as fast but so artistically well done that I feel it more than makes up for the lack of speed.

In many ways this car is building itself and I am just an instrument to shape the metal. It is this peaceful zen state I find myself in tinkering with various ideas and learning why engineers made design choices that they did.

This thread weirdly feels like more of a journal than a build doc. Perhaps I isolate too much, either in a UNIX shell during the day, or the warehouse at night. Once the E28 is done I can grant myself a license to go out and finally meet some of you fine people.

[r.martin]

What do you guys do for cams on turbo M30s? Do you just run the stock one or is there a common go-to profile?

[tschultz]

Stock is good. I would start there personally since you don't really know how poerr delivery will be. If you have the rockers and the intake aimed for higher rpm, then maybe look at one but I don't know the best source

[r.martin]

I had heard a rumor that cam modifications made relatively little difference on forced induction motors.

I have found steel rocker arms with scre adjusters from Cat CAMs, they are pricey but my concern is pushing this motor too hard on "HD" rockers, which look visually thicker than stock rockers, but I have no way of knowing what their actual strength is.

Do the rockers only become a failure point if you try to push RPM?

Also do we have any approximation of what stock connecting rods can handle?

[r.martin]

Additionally, I am now concerned with the strength of the getrag 265 gearbox. I have identified the box as a 265 due to the removable bellhousing, and it being a 5 speed. This may be total error, but from looking at photos of G265 gearboxes, I have come to the belief that I have the same transmission.

I am not planning on dropping the clutch or beating on this transmission, but with how absurd this project is getting I feel that it is something not to overlook.

What is a recognized HP/torque limit of a G265? and I suppose with a removable bellhousing a range of other gearboxes could be adapted to fit. The idea of switching transmissions now mildly irritates me as I had to have a custom 6 puck clutch made to match the input splines of this box.

[iamcreepingdeath]

I seem to remember reading somewhere that the G265 is the strongest trans you can get in an E28... Lemme go find it.

EDIT: Here ya go: viewtopic.php?f=3&t=61367

G265/6
Used on '82 528e and '85 535i. Also on '85 6/735, '80-'81 E12, '81-'82 6/733s and E30 M3s. Considered the strongest transmission available for E28’s. Is easily identified by a removable bell housing which allows use on M10, M20 and M30 engines, as long as the correct bell housing is used or reused. All except M3s have a sheet metal shifter support. It’s use in '85 M30 powered cars was reportedly dictated by the short life of the 260/5 in those cars. Overdrive 5th.

That said, I have that trans in my E28, and I am shooting for 500 bhp, I sure hope it survives...

[r.martin]

Thank you Sam for taking the time to find that thread. I was really questioning how close I am to getting the driveline sorted.

From that thread it would seem that my car (84' 533) should have a G260/5 as I thought this was the original gearbox. However the description of this transmission says "non-removable bellhousing" which makes me curious. I feel this is a relatively important aspect for me to understand as a new driveshaft has to get made and I'd hate to buy that twice.

However it seems that high HP gearboxes are limited unless I want to get into some absurdly expensive stuff.

Also it seems that replacement parts/service manuals for the g265 are no longer obtainable? I saw a price to upgrade the g265 from metric mechanic for $4000+ which is inching toward the cost of making another gearbox fit.

I'm going to dig around as I have a sneaking suspicion I may end up pushing too much through this box.

[r.martin]

I can't seem to find clear information on this. As far as conrods go, does anyone have any guidance on which design is more appropriate for high power forced induction?

I have read that H-beam used to be considerd the go-to for turbo builds and that I-beam was thought of as a high-rpm design and easier to manufacture.

These ZRP rods from Greece look decent.


But these Arrow rods look even stronger at 2x the cost

[r.martin]

I was just looking through some old photos and realized the factory conrods are actually pretty strong looking. Do we have any information on how much power stock rods are good for?

Rods from my motor, the machinist said they looked good and he thought they would be pretty strong.


Custom RaceTec pistons designed from the original ones

[mitch5]

There was a guy in Florida by goodntight? He had a hx55 hooked up to a stock b34 bottom end pushing a claimed 600whp. His 1/4 times seem to back that up. I think his issue along with a lot of others running boosted m30 engines was the head gasket.

I ran arp studs and had to go through a learning curve to stop blowing head gaskets. Any detonation will blow a stock style head gasket. All my abuse on these motors has only damaged rockers and head gaskets. I think you will hit restrictions in head flow before you ever reach the limits of the stock bottom end.

Since you seem to be In The machining phase you might want to consider surfacing to a proper ra finish to run a full mls gasket. Even more extreme would be to print the deck.

[r.martin]

There was a guy in Florida by goodntight? He had a hx55 hooked up to a stock b34 bottom end pushing a claimed 600whp. His 1/4 times seem to back that up. I think his issue along with a lot of others running boosted m30 engines was the head gasket.

I ran arp studs and had to go through a learning curve to stop blowing head gaskets. Any detonation will blow a stock style head gasket. All my abuse on these motors has only damaged rockers and head gaskets. I think you will hit restrictions in head flow before you ever reach the limits of the stock bottom end.

Since you seem to be In The machining phase you might want to consider surfacing to a proper ra finish to run a full mls gasket. Even more extreme would be to print the deck.

Thank you Mitch, and yes we went with a cometic MLS head gasket, it was on the thicker side, almost as thick as I could get. My idea was to lower compression ratio and then "compensate" for the low comp ratio with the blower. Kind of absurd, but I was trying to figure out how I could justify the fashion of the compound setup.

Cometic MLS gasket


Intake side ported to match the manifold flange



I did "upgrade" the rockers to the HD variant I think from IE, but I don't recall. However this was when I was expecting ~500whp. Now I am more comfortable with building the motor to a 7/10 level and fortifying for more strength. I am thinking that a steel version of these screw type rockers will be plenty strong, and make maintenance simpler. Do you think this may be overkill?

My current rocker upgrade:

The screw-type in ALU


I did also get the suggested lock collars


I just got a DM back from Euroshark90, and he says his Chinese H-beams are still going strong after 7000 miles and 650tq/700whp @26 psi. I'd love to know more about how he accommodates those numbers as far as rockers and head work.

[marc79euro645]

I'm using the IE hd rockers and locks as well. I'd broken 2 stockers on different times, prior to the upgrade.
Who is making the steel ones?

[r.martin]

It seems that CatCams can make steel rockers. Are you HD rockers holding strong?

Does anyone know if the bellhousing bolt patterns for the M30 and the M52 are the same? This would open up a several options for transmissions without having to go to a full race box.

[marc79euro645]

Yes, so far 10years of abuse, no problems.

[tschultz]

Regarding transmissions, the options are M30 G260, G265 with M30/S14 bell housing and G245 from M10 cars.

Obviously you want the beefier G265 or G260. Some people make adapter plates to use other transmissions but not a ton of this information here ony mye28.

https://www.domi-works.com/products/bmw ... dapter-kit

[r.martin]

Regarding transmissions, the options are M30 G260, G265 with M30/S14 bell housing and G245 from M10 cars.

Obviously you want the beefier G265 or G260. Some people make adapter plates to use other transmissions but not a ton of this information here ony mye28.

https://www.domi-works.com/products/bmw ... dapter-kit

Thanks Shultz,

you kind of just jammed me up pretty bad.... I am now seeing that a DCT can be adapted to the M30, and in doing some research it seems that these gearboxes can hold a TON of power and give really fast shifts. Also they can be had with the price of this adapter plate for the price of a T56 magnum.

I am going to need to think seriously about the experience of driving this car as I had previously had that unsprung 6-puck clutch made to give a more aggressive driving experience. But DCT seems similar to sequential shit times and cheaper.

Now I need to see if the F80 manual and F80 DCT have the same bellhousing pattern. that may end up being a cheaper and more robust manual upgrade path for our cars.