BMW Forums : BimmerForums banner

Turbocharged E36 M3

25K views 54 replies 16 participants last post by  Amos 
#1 ·
Hi All,

I'm new to this forum however I have posted on a couple of other forums. Thought id share what I have done to my E36 M3 with you guys as I have ties in the UK as I am a strong Cossie fan having owned one and worked on plenty here in Aus including having a customer that is a current record holder for the fastest street registered Cosworth powered car in the country.

After receiving an offer I couldn't refuse for my Cosworth I decided to purchase an E36 M3 and turbocharge it. The car is to be kept as a genuine street car and therefore I wanted to retain my air con etc etc. I really wanted to mount the turbo on the Exh side however being right hand drive (Same as the UK) this limited us on space. Regrettably after hours and hours of thought I decided that I would have to mount the turbo on the intake side. I am well aware of the down side of mounting the turbo on the Intake side (Heat soak into the plenum, excessive Exh pipe work, added turbo lag etc etc).

The car had to pass an IM240 Emission test for the Engineers certificate, as I wanted the car 100% street legal. The IM240 is similar to a MOT on steroids!!!! Basically the car is driven on a dyno in a sealed room at the RTA/EPA (Roads and Traffic Authority/ Environmental Protection Agency) for a period of 240 seconds (Hence IM240) simulating road conditions. They set a specific target and if the car fails on the emission test the Engineers certificate is null and void and the modifications are rejected. The car passed without a problem.

The OEM fuel pump has been removed and replaced with a pair of 500hp Walbro pumps. Im using an Autronic SM4 to control both fuel and ignition as well as running an Autronic R500 CDI. The Autronic also controls the VANOS system as well. I initially had the Autronic running the engine in N/A form prior to turbo-ing the engine so I could iron out any problems (if any) as I didn't want to figure out if any problems with the tuning occurred after being turbo-ed or if it was related to the Autronic.


The turbo is a GT35 running a 44mm Tial external gate. The Turbos down pipe is 3.5" in diameter and then spilts into a twin 2.5" system all the way to the rear of the car. We had to make up a custom made intake manifold to allow the turbo to fit. It goes without saying that a intercooler was also added.

I have a fuel pressure input into the SM4 that constantly compares fuel pressure vs manifold pressure that I set up in a GPC table. If the fuel pressure is 3psi down on its set target the ECU will turn off the CDI preventing a lean out in the event of a single fuel pump failure or any other reason to why the fuel pressure drops.

All the work was carried out by a friend and myself. Between us we have over 45 years in the performance industry so we are not new to the game. I have been Pro tuning Autronic/Motec ecu's for over 15 years so managed to complete the whole project ourselves without having to rely on out sourcing any work other than purchasing of materials.

I have intentions of rebuilding the engine to be able to handle up to 30psi boost and have thus far collected a few parts (Carrillo Rods, JE Pistons etc etc). Currently the engine runs 8 psi boost and on a safe tune the engine made 397 rwhp on a dyno dynamics dyno and 385 rwhp on the same dyno
however in "Shoot Out Mode". I mentioned this as it was brought to my attention (on another forum) that any figure claimed in non "Shoot Out Mode" can be fudged..

Twin In tank Walbros.


Tial External Gate.


3.5" Inch down/dump pipe.


Twin 2.5" Exh System.


Intake Manifold.


AP Brakes.


Intercooler.


Engine Bay.




Shoot Out Mode.


Normal Mode.
 
See less See more
11
#2 ·
Very very nice mate, and so really good information there. As you know I am looking into doing this to my 3.0 M3 over here in the UK, I also love Cosworths (Just sold my Escort RS Cosworth, what a car)

Have you done any internal work on the engine? or is it totally stock?
 
#5 ·
No, the engine has not been rebuilt. It is totally stock. When i purchased it just on a year ago it had a genuine 57,000km (approx 35,600miles) on it and now has 66,000km (approx 41,200miles) on it. This is one of the reasons to why i purchased this car. I didnt want to throw plenty of money at a high milage car.

I had the car on the weigh bridge for the Engineers certificate. With no driver and 1/2 a tank of fuel it was 1503kg :eek I thought it was heavy as i was hoping for low 1400's.

No videos as yet. Sorry.
 
#7 ·
Thanks. The next plan is to give the car a run over the 1/4 mile to see what it will run. I'm just in the process of fitting the 3.2lt diff/rear end. Once i give the car a run down the 1/4 i will try and get some video footage to post.
 
#13 ·
Thanks jons.

Nat,

I’m not sure what you mean by “3.2 Evo”? Yes my car is the 3.0lt. I’m new to the BMW scene only purchasing this car in March of 08 so am not aware of the correct terminology used to distinguish one model from the other. To the best of my knowledge here in Aus we had the E36 M3 3.0lt (S50B30) and the 3.2lt (S50B32). I don’t believe we here call the 3.2 an “Evo” but if that is what your question is in regard to I simply don’t know how much more power a 3.2 can handle over a 3.0. TBH I don’t even know the limits of the 3.0 hence why I am very cautious in my tuning.

I have asked on other forums and where ever I can to try and know what is the most boost anyone has run threw a totally standard engine with reliability and longevity and have never had a straight up answer. The people that have passed on there “thoughts” never seem to have any data to back up there claims! One “Performance Workshop” have told me that they ran 1 Bar boost threw a totally standard 3.2lt engine however there onto their third engine already!!!!! On another forum I read that someone claims that they had machined down the stock pistons and ran 22psi and Nos and made some stupid amount of HP however I don’t believe that for a second. They mostly speak out of their behinds and it seems that the only method to finding the internal strength limits of these engines in standard form (head gaskets, pistons, rods etc) is by careful tuning and self-education/experimentation.

mboon,

Getting the turbo to work with these engines isn’t really as difficult as what is originally thought nor do I believe it is the tricky part regardless of it being a 3.0 or 3.2. The 3.2 engines would require far more tuning time due to the double vanos no doubt. I suppose it does come down to the people/talent (Fabricators, welders etc) that you know. I personally think the trick part is the engine management/tuning. The tuning will always literally “make or break” the engine.

I have a friend that is trying to decide wether to turbo or supercharge his 3.0. We have considered purchasing a second hand engine and dropping that into my car and tuning it properly (No Det, correct A/F etc ) and at first pushing it to 10psi and drive it around “enthusiastically” for a while and then elevate the tune to 12psi and repeat to really find out if any internal engine parts will fatigue. Personally I don’t think that there would be any sense in running any more than 12psi boost on (Australian) Pump fuel as with the high comp theses engines run there would be barley any Ign advance left and you would be simply elevating the EGT’s with very little yield in power. The ONLY thing preventing our “testing and evaluation” is the fact that a 3.0lt will set us back an expensive $10K from a wreaker/breaker! :frown
 
#14 ·
Looking good so far. I like you have added turbo chargers to everything from a pencil sharpener to a bread slicer over the years. Great way to get a bit more grunt out :D

How do you cope with fuel pressure though with the twin in tank pumps? I have always had it surging and rising at low engine speeds with twin 044s on stock end feed rails, found a better way was to run a 15-22v inverter rising voltage on boost pressure, this give support for just over 800bhp from a single 044. It also protects the AFRs from fluctuating as electrical load rises on the alternator, or the alt fails, not to mention keeping the fuel cooler. Not many people get to the bottom of the idea that a fuel pressure regulator is a proportional amplification device and many engines die as a result. Your ecu monitoring 2 pressure transducers will be a safe guard against it though.

How did you calculate the pipe sizes for exhaust and shape/volume for the inlet manifold? Boyels law and a few hours with a sheet of paper found me an extra 20bhp and 50lbs on the same set up by simple tuning the sizes to create the best scavenging effects. Intake temps also dropped from 35 to 29 degrees C on a day when the ambient temps where identical.

Do you map your own stuff too? I have used a load of Motec ecus but not autronic, how do you rate the 2 side by side?

Oh, stock sods are fine for big power by the way as long as you don’t go silly with the revs. Crank I don’t know. Pistons depend how well the map is set, but a nice set of Mahle pistons would be a nice addition to any engine IMHO :D


Good to see some one doing it not chatting for a change too :thumbsup
 
#16 ·
Danm54,

Thanks. Costing isn’t so bad as far as labour content is of concern as I do most of the work myself. :cheerful

Mrbarry,

Funny you mentioned about the fuel pressure oscillation problem. Clearly you too have been there and done that. I’m running a SARD fuel pressure reg and initially I thought that the Sard was causing some sort of a restriction and was not able to return all the fuel that the twin pumps were flowing as no matter what I did I could not reduce the fuel pressure below 3.6 to 4 bar at idle and like you mentioned the fuel pressure was not consistent. I removed the reg all together as a test and the fuel pressure would still remain at 3.6 bar. I traced the problem to the “Syphoning” tube BMW use in the fuel tank. The purpose of the syphoning tube as the name suggests is to siphon the fuel from the passengers side to the drivers side due to the fuel return line draining back into the passenger side however the fuel pump/s are in the drivers side. The fuel flow from the twin pumps simply over powered the syphoning tube and therefore created a restriction and would randomly elevate the fuel pressure dependant of the level of fuel in the tank. To counter react this problem I simply added a “Tee piece” just prior to returning the fuel into the passenger side and the Tee piece’s other side would drain any excess pressure back into the drivers side. I monitored this set up for a number of weeks to make sure that I didn’t end up with a full tank on the passengers side and an empty tank on the drivers side (The pumps side). I monitored this by measuring the resistance on each fuel level sender with a multimeter. In the end I required a 3mm restrictor in the “relief line” that was draining back into the drivers side. The set up works brilliantly and once I’m down to approx 20lt of fuel the passengers side is almost empty and the drivers side contains the 20litres.

Yes I have been Pro mapping/tuning Autronic/Motec ecu’s here for well over a decade. Both ecu’s were originally developed by the same Man. I personally believe it is 6 of one and half a dozen of the other between them. Motec does tend to have a few more “bells and whistles” however you pay handsomely for this luxury.

I know exactly by what you mean by the people that seem to talk the talk but not walk the walk and further more seem to have claims that they can not give a definitive answer to. Thanks for your compliment. :thumbsup
 
#18 ·
Fault finding is fun I always find :lol Going thru data logs and looking to see annomolies can take seconds or days. I also had issues with twin pumps over flowing at low injector duty and the fuel rail restricting the flow as well as the return lines, this sends fuel pressure high. But each problem is individual. I have a background of this sort of work and do it all day every day for a living. You just get used to it.
I find it very rewarding truth be told, overcoming problems is like a pat on the back :D I am sure BMEP agrees.

So the Autronic ecus then, 32x32 maps, trim tables for boost control etc? Would be a new alternative for my builds, Motec ecu is £1100 plus sensors. Only thing with Motec I know how well it performs, reliability is perfect. Even things like the dash displays and AEM displays all just plug in via CAN or RS232 and work first time every time, never comes up data errors, connection errors..... so many others fall over once you want more than a fuel and timing table.
 
#19 ·
Mboon,

Realistically fault finding will be dependant on the experience of the operator. To narrow down on the irregular fuel pressure problem I simply used a process of elimination to find the fault. By bypassing the fuel pressure reg totally I knew the problem wasn’t there and just continued following the path of the fuel till it became obvious that the fuel flow with the dual pumps was over powering the OEM syphoning tubes flow capabilities.

When we modify cars it comes as second nature to always expect something to give you a little grief. It comes with the territory and as Mrbarry mentioned, it is fun as well as frustrating at the same time!

Mrbarry,

The Autronic SM4 has 32x16 user defined fuel and ignition tables with multiple user defined correction tables for fuel and ignition as well. The Autronic is also compatible with plenty of after market Dash Displays. I am using a DashDAQ with the SM4, this dash displays all the data you can normally see on the laptop. Intake/Water temps, Throttle Position, Battery Voltage, Road Speed, Vanos Cam position, Manifold Pressure/Vacuum, Ignition Advance, Injection Time….etc etc etc. I have also fitted an Innovate A/F gauge where the OEM temp gauge was as I now have water temp displayed on the DashDAQ and the Innovate A/F meters second analogue output feeds back into the SM4’s O2 input allowing closed loop operation at light throttle as well as data logging the A/F ratios into the SM4’s internal data logger.



 
#20 ·
Superb work mate :thumbsup

Cant believe I missed this thread.

To answer a few others questions, the 3.2 evo (S50B32) can certainly be turbocharged but there is much less space between the bore walls for cylinder head gasket sealing and this is its achilles heel.

BMEP - have you run it down the 1/4 yet?

Looks a great project...

ps// How did you calculate the volume of the intake plenum? it looks a fair bit smaller than the stock item :confused
 
#21 ·
Agreed and it will not flow equaly to each runner according to the CFM data I have for a manifold off a supra of almost identical design. Manifold design is a night mare, did hundreds of mock ups and idealy you want a huge plenium chamber fed from inversly exponential curved walls from throttle body to runner entrys to equalise flow, but not so big as to give silly fill times, a FMIC can add 20ms to the full time of the pressure side of the engine, and a bigger plenium could double that. Bottom line tho as you make one area better another suffers. Idealistic is unobtainable.

What sensor do you use with the wide band btw? 6 or 7 wire, Bosh, Motec etc etc etc. And where are those dash displays available from?

Cheers :D
 
#22 ·
Boostist,

Thanks for the compliments. :thumbsup

I have not run the car down the ¼ but do intent to. I have been on the lookout for some 16x8 wheels to fit some decent street slick (Mickey Thompson/Hossier etc) type tyres too when I go race. Do you or anyone reading this know of any OEM BMW wheels that are 16x8 with a 38-41mm offset that will fit the M3?

You had mentioned in another post that the limits for the standard pistons are 500rwhp. Could you please elaborate on this? How did you come to this limit? Do you know someone that has actually made this rwhp on stock pistons? The reason to why I ask is if I have a definitive answer I will be inspired to up the boost to 10psi.

To answer your and mrbarrys similar question Re the intake manifold. Personally I believe it is far more important to concentrate on “runner volume” rather than “plenum volume”. I consider “runner volume” the volume from the valve seat to the trumpet mouth.

No doubt whatever type of intake manifold is designed there will always be some form of flow difference between runners. Currently at 8psi I am not too concerned however once the engine is built with stronger internals/elevated boost I will be running 6 individual pyros as one of the nice things about the Autronic is that I have individual Cyl fuel trimming based on load/rpm so in effect its like tuning 6 individual engines rather than a single 6 Cyl engine. Tuning in this manner will correct any uneven flow characteristic of the intake manifold.


Mrbarry,

The O2 sensor is the one that was supplied with the innovate kit. It is a 5-wire sensor. I purchased the DashDaq here in Australia by their DashDaq dealer. I believe the UK dealer is … http://www.dashdaq.co.uk/
 
#28 ·
Sorry for the slow reply!
Somehow missed this thread again :eek

ok, 500 RWHP so ~~ approx ~ 575 engine hp is a figure that the mad Swedes have come to when turboing the S50B30 and B32 engine. That is with a good tune and no det or knock at all.

The cast pistons apparently can fracture under this kind of horsepower and they basically go to fully forged pistons and Carillo or similar forged conrods if they are trying to get anything above 450 rwhp (safety margin).

But that is the MOST hp they have made semi long term and not killed the engine with... and they are gods of turbo BMW!!


You are making great numbers for a dyno dynamics and low boost.

10 psi is the highest I would risk with such a high compression ratio... though you could always add water injection to the intake charge and squeeze a couple more psi out :D
 
#23 ·
This week end I ran 30psi on standard pistons in a toyota, they are very poor quality for high boost being cemi forged with a high silica content, but a good map makes that unimportant, the engine is still absolutely fine. 3-psi is circa 460bhp on my Garrett GT3076r with the 256 cams and other hardware I used. So I would be more than happy to run 10psi on an M3 but then again you have to pay the rebuild if it goes wrong.

What comp ratio is aht M3? I run 9.2:1 at 30psi, high but the ecu is mapped to compensate, the retarded ignition costs a few bhp but the extra efficiency on the higher comp ratio replaces that bhp, plus the turbo is spooled faster and earlier.

My center feed manifold flows 1.5-2% more air into the center banks than the ends, I have flow bench results to suport that, best end feed I have seen for a 4 pot is 6% different to the far runner, a 6 pot would be worse by far. OK you can map that out using individual cylender trim compensation but allthough the AFR is the same in all ports the ones running less air will make less power and ultimately the engine will be down on power. The percentage also changes with flow volume, so as rpm rises the cylender trim must alter to suit, M400 custom 4d maps are supurb for this purpose.
 
#24 ·
Mrbarry,

The stock M3 compression ratio is 10.8 to 1.

My comment regarding individual Cyl trimming is to first and foremost keep the engines target A/F “safe” between Cyl’s. I am well aware that flow to each runner can vary with rpm/load hence why I mentioned that the SM4 is capable of trimming each Cyl based on rpm/load.

As for the varying power output between Cyl’s once the fuel trimming has been preformed, that’s simply the way it is when it comes to the world of modified engines. Assuming 550rwhp out of a 6 Cyl, then that is approx 92rwhp per Cyl. If we have a deviation of 5% airflow between the lowest flowing vs highest flowing runners then it works out to approx 5rwhp difference per Cyl. Knowing that each Cyl is now running the correct A/F and that there would be approx 5rwhp deviation between Cyl’s, id be more than happy with that result. Wouldn’t you? If not, how would you go about trying to yield perfect airflow between Cyl’s given the M3’s constraints within its engine bay?

Any chance you can share/post your flow sheets/data you have on your intake manifolds? With 9.2 to 1 comp and 30psi, what fuel were you using? What ignition advance would the engine tolerate at peak torque and power? I would find it very difficult to believe that 9.2 comp with 30psi boost will tolerate more than a 4-6 Deg of ignition advance on pump fuel.
 
#25 ·
at 30psi I use 102 Sunoco fuel, and it takes 3 or 4 degrees advance on top of a 10 degree base setting. On 98 road fuel I run 22 or 24psi. Those settings ar 2 degrees shy of det to allow for slight variations in fuel, humidity, all the rest of it.

I wasnt having a pop with my post tho, just sharing some findings with the people reading the thread. I have been playing for years but am no expert. not even close. Just an enthusiast who has blown some stuff up, and had some fun :D

I have been looking at data logs the last few days, what traces are you interested in seeing and I will post a section at 22psi. RPM, Boost, Ignition over base and throttle possition perhaps? I will get a snap shot and pop it on photo bucket then post it here.

I also have some CFD work being done for manifolds and as soon as I get some nice results will post the pictures for people to look at.
 
#26 ·
Mrbarry,

To save any confusion/misunderstanding I didn’t take it as you were having a pop. I do understand that sometimes posts can be read incorrectly. Also understand that my reply question was of genuine interest. Even though there are plenty of “internet experts” out there, there are also people out there with plenty of talent/knowledge. My questions Re; what method you would have used to help equalize airflow between the 6 Cyl’s via intake manifold design was of genuine interest and not intended any other way.

I’ll grant it to you that you admit that you have “blown some stuff up” via the learning process :thumbsup. Sometimes you need to take three steps back to make one step forward.

As a matter of interest I managed to get my hands on a stock S50B30 piston. I cut the piston in half to have a better look. I was surprised at how soft/easy it was to machine the piston on the mill after cutting it on the saw.



Mboon,

There is a company called CA Automotive in the UK. I’m not sure if you have heard of them or not nor do I know if they are near you but its worth having a look at their site. Here is a link that may be of interest to you……

http://www.ca-automotive.co.uk/prod...eXBlPWUzNm0mY2F0X2lkPTM5JmNvbXBfaWQ9OA==.html
 
#27 ·
OK bud thats good, I just dont want to rock any boats. I am no expert, I am interested in tuning, enjoy playing and finding out whatworks and what doesnt, still early days for me but it still shocks me some of the comments people make about tuning cars. Best being a guy who swore you can get 700bhp from a Skyline by simply doing exhaust, decat and air filter. Mmmm.

I have some CDF stuff being done as I say, will post info once I have it in a new thread and here. It may be of use to some one, even if only to let them know what not to do :lol:

a Motec log from some testing last week, very gentle ignition and boost its a standard 257.000 mile engine.
To upload it I had to change the file extention to a .txt, you need to change it back to a .LOG to open it in interpret Motec logging software. This map runs 10.7:1 AFRs and 5 degrees from Det, it was a rapid mapping session ready for a testing day. Speed is not logged as I use an aux input to sense speed not the dedicated SPD chanels, allows me to use a digital speed sensor and still enable launch control (Only used in the last few runs, runs are clear to see between taxi'ing) :)
 
#29 ·
That’s cool Mrbarry, I’m not the type to rock boats either. No need for the added stress!

The guy with the Skyline forgot to mention that the last 100 HP comes from the white stickers plastered down the front guards that says…. HKS, TRUST, SARD, even though the stickers add weight, they also add approx 100 HP! :laughing Lol

Boostist,

Id agree with the 575 flywheel power limit on these engine however with a limited life. I figure my 397rwhp is “probably” approx 490hp at the flywheel and have no doubt that 575 flywheel hp is easily attainable with a little more boost. For now though I will leave it all safe at 8psi. Oh, and the car already has water injection fitted.

I already have a set of JE pistons and Carrillo rods as well as head and mains studs sitting on the shelf waiting to go into the engine when I rebuild it. Just waiting on some upgraded valve springs and stainless intake/inconel exhaust valves to arrive as well as a complete engine rebuild kit (Gaskets, seals etc) from a BMW dealer.

Once I have all the parts required, there are a few things I need to do prior to going ahead with the remove/rebuild/replace of the engine. I’m soooooo looking forward to some more boost! :yes
 
#31 ·
Late reply/Update......



I haven't been on this site for a while so sorry for my late reply Jankes-BMW. Here are a few pics......

We simply continued from the OEM header pipes into a single pipe that swings up and under the sump and then into the turbo......



Then the dump/down pipe off the turbo is 3.5” and then splits into a twin 2.5” system......



Then it continues into a twin 2.5” system with twin High Flow Cats......



UPDATE;
I have since increased the boost to 10psi still on a totally standard engine. With careful tuning the end result was 433rwhp on a Dyno Dynamic Dyno......



As can be seen from the date on the dyno sheet, this tune was done a few months ago and the stock engine is still going strong. Here is a video of a power run......



Current pic of engine bay......



A few weeks ago I entered a Dyno Day. Unknown to me someone took a video of the M3 during a power run at the dyno day and put it up on youtube. As a matter of interest the M3 made 437rwhp at the dyno day on 10psi boost on a different dyno to the one I normally use so its good to know that the M3 is within a few rwhp between dynos......

YouTube - Turbo E36
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top