BMW E30 323 with M20B25 Turbo


186.jpg 187.jpg 188.jpg 189.jpg harness_build_1.jpg motor_testad.jpg

The engine

The turbo system

The config and tables will be updated on this page, and documented properly given some free time. Here they are: [config.txt] and [tables.txt]

Bear in mind that fuel has not been set properly, including afterstart and other enrichments.

Update 2007-01-20

This is the off-season for cars like this unless you enjoy ice racing. The engine has been dismantled and revealed broken ring lands and signs of detonation on cyl 1 and 6. The other pistons looked ok, but could've run a little cooler. The reason for the broken ring lands is something I need help to resolve, I think I already know why!

The rpm in the logs below reveal that all is not well due to the jagged appearance of the rpm signal. Unfortunately the opportunity to make a trigger log on this particular setup is lost as the engine is in pieces and so is the car. We had no wheel errors but something else was off. I will test the particular VEMS Genboard that was used on my "test" engine later this year to see if the problem persists. The exact same settings for the trigger has been used on several other BMW engines of the same model and trigger setup and revealed no problem with the rpm signal.

My belief is that what has happened is too much ignition advance on cyl 1 and 6 only. The other cylinders probably had the correct advance.

We measured EGT at the cyl 6 exhaust port and could never reach very high EGT (700C, no more) and this is probably due to the high ignition advance. Had we measured in some other cylinders the problem might have been revealed as the other cylinders have signs on the piston tops of running hot as the surface around the intake valve is very clean from deposits - or so people tell me.

Could I get help to confirm my suspicion?

Update 2006-08-28

This past weekend we were at Piteň raceway in northern Sweden to see how it compares. It pulled [11.009 @ 203 km/h] and the fastest trap speed was 208 km/h,this was at only 1.3 bar boost in a car that most certainly weighed more than 1250 kg with driver.

We still have not dug into the trigger problem, but the car runs well so other things have been given priority.


Update 2006-07-22

This thing is a rocket ship! But there is a couple of mechanical issues:

The two VEMS Genboard issues are:

We have no wheel errors and the engine seems to run just fine. So how serious is this ? What is this "up and down" behavious that I can't seem to see on engines with the simple trigger setup?

We use firmware 1.0.38 and MegaTune with an upgrade vemvsv3.ini for that firmware.

Did you take InputTrigger/TriggerLog to verify the VR polarity with 60-2 wheel ?

Changes from 2005

Ignition coils

Above 6000 rpm and some boost this engine ran into a problem with the ignition dwell. The Ford EDIS coils need 4 ms dwell but that was simply not possible since the method used in firmware at the time did not make it possible to use overlapping dwell. At 7000 rpm one revolution of the engine takes 8.57 ms and there are three ignition events. The time between each ignition event is 2.86 ms on a 6 cyl engine, and there is currently 400 us "coil off time" which means an effective 2.46 ms dwell on coils that need 4 ms. The problem would be more serious on a 8 cylinder engine.

The coil dwell limitation had not been resolved by spring 2006 when the car was on the road again. The [high power ignition coils] from Swedish company "Biltema" were chosen and driven in "dual out" mode, which means wasted spark. This has worked well, we have not found their limit yet.

Injector FETs

We have established that poor quality fuse holders caused 2 injector drivers to break. These holders were replaced with more reliable holders with spade connectors that attach directly to the fuse.

I have replaced the damaged FETs, unfortunately the P259 was also damaged and removed, testing on the engine shows the board to be healthy again.

Injector Flyback

The GenBoard/Manual/PowerFlyback has been installed and made a tremendous difference with the injectors. We have found a setting for the injector characteristics that work, more tuning will probably change this but it runs well at idle and the map seems properly flat and uninteresting in this area.

injopen = 0
injocfuel = 400 usec

Extra pins

No stepper motor chip in place, leaves many EC18 pins free for use.

EC18-5Ground for SCL and SDA
EC18-10SDA (PD1) - on JP1 connector (current default for AntiLagSystem)
EC18-11SCL (PD0) - on JP1 connector (current default for shift cut)