E36 3.0 ITB Race Build

[Rob]

I'll aim to put something together once the turbo is done and tuned (in a few weeks), i'll post in here wherever it ends up being

Packaging has been interesting to say the least, plenty of thought required but everything worked out...

View attachment 1749530

View attachment 1749531

View attachment 1749532

View attachment 1749533

If you want me to move a load of posts out of here into a new thread, just give me a shout.

That pipe fit is toight!

 

[BIFCAIDS]

If you want me to move a load of posts out of here into a new thread, just give me a shout.

That pipe fit is toight!

 

[Touring_Rob]

I kind of like the evolution from E36 to this, makes for a good story

 

[f0xy]

Mapping happened

Track day happened

Everythings been hot

Changes have taken place - now for a couple more days over the next few weeks

 

[Matt Cup]

Any vids?

 

[Touring_Rob]

Appreciate the time you took to create that update mate

 

[JamesBryan]

Appreciate the time you took to create that update mate

325bhp at 4psi and 385bhp at 7psi it made.

 

[f0xy]

Any vids?

Ill probably upload a clip this week

Appreciate the time you took to create that update mate

325bhp at 4psi and 385bhp at 7psi it made.

Stock it did 242hp, esentially the book figure for the engine.
Then as above, here is stock vs 4psi, albeit with 50C, intake air temperature.

Since spent the past two weeks putting together a big cooling system for the charge cooler, with two more radiators up front, bigger water capacity, high flow pump to get some more realistic temperatures.

 

[NorthloopCup]

Very nice work indeed Jord!

 

[f0xy]

Any vids?

Done 3x days with the turbo now. These were the most recent two. Been mainly running it on 3psi as its a good match with the E46 (well I keep going past ~6000rpm), with 7-8psi on the button if I want it.





Have made more changes, just two more days left this year, but all seems well at the moment. I've started sorting images out to put a thread together

 

[f0xy]

I'll be doing chronological updates on a new PH thread as and when I get the time in the coming weeks

When a standard 987 just won't do... Track/Race/Turbo Build - Page 1 - Readers' Cars - PistonHeads UK

 

[Touring_Rob]

Piston Heads

 

[f0xy]

Piston Heads

View attachment 1765291

Sorry I cant subject the general public to the adverts on here It was a different place when I first came here 20 years ago

 

[Rob]

Sorry I cant subject the general public to the adverts on here It was a different place when I first came here 20 years ago

You could just post it to both… y’know, for the old school among us that remember your old grey Clio?

 

[f0xy]

You could just post it to both… y’know, for the old school among us that remember your old grey Clio?

I know who the OGs are Mad that thread has over a million views now.

 

[f0xy]

You could just post it to both… y’know, for the old school among us that remember your old grey Clio?

This thread is ahead of what I've managed to write up so far anyway.

Handy it can all be dropped out as a complete setup, few niggles resolved and going back together for 2026

 

[f0xy]

No idea why they went as attachments and didn't insert

 

[f0xy]

2026 started well

 

[Touring_Rob]

Ouch, worked out why yet?

 

[f0xy]

Ouch, worked out why yet?

Well DET killed it, due to oil ingestion, but also an issue with knock control (which would have prevented it). It's a long story and probably easier to explain at Cadwell in July if you're interested!

Spare engine is built up and ready to go in this weekend

Before Oulton, I got it on the scales post-turbo to compare the weight to pre-turbo - only added circa 25kg overall.

 

[Touring_Rob]

Look forward to it. I couldn't see the piston tops but assumed ringland failure to stock ring gap.

See you at Cadwell!

 

[Touring_Rob]

Done 3x days with the turbo now. These were the most recent two. Been mainly running it on 3psi as its a good match with the E46 (well I keep going past ~6000rpm), with 7-8psi on the button if I want it.





Have made more changes, just two more days left this year, but all seems well at the moment. I've started sorting images out to put a thread together

Just watched these again. It's properly quick!

 

[f0xy]

Just watched these again. It's properly quick!

It's a good level on both base (wastegate) and 7-8psi, I've added flex fuel over winter so I can run a blended fuel, but ran out of time to get back to the dyno to tune that before the above. My plan this year was to concentrate on the set up more, as it had been good for 6 days in the back half of 2025 with the turbo.

At Oulton it happend mid afternoon, now I know, the damage was done earlier in the day - I was already bleeding boost into the crankcase (via a cracked piston) on an out lap, this then rose rapidly and blew the cam seal out when the piston finally gave way on cylinder 6

 

[Touring_Rob]

What a useful sensor on a boosted engine!

 

[f0xy]

Great success at Cadwell last week on the replacement engine

 

[f0xy]

 

[f0xy]

 

[f0xy]

View attachment 1791404

PSA... don't use junk... use the real deal. Normal service has resumed

 

[f0xy]

Fun day out, zero issues, perfect weather

 

[YorkshireKyle]

How brown did your pants go coming up the mountain at the end?


Good save that!

 

[f0xy]

How brown did your pants go coming up the mountain at the end?


Good save that!

I've had some big moments everywhere at Cadwell over the years, and still not fully confident around Charlies where I crashed/rolled the E36..

Trying to use up old tyres last week didn't help either!

Back to AR-1 on the rear for CSF.

 

[Touring_Rob]

Looks brilliant. Is ths M2 standard? Given how mentally quick your Cayman must be on track the M2 is mighty impressive.

 

[f0xy]

Looks brilliant. Is ths M2 standard? Given how mentally quick your Cayman must be on track the M2 is mighty impressive.

The M2 is on Nitrons and CR-S, with a map, approx 600hp - however it was his first time at Cadwell... and being a confidence track he was not pushing it at all. There is still lots of time in the Cayman, its consistent high 1.3x's round there though. It was the first outing on the Garrett and running ethanol, so boost vs target was all over the place, however back to the dyno to sort that next week

 

[Touring_Rob]

The M2 is on Nitrons and CR-S, with a map, approx 600hp - however it was his first time at Cadwell... and being a confidence track he was not pushing it at all. There is still lots of time in the Cayman, its consistent high 1.3x's round there though. It was the first outing on the Garrett and running ethanol, so boost vs target was all over the place, however back to the dyno to sort that next week

Yes I noticed the boost target was off, seemed stable though.

What are the red orange and blue markers showing re oil pressure?

 

[Touring_Rob]

PS - i find it a bit nuts that a 600bhp can head on on track and a) be driveable and b) not overheat. Shows how much extra cooling capacity BMW included... I guess they have setup for track days in Texas or something.

 

[f0xy]

Yes I noticed the boost target was off, seemed stable though.

What are the red orange and blue markers showing re oil pressure?

View attachment 1793960

Yes it's stable, just the whole 3D base duty cycle table for the solenoid is now off on the Garrett. I was running ~E35 last week, and last time on the dyno tuning ethanol the old turbo wouldn't make more than 5psi on 100% duty cycle (now know it was f*cked).

I don't use closed loop boost control - for good reason, if there is an issue (charge cooler restriction for example, or otherwise), the ECU does not just ramp in more boost, and I know to look for a problem.

Hindsight I should have never used a 3 port solenoid, as on E50 I couldnt keep the gate shut on 100% duty cycle. Reason being I want to keep my lowest setting at ~3psi flat, which is the gate spring (300hp on 99RON, 330hp on E50) as its a good level for low. I've now swapped to a 4 port solenoid now and just going to have an hour next week to set up the duty cycle tables from scratch on petrol and max blend.

Red = Oil pump outlet
Orange = Cylinder head (furtherst point)
Blue = Turbo oil feed inlet (It's regulated at 2.7bar with a TurboSmart OPR)

PS - i find it a bit nuts that a 600bhp can head on on track and a) be driveable and b) not overheat. Shows how much extra cooling capacity BMW included... I guess they have setup for track days in Texas or something.

Yeah they are solid, however temperatures are an issue after a few hard laps even on "stage 1", and they do pull power out. Being so new it will take time for people to develop them properly I guess, at least starting from a road car. You arent running it flat out for 45min like I reguarly do put it that way...

It's quite common on that whole platform to run blended fuels now too on an unlocked ECU. The power isn't really needed for any UK circuit IMO

 

[Touring_Rob]

I've always run closed loop boost and once setup have found it to be so much better than open. Boost changing with temperature did my nut in. I guess this would be a good use of the borgwarner turbo speed sensor or looking at a pressure sensor on the outlet of the turbo with some rules to cut boost or something if its gone crazy.

 

[f0xy]

I've always run closed loop boost and once setup have found it to be so much better than open. Boost changing with temperature did my nut in. I guess this would be a good use of the borgwarner turbo speed sensor or looking at a pressure sensor on the outlet of the turbo with some rules to cut boost or something if its gone crazy.

I didn't see too much variance in boost on the previous setup, maybe half a psi at most - but it's mostly used in these months and never on the road I guess. The closed loop does work well on ECUMaster - it's something to look at, as my resolution is going to drop with the 4 port, so the gate will be harder to control with the 3D table...

I already log charge cooler inlet (compressor) and outlet with a pair of TMAP sensors, added on bosses for those, to see the delta... normally around 1psi drop across the core at most, so can easily just limit RPM/cut boost duty from there if needed..

 

[Touring_Rob]

Impressed theres such little variation, 0.5psi is tight, some setups don't get that close even with closed loop.

My issue was that on open loop you would set something up perfectly on the dyno, then on the road under different conditions (temp/load etc) you would make more boost, enough to hit the over boost threshold, or other times the car would feel unexplainably flat. With closed loop (well setup) power delivery is just consistent.

I do agree with your logic though. A boost leak or collapsed intercooler would mean working the turbo much harder to hit target, thinking on it I would expect there to be a table to limit duty, I would log valve duty over several pulls then set the max duty table to be x% higher, even better if you are already logging pressure near the turbo outlet, imagine with the ECUMaster you can create custom tables or fault conditions to mitigate this?

 

[f0xy]

Impressed theres such little variation, 0.5psi is tight, some setups don't get that close even with closed loop.

My issue was that on open loop you would set something up perfectly on the dyno, then on the road under different conditions (temp/load etc) you would make more boost, enough to hit the over boost threshold, or other times the car would feel unexplainably flat. With closed loop (well setup) power delivery is just consistent.

I do agree with your logic though. A boost leak or collapsed intercooler would mean working the turbo much harder to hit target, thinking on it I would expect there to be a table to limit duty, I would log valve duty over several pulls then set the max duty table to be x% higher, even better if you are already logging pressure near the turbo outlet, imagine with the ECUMaster you can create custom tables or fault conditions to mitigate this?

I also guess it depends on the setup, wastegate priority and how well the PID is set up for the solenoid first too... so many variables even down to different ECU vendors. With me using a 60mm gate, it has a massive frontal surface area and is basically straight in the flow of both banks - to make ~3psi base on spring it's actually got ~8-9psi worth of spring stacked in it. Quite stable due to the actual hardware more than anything I think.

What it showed for me was when the previous engine was ingesting oil (from the crankcase/turbo via the charge cooler), I would drift away from boost target as the charge cooler was saturated with liquid oil, once it had been cleared/burnt off, boost would return to target.

On ECUMaster the closed loop boost control has upper and lower hard limits, same as closed loop wideband does etc. So you can never truly run away, just need to have your base duty cycle table somewhere close and give it a small window. You can do custom corrections on boost, and custom functions for any variable (which can then trigger an RPM limit, or whatever else etc).

The issue is the EMU Black is hardware limited now - they developed a full firmware rewrite from scratch beginning in 2024 with all new functionality and memory management, I was involved with the early beta and have only ever ran this car on the latest version. It was released as 'V3' earlier this year, however with lots of new suggestions, changes and even more features added, they are on the edge of the hardware limits for memory in regards to more corrections and custom tables.

Because of this you only have one custom boost 2D correction table for duty cycle, and I use this for a correction based on ethanol content. When on blended fuel (and more ignition timing), I need a percentage more duty cycle on the gate as the RPM increases (above what the base duty cycle is on petrol) to maintain the target boost. Closed loop would correct this, but at peak RPM and boost it wants about 25% more duty cycle on E50 than petrol, which is a massive allowance to let it control. If I were to use closed loop now, it would take that correction into account, and ideally you'd only be trimming a few percent all things considered

I have the overall boost cut set at 0.5psi above my peak and I think i've only hit it once in a year, so that gives an idea how much it varies and is the ultimate protection regardless of whatever it tried to add