Summary of how to engineer a 3.2 N/A VR6 24v into a turbocharged engine. I have always written informative ‘how to’ documents for this Club GTI group. It is a group I have remained loyal to since 2006, when I first wrote ‘How to ABF your MK2 Golf’. This ability to share my own work, is born out of my desire to tailor a vehicle to specific requirements, then report on how the task was completed. As a result activity like this has earned the respect of many an enthusiast. I believe the distribution of automotive information, empowers ambitious folks to have a go themselves, which leads to further enjoyment of their hobby. In this day of Facebook, where everything is perceived instant and folks want information on demand, with no time to read a forum anymore, I will take time to give back to this group using this medium. While I will not be expected to chronicle every detail, this information should enlighten what is involved when building up an engine for a specific requirement. It is no secret that I own an Audi 8L S3. As a Club GTI supporter and enthusiast, I have shared many a post, exclusively with this group on the Club GTI Facebook page, of that car, since it was purchased back in late 2014 with the original AMK 1.8T. Currently that powertrain is no more resident in that vehicle, which has generated this post on turbocharging the VR6 24v engine. As I get familiar with the intended character engineered into the current powertrain and display the car to enthusiasts, and I have been asked many questions in private on the ingredients required to build similar. While I accept, many folks can be lazy and what information on a plate, and despite the costs to carry out this type of work, as an enthusiast who has nothing to gain or lose, I will share details from my work of the net components that define my S3 3.2T's character. A few notes As this was a piece of research and learning exercise like my WOLF R, involving many potential error states in the modified hardware, and many other unknowns, all mechanical, electrical, integration, component selection and calibration work was either done by me, or where outsourced directed by me. My experience is, this work can be done without ramps with some pre planning. While there are a few outfits here in the UK that can carry out this type of work, the outcome is limited by their own experience and at your expense. As one fully skilled in the trade as well as one exposed daily to the OE product development environment, I wanted to be fully in control of my success, failures and learning, so the ethos of the car could be built to be exact requirements, as well in my spare time as I managed the finances to fund it. While I am happy to share what I have done over the past year, here on Club GTI freely to the public and consider myself suitably qualified and resourced, if you do wish to take points from this exercise to build your own, you do so at your own risk. Aspects on the mapping methods I will not detail. Introduction from Dragon green to Goodwood Green. After 5 years driving a trusty and rugged dragon green VR6 MK3 aka "Der Ankor", it was time for a change. The MK3 was brilliant and served its purpose, allowing me to learn more about 12v VR6 engines evolution and vehicle control. I sold it on for its 200bhp engine. The tone of that 12v VR engine and the memories was missed. In October 2014. I picked up a Goodwood green Audi 8L S3 and began to understand the vagueness and quirkiness of this type of vehicle. A Mk4 R32, which is the same PQ34 platform, would have been nice for the just the pure sound, but the price for exclusivity and the lack of pace, put me off. To my disappointment, the 8L S3, despite the Rennsport wide track and flared fenders, had even more boring performance, so being one blessed with professional mapping gear, I set about to recalibrate the engine map. That exercise made a stepped change to the car’s performance. Noticeably from 3000rpm, where regulated boost pressure of 1.3bar kicked in, the car kept accelerating just as hard till 5000rpm and would soon be in the new red line at 7200rpm, where boost ramped off to ~ 1bar. I drove the car like that for a few thousand miles, including a trip to Wolfsburg until inevitably the stock 5304 -23 turbo developed running issues and there was also light blue smoke that could be seen in heavy traffic. I wanted more performance. So the cylinder head was refurbished with new valve guides to get rid of the smoke, H sectioned rods were fitted, the block was tidied with new rings and a F23 hybrid turbo and ported ‘chinafold’ was added. As the stock injectors were past their thresholds, new 550cc jobs were purchased and the map tweaked for the extra flow. A TIP was purchased from Bill's Badger 5 also. The car was OK with a tad more lag than stock. Peak torque was now from 3500rpm until 6000rpm at 1.2 bar. I also added a SRE clutch disk which was carried over to the current powertrain. These new changes had a slight improvement on the car’s performance from stock + a remap and can be subjectively rated as OK. It was not what I wanted ultimately. I drove the car for a year and bit. In 2016 and I wanted a car with performance as a E92 M3 V8 or B7 RS4 without the high running costs. As with all my projects, I did some benchmarking of such vehicles, to blueprint in my mind what I wanted my near 1500kg 8L S3 to feel like. The response and torque of the 4.2 l RS4 V8 left a lasting impression that I wanted to capture in my S3. Building a ‘big’ turbo 4 cylinder was ruled out, as the required power and torque to deliver the desired drive feel meant, an engine that would be less durable and highly stressed. On such a 4 cylinder, the range of full load torque would be too narrow and commence at post 4500, if it was stroked at greater expense or at near 5500 rpm as a 1.8 . Projected torque and the costs involved were not desired. I always have had a soft spot for the VR6 engine and missed the 'wabble' noise of my MK3 VR6. As stated before I really liked the sound of a 3.2 VR6. For the S3, I was prepared for a large project and the addition of a turbo would be desirable option to a V6, considering the platform of my Audi was similar to a 4 Motion 24v Golf. But there was a balance to this, as complexities with engine controls or lack of vehicle compatibility at the time with SEMs, initially made that option less desirable. Determined that something had to be done, I kept looking around for justification. Then I started to watch several and study videos of many 24v 3.2, 3.0, 2.9 vehicles seen in Germany with their 700 bhp to 900 bhp engines. I liked those a lot. It was videos of Alex Bomann’s Rallye that finally convinced it was going to be done. I reviewed my resources and skills and as one who is known to put his mind to a task and complete, I decided to take the plunge and build a road going 3.2 turbo 8L S3. There was going to be no turning back! One of my main challenges was, I decided to also use the OE Bosch ME7.1.1. ECU to control the lot correctly. Despite being prolific in the knowledge of NA and 20vT me7 stuff, I took the attitude where I lacked in knowledge for a ME7.1.1 NA to turbo control strategy, I would be forced to learn. Objectives of this exercise. Engineering a 3.2 N/A VR6 to work with a turbocharger. Fitting the new powertrain to an 8L S3. Adapting the ME7.1.1 to control this bespoke engine and gearbox combination, adopting methods used on 12v M3.8.1 to run a boosted formally NA engine, as well using familiarity of the ME7.5 and MED9 ECUs found in other VW brands to know how to compensate for missing features.( I do own all legit mapping tools and editing suites). To create this vehicle, without modifying it's external appearance (i.e a sleeper). To use the creation of this vehicle as a test rig and learning experience of the 24v Vee Rheinmotor VW/Audi engine and controls. To create an exhaust note that will not be intrusive in the cabin at any road speed, but with a deep muted sound track at idle. To meet drive feel requirements translated to an engine performance at *500bhp@6400rpm and 450-480lbft@3500 to 5500rpm. To create the above to exceed durability beyond 50k miles. To be compliant with UK MOT inspections. To create a vehicle that anyone could drive, similar to a E92 M3 or B7 RS4. *dyno variability can have an effect on measured the result from place to place. Just take as guidance only. Creating the 3.2 VR6 Turbocharged engine. This essay focuses on turbocharging a 3.2 VR6 engine and some additional work to allow fitment to a facelift 8L S3. Some of what is written here can apply to a MK4 Golf AWD with a V6 engine, which includes the 2.8 4Motion (BDE/BDF) and R32, and Audi TT V6. . On the surface, general base engine prep methods also applies MK5 Golf R32, 8P A3 3.2 and 8J TT V6. However the main focus is fitment to an 8L S3. The 8L S3 is a PQ34 vehicle with Haldex controlled AWD. This means it shares a platform and some powertrain components with the Golf, Beetle, Leon, Toledo and Octavia. If you have acquired a basic 3.2 head and block was done in this exercise, converting a 1.8T AWD vehicle like a 2001 EU3 S3 with it’s 02M gearbox to a V6 vehicle is made easier if there was a doner car for the main V6 components such as: Power steering lines. Both engine control looms complete with small modification for after run water pump supply. Relay 409 added for door activated fuel pump prime (already on 4 motion, V5 20v and R32 Golf) 022906032 ** ME7.1.1 ECU HFM6 MAF (with 5th pin IAT input) Right Side V6 Engine bracket 066 VR accessory bracket with A/C pump, alternator and PAS pump HEGO junction box cover 02M V6 Gearbox + DMF Coolant flow hoses and pipes and after run pump V6/V5 twin fan and cowling. Depending on the donor and your base engine, you can sell off the rest of the donor vehicle to recover some of the cost. In my case, a late 2001 European MK4 Golf 2.8 4 motion vehicle, running a BDE engine, was sourced then broken up to donate the above components. Base engine selection. The base engine for the exercise was a 3.2 NA VR6 engine, which came in many VW/Audi/Porsche products between the EU3 and EU4 era in Europe. All transverse mounted EU3 3.2 VR6 engines share similarities and basic spec. The 8l S3 in this writing is an EU3 spec vehicle. So to reduce complexity, a 3.2 engine from the same era was used. BFH engines from the MK4 R32 at the time carried an unnecessarily high price tag. So wisdom prevailed and the exact spec long engine from an 8N Audi TT V6 aka the BHE was acquired. EU3 engines had the same heads, cams, blocks, cranks and pistons to run either 237bhp@6400rpm or 250ps@6400 from 11.25:1 CR. Power was capped via calibration settings. Unlike later 250ps EU4 engines with typical codes of BDB,BMJ and BUB, EU3 engines ran an older forged 021 crankshaft vs a 022 casting. Reports from the aftermarket community with boosted 3.2 V6 engines, suggest the use of a 021 3.2 crank. No test data exists to determine at what cylinder pressure, torque, cranks oscillation or rpm, the crank is likely to fatigue and fail. Fortunately, many 800 to 1300 bhp drag racing engines are known to use the standard crankshaft. While the exact durability of such components from an event cannot be always tracked, it was decided to use the 021 crank, as standard in the BHE engine, on the basis of common sense and projected durability for an engine that will be modified for at least half the power as the European drag racers. Without an analysis of crankshaft harmonics it is believed a 022 crank might just be suitable for this sort of project, which would open up the possibility of using later BDB, BMJ, BUB base hardware. The OE crank damper was going to be used as no engineering on in field data exists for fatigue failure on boosted engines. As per the original 1.8T AMK, the OE DMF (Mk4 R32 items) will be utilised to retain vehicle NVH refinement along with a stock clutch cover for acceptable pedal effort, reusing the SRE clutch disc from already in hand from 4 cylinder ownership. The engine build Bottom end The 24v 3.2 engine has pretty robust thick piston crowns and I sectioned conrods that could sustain the forces from increased cylinder pressure and deceleration and acceleration from piston speeds. I can attest to their robustness after 2 two exhaust valves were dropped into a cylinder at 7000rpm. The rod and bearings were undamaged and the piston though thoroughly beaten was not punctured with little marks surface marks left on the cylinder bore. The out come of such apparent carnage meant, only 1 piston was replaced and the the engine lives on. A testament to the strength of the 3.2 VR engines, has been demonstrated on drag racing engines seen in the EU, which use many standard components. Due to the relatively low desired power target and the torque to be developed, the stock pistons and rods were to be retained. For more robustness to bearing shell distortion, later bearings for the FSI V6 engines aka “03H” type sputter bearings were used. Specific ARP fasteners for the big ends were also acquired. As more air was expected to be packed into the cylinders, there was the risk of approaching cylinder pressures that could incur dieseling, hotspots, and detonation, inhibiting torque development,. So an experienced decision was made to reduce the compression ratio from a base 11.25:1. This was set with the use of a spacer plate, sandwiched between an OEM MLS gasket. Further work is ongoing to determine any revision in compression ratio. Cylinder head. No performance enhancing work is required on the cylinder head, if it is stock for this stage of performance. However with engines of unknown history, it is advised to have the valve guides checked/replaced and the deck lightly skimmed. The chain upper cover also will need to be skimmed to match any material moved from the cylinder head. In this build, the control block for fluted cam phasers was found to be seized from oil residue, causing substantial running issues. So it advised to inspect this area, if the engine is unknown. Ensure cam sensors G40 for intake "bank1" and G163 for exhaust "bank2" are in reasonable shape with no visible dents. If both sensors fail or are inoperable, the engine will not start. Due to lessons learnt during exhaust valve drop at high valve bridge EGTs, where lambda 1 was being sustained without knock, MK5 022109611P/N solid exhaust valves now used, replacing 022109611L/M sodium filled items. These are now superseded 022109611AA/AB solid valves. This operation makes the valve train pretty robust.