16v boxer with turbo?
Posted: Thu Jul 09, 2020 8:39 am
Has anyone here done a turbo conversion for a 16v boxer engine?
I did for mine couple of years ago, and back then I wasn't able to find any project threads. There were some photos and videos of 16v turbo boxers, but that didn't get me very far with thinking how should I do it and what parts should I use. I know a guy who had built a 8v Sprint turbo, and was able to steal some of his ideas.
So... If you have installed a turbo to a 16v boxer I'd be interested to hear how you did it, why you did it like that, what problems there were, and what was the purpose of the build.
I built mine for road use, and the true purpose was my interest in how stuff works. It's my first turbo build, and also my first turbo car.
Parts I used:
KKK K26 turbocharger from Audi 200 Turbo
Wastegate from Audi 200 Turbo
140x550x65 intercooler
Generic Bosch 044 style fuel pump
Generic pod filter
470cc Injectors from Astra sportymodel (Z20LEH engine)
Megasquirt 3
MAF sensor from 156 V6
Wasted spark coil and external ignition module
TPS from Volvo 850 (Bosch 0 280 122 001)
Remote oil filter
Innovate MTX-L+ wideband lambda
And all this on a quarter million km driven stock engine with whopping 0,4 bar boost.
Charge piping was done from 51mm exhaust pipe bends. Oil dipstick pipe and engine side support rod had to be modified to get enough room for turbo stuff. Catalyst is stock, and so is rest of the exhaust system from there. Megasquirt is attached to stock engine wiring harness with an adapter harness. There is also an additional wiring harness for the wires that I didn't have on the stock engine wiring harness.
It was surprisingly difficult to find a good selection of compressor maps. From the maps I was able to find, I figured that the turbo on Audi 200 (and some other Audi models) would fit to my setup. Sadly it had external wastegate which turned out to be quite difficult to fit in the engine bay. Now it's installed horizontally next to chassis rail and has almost total of 2cm play on full suspension travel between the tie rod and drive shaft. If I would do this again, I'd try to find a compressor map from a turbo which is used on the 1.8T VAG engines and see if it would be any good. It has internal wastegate and I expect it be technically better than the turbos from 80's. There are probably other more modern (cheap and well available) options too.
At first my intention was to use speed density algorithm for engine control (=MAP as engine load). That's what majority of car people are using with their aftermarket ECUs. Of course I had the awesome individual throttle bodies, and because of that I wasn't able to get a usable MAP signal for engine control. MAP signal was usable once the engine got on boost. When the engine was not on boost the MAP signal was close to atmospheric air pressure even on a slight throttle. In practice it was not really possible to scale the fuel map for this. The dyno operator told me it would be possible to make a blended fuel map. When not on boost, the engine load would be based on TPS, and when on boost it would be based on MAP. As you can imagine, tuning that kind of fuel map is time consuming. Especially if you have flexfuel-plans. One option would have been to gut the individual throttle bodies and install a normal throttle body. Dyno guy also told me that MAF sensors are sh*t and he wouldn't tune such car. I did some research on internet, the internet too was saying that MAF sensors are sh*t and nobody on their right mind uses MAF sensor on an aftermarket ECU.
I did the reasonable thing and started investigating how OEM manufacturers have tackled the problem. It came to my attention that Nissan had built turbo engines with individual throttle bodies (SR20 and RB26DETT). Both were equipped with MAF sensor, like many OEM engine managements are. I installed a modern hot film mass flow sensor instead the ancient flap sensor and got my engine running very well on all engine loads. I called the dyno operator and told him what I had done, and that I had driven some 200 km or so and found out it works fine. He thought about tuning a MAF car for a short time and decided it might be interesting after all.
Future plans include adding an oil cooler to the system and making air intake from some OEM air filter housing. And flexfuel and boost control.
Was it worth the trouble? Yes, I think so. It made only 151hp / 189Nm on dyno, which is not a huge increase over stock power, but still nice figure for a lightweight car. Later I found out I was missing up to 2mm lift on some camshaft lobes. After fixing that and building a new exhaust I think there is a little bit more power available. On positive side now the car makes cool whistling and choo choo noises which I didn't have before. There is also potential for increasing the power from what I have now.
Let me know if you guys have experience on 16v turbo boxers! It will be interesting to hear if you did it like I did, or in a some other way.
I did for mine couple of years ago, and back then I wasn't able to find any project threads. There were some photos and videos of 16v turbo boxers, but that didn't get me very far with thinking how should I do it and what parts should I use. I know a guy who had built a 8v Sprint turbo, and was able to steal some of his ideas.
So... If you have installed a turbo to a 16v boxer I'd be interested to hear how you did it, why you did it like that, what problems there were, and what was the purpose of the build.
I built mine for road use, and the true purpose was my interest in how stuff works. It's my first turbo build, and also my first turbo car.
Parts I used:
KKK K26 turbocharger from Audi 200 Turbo
Wastegate from Audi 200 Turbo
140x550x65 intercooler
Generic Bosch 044 style fuel pump
Generic pod filter
470cc Injectors from Astra sportymodel (Z20LEH engine)
Megasquirt 3
MAF sensor from 156 V6
Wasted spark coil and external ignition module
TPS from Volvo 850 (Bosch 0 280 122 001)
Remote oil filter
Innovate MTX-L+ wideband lambda
And all this on a quarter million km driven stock engine with whopping 0,4 bar boost.
Charge piping was done from 51mm exhaust pipe bends. Oil dipstick pipe and engine side support rod had to be modified to get enough room for turbo stuff. Catalyst is stock, and so is rest of the exhaust system from there. Megasquirt is attached to stock engine wiring harness with an adapter harness. There is also an additional wiring harness for the wires that I didn't have on the stock engine wiring harness.
It was surprisingly difficult to find a good selection of compressor maps. From the maps I was able to find, I figured that the turbo on Audi 200 (and some other Audi models) would fit to my setup. Sadly it had external wastegate which turned out to be quite difficult to fit in the engine bay. Now it's installed horizontally next to chassis rail and has almost total of 2cm play on full suspension travel between the tie rod and drive shaft. If I would do this again, I'd try to find a compressor map from a turbo which is used on the 1.8T VAG engines and see if it would be any good. It has internal wastegate and I expect it be technically better than the turbos from 80's. There are probably other more modern (cheap and well available) options too.
At first my intention was to use speed density algorithm for engine control (=MAP as engine load). That's what majority of car people are using with their aftermarket ECUs. Of course I had the awesome individual throttle bodies, and because of that I wasn't able to get a usable MAP signal for engine control. MAP signal was usable once the engine got on boost. When the engine was not on boost the MAP signal was close to atmospheric air pressure even on a slight throttle. In practice it was not really possible to scale the fuel map for this. The dyno operator told me it would be possible to make a blended fuel map. When not on boost, the engine load would be based on TPS, and when on boost it would be based on MAP. As you can imagine, tuning that kind of fuel map is time consuming. Especially if you have flexfuel-plans. One option would have been to gut the individual throttle bodies and install a normal throttle body. Dyno guy also told me that MAF sensors are sh*t and he wouldn't tune such car. I did some research on internet, the internet too was saying that MAF sensors are sh*t and nobody on their right mind uses MAF sensor on an aftermarket ECU.
I did the reasonable thing and started investigating how OEM manufacturers have tackled the problem. It came to my attention that Nissan had built turbo engines with individual throttle bodies (SR20 and RB26DETT). Both were equipped with MAF sensor, like many OEM engine managements are. I installed a modern hot film mass flow sensor instead the ancient flap sensor and got my engine running very well on all engine loads. I called the dyno operator and told him what I had done, and that I had driven some 200 km or so and found out it works fine. He thought about tuning a MAF car for a short time and decided it might be interesting after all.
Future plans include adding an oil cooler to the system and making air intake from some OEM air filter housing. And flexfuel and boost control.
Was it worth the trouble? Yes, I think so. It made only 151hp / 189Nm on dyno, which is not a huge increase over stock power, but still nice figure for a lightweight car. Later I found out I was missing up to 2mm lift on some camshaft lobes. After fixing that and building a new exhaust I think there is a little bit more power available. On positive side now the car makes cool whistling and choo choo noises which I didn't have before. There is also potential for increasing the power from what I have now.
Let me know if you guys have experience on 16v turbo boxers! It will be interesting to hear if you did it like I did, or in a some other way.