Hi guys. I am writing this topic to share my experiences with my university thesis. Maybe for someone will that info usefull. I decided to turbocharging my peugeot 106 1.4 kfx. I did it with stock ECU which is controling fuel according to MAP sensor. Ecu is Magnetti Marreli IAW 1AP.40. So i bought 2bar Map sensor but That ECU immadiately evaluate that new sensor as faulty and switch to Throttle angle controll mode. I was looking for some MAP sensor calibrating map/ linearisation but nothing found. I tried to change a lot of values which looks like MAP linearisation but without succes. Probably that ecu have not map for linearisation Map sensor but it is based on some calculations? So i decided to try it with fueling according to throttle angle. I know its not ideal for Turbo engines but someone have to try it for group succes. So i install KP35 turbo on that my little 106. After little wastegate hole enlargement with rasp i got required boost 0,5bar. Turbo start boosting @1500rpm and at 1800rpm is boost 0,5 bar to maximum rpm. But with new air quantity old injectors are too small. So i bought some from 1.8t VAG engine. Old injector IW155 are 115cc and that new ones are 224cc. So i had to change injector calibration constant. Main fuel correction map had to be changed too. So fuel is controled by VE map and Main fuel correction map. After a few attempts (approximately 40 remaps) engine run quite good in full throttle. Problem with knock was solved after ignition angle reduction by about 10 degrees. Unforrtunately i am poor guy so i cant afford wideband gauge... So my only option is to tune fueling according to narrowband O2 sensor, engine sound and intuition. I know... doing turbo engine with stock Ecu and narrowband O2 i am very brave... But i get interesting results from measuring.
Both measurements at full throttle:
When is Lambda sensor disconected from ecu- Only Open loop- Is mixture very rich- 0,8-0,9V O2 senzor.
When i Connect Lambda sensor in Ecu- Closed/open loop - mixture looks like stoich... Voltage of O2 sensor oscilating in rich/lean range.

From that measurement i guess that ECU works only in closed loop mode.
Now engine runs quite fine in full throttle. But when i press accelerator pedal too fast engine will get too rich mixture for moment and when turbo pushes the air mixture its ok. I think that is main disadvantage of controlling turbo engine according to throttle angle. It is late response of turbocharger effect.
With that option of my car i get 98hp and 143 Nm on dyno. Stock values are 75hp and 111Nm. So increase of power and torque is exactly 30%.

All that components cost me 300eur.( turbo, injectors, fuel press sensor, boost sensor, fuel press regulator, blow off valve and some neccesary equipment). I think this is cheapest turbo addition in NA engine in world :D Maybe in future i will buy Standalone ecu and wideband sensor and it will be done properly. In attachment my ori soft, remaped soft and mappack. Note: Map named like Lambda map is probably bad named. It have no sense in full throttle lambda 1,3? too lean i think. So far i have driven about 1500km and engine is still alive so i hope it stays in that way.

How is it holding up now 3 years later?
Wanting to try something similair but i have my doubts about the strength of the engine.