Tuning Strategy:
There is a bit of a cloud of mystery surrounding engine tuning. It seems to be considered as some mysterious secret black art, understood by few and mastered by even less. The reality is that tuning an engine involves no magic or witch craft. It requires a solid understanding of engine operation as well as what we are hoping to achieve, but beyond that it requires experience and the right equipment to get the job done.
We approached this project with the aim of making 1000 whp on Q16 race fuel, and we expected to achieve this goal with around 40 psi boost. The principles of tuning don’t change regardless of how much power the engine is making, but it can become a bit stressful as the boost and power levels climb.
The initial tuning is done while running the engine in using the dyno in steady-state mode. Here we will map individual fuel and ignition sites to optimise the tune. We use the dyno to hold the engine rpm constant and start with the minimum load we can use. The load is gradually increased and each site is mapped for fuel and ignition. Once we have reached full throttle, the rpm is increased by 500, and the process is repeated. We use this technique to build up the low boost and vacuum areas of the map, and it produces a very accurate and smooth tune in all areas.
Once we have finished steady-state tuning, we will start full load runs. In this instance the dyno is used to simulate acceleration. The Dynapack is very sensitive, and we can enter a start rpm, finish rpm and how many seconds the run should take. We use data from the track to ensure that the run length is the same as how the engine accelerates on the drag strip, ensuring our tune is as accurate and realistic as possible.
We always start tuning on the lowest boost level we can achieve (wastegate spring pressure), and start with a very conservative timing curve. This timing curve is something that we know from previous experience with similar motors. Initially we start by doing short runs from 3000 rpm out to 5000 rpm, gradually building up the map and getting the fuel curve where we want it. Once the tune is correct, we will add another 500 rpm to the run and repeat the process. In the end we have a run out to whatever rpm we intend using with a suitable mixture. Now we can confirm that the boost curve is doing what we want.
In this case the boost curve was perfectly flat which is exactly what we expected given the way the wastegate was integrated into the collector of the manifold. We ended up with 21 psi (1.4 bar) boost on the wastegate spring pressure, which is low enough to allow the car to launch properly, but will still allow us to achieve 40-50 psi with the boost control. At this point we will start slowly advancing the timing curve and checking how the engine torque responds. What we are aiming to do is find the ignition advance that maximises torque at each point in the rev range. As we start tuning with a very conservative timing map, we normally see quite large gains in torque as we begin adding timing. During this process we use a special knock detection unit that allows us to listen for detonation.
Once we get near the optimum advance angle, the gains in torque start tapering off and we know this is the timing the engine wants to see. This technique is known as MBT tuning and provides the minimum advance required to achieve best torque. If you can understand how this technique works, it should be pretty obvious that there is no way to do this without a proper steady-state loading dyno. Road tuning and inertia dynos simply can’t achieve these results.
While tuning the engine, the data from the ECU is downloaded after each pull. We can then make sure that the exhaust gas temp is even on all cylinders, and confirm that all the other logged data is within safe bounds. Based on the datalogging and the dyno information, changes are then made to the mapping to suit.
Once the baseline has been established on wastegate boost, it is time to start increasing the boost pressure. We proceed by making small changes to the wastegate duty cycle and confirming the tune at each boost increment. The aim is to make small changes in boost and gradually creep up on a complete tune. There is no future in jumping from 20 psi straight to 40 as it doesn’t take very long to damage a motor at this sort of power level. By making small changes we can start filling in the map beyond the area that is currently tuned as we can see a trend for the fuel and ignition maps. This means that when we step up to a higher boost level, the map is probably close to perfect already.
Results:
Ultimately everything went pretty much to plan, with the engine producing 1001 whp at 42 psi boost. We had an annoying gremlin in the form of the ECU shutting down that caused us a bit of grief, but we believe this was the result of electrical noise from the ignition system. Fortunately the issue didn’t rare its head at the track. At some time we will get back on the dyno for some full runs out to 10,000 rpm but for now that can wait.
The engine was always going to be laggy with such a large turbo, and it was a calculated risk. We had estimated that we would see full boost by 8000 rpm, and this proved to almost exactly right. With a rev limit of 11,000 rpm, and clutchless shifting, this provides a useable powerband for drag racing. We can still see upwards of 20 psi on the 2 step launch control, and as we saw at Masterton, the car is still capable of 1.32 60’ times so it all looks pretty promising.
We are looking forward to the next outing when we can start leaning on the engine a little harder to see if we can crack into the 8’s. Ultimately we expect to see mid 8’s at around 170mph. If you see us at the track, make sure you come up and say hi, and feel free to ask any questions you have. As long as we aren’t in the middle of some rushed repair between rounds, we are always happy to have a chat.
This is a screenshot of the Motec ECU Manager software that is used for tuning the engine. The layout can be easily altered to show the tuner any specific data they are interested in. Viewing the maps graphically is an easy way to ensure that there are no accidental holes in the map. A correctly tuned engine should always result in a relatively smooth graph.
One of the best aspects of the Motec system is its data analysis software i2. The data shown above is from Dave's 9.52 pass at Masterton. The logged data in this instance shows Engine rpm, lambda, boost pressure, oil pressure, engine temp, air temp, sync position, ignition advance, throttle position and gear. For analysing drag runs, we can enter all the run data including 60’, 330’, 660’, 1000’ and 1320’ times and speed. The run data will then be overlaid with these distance points which are shown by the vertical red lines on the screen. It makes analysis of run-to-run variations very easy.
Here the i2 software is showing us the EGT data from Dave’s 9.52 run at Masterton. We can see that the EGT’s are very closely matched for the entire run length, meaning that all 4 cylinders are running correctly. This data can also be used to detect which cylinder is misfiring as the EGT will go cold.
Note that this was only at 8000 rpm and the engine has been making peak power on full pulls closer to 9000 rpm. 
