I'm very late here, but nice work FUB. If I can summarize:
Background:
FUB modified a scroll-type SDI intake and fitted it to his modified ALH.
This ALH runs tunes that calculate theoretical air mass flow and ignore the MAF signal.
FUB then took his car to a dyno shop, and tested one intake against the other. He ran the car to pre-warm it and tried to keep engine coolant temps consistent by keeping the heater on full blast for all runs. First he tested the SDI. About 1.5 hours later he rolled onto the dyno again with stock setup.
The results appear to show SDI costing significant power and torque at every rpm.
Thread commenters analyzed various logged data and can't make sense of results.
2 notable questions thread commenters brought up though:
1. If the EDC15 ECU uses 'fuzzy logic' and gets more powerful the harder you beat on it (to my knowledge, this is true for stock tunes, but I don't know what effect the tuners have on this), then a most useful comparison is to switch the run order.
2. If the tune is ignoring air mass flow, and if the SDI has increased system volumetric efficiency, then the tune may actually be commanding too high of a boost, moving and compressing more air than is required for a given power output at a particular smoke opacity; thus the engine has more parasitic load. If there is merit to this hypothesis, then running the same dynometer test with MAF based tunes may provide for a most useful comparison.
IMO, FUB's data is an excellent discussion start to a new experimental design:
Repeat this test in reverse run order with MAF based tunes. Record smoke opacity. Compare dynometer power and torque data against smoke opacity. Perhaps smoke opacity can be gauged with something like this tool
http://www.machinebuilding.net/p/p4197.htm although it would be best to plot power/smoke opacity at each rpm point.
W
hat this data tells me is that TdiClub has some excellent engineers and scientists, and we are in need of an open-source designed, home made DIY dynometer that users can build for the $550-$750 range from easily-sourced materials. Is there any club interest in starting a thread to gauge and collect user knowledge for such a project?
I would request design requirements list starting with:
-frame built of easily weldable (using 110V welder) material like box section or round tubing.
-roller set sufficiently large for low drag, reliable bearing set with zerk fittings and sight windows
-chain or toothed belt driven fan load
-fan load has variable pitch blades or variable pitch obstructions for calibrating
-preference for off-the-shelf industrial fan or fans requiring minor modifications (+ off-the-shelf industrial fan will already have safety guards).
-fan load has keyed input shaft or universal coupling on backside so you can attach direct-drive motors of known power for calibration modes
-fan large enough to push volumes of air at velocities similar to what the car would actually experience on the open road to prevent over-heating, it also blows the smoke away.
-large rubber wheel chock integrated with physical safety INTERLOCKS such that fan load can not be driven unless safety chocks are properly deployed.
-bright mark and USB optical rpm sensor
-a non-skid shelf for laptop data capture: laptop logs USB rpm data and smoke opacity data using serial input logging. Laptop also connected to VCDS for capturing engine data. Raw text files can be merged in Excel.
-community written, open-sourced text data merging Excel macro.
-community written, open-sourced data analysis using Excel macros.
-A thread here for comparing data.
-community generated, open-source Excel file for temperature and pressure corrections.
-community based design plan, design FMEA risk assessment, warning stickers, and legal disclaimers