31 Oct 2014

A dry sump system is on hold because there is a port in the engine block that has to be drilled, threaded, and plugged. Depending where it is, the concern is introducing aluminum chips into the oil galley. Balance that with: what’s peace of mind worth? Dry sumps are run on any car that sees high G loads, which Midlana certainly will on sticky tires, so this needs more digging.

Spent a few hours testing boost-by-gear; the good news is that it works pretty well. The bad news is that the boost control valve everyone uses isn’t that great (or it’s not being controlled well); it’s an industrial vacuum/pressure control valve that isn’t intended to be used proportionally. The issue is that the exact opening depends upon a lot of things, engine temperature and weather can really mess with an open-loop setup, resulting in boost that varies by as much as 20% – that’s a lot.

Lastly, I found out why the air/fuel ratio (AFR) was so stable – the O2 sensor’s bottomed at the low end of its range! It turns out that the range of the Honda (Denso) AFR sensor is from 11.5 to 20:1, great for a street car, not great for a turbo car because they’re tuned to run right at 11.5:1 at full throttle – basically pegged at the low end. The somewhat good news is that I have an Innovate handheld unit that uses a Bosch AFR sensor that works down to 10.5:1 (and no, I can’t just swap that sensor for the Denso – it requires custom circuitry in the ECU). Just for knowledge sake it might be interesting to know what the mixture really is, but what would I do about it once I know? The AFR was originally set to wherever the tuner felt was best. With water/meth, it’s richer, but so what, because what do I want the AFR to be if the water/meth quits working? The answer is I want it to be the AFR that my tuner originally set it for, which means that I shouldn’t mess with it. A plus for this argument is that the car runs great right now – rich or otherwise – so I’m not real motivated to lean it out any.

As I learn more about turbo engines I’m becoming less impressed with the Hondata KPro ECU. It makes the best of the hardware that it has, the existing Honda OEM interface circuit upon which it’s based, but there’s only so much Hondata can do with it. Imprecise boost control, an inability to do flex-fuel, and traction control being an expensive add-on, I’ve started looking into better standalone ECUs. I’ve got one in mind – we’ll see how it goes.