Bottom end of the VE table and cold starts

I am interested in improving cold starts on an EJ253 (ex 2010 Forester) and am trying to conceptualise what the bottom of the Motec M130 VE table should look like.

I have a mechanical throttle which results in MAP of around 25 kpa at warmed up idle of a bit under 2,000 RPM.

However, prior to warm up, at 1,000 rpm the same throttle setting gives around 35 - 40 kpa MAP.

At cranking speed of 300 RPM (or even less if the battery is very cold) I imagine that MAP is going to be much higher so would need to ensure that the 30 kpa + MAP cells contain much higher values to reflect the additional air entering the cylinders and correctly accomodates the % enrichment called up by cranking trims. the 0 RPM column probably needs to be pretty close to ambient but would appreciate how others have managed this.

I think that I have addressed the coolant and fuel compensations contained in the M130 but am still looking at a 200+% cranking trim on cold mornings. Probably the result of insufficient rise in the second (500 RPM) column of the VE table.

Any advice, rules of thumb or illustrative tables would be much appreciated. For the record I am using the first injection for a priming pulse (up to 1100 % @ -20C) and repeating the temperature based % enrichment value without change for the remaining cranking cycles.

Regards to all.

The reality is that it's just not possible to accurately calibrate a VE table in the 60-100 kpa and 0-500 rpm range that you'll operate in during cranking. Likewise the very low load regions have the same issue although since you effectively can't access these in operation they're less of a real issue. My advice is to follow the trends/shape of the VE table from the properly tuned areas and extrapolate this shape into the low load and lower rpm regions. It won't be perfect but it'll be there or thereabouts and certainly good enough. From there you can address startup with your cranking/cold start adjustments.

My advice would be not to overthink this too much. If you've got a consistent table shape to your VE and the engine starts quickly and crisply then that's about what you can expect or hope for.

Thanks Andre. Will do.

It was cold enough this morning at 1° C and not easy to get the ECU recognised by the PC but eventually came good.

Anything less than a fast crank does not cut it with the Cycle Lock either but I guess the days of the old carbie springing to life on a terminal crank are long gone.

One thing that does interest me is after start, while still cold, the idle with closed throttle is slower and MAP is higher but Quick Lambda results in a much lower VE number (45 v 68) than warmed up faster idle with lower MAP. I thought that the higher MAP would result in more air flowing through the engine.

I have an airbox MAF sensor so might see what is really happening.

Thanks again.

Rupert, Do not use Quick Lambda on the Efficiency table except when the engine is fully warmed up. The idea is that that table represents the normal operating conditions, and various compensations deal with conditions where that need adjustment. So while the engine is warming up, you should look in the Tuning->Fuel Volume Trim tab at the "Coolant Temperature Fuel Volume Compensation" table. This table is where you make a trim adjustment while the engine is cold. That table also has the Q key "Quick Adjust", that will enter the necessary compensation required to reach the current Fuel Mixture Aim.

There is another table "Engine Post Start Fuel Volume Compensation" that can be used if you just need extra fuel for a few seconds following startup to wet your manifold, etc. Having your LTC configured to be active at crank time make it possible to tune that table as well -- but do the Coolant Temp table first.

BTW - more voltage increases cranking speed. Would it be possible to have a ground-based auxiliary battery in series so you crank with an additional 12V. We've done two-batteries in series on some high-compression motorcycle engines. They start a lot easier at 275 - 300 RPM!

Thanks David. Very helpful as usual.