Heat, intake temperatures and summer pulls on an EA888 Gen3

The same car, the same file, the same fuel: a log taken on a cool evening shows zero timing pull and one taken at midday in August shows a few degrees on every cylinder. Neither log is wrong. They are measuring different engines, because temperature is part of the engine.

What heat does in the cylinder

Knock is the charge igniting from pressure and heat before the spark. Boost raises the pressure; intake temperature raises the heat. A charge that enters the cylinder at 50 °C is much closer to lighting itself than one that enters at 25 °C, at the same boost and timing. The factory file knows this and retards timing as intake temperature climbs; a Stage 1 file does the same, on a slightly higher baseline. The ECU's knock control does the rest. What the owner sees is a car that is noticeably quicker on a cold night and a log with more timing pull on a hot day.

Where the heat comes from

The turbo compresses the air and compression heats it; the harder the turbo works, the hotter the air leaving it. The charge cooler brings that temperature down, but only toward the temperature of the air passing through it, and only when air is passing through it. Sit in traffic for ten minutes and the cooler soaks up engine-bay heat; the first pull afterwards feeds the engine air far hotter than ambient. This is heat soak, and it is the reason the datalog instructions say to drive at speed for a few minutes before the pull.

Reading a hot log

The tuner reads timing pull alongside intake temperature, never alone. Pull that grows as the temperature column climbs through the pull is the ECU protecting the engine from heat, and the correct response is a cooler log, not a revised file. Pull that is there from the first sample on a log where the intake temperature is reasonable is the file asking too much, and that one does get revised.

Boost also reads differently hot. The ECU corrects its airflow model for temperature, so a hot log at the same boost pressure is delivering less air, and less power, than a cool one at the same number. Dyno operators correct for this; the seat of the pants does not.

Taking a log that shows the tune

Cool the car: five minutes of steady driving with the bonnet's airflow doing its work, not idling. Then the pull, once, in third gear, from low rpm. Note the ambient temperature when you send it. If it is the hottest day of the year, say so, and expect to be asked for a second log when it is not.

What the file does about summer

A conservative Stage 1 leaves margin precisely so that a hot day uses the margin rather than the engine. On the EA888 Gen3, the simulator on this site models the knock side of this without the temperature side, which is a fair reminder that a real log has one more column than a model does.

If the car's summer logs are consistently worse than its winter ones by more than a couple of degrees of pull, the charge cooler is worth a look before the file is. On a stock Gen3, that usually means it is dirty, not undersized.