topic Re: mpc5746 Etpu crank errors in MPC5xxx

mpc5746 Etpu crank errors

nickbouwknecht — Mon, 13 Feb 2023 15:19:55 GMT

Hello,

I'm having problems with the crank signal measurement. When running the engine a timeout error keeps coming. Changing window ratio's doesn't make any difference.

We expect that the crank pattern is the problem ( see attachment ). Because an additional signal is present in the gap. How can we configure the ETPU correctly for this pattern?

Re: mpc5746 Etpu crank errors

johndiener — Tue, 14 Feb 2023 05:13:06 GMT

Your pattern looks like a pretty standard 60-1. Does the engine synchronization state ever get beyond seek (0)? Are you able to watch the crank state parameter? If the eTPU software can identify the pattern the crank state will spend most of its time with a value of 7, iterating through 9, 11 and 8 at each gap. The thrust of these questions is: does it ever synchronize? If so, does it then frequently lose sync ("timeout" per your message above)? Is it losing sync under steady-state conditions? What are your window ratio values and other configuration? Is it possible your signal is noisy or inconsistent? A lot of questions, but your attached snapshot of the crank signal looks like something the eTPU should be able to process.

Re: mpc5746 Etpu crank errors

nickbouwknecht — Tue, 14 Feb 2023 08:03:18 GMT

Thanks for the quick reply !

The ETPU is able to synchronize. But when running, a hick up appears with the fault code 4 ( timeout ). When the engine is running steady state on a certain load. The timeout error will stay away for aproximatly 10 minutes before it comes back. The signal is verry clean and doesn't contain any spikes.

The window ratios are taken from the Excel sheet. But changing them won't make any difference.

Win ratio normal: 0.05
win ratio after timeout: 0.1
Win ratio across gap: 0.3
Win ratio after gap : 0.3
Gap ratio : 0.33

TCR2 Ticks per tooth : 100

We are measuring the falling edge so the ETPU ( Hopefully ) doesn't recognize the rising edge in the gap. But when looking at tooth times ( see attachment ) the gap is ± 2.5 tooth wide. On 1 tooth in gap and 59 teeth til gap, the ETPU won't sync. But on 58 teeth til gap and 2 in gap the ETPU will sync, with the described error.

Nick.

Re: mpc5746 Etpu crank errors

KarlP — Tue, 14 Feb 2023 20:15:00 GMT

If that is the signal at the eTPU input pin and it is set to trigger on the falling edge then it is a 60-2 pattern. The rising edges are ignored.

You could try widening the windows by increasing the ratios significantly, e.g.

win ratio normal = 0.75

win ratio after timeout = 0.95

win ratio across gap = 0.95

win ratio after gap = 0.95

gap ratio = 0.4

If that helps then you could narrow them back down again one at a time to find out which triggers the timeout.

The Ashware eTPU simulator (https://www.ashware.com/product-page/etpu2-development-tool) is hugely useful for checking/testing/proving these settings. You can replicate a flywheel signal from logged data when the error occurs and then test the crank function to see which setting is causing a timeout/sync problem.