Syvecs Forums

The other day I suddenly saw 2 bar boost peaks on a 1.6 bar map. Toucan did its job and gave a boost alarm, but when I looked at the Syvecs logs it was not immediately obvious what was up.

The fault was simply that the pipe from the turbo boost pick-off to the wastegate solenoid had dropped off, and once I'd found that, I saw that the ecu had pulled fuelling as a result of the MAP limit fuel cut, but there were no warnings or errors in logs or anything.

Am I missing something that I should have set up?

Or maybe this is something that could be added to the logbook entries, along with other safety/limp/trip events?

I might actually have answered my own question as I now see a MAP sensor failure in the error log that I maybe missed before - says the voltage was out of range, which was a result of the overboost

I then found that I hadn't configured sensor errors correctly on the CAN datastream, so Toucan should now show a sensor error if this happens again...at least it will prompt me to investigate even though the error wasn't the MAP sensor itself.

So you want a "spanner monkey fail" error in the datastream?

The ECU can only deal with what it's been shown. If you want an error to show up in the case of boost way above target then simply bring the WgMapErr1 and if it's sufficiently positive then you can use the Toucan to thrown an error. You can't rely rely on trips with the boost control as boost control is often out in transient moments. You could add the logic into the toucan to decide if the boost is okay, but if the boost control isn't very tight I can imagine people complaining that their Toucan says the boost is wrong all the time, much to the frustration of the mappers.

I'm a bit confused as to why you didn't hit the boost cut though, sounds like it hasn't been set up correctly. If you over volt the sensor it will just revert to default value, and reduce fuel, whereas an overboost cut will cut all the fuel for a %age of engine cycles.

pavlo wrote:So you want a "spanner monkey fail" error in the datastream?

Pipes do come off in the real world, so I was looking to see how Syvecs "deals" with it, and whether there was anything useful Toucan could display. Said pipe has been on for at least a year without problem, and it popped off as part of gear cut testing I would think...so keyboard monkey fail really

pavlo wrote:I'm a bit confused as to why you didn't hit the boost cut though, sounds like it hasn't been set up correctly. If you over volt the sensor it will just revert to default value, and reduce fuel, whereas an overboost cut will cut all the fuel for a %age of engine cycles.

Looking at the CAL, it would appear that the pressure limit base is 3 bar to 5000 rpm, dropping to 2.9 bar at 6000 rpm.

I can see fuel being pulled out once it exceeded 3 bar - it's set to pull 10% at 30mbar overboost, rising to 50% at 100mbar.

I'm thinking I should probably drop the limit base to perhaps 2.85bar.

Now I'm understanding it better, I don't think I really need to add anything extra to Toucan: it alarmed on the overboost and that prompted me to look at the logs. Inexperience meant I didn't immediately see what was the cause, but it was obviously pipework related and a quick inspection of the plumbing found it in no time.

Just to be clear, those percentage values are percentage of fuel CUT not a scaling of the fuel. So at 50% for 100mbar overboost it will cut 100% of the fuel for 50% of the cycles (cue Anchorman joke!)

Sounds like you're running a 3 bar map sensor though without sufficient headroom for the boost cut to be effective in it's current setup. I would be tempted to steepen the cut curve so you get more cut earlier on in the overboost condition.

As for the overboost detection, the syvecs can output target boost, boost error and all the relavent boost control duty cycle values, so it would be entirely possible to create an error condition within the Toucan to detect if the boost control struggling to keep the boost down ("danger to intake manifold" error). Just needs to be done in such a way as you're not going to get lots of false errors, so beyond a TPS threshold, time threshold etc.

Yes, I think it's a 3 bar MAP sensor - whatever is standard on an '05 JDM STi.

Many thanks Paul for the suggestions for a "danger of engine death" warning based on various parameters - I'll have a play around and see what I can come up with and, in the mean time, will steepen the curve and lower the threshold as well as you suggest

Just to add, I thought that might be good as overboost fuel cut is really a last resort, and if you're target is 1 bar but you're running 1.5 bar with a 1.8 fuel cut for the higher boost settings you may never know there is a problem, which is where I suppose you're coming from.

Not sure if you guys have seen but on the latest firmwares you can adjust Map Limit cut based on Cal Switch which would enable you to stop large errors in WGC.

Also its VERY useful for cars which use different Fuel types