Boost Control - INJ5 open circuit error

[mrsaturn7085]

Having a random error pop up under boost lately - INJ5 (boost control solenoid) is showing open circuit, though no driver warning is being set. What's even more strange is that the system is still working 100% correctly...

The error randomly sets after boosting above the boost control threshold, then randomly clears itself at a later time.

Sign of a failing solenoid, perhaps? The solenoid probably has 95k miles on it, so it wouldn't be a surprise. Solenoid is a factory part, 2006 USDM STI with factory wiring and a patch loom connecting the WRX9-10 M800 board.

[Rude Engineering]

Are you sure that this is a recent issue with your car? The Inj 5 error will be activated based on the injector settings. This can be a confusing consequence of injector outputs being used for alternate functions.

[mrsaturn7085]

The car has only been running since Sept. 20th and the error only presented itself regularly in recent data logs; the first occurrence on Oct. 7th went completely unnoticed until I reviewed logs today (error flag only lasted ~1 sec). The next event was Oct. 26th and, again, the error lasted approximately 1 sec before the flag was cleared on it's own. More recent logs (Nov. 11th onward) show the error appearing after almost every WOT event (longest set error flag was 2 min 58 sec).

It always occurs after WOT boost, then clears itself after some time after leaving WOT. If I enter WOT boost again, sometimes this clears the flag almost immediately (but not always).

Again - all closed loop control is working as intended... the only issue at hand is the error flag randomly showing up.

[Rude Engineering]

I would ignore this error as it is not connected with the boost control function. 1 second is the hold time for errors in M800 (unless you have ECU Manager connected in which case the error will be present until you manually clear ). These errors do appear sometimes when Injector outputs are used for other functions.

[mrsaturn7085]

Will do - thanks!

[Scott@FP]

Most boost control solenoids do not do anything above approx. 80-85% DC (depending on freq) if you are allowing it to reach 100% DC on the output try limiting it to a lower %.

[mrsaturn7085]

Clamped at 95% duty but haven't seen above 90%. The current aim boost table is quite rough and I'm working on refining it as time allows. Throttle-clamped 15% and 20% TP 'strips' of the boost vs. duty are logged but I still need to test the higher TP areas.

Testing has been following my usual way of filling out the boost control tables on a Subaru:

15/20/30/40/50/60/80% TP each tested in duty cycle increments of 15% (or less, if I'm close to target) until I'm hitting desired peaks. ~7 pulls at each clamped throttle position takes a while but it's always given me pretty accurate results before switching to the M800.

Finding a nice, long, dry strip of tarmac without traffic on a weekend hasn't been the easiest thing to do as we head into Winter... and ~1 hour of dyno time seems a little silly given the time it takes to get the car to the dyno and strapped down.

[mrsaturn7085]

Follow-up: for no apparent reason, the [non?]issue is no longer presenting itself.

[mrsaturn7085]

Another follow-up:

After about 3 seconds of WOT, today, the error flag came back (with some friends, this time).

I had ECU manager connected and running, and the errors were:

Error Mass Air Flow - Mass Air Flow Sensor
Error Inj Max Duty - Injector Maximum Duty Exceeded
Error Inj 5 Combined - INJ5 Open, Short or Peak

This was also enough to set the driver warning light, however with ECU manager connected, this may have been triggered before 1 sec delay.

The MAF error is confusing as the MAF is not used for anything on my setup (set to AV3, base settings from MOTEC).

The Max Duty error is also confusing as my peak INJ5 duty was 66% (MAX = 95% in boost control settings)... however fuel injector duty exceeded 100% for about 0.25s (peak = 101.8%) - is this enough to trigger the warning?

Fuel was pretty rich in this area (0.75 lambda vs. 0.78 lambda target) so the only reason the injectors are maxing is the tune needs a little tweaking, but would this been the trigger that set ALL these error flags?!

[mrsaturn7085]

Update:

MAF sensor was not logged, so I'll just let that error slide for now.
INJ5 error was triggered at the same time as max duty error and INJ5 is not an injector so I'll ignore that, as well.

Tuner confirmed that injectors were about 95% duty cycle on the dyno during tuning @ red-line (this was AFTER a new pump to fix a previous issue w/ fuel pressure). Hitting ~103% duty on the road triggered the flag so it's time for larger injectors; as an alternative, the car will be restricted for rally eventually, so I may just add the restrictor to clamp the high end fuel load (and prevent the need to fully remap the fuel).