04 WRX MS3pnp - document errors, aux inputs, etc.

[michaelwoodcock]

hey guys! I am wrapping up a MS3PNP install on a 2004 Subaru WRX, and I have run into a few issues with the documentation.

It started right up on the first key turn, very good! Even though I have installed 850cc injectors. I tuned a little at idle and made too many changes before buring (some changes made off line), then the car would not start so I flashed a base map I downloaded from the website, but it seemed to run differently than the base map that came with the unit. I should have backed up my map, my mistake. I will do that this time for sure!!!

Issue #1 - stock dash on tuning software shows O2 ratio always at 10.2. I check all settings but I cannot figure this out. I opened the ECU and verified jumper seems to be installed correctly to make use of the stock O2 sensor. I presume the stock O2 sensor on the WRX for the front is the 4 wire O2 sensor, and the rear one is the 2 wire one, I am not sure if this sensor is broken. Also both O2 sensors may be 4 wires, one was missing when I bought the car (the rear one) so it's possible I have it hooked up in the wrong place but very unlikely. I have it hooked up in the right hand side of the engine bay to the only compatible connector in that location, somewhat near the boost control solenoid. I may or may not replace it. It is in a very strange location on these Chinese headers, it's right off the right bank, and only reads good from one cylinder. I will probably use an innovate O2 sensor, collecting data from after the turbo, with brings me to my second issue to far:

Issue #2: Documentation is very difficult to understand for assembling the auxilary connector. The connector is numbered, and I wish they would have used these numbers instead. Here's a few pictures illustrating this:
This is the connector supplied with the kit, and incorrectly detailed in the manual

ECU CONNECTOR.jpg (107.51 KiB) Viewed 553 times

Here's a picture showing how it is detailed in the manual:

20180129_182608 (Medium).jpg (98.04 KiB) Viewed 553 times

It is my understanding that if I hook up an analog input into the logic level output, I can cause some serious damage! so I want to be sure I do this right.

The supplied connector does not match the documentation, it has more terminal locations. Also, the plastic has different tab locations, so I am not sure if it's labeled as viewed from the back of from the front. The supplied connector is numbered 1-26. It's at work right now, but IIRC, it's labeled with the connector release being on the top, from left to right if you view it from the back. I will take another picture to illustrate and confirm this later as I'm sure this is a difficult thing to understand from my description.

[michaelwoodcock]

ok I just figured the connector out. The wires, and locations are labeled as you insert them from the rear. Now on to getting this thing programmed & tuned!

[snowrx]

Double check the tune with the MSPPNP MAP sensor.
My MS3PNP Pro came with a different MAP sensor built in than was listed in the base tune provided with the unit.
I ignored the weird MAP reading at first in my hurry to auto-tune some basic fuel numbers, and got knock that took out the ring lands on two pistons.
The two sensors had different calibrations, and in my case did not report the full boost the engine was seeing, so the timing was over advanced.
You might want to crack the case and look at the sensor in yours, or check reported MAP against a boost gauge & compare engine-off MAP with what you should see at your elevation.

The pin-out diagrams fooled me too

[michaelwoodcock]

Double check the tune with the MSPPNP MAP sensor.
My MS3PNP Pro came with a different MAP sensor built in than was listed in the base tune provided with the unit.
I ignored the weird MAP reading at first in my hurry to auto-tune some basic fuel numbers, and got knock that took out the ring lands on two pistons.
The two sensors had different calibrations, and in my case did not report the full boost the engine was seeing, so the timing was over advanced.
You might want to crack the case and look at the sensor in yours, or check reported MAP against a boost gauge & compare engine-off MAP with what you should see at your elevation.

The pin-out diagrams fooled me too

Thanks for the heads up! I updated my firmware and worked out some bugs in the map/tune after doing so. I currently do not have the boost control solenoid hooked up so I verified the sensor under boost only, it reads correctly to the boost level of 7 PSI for the BLOUCH turbo waste gate spring. Do you remember if the MS3PNP controlled the boost control solenoid accurately through the stock wiring? I see some people use the auxilary out to control boost. I'm not sure why that would be beneficial at this point, I may find out soon enough.

The car seems very slow at the current 7 PSI boost pressure, and the engine seems to run a little hot. No overheating, but the cooling fans come on more often than before. Almost as if the timing is retarded in the base map. Unfortunately we do not have an AWD dyno in town, that would be ideal in my opinion to figure out good ignition settings. Have you rebuilt the engine yet? If so, how do you find the ignigion map?

Also before messing with the ignition settings I was going to enable the knock sensor in safe mode, are the default values in tuner studio good for the OEM knock sensor, or are they just generic values recommended by the software for any car? Either way I guess I'll find out before next week end.

[snowrx]

On my SW0205 PnP, I did not find settings the stock solenoid liked, so I used the same output wiring to run a VW solenoid, which seemed to respond fine.

Engine's been rebuilt for a while and running fine except for some minor starting / idling issues. I used the ignition map provided with the MS3, but had to pull some timing due to boost knock. But my engine is a hybrid using machined 2.0 heads on a 2.5 block, so my compression ratio is higher than stock and I'm running stock or less boost at present.

I've not compared the knock settings to any other tunes, can't answer about Tuner defaults. I have been in there after looking at a bunch of datalogs, and changed (lowered) some of the knock input threshold values to more accurately reflect "normal" engine noise.

[michaelwoodcock]

Thanks for the info! the car is running at a decent state now, A few issues I'm working on now include:
Idle: perfect when cold, it goes up and over 1800 when warm. very strange.

ECU's Air Temerature reading fluctuates from a reading to minimum, I guess there's a bad wire or something somewhere. But I go to reading about where to find the Intake Air Temperature Sensor and I find that on the WRX it's built into the MAF, and my MAF is disconnected, leaving me wondering where the MS3PNP is getting it's reading.
I also read that it's best to have an IAT after the intercooler. Time to weld another bung I suppose

[snowrx]

Yes, add the IAT sensor before you chase much else. Warm hot idle- look for a vacuum leak due to expansion of manifold joints w/ spray fuel or listening hose.
What happens when you lower the target idle at warm CLT's?

[michaelwoodcock]

ok here's one pull i did. the car felt like it was hickuping a few times. I suspect this is due to the MAT (Manifold Air Temperature I presume) if you view the logs, the reading fluctuates when there is a stop in RPM climb. Also there's little chance when this sensor is reading that it is actual air temp going into the engine. I have the stock intercooler setup with no sprayer or methanol yet. It's also 70 ambient temperature right now, so it's just reading ambient temperature wherever the sensor is.

I initially thought the problem was knock, preignition, etc. so i tried to enable the knock sensor with the default values and the car thought it was knocking all the time, so I disabled it again to finish my test drive

[michaelwoodcock]

Yes, add the IAT sensor before you chase much else. Warm hot idle- look for a vacuum leak due to expansion of manifold joints w/ spray fuel or listening hose.
What happens when you lower the target idle at warm CLT's?

Thanks for the help,
I'll do that. I was considering a vacuum leak but itidles perfectly every time when it's cold. Or warm. Just after driving it around does it have idle issues. A lowered target has no effect when it has the issue after driving. If it's a warmed engine, and not hot enough it will get to target idle no problem

[michaelwoodcock]

ok on an interesting note, I found this quote:

JBPerf.com has a good VR conditioner module.

To use an injector or high current output for PWM, you'll need to add a 1N4001 diode across the solenoid, banded end to the 12 volt feed. This is built into the PWM outputs, but absent on the high current and injector outputs. The diode slows down the valve closing, letting you "park" the valve in the middle of its travel instead of snapping it rapidly open and closed.

from this thread:
http://www.msextra.com/forums/viewtopic.php?t=64045

I am thinking the Subaru OEM boost control solenoid needs this to be better controlled. Right now it's hard to get it to work. Ordering the diode to give it a try