Subaru Conversion Won't Idle Options

[turnaround89]

Hi everyone, so December 2018 was the first time I got to hear the WRX STi motor I purchased run. It ran rough because the intake wasn't connected, hardly any sensors hooked up, etc. However, it ran and that made me happy knowing my investment wasn't useless.

The engine that was purchased is supposed to be a subaru wrx gc8 sti v4 turbo ej20

Cut to a few months later, now I have all the required sensors (MAF, purge solenoid, ignitor, etc) that I was missing purchased and hooked up. Vacuum hoses hooked up and I thought now the engine should run and idle beautifully. Nope, it will fire and run for maybe 3 seconds then shut off. Check out the video in the link.

Subaru Engine Video

The only thing I dont have hooked up are to the non-pressure side of the intake. I tried plugging these holes today and the car did t idle any better. Now, I need to find a permanent solution to these holes, but I swear the engine is running worse now with everything hooked up then in December with nothing hooked up.

Any ideas where I could start?

I was thinking fuel pressure and timing would be my next step. In the video you can hear the fuel pump prime the system before the engine kicks over.

The wiring harness was sent out to iwire services to be redone and that appears to be functioning properly.

Thank you everyone for your help.

[904svo]

Just a WAG, Make sure the MAF is installed correctly if its in backward, that is what it
will do start and then stop.

[Chi-town]

You need to have everything connected on vacuum and pressure side of the MAF

The early systems are very finicky. Are you sure you got the right MAF sensor?

[turnaround89]

One of my concerns was if I got the correct MAF sensor. I did as much researching as possible and believe I purchased the correct one for this particular engine. The MAF is installed correctly as the flow arrow is pointing into the intake and its bracket only allows for a one way install. I'll double check part numbers again to be sure U purchased the correct one. It is a jdm engine, if that helps at all.

For the blow off valve, does this have to route back to the non-pressure side of the intake? Or could I route it to atmosphere and plug the hole in the intake?

One item I know the engine does not have is the PCV hoses. The engine simply has filters on the breather ports. Normally these would feed back into the engine. Is this absolutely necessary for the engine to operate properly?

Keep the ideas coming, thank you everyone

[Costa05]

Just a thought here. My mazda started doing the same thing. I unplugged the MAF sensor and it ran better but not 100%. Mine went out. New one solved it.

[Chi-town]

The factory bypass valve cannot be vented to atmosphere, it needs to be recirculated.

I've never agreed with removing the crankcase breather hoses or PVC for a street car. Full race car yes but vacuum helps alleviate crankcase pressure and helps rings seal, race cars use an evacuation system that isn't viable for street cars.

It's the nipple on the intake manifold that goes to the PCV capped?

[falcor75]

Is there any immobilizer in the ECU that you need to have defeated or is that done. Sounds like a typical immobilizer thing that it lets the engine start but misses other inputs from the original car and then shuts the engine down.

[turnaround89]

I dont believe there is an immobilizer. I can hit the throttle and get the car to accelerate, although not smoothly in anyway. You can sort of keep the engine running by hitting the throttle, but again its rough running.

I will work on routing the bypass into the intake and also see if i can route the breathers back into the intake. I believe the PCV is still in the intake. Is all I have to do is route the hoses to the PCV?

I've unplugged the MAF and a few other sensors earlier today such as TPS and the idle air control valve to see if that had any affect. Not any real drastic affect when any of these were disconnected.

I looked into the MAF and double checked my part numbers, I ordered one that should have been pulled off a gc8 ej20k.

appreciate the ideas, thanks everyone

[pploco]

The PCV vents with filters on shouldn't cause the motor to die like that. On a previous conversion, I vented mine with filters with no issue. When you come up on boost, you run the risk of having positive pressure pushing through the head vents that can cause oiling problems, but it should start and run fine as you have it now.

You can vent the blow off to atmosphere without issue, so that's not the cause.

What ECU are you running? Do you have an OBD port you can grab a code from?

[lierofox]

Mine's the exact same EJ20K engine for my swap, I went with a front-cut and in my case it came with a blow-off valve and the original recirculation valve port was capped off, but yes, in stock configuration the recirculation valve goes from the intercooler to the turbo inlet tube that runs under the intake manifold, between the turbo and the MAF (though at idle the presence of a blow-off valve or recirculation valve shouldn't matter as long as there's no vacuum leaks either way.)

Do you have both the MAF and the MAP/Boost/Barometric Pressure Sensor hooked up? These engines use both.

Otherwise I'd check the fuel pressure while cranking/running (Those engines will start on basically fumes but won't run well after starting.) Check the intake for any vacuum leaks.

Let's see what else... I've never gotten my EJ20K swap running 100% (particularly when it rains) but I'd double check to make sure your sensors are getting proper power, ground, and signal levels, I know on my MAP sensor the ground wire went down to a junction connector that bridges all the sensor grounds together, but for the MAP sensor it wasn't making any contact, so the voltage divider wasn't able to work properly and the sensor was spitting out a constant +5v down the signal line making the ECU think the car was constantly running at ~30psi of boost.


EDIT: Waaaaaaaaaaaait a tick. I was watching the video a bit closer admiring how good it looks with the intake manifold flipped around and I think I see the problem. Your Idle Air Control valve looks like it's just pulling in through a small filter to the top left, and not plugged into the turbo inlet tube. If so it's not drawing in any air through the MAF sensor at the bottom right of the video at all, so the MAF doesn't think there's any air flowing into the engine, that's all unmetered air that'll bypass the throttle and so the ECU gives it basically zero fuel to keep it running.

(IMG:http://www.914world.com/bbs2/uploads_offsite/i.imgur.com-18880-1556610543.1.png)

Digging through some of my pictures from when I did the timing belt and head gasket on my engine it looks like you're using a turbo inlet from a different engine as well. I don't actually see the port that the idle air valve normally plugs into, which should be right where the turbo inlet tube turns 90 degrees and goes under the intake:

(IMG:http://www.914world.com/bbs2/uploads_offsite/i.imgur.com-18880-1556612318.1.jpg)


It's not the best picture of it, but it's all I could find of the engine as it was in the front-clip before I removed it for the swap, you can see the path the IAC tube normally takes over the inlet to the turbo inlet tube:

(IMG:http://www.914world.com/bbs2/uploads_offsite/i.imgur.com-18880-1556612319.2.jpg)

[Chi-town]

An aftermarket bypass valve can be vented to atmosphere just not the factory unit.

Good catch on the idle control not being hooked up, that needs reconnected

[turnaround89]

You guys are the best! Greatly appreciate the help.

I'm the idiot that ran the idle air control through the filter. I'll need to see if I can source a new turbo inlet tube with the proper port or modify my current one to have a port for the IAC to connect to.

I believe I have the boost and barometric pressure sensors all hooked up. I'll snap a picture tonight to show the sensors i have and the lines running to the engine. Based on some vacuum/ boost line diagrams I found I've run the sensors to the intake ports in the diagram.

Out of curiosity, why does the stock bypass valve have to go back to the turbo inlet where an aftermarket one can run to atmosphere?

[zipedadoo]

Do you have the speed sensor hooked up? My ej22 ran like crap until I hooked it up. But it was a much older engine.

[Chi-town]

Stock BOV needs a little but of vacuum on the back side (inlet vacuum) to seal. Aftermarket BOV run a stiffer spring pressure so they don't need it.

[turnaround89]

Yes, the speed sensor is hooked up. When the ECU wiring was cleaned up I had to purchase a new speed sensor. The stock speed sensor was mechanical, the new one is electrical.

As for sensors boost, barometric pressure, and I believe purge control solenoid are all in the image below. The last image is the diagram I used to hook up all the boost/vacuum hoses.

(IMG:http://www.914world.com/bbs2/uploads_offsite/i.imgur.com-9067-1556676714.1.jpg)
(IMG:http://www.914world.com/bbs2/uploads_offsite/i.imgur.com-9067-1556676715.2.jpg)
(IMG:http://www.914world.com/bbs2/uploads_offsite/static.opposedforces.com-9067-1556676716.3.png)

Ignore the mess of hoses, it will all be cleaned up once the engine is running properly. Just getting everything together

As always, thank you to everyone!

[lierofox]

I ended up getting an aftermarket silicone intake for my Subaru (that I still have yet to install...d'oh) that has all the proper ports on it, though it's designed for use with the stock, non-flipped intake manifold position.

https://www.ebay.com/itm/Silicone-Turbo-Int...;frcectupt=true

That one's basically the same as the one I got, the long thin snorkel looking tube would just need a nipple fitting to allow you to plug the IAC inlet hose into it.

As for stock vs aftermarket recirculation/BOVs... there's a lot of discussion about that regarding why it vents back to the intake in stock configuration. A lot of people cite that the AFRs will be messed up if it vents to atmosphere because the MAF already measured for that air. I don't know if I buy that as being THAT important since snapping the throttle closed would change the AFR regardless (plus the massive volume of compressed air would just blow back up the turbo inlet, past the MAF, and towards the airbox. Does the MAF have a way of telling what direction the air is flowing past it?)

Personally, I think the MAIN reason it vents back to the intake is just because the stock Subaru airbox and intake resonators will help muffle the sound of it.

[pploco]

I ended up getting an aftermarket silicone intake for my Subaru (that I still have yet to install...d'oh) that has all the proper ports on it, though it's designed for use with the stock, non-flipped intake manifold position.

https://www.ebay.com/itm/Silicone-Turbo-Int...;frcectupt=true

That one's basically the same as the one I got, the long thin snorkel looking tube would just need a nipple fitting to allow you to plug the IAC inlet hose into it.

As for stock vs aftermarket recirculation/BOVs... there's a lot of discussion about that regarding why it vents back to the intake in stock configuration. A lot of people cite that the AFRs will be messed up if it vents to atmosphere because the MAF already measured for that air. I don't know if I buy that as being THAT important since snapping the throttle closed would change the AFR regardless (plus the massive volume of compressed air would just blow back up the turbo inlet, past the MAF, and towards the airbox. Does the MAF have a way of telling what direction the air is flowing past it?)

Personally, I think the MAIN reason it vents back to the intake is just because the stock Subaru airbox and intake resonators will help muffle the sound of it.

Yep

[turnaround89]

Hi again everyone,

So I finally have the bypass valve recirculating back into the non-pressurized side of the intake. The idle air control valve is now correctly pulling from the intake as well.

Fuel pressure is roughly 40-42psi on start-up and holds pretty constant while trying to idle. Fuel pressure does increase with increased revs which makes sense due to the boost controlled fuel pressure regulator on the car. I don't believe I am having a fuel pressure issue.

I believe something else is going on related to a sensor, just not really sure where to start. The MAF is newer (used, but from a 96 EJ20 GC8). From what I can see all the sensors I need are hooked up as I don't have any extra connectors on the ECU harness.

Below are two videos, 1 where the car fires up and attempts to idle, but shuts down quickly. The second is my hitting the throttle showing the car revs strongly, but once it returns to idle, the engine shuts off.

Subaru WRX Idle Issue
Subaru WRX Engine Revving

The engine is pulled from the vehicle to make tracing this issue, hopefully, a little easier. This also makes working on the engine easier overall. That is the reason for the mess of wires going everywhere on top of the engine, I don't have anywhere to organize them except to sit on top of the intake.

Still trying to track down this idle issue, any ideas are greatly appreciated. Thank you all!

[Chi-town]

Pull the idle air control motor off and watch it as you power up the system.
It should move in and out. If it doesn't it's bad. Also make sure it's clean in the passage.

Do you show any codes in the ECU?

[904svo]

If you don't have a program to check your engine. I would suggest you try FreeSSM which can be downloaded from the internet. This will allow you to check all your sensors
for data and also check for ecu codes.