Need some Subie expertise

[65 Cobra Dude]

Guys,

I've been chasing this a while to no avail. My 818 sounds like it is hitting a rev limiter at 3500 rpm. I've checked the timing belt (was off by 2 teeth but is fixed now), the compression is 150 at each cylinder, vacuum is 22 lbs at idle and drops to 5 at 3500 rpm. Fuel pressure is 50 at idle and while revving. I checked fuel volume and it seems to be good as well. I'm using the Boyd aluminum fuel tank, sender and pump and Wayne's aftermarket wiring harness with engine control from Electromotive. New plugs and the coils all read about the same resistance across the same terminals. New fuel filter and fuel pressure and idle rises when pulling vacuum off fuel pressure regulator.

Any ideas? I am new to Subies and could use help please.

Thx,

Henry

[STiPWRD]

This sounds like it may be caused by the ECU, have you contacted Wayne?

[longislandwrx]

closed loop to open loop transition? are you still using the maf?

[65 Cobra Dude]

Thanks for the feedback. Yes, Wayne sent me a new program and the same symptoms. I'll contact him again but thought maybe it was something mechanical. MAF is not used with the Electromotive system, nor are the O2 sensors.

Thx,

Henry

[STiPWRD]

Maybe it's a boost cut off? 3500 rpm is about when the turbo spools up - have you check how much boost your producing as the revs climb? Are you running a wideband O2 sensor at least - any AFR measurements? Any knock retard or misfire codes? Are you running the factory boost controller - all the air lines hooked up correctly?

[65 Cobra Dude]

STi,

Thanks for the response. No, I have not checked boost and not sure how to check it? I do have an external WB O2 sensor that I can hook up and check the AFR. No codes as Electromotive does away with the OBDII. All factory as far as I know. When you say air lines, is that vacuum lines?

Thx,

Henry

[STiPWRD]

No, I have not checked boost and not sure how to check it?
You can buy a cheap boost gauge and run an air hose to it off of one of the intake manifold nipples (or T off of an existing IM hose). This will give you a good reading of the manifold pressure/vacuum. Keeping an eye on boost isn't just good for troubleshooting, it's an important diagnostic. I've used this one before since aesthetically it matched up well with my sti cluster (red numbers with black background):
http://www.ebay.com/itm/52mm-GLOWSHIFT-TINTED-7-COLOR-LED-TURBO-BOOST-VACUUM-GAUGE-METER-/290567907999?hash=item43a7326a9f&vxp=mtr


I do have an external WB O2 sensor that I can hook up and check the AFR. No codes as Electromotive does away with the OBDII. All factory as far as I know. When you say air lines, is that vacuum lines?
Right, I realized the electromotive ECU probably doesn't show engine codes, but I would definitely suggest measuring your AFR. It can help you diagnose a vacuum leak or boost leak. By air lines I mean vacuum and positive pressure - there's 3 hoses that connect to the stock boost control solenoid: 1. a vacuum hose from the turbo inlet, 2. a positive pressure hose from the compressor, and 3. an air hose that runs to the wastegate to provide the boost control.

Are you able to data log with the electromotive? That may also help shed some light.

[305mouse]

Have you checked for leaks? I had a boost leak at the tmic and my car wouldn't pull, couldn't get above 40 and just sounded like crap.

[Evan78]

Can you rev past 3500 with the clutch in?

[65 Cobra Dude]

Thanks guys, I'll give these ideas a try tomorrow. Evan, curious why you ask about the clutch in? I did not try that.

Thx,

Henry

[Evan78]

I wouldn't be able to tell you anything based on the result, but I was just curious if load had any relevance to the issue. I think if I couldn't rev over 3500 while driving, I'd probably pull over, clutch in, and see if I could rev beyond 3500.

[65 Cobra Dude]

Ah, got it Evan. I have not driven the car on the road yet, all in my garage. I'll try anything ;)

Thx,

Henry

[TrickyPete]

Funny thing is mine is reving at 3000 as well at idle and can't figure it out either. I have a N/A motor so I'm wondering if its not the turbo... Not trying to hijack your thread, I am curious to what you find out though. If I figure mine out I will post it.

[C.Plavan]

What Motor? Do you have a 2.5L Drive by wire?

[Bob_n_Cincy]

Henry,
Is it possible that the pulse wheel crank sensor clearance is a little to big.
At high speeds it doesn't pick up the pulses.
I have no experience with Electromotive systems. But this seems logical to me.
Bob

[Desertrunner]

Henry,
Cam angle sensor generally does that, it maybe not seated right or the signal is turning bad at revs. Also measure the out put of the TPS on the throttle to confirm its not bad.
Tony

[Scargo]

For the moment, all I can say is it is reminiscent of "limp mode" when the ECU doesn't see any boost pressure.

[Bob_n_Cincy]

Henry,
Are you getting any trouble codes.
You can find them on page 55-56 of this document.

http://www.electromotive-inc.com/wp-content/uploads/2015/01/Tec-S-HardwareManual-1.2.pdf

[65 Cobra Dude]

Thanks again guys. On my way out to garage to troubleshoot. Bob - thanks for the trouble codes, never saw that before. Will keep all posted.

Henry

[65 Cobra Dude]

OK, I looked at the boost and it never got above 2 lbs or so but I could not go over 3500 RPM. I hooked up a light per the electromotive manual and did not get any codes. I looked at the two lights on the ECU and they were colored as expected per the manual. I checked all my air and vacuum lines and did not see any issues but I am no expert here. It's a 2004 2.0 WRX motor with cable throttle. I am away through Sunday for Uncle Sam but will continue troubleshooting on Monday. Thinking about calling Electromotive directly. The car was running fine in the donor with the OEM computer so I am thinking it's the ECM. I have not checked AFR yet but can after the weekend. I just wanted to make sure I am not doing something stupid before bringing in the pros.

Any more ideas?

Thx,

Henry

[Slatt]

Sorry, it's very very late tonight for me and I might be way off here, but there's a tiny orifice 'plug' in the line from the manifold to the boost controller, ya? If it's plugged with grit you get excess boost(?) if it's missing then maybe you get???

[longislandwrx]

stock map sensor? have you tested it?



some more things, are you running a catalytic converter in the uppipe/downpipe? is it mega clogged? is your turbo smoking/making noise?

did you check your turbo inlet for leaks/cracks? get some logs from your sensors.

[STiPWRD]

You should be hitting well over 2 psi of boost by 3500 rpm, which location did you measure boost from? I'd suggest making the call to electromotive. Is the car idling funny? Also, how many miles were on the donor motor? Just curious, it could be a cracked hose somewhere.

[65 Cobra Dude]

Yes, stock Map and no cat. I am running Mike Everson's exhaust with just a muffler. I'll check the small orifice when I return. Yes, the car is idling funny (1500 rpm) and had 60k miles on the donor. No noise or smoke from turbo. Measuring boost from line that goes to over boost solenoid. Can't wait to get home and get this solved.

Thanks for all the inputs,

Henry

[305mouse]

This sounds just like my problem years ago. You need to check for leaks under pressure. We had a pvc pipe fitted with a cap on the end and a valve stem mounted in the end of that, all nice and airtight. You connect that to the TMIC between the TB and TMIC. Then connect an air compressor and add some air. Take a spray bottle with soapy water and start spraying all your hoses and connections while listening for leaks.

[longislandwrx]

what year donor?

[STiPWRD]

I also think it's an air leak if the car is idling funny. Measuring AFR will give you a quick answer if you're losing air somewhere you'll be running rich. I'd suggest measuring boost off of the intake manifold instead of lines going to the boost control solenoid.

Do you have any bends in the are intake before the MAF sensor? This will mess up the air measurement and give you a very rich idle. The air flow has to be straight and laminar at the MAF.

[Evan78]

Do you have any bends in the are intake before the MAF sensor? This will mess up the air measurement and give you a very rich idle. The air flow has to be straight and laminar at the MAF.He's using an Electromotive ECU without MAF sensor.

[STiPWRD]

He's using an Electromotive ECU without MAF sensor.

D'oh (smacks forehead) nevermind

[65 Cobra Dude]

Think I may have found the issue. Bear with me because I am probably using the wrong terms. The plastic intake tube that has the PCV hooked to it is leaking. It's connected to the 2 pipes that go in front of the intercooler. Would that cause my issue? Remember, there is no MAF because of the Electromotive system.

Thanks,

Henry

[STiPWRD]

Henry, can you take a picture and point out the tube you're talking about? Are you referring to the turbo inlet tube (about 2" in diameter)? The PCV valve is screwed into the intake manifold right below the throttle body and draws vacuum from the turbo inlet. Since you don't have a MAF, a leak in the inlet tube should not affect you. Have you heard back from Electromotive?

[Scargo]

Eenie meenie chili beanie, the spirits are about to speak. I need a picture because the description is not clear. Perhaps because I have been away from stock setups for so long?
As I suggested earlier, if you have a leak in the pressure side the stock system goes into limp mode, which is what it sounds like is happening. Is this leak you found after the turbo? Sounds like it if it is related to the intercooler.

[65 Cobra Dude]

Picture below:

41708

I did not call EM yet - was checking other suggestions first so I don't look stupid (er) :)

Thanks for all the help,

Henry

[STiPWRD]

Picture below:

41708

I did not call EM yet - was checking other suggestions first so I don't look stupid (er) :)

Thanks for all the help,

Henry
That is indeed the turbo inlet tube and since you're not running a MAF sensor it will not matter that the tube has a leak. There's no shame in calling EM, in fact they may have a quick 5 minute solution for you - I'm sure they get tech calls all the time. This just really sounds ECU related.

[65 Cobra Dude]

OK, latest update. I replaced the inlet tube with a silicone one just in case. Vacuum stays at 20 PSI and drops to 0 upon gunning it and when letting off, it goes to up to 23 and drops to 20 and stays rock steady. I also hooked up my wideband AFR meter and it reads 10 when idling but goes up to 15 when I gas it like it. Timing, compression and fuel pressure are fine. I checked all the vacuum lines and replaced a few. I called EM and spoke to one guy who said this other guy is the expert with these systems on the 818. The expert will call me back, hopefully tomorrow.

Thx,

Henry

[STiPWRD]

An AFR of 10 at idle is very very rich, this makes sense that the car would idle funny. Did you upgrade your injectors to larger ones by any chance? OR in the EM software, are your injectors scaled to a smaller size than they are rated to flow? Also, would you mind taking a picture of where you're measuring the boost from? In post 24 I wasn't sure what you meant by over boost solenoid. It could be that your measuring what you believe is boost from a line that doesn't actually carry positive boost pressure. I hope the expert guy helps.

[Evan78]

Cobra Dude, are you sure you didn't reverse those numbers? It would be much more typical to have an idle AFR of 15 and go rich (10) when on the throttle rather than the other way around as you described.

[65 Cobra Dude]

Sorry guys, Uncle Sam has been hoarding my time. I called EM on Tuesday and the expert was supposed to call me back. I called again on Friday and left a message. Hopefully I'll hear from him on Monday.

Sti,

First of all, thanks for staying engaged. Injectors are stock as far as I know and not sure if the EM program has them scaled smaller. I checked the boost again today and only got 2 lbs. I am tapping right off the outlet of the turbo. Picture below:
41964

Evan, I am sure I did not reverse the numbers (wish I did:>))

Any chance it could be the fuel pressure regulator giving too much fuel?

Thanks again for all the help,

Henry

[Scargo]

Lookit all dem hoses! You need clamps or zip-ties on all those connections.

[65 Cobra Dude]

Scargo,

I can clamp or zip tie them all. Are you saying this may be part of my issue?

Thx,

Henry

[Scargo]

Don't know. If they fit well and are fresh connections, with new hose, then I doubt it. It is common practice in racing to leave nothing to chance. Everything is exposed to extremes. This is another one of those "ask me how I know" comments. I lost a little hose on my boost controller. That equaled running like a scalded ape till the pistons broke their ring-lands. Oops!
Just because it is sucking, and not normally under pressure, doesn't mean it can't fall off. I watch the "big boys" zip-tie any hose stuck on a barb and make sure the are not flopping around.

[STiPWRD]

I am tapping right off the outlet of the turbo.

I think that's an ok spot to measure boost so you should be good there. If the injectors are stock, this circles me back to the ECU again. I'd suggest reading up all you can find on the EM ECU and tuning software. This will hopefully show how to look up injector scaling and other settings that are contributing to the rich condition and boost problem. Unfortunately I'm just not familiar with electromotive but keep at it and we'll figure it out. If you can figure out how to post a data log, that may also help. I have a feeling it's something related to the internal ECU settings.

[65 Cobra Dude]

I talked to the expert at EM today a couple of times. We updated the software and tried to update the firmware but the ECM would not communicate with my laptop. Bottom line is he asked me to send the ECM to him and he would check it for HW and SW issues and get it back to me in a couple of days. I shipped it to him this morning and they will have it Thursday. Hopefully they'll find an issue with it, fix it and send it back quickly. Will keep you up to date and thanks for keeping engaged.

Henry

[65 Cobra Dude]

Got the ECM back yesterday from EM and was really excited to hook it up. I called the expert to make sure I was doing everything right before hooking it up. Disconnected battery, hooked up ECM, connected battery, connected ECM to laptop, opened latest software and uploaded it to ECM. Started her up and the same exact thing happens, limits revving to 3500 RPM. Getting pretty frustrated. I am going to reach out to Wayne again for some help and then back to EM.

More to come,

Henry

[STiPWRD]

That sux man, stick with it, there's a light at the end of the tunnel. Sometimes it can help to have a second set of eyes look over your setup. Is there any tuning shops in your area - perhaps that specialize in Subarus? Or any knowledgeable friends that can come over? I hope Wayne can help. There's multiple ways to skin this cat but keep trouble shooting and marking things off the list. If it were me, I'd learn how to tune this ECM. Another option is to just scarp the whole thing and start fresh with a OEM ECU and harness.

[65 Cobra Dude]

Just came up with another idea, a real wild shot but does not hurt to ask - anyone fairly close to me in Port Charlotte, FL that has the EM system on a 2004 WRX donor 818 that I can plug my computer into to see if they have the same symptoms?

Henry

[Scargo]

As I've said before, it is extremely reminiscent of a boost leak/no boost. What if the sensor, or wiring to the pressure sensor were bad and it "thinks" it has no boost?

[65 Cobra Dude]

Thanks STi, I'll get it eventually and will look for a Subaru knowledgeable person in this area.

Scargo, Not sure if the Electromotive ECM is smart enough for that but I will ask Wayne. I am thinking the OEM ECU would look for that (If I am wrong, please tell me?). I'll put tie wraps on all the boost and vacuum lines that don't have clamps as you said above.

Thx,

Henry

[Scargo]

I'm positive Wayne if far more knowledgeable about it than me, but any system has to measure boost and regulate it. I am suggesting a bad MAP sensor or the wiring to the sensor from the ECU. I don't know if there's more to it than that. If you are using a stock MAP sensor perhaps you could borrow one for a few minutes. They can also be cleaned if they are fouled and if you have repositioned it by using an aftermarket CAI then it could be way off if it's in the wrong spot.

[Evan78]

Is it possible to post some audio of what it sounds like? Either audio only or a video w/ audio. Perhaps the experienced guys here will recognized something.

[longislandwrx]

stock map sensor? have you tested it?




again, did you test your MAP or log your MAP signal at the ecu?

[65 Cobra Dude]

Wohoo,

Thanks to Wayne, we have the solution. When using the Electromotive ECM, you do not have control of the TGV (Tumble Generator Valves). Upon startup in normal configuration, the TGV is closed which restricts the air. He had me pull the motor off which causes the TGV to go wide open which gives it the air it needs to breath. I then put the motor back on and it locks it in open position, maximizing air. It's very similar to an automatic choke which closes down air and gives fuel until it warms up. It then opens and gives it the air it needs once at temperature. I was able to bring it up to 6000 RPM and it sounded great. Still needs a little massage because it idles at 1250 RPM and has a hesitation before warming up which is expected but also has a hesitation when warm. I am working with Wayne to solve this issue. I am excited that this problem has been solved. Thank you to Electromotive for your help and thanks to Wayne for getting it solved. A special thanks to those on this forum that helped a Subie idiot (me) work on the issue.

Thank you all,

Henry

[Wayne Presley]

Glad I could help

[65 Cobra Dude]

Couldn't have done it without you Wayne. I never would have figured that out.

Thx,

Henry

[Scargo]

I guess Wayne has suggested doing the TGV delete? 4-10 horsepower gain minimums at various RPMs seems well documented. Not supposed to have those nasty things on the motor!
Glad he found the problem.

[305mouse]

TGV delete is well worth it, didn't realize you hadn't done it. Minimal work for better flow.