ScoobyNet.com - Subaru Enthusiast Forum - 95 WRX RA hesitates & won't rev over 5.5K

ScoobyNet.com - Subaru Enthusiast Forum (https://www.scoobynet.com/)

- General Technical (https://www.scoobynet.com/general-technical-10/)

- - 95 WRX RA hesitates & won't rev over 5.5K (https://www.scoobynet.com/general-technical-10/946346-95-wrx-ra-hesitates-and-wont-rev-over-5-5k.html)

95 WRX RA hesitates & won't rev over 5.5K

This problem has been solved!

It turned out to be a bad earth to (or from) the ECU (B43 Pin 22 -Control Systems).

When logging with EVOScan I noticed that the ECU said the voltage was 12.24 at idle and 11.92 during the misfire.
I checked the voltage at the battery and it was 13.5
Then I ran a lead from the +ve battery to the ECU power. It didn't cure the problem, but the misfire was a little smoother, so I knew I was getting close.
Then I ran an earth wire from the battery to both B43 Pin 22 and Pin 11, and just like magic, the problem disappeared. Brilliant!
EVOScan now says the voltage is 13.2 at idle and drops to 12.8 at full chat. May need to check the main earth B44 Pin14.

I note that on page 80 of the manual, it says the first thing to do for many of these issues is to check the power to the ECU, including the earths! Doh!

Thanks to those who helped with this issue.

Here is the first post:

I'm new to Subarus and have read the forums loads to try and solve this problem, but alas, so far I haven't put my finger on it. Perhaps someone could have a look and point me in the right direction.

Car:
95 WRX Type RA JDM

Symptoms:

Won't rev over 6K in neutral if throttle is stamped down, but does go to 7.5K if pressed medium or slow.
Will not go over 0.6/0.7 Bar.
Will not rev over 5.5K in second gear, hesitates at this point.
There is power below 5K.

Fault codes:

31 Throttle Position Sensor Circuit
24 Open/Short in ISC Valve Circuit

What I've looked at:

MAF replaced.
TPS checked at ECU - 4.88V & 0.9V These values are okay.
MAP sensor (pressure sensor) replaced. Hasn't threw up code since.
3 Coils replaced with known low mileage ones, 4th is one of the ones that came with her. Has been swapped with other ones and made no difference.
All four coils now replaced with Newage ones.
Boost pressure tested - A few minor leaks, boost gradually drops.
Ignitor replaced. (Edit: Bypassed by Newage coils).
Waste gate opened at 0.5 Bar during pressure test.
Speedo is working (so lower rev limiter is not active).
Running higher ron petrol (currently on 97 + booster).
New spark plugs.
Timing checked.
Earths cleaned.
Compression test done - 150 on left side, 160 on right side. All okay.
Replaced dump valve with OE recirculating valve. (Edit: Reinstalled dump valve as OE valve was leaking).
Tried different ECU & dash (thanks Scubbay), but no difference.
Removed restrictor from turbo inlet.

Ideas:

Bypass wiring from ECU to where Newage coils taps into wiring.
Spoke to previous owner who said all this started when the manifold was changed from a standard one, to a four branch. Something to do with O2 sensor?
Looking at the logs below, I see that when reving in neutral the signal from the TPS jumps about from 4.7 to 0.8. It appears to rev out while telling the ECU the throttle is shut. This combined with the fault code 31 Throttle Position Sensor Circuit may be the issue.
I disconnected the TPS and the CEL came on, but she still reved up and splittered around 5.5K, so that would tell me it has nothing to do with the TPS.

I bought the EVOScan software and cable and have done a few runs in first gear.

The logs show a series of 3 runs:

1) First gear full throttle
2) First, second and third up a hill
3) Reving in neutral

Here is the link:

http://www.filedropper.com/secondruns Scroll down a little and press 'Download this file'.

Knock reaches a max of 40
Injector pulse width has values around 15 being shown under full throttle when the revs reach around 5K
Speed sensor/circuit is clearly working.
Is O2 sensor value normal at around 59?
AtmosphericPressure is constant as the altitude didn't change.
AirFuelCorrection is static at -128, maybe not being measured?
Reving in neutral shows that the ECU is being told the throttle is closed when it is actually full open. This sounds like a problem (thee problem?), but when the car is moving, the throttle sensor behaves okay. Edit: This is not the case - When the throttle is closed the value is 4.7, but 3/4 open the value is 4.04, but full open it is 0.8, so don't let that catch you out!

When driving, the car builds boost and pulls up to around 4.5/5K and then acts as if it is on a rev limiter, won't go past 6K.
In neutral it will not go past 5.5/6K if the throttle is whipped open, but will go to 7.5K if it is pressed slowly/medium speed.

Some values from the reving in neutral log:
http://i219.photobucket.com/albums/c...ginNeutral.jpg

get ready for it to go bang.....your running a jap import on 95 ron fuel when it needs 99 minimum.thats where your problem lies.the ecu is seeing loads of det and retarding timing and reducing boost as fuel crap for the car.empty tank and fill with tesco 99 momentum or shell v-power and reset ecu.
you cant use 95 ron fuel in that car unless its been mapped for it.

95 ron in an import....................
kabooooommmm

hopefully the ecu has saved you

Will an octane booster do the job, or is the 99 ron the only way?
I'm pretty sure we can get 97 ron around here, but not so sure about 99 ron.

If the ECU was swapped to a UK ECU, or a daugherboard installed that would be fine right? Or would it need at least 97 ron?

Quote:

Originally Posted by Stefen1 (Post 10748817)

tesco momentum or v power are both 99 ron.
maybe another option is to get it mapped for 97.
octane booster is ok but not ideal

no uk ecu swap will work.tesco do 99 everywhere.either that or get it mapped.octane booster is not worth it,just put the right fuel in it.

Quote:

Originally Posted by marcevs72 (Post 10748832)

no uk ecu swap will work.tesco do 99 everywhere.either that or get it mapped.octane booster is not worth it,just put the right fuel in it.

It looks like the original poster, like myself, is from Northern Ireland. The best fuel we get here is 97ron, tesco do not do 99ron here and shell sold all their stations and didnt do VPower here anyway.

I've just been talking to Tesco and they checked on http://www.tesco.com/momentum99/ to see if there are any nearby, but as Scubbay as said, there are none in Northern Ireland. The nearest one is in the south of Scotland, across the Irish Sea!

So I guess the only option to run as it is, is to use 97 ron plus octane booster.

I'll try this and report back.

Thanks to all for the info and help!

use 97 plus try to get a decent octane booster,try millers its expensive but raises ron level properly.best bet is to get it mapped with 2 maps one for 95 and one for 97 so covered whatever fuel is available.just take it easy now as will go bang very quickly.

Its came from Japan mapped for their fuel which is probably around 102ron, get it mapped ASAP!

Don't just think "ah it will be alright for now". Thats why theres so many Imprezas that have needed rebuilds.

You can have it mapped for whichever fuel you want, if i where you i'd go 97ron plus meth.

Seriously people stop with this rubbish about how Imprezas need 99 RON+ and he needs to get it mapped straight away.

Early WRXs will run perfectly fine on 97 (Japan has regular (89),hi octane (96) and premium (100+) and 100 wasn't released till nearly the end of the 90s or at least widely avaliable till the end of the 90s).

Also quite hard to tell what actual RON you get as last time I was over there they don't put the ratings on the pump just ,regular,hi octane and premium

Personally I would try to get some proper coil packs or at least inspect the ones you have as they are known to crack at the ends,check them carefully or you may actually crack them.

Yeah and theres not that many Subarus out there needed rebuilds because of it so yeah, no big deal. Not point getting a remap, just fit an induction kit, chav valve, 95ron it will be fine no matter what people say.

he is right to a point though as there wasnt 99+ ron fuel when the early sccobs were made so will run on 97...BUT once over here with a few goodies bolted on etc the chance of det is high so using better fuel is just sensible.plus he needs to check what ecu he has as my 93wrx has the later z4 ecu with better fuelling map and will be fine on 97 but i still run it on 99.

Quote:

Originally Posted by Adam-M (Post 10749290)

Oh grow up
Neither mod was mentioned and I didn't say 95 ron was good enough.
People who perpetuate the fuel myth are not helping and in this case could cost this bloke a lot of money if he did take the 'It needs a remap NOW!!!' advice.

To the OP there is no need to rev so high in any of the gears,so your power lose is no surprise.
Yes the fuel may be causing an issue when you're giving it WOT and full boost so do as marcevs72 advised ,fill up with 97 rest ecu,may take a couple of tanks.

The car came with a full stainless steel exhaust, and a dump valve.
It has a TD05 turbo and runs with a Z5 ECU.
There is a standard airbox and no sign of a MBC so I believe it is (rather, should be) running standard boost.
I understand that the Z5 has a more aggressive map than the Z4, so it most definitely needs at least 97 ron, if not more.

I wouldn't normally rev as high, but I was trying to get it to log fault codes to get to the bottom of this. Not a good idea with 95 ron :nono:

I'll pump the old fuel out and get some super unleaded and booster tomorrow. Thanks again for the info.

Remove the dumpvalve and replace with an OE recirculating valve (any subaru breakers should have them).

yep replace the dump valve with standard re-circ one.the car will run a lot smoother and will stop it over fuelling as well.first thing i did on mine and i actually like the chav noise but better that car runs properly.

I ran the car until it was dry, and put £15 of Super Unleaded in it, along with the correct amount of Lucas Octane Booster. The battery was then disconnected for a couple of hours to reset the ECU. The performance is the same though, it hasn't made a difference.

She also threw up this fault code again - 24 Open/Short in ISC Valve Circuit.

On the advice of a fellow member, I checked the exhaust manifold flanges. Around the V-band is also sooty:

http://i219.photobucket.com/albums/c...h/Untitled.jpg

The turbo has a restrictor on the inlet:

http://i219.photobucket.com/albums/c...h/IMG_9577.jpg

So perhaps there is essentially a small blow out and the lambda sensor is feeding the ECU incorrect data? Would this cause the listed symptoms?
It would have to be really big blow out, not to be able to spool the turbo though, right?.

I also need to pull all the spark plugs and check or replace them. There is hopefully another set of coils on the way, so those two jobs will be done at the one time. Is there a particular grade of plug that is best?

Haven't managed to source a recirc dump valve yet, but one will turn up.

NGK PRF6B or PRF7B are both good.
Gapped to 0.7mm.

Plugs changed to NGK PRF6B
Coils changed.
First post updated.

We had a similar misfire on our RA, changed coils etc, finally fitted newage coils doing away with the igniter and its ran 2 years since without issue.
If your signal wire is broken to the ecu the limp mode will kick in regardless of speedo working or not. Seen a few jdm cars where the wiring was altered to try and remove the limiter. Worth tracing from dashclock to ecu. I have spare clocks and ecu if you wish to borrow them.
Its also worth opening your ecu to check a daughterboard or chip hasn't been fitted which can also cause problems.

That restrictor on the turbo is unusual, the vacum to actuator should be restricted not the turbo induction.

Good idea - I'll check that speed sensor wire and report back.
Thanks for the offer of the ECU/dash.
I checked the ECU tonight and there is no daughterboard present.
I'll try this test loom as well to try and sort the issues. If that doesn't work I'll probably have to get a set of newage coils.

The restrictor - I did wonder about that, but the turbo inlet is made for the restrictor. It may affect the flow at high boost, but I don't think it's part of this current problem.

Checked the signal wire from the dash to the ECU with a multimeter. There seems to be a good connection.
These early Imprezas have a cable operated speedo, which if I understand correctly uses the dash to generate the speed signal for the ECU. This signal is made up of a series of 0V and 5V pulses. I guess I'd probably need an oscilloscope to check this?

There does not seem to be anything in place (either in the loom, cable or dash) to adjust the dash to MPH, but it does display in MPH. How can this be you may ask.... Today I discovered that rather than the close ratio box that the Type RA should have, it has a gearbox (and diff I hope!) from a 98 UK impreza, which has been fitted with the cable speedo. So now the cable is spun at a MPH rate.
I only checked the code as I noticed there was a connector on the top of the gearbox that wasn't plugged in, and it didn't fit into the corresponding plug on the loom.
So there are two sensors on the gearbox not being fed into the car's electrics - reverse switch and I can't think what the other one is...

Questions:
The ECU is now getting the speed reading in MPH rather than KPH, is this a problem in itself?
How can I test the ECU is actually getting the signal at all? Swap dashes? Multimeter?

There should be a geared adaptor probably between the short cable from the gearbox and the longer cable to the dashclocks. This wont be your problem. Either that or its between the cable and speedo. Regardless its the speedometer thats calibrated differently not the gearbox speedo drive. Have you identified the speed sensor wire at the ecu?
Where are you located? I may be able to direct you to your nearest specialist with the correct software to identify the issue i.e select monitor. pm if you prefer.

Updated first post with new info and logging data.

your clocks simply have a 110mph facia fitted.

First post updated with solution. :-)