utter loss as to whats wrong.
 

having trouble diagnosing what could be wrong with my car.

87 TII with 87.3k on the clock. pulls 17-19in/hg of vac at idle warmed up.

problem started after i got it back from the shop recently. it runs pig rich at idle and vac, and perfect at transition, and then what feels like lean at WOT/boost.

i've checked the maf, map, tps(and adjusted it, was at 620ohm and now its sitting spot on at 1k and maxes out at 4.25k), plugs(those were replaced around 2k miles ago), wires, changed the coils, i've checked the IAT sensor(ordered a replacement, its supposed to read 3.5k ohm and mine is 30.6k ohm), o2 sensor was replaced not long ago(maybe 7-10k miles, so it should be good).

the only thing i can think of is before this all happened, i had a faint smell of gas when shifting off of redline at WOT, it would rapidly go away and wasnt there at idle or cruising so i doubt its pulsation dampener. oddly enough the smell is no longer there even when boosting.

could it still be the MAF or MAP sensor? or is it more likely a vac leak? or possibly a harness issue with the injectors?

when i'm crusing around in vac, it sounds and feels like the car is running on one rotor. i also checked each rotor just audibly with a plug pulled on front and rear, three strong pulses equally spaced in timing and equal in strength.

i'm at an utter loss.

here are pictures of each plug,

leading 1 https://i.imgur.com/EPekVXw.jpg

trailing 1 https://i.imgur.com/so4zQxE.jpg

leading 2 https://i.imgur.com/1PtUsvA.jpg

trailing 2 https://i.imgur.com/h4eZTqw.jpg

boost gauge showing vac at idle.

https://i.imgur.com/fjel2Fj.jpg

a vacuum leak would generally cause a lean idle and a rich boost AFR. the fuel pump wiring is usually an issue for lean in boost issues where the 2 step relay doesn't provide proper 11+v to the pump.

without a wideband you're just guessing, the plugs actually look just about right or maybe slightly rich which is better than lean.

Right so I'm checking the voltage to the pump, I think I'm doing it right, after letting the car run for a little, shutting it off. The four pin connector going to the pump gives me 10.9v on one connection.

Here's the thing, when I start the car the same pins only give me 2.3v when the car is running. Another set of pins pull a solid 8.8-9v.

I'll also say, the shop that drove my car drove it out on a pretty near empty tank of gas. You think it could be a clogged fuel filter?

That could have very well clogged up the fuel pump intake sock since it was sucking up all the garbage in the tank.

I work in the automotive industry as a technician and I please ask you not to bring us cars low on gas.

the wire you need to check for the voltage to the fuel pump is at the LR strut tower under the carpet, a 4 pin connector. the fuel pump is blue on the front harness side and black with a white stripe heading to the top of the fuel tank. the wire does change back to blue in color again once inside the tank to the pump. so if you're checking the blue wire between the LR speaker tower and the top of the tank it is incorrect.

the blue wire is for the fuel level sender which feeds a very low voltage 2-3v low amperage signal.

i should also correct myself, a vacuum leak before the turbo would result in every range of driving being lean. a leak after the turbo would result in lean out of boost and rich in boost conditions.

Yeah I'm having the opposite, rich in vac and what seems like lean in boost.

I disconnect the maf, car dies, disconnect the map, it runs as if it was never really there(voltages check out near spec of the dealership service book), IAT is bad but I ordered a new one of those, it'll be here on Thursday. I'll replace it then.

to me it sounds like a bum fuel injector. easy to swap the primary for the 2nds and see if it runs better.

you might want to invest in a wideband gauge just for peace of mind, it will also greatly aid you in diagnosing issues like this so as to take the guesswork out of things.

should be the first addition with any rotary :nod:
considering how detrimental afr is on these, and how cheap wb kits are nowadays

even easier is to de pin the injector wires at the ecu, and put the wiring going to the 2nd injectors in the primary holes... you can do one rotor at a time to tell which injector is gone bad.

had the same issue, and this was what i used.

fixed for you :)

well it is in my opinion anyways, i wouldn't really put a wideband onto an n/a car unless there was already a way to modify the injection duty cycle for tuning purposes. for turbo cars it is a tuning and diagnostic tool, for non turbo cars it would most likely just be a diagnostic tool.

Definitely a warranted repair to my statement! :) Yea I think on an na a wb would probably only be utilized after a stand alone or piggy back unit

Ah well, I have some 750's I've been holding onto for some time. Guess I can send them off to RC engineering to be cleaned and benched. Send my spare turbo ecu off for rtek 1.8 modifying and just put it all together. I've got a new fuel rail to go along with it all as well, and then I'm picking up a walbro pump and most likely an aeromotive pressure reg as well. Update the fuel system.

Have to put that all on hold as I'm trading my blob eye wrx wagon for an FD very soon...

first and foremost, what did the shop you had it at do before this started? do you feel that the problem could be a result of work they performed? if so, have you tried contacting them to make reparation?

how did you test the components? backprobing voltage at the ecm is much more accurate than testing resistance at the component. this way you're testing the entire circuit, up to the ecm, and it's live. you see what the ecm sees. if you want to see values while driving, a graphing meter would be your best option to view the signal. actually a scope would be best, but a graphing meter will suffice. how is the idle?
i see you're planning on installing a bunch of new parts without first resolving the issue...but still no wideband?!

Wideband would come after I trade off the wagon.

Actually I'm trying to solve this problem without spending money currently.

No this issue was peeking its head up before I brought the car in, but it only really started happening after they drove it with a low tank.

The issue really started happening about two months ago. I ran the tank dangerously low on the way back from somewhere highway cruising(the fuel light actually came on in fact), but its only really gotten real bad after they ran it low(which leads me to believe a fuel issue if anything.).

I'm not going to buy a wideband yet as it'd require spending 200-300 on labor to modify the downpipe on top of spending 300-400 on a good gauge and sensor. When I can just as easily have a local shop throw the car on their dunk with a far more accurate wideband any gauge will have for a fraction of the cost.(my car has no emissions, so tailpipe reading will be just as accurate as downpipe really)

i picked up an innovate kit for 150$. and that's cdn, so you'd probably get it cheaper. it comes with the bung. just need to drill and weld the bung on. shouldn't take more than 20 minutes. if this all started happening after running on e, and then amplified by doing it again, the pump would be my first suspicion. pumps use fuel to keep them cool and lubricated. have you removed the access panel and pulled it out to inspect? what you really want to see is amp draw when testing a pump. most pumps usually draw around 3amps. a resistance, ie blockage, to the pump will cause it to draw high amperage. if you don't have an ammeter, pull the pump out and have a look at the sock, and bottom of the tank. it's easy enough to do. if you already have the walbro, may as well install that now anyways.

I don't have a welder, and no shop around here will do it with it attached to the car. The bolts/nuts are also rusty so they'd have to cut them off to replace them.

No walbro yet, having a guy send it to me. Used, but only for a few months he claims.

No, sadly no ammeter sadly, wish I did. The screws holding the panel to the tank are also rusty and will probably break if I take them off. Any idea what size/thread they are?

ohhh yea. those screws like to do that lol. you might be surprised tho. mine were pretty rusty too but they actually all broke free. you can try using a hammer while twisting lightly. like an impact driver (assuming you don't have an impact driver). it may help shock them loose.
they won't weld the bung on vehicle? preposterous! it's not even hard to get at a good spot. right after the elbow, out the side, is fine. take it to someone who knows what they're doing. it is honestly a 20 minute job. and that includes setting it up on the hoist, prepping the surfaces, dragging the welder out and setting it up.

I should make it known that I have the injectors, and I have the fuel rail. I'd just need the regulator, lines, pump and ecu modified to handle them if it comes down to the problem being the injectors.

Max of what, 300-400 dollars in parts and I can do it all myself on a weekend with good weather.

I can honestly say*knock on wood* that I've not had a single major electrical problem with my FC. The only issues I have right now electrical wise is that my passenger window doesn't roll down with the controls on the passenger side, and the power mirrors don't work....but yet again that's not even electrical, that's a mechanical issue.

so you swapped your injectors( or at least the wires going to them at the ecu plgu) yet?

you gonna get off ur ass and do it so we can confirm these symtoms match the diagnoses for future people, are you just going to replace the whole shebang and claim that you "fixed it" yeah.. right.. and buying a car to replace it is "fixing the problem" too

you have an issue, solve your issue, you dont need a wideband, or pump or any other bullshit, just swap your injectors.

Well I'm replacing the intake air temp today(way too much resistance on it than what's acceptable). I'll pull the carpeting and hopefully not fuck up the ecu connector attempting to change the pins. I thought there was a tool you used to change them?

Here's my common sense question though, if only one injector was stuck, wouldn't the car run fine in boost if it was getting too much fuel at vac?

I'm going to pull the fuel pump as well today and take a look at the sock before going to the ecu.

Corrected.