Jump to content

TPS/ISC Reset 88 TSI


SCillini
 Share

Recommended Posts

I am working through a poor idle/ shutoff condition on my Conquest. I verified the MAF is working tonight and was going to the next step of resetting the TPS/ISC.

 

A bit of history, this engine was rebuild prior to me owning the car, and I purchased it as "non running" condition. The new engine had never seen combustion until I fixed the one tooth off timing issue. Now it will start, run for a few seconds and die. I can tease it to stay running by playing with the throttle plate, but eventually the issue will catch up to me and the engine will finally stop. Initial suspect was the MAF, but after testing the ambient temp sensor good, and verifying when the MAF is unplugged the engine ran worse and literally flooded the intake with fuel so the ECU is getting a signal, the next stop would be to verify the TPS and ISC motor are working ok.

 

I found the reset instructions for this on the forum but I cannot run the car long enough to get it to "operating temperature". Do I just struggle with the engine until it seems hot enough? Is there a method to use without the engine at temp?

 

Anyone have any other suggestions on the idle issue?

 

Thanks in advance for your assistance.

Link to comment
Share on other sites

You can test the TPS and ISC without a running engine. You'll need a multimeter (analog needle-pointer type is easier for this test but a digital one will work). Unplug the connector at the TPS. Connect the multimeter to any 2 of the 3 wires, set the meter to measure resistance on a scale that can handle 5000 ohms or so. (often the "20K" setting on digital meters) You should see some resistance as soon as the mter is connected. Now SLOWLY move the throttle linkage, opening the throttle, as you watch the meter. The resistance should smoothly change (it could increase or decrease depending on which wires you picked); if it holds constant move one meter probe to the other TPS and re-try. As long as you get smooth motion (no drop-outs or jerkiness in the resistance vs. throttle position) the TPS is okay.

 

For the ISC, it's really three pieces: a small electric motor, another position sensor (the MPS) that you'll test with the ohmmeter, and the "idle" or "nose" switch. Unplug both connectors at the ISC assembly. One will have wires going to the motor itself (which is inside the black plastic/rubber housing)... the other is the MPS. Hook the ohmmeter to 2 of the MPS wires - the two that match up to the green+red and blue+red wires on the harness side of the connectors. You'll need a source of 6volts; those plastic "AA" battery holders work great for this. Do not use 12 volts or the car battery unless you want to know for sure your ISC is bad as 12 volts will fry the motor and jam/deform the internal mechanism. With the ohmmeter connected and ready to read, hook the battery pack to the motor wires - they're the wires that line up with the yellow+blue and yellow+black wires on the harness side of the connectors. The little rubber bellows pointing downwards from the ISC should extend or retract and you'll see the throttle linkage move a little bit. Watch for the motion to be smooth, no signs of jamming or other mechanical maladies as it moves. When it stops moving, reverse the battery wires and watch the ISC move in the other direction. Do this a few times to make sure the ISC motor and gearbox work smoothly. Then repeat the polarity swaps while you watch the ohmmeter for smoothly changing resistance... back and forth again with the polarity.

 

Last test: the idle/nose switch. Hook one ohmmeter lead to the throttle body or a good ground. Hook the other meter lead to the third pin in the ISC connector - the pin you didn't hook batteries to. With the throttle closed you should read pretty close to zero ohms. Turn the throttle linkage a bit to simulate pressing the gas pedal; the meter should jump to infinite ohms or open circuit quickly. Release the throttle slowly and verify the resistance returns to zero ohms. If you get infinite ohms at all times, look at the base of the throttle body from the exhaust side of the engine - look over the engine - for what looks like a carb style idle adjust screw. Turn that screw 1 full turn counterclockwise and watch the throttle linkage for any signs of motion... if the linkage moved (the screw was turned in too far) check the meter... if you now have zero ohms then this "fixed SAS" screw was mis-adjusted and was preventing the nose switch from functioning. That switch tells the ECU when the throttle plates are at the idle position... if it always reads infinite ohms then the ECU never goes into idle mode which makes for a lousy idling engine.

 

Standing at the passenger side fender again, look across the motor to the throttle linkage at the rear of the throttle body. In the linkage you'll see a hex-headed bolt with a screwdriver slot (about 8mm bolt head if I remember correctly). This is the one you adjust during the ISC/MPS idle adjustment procedure outlined in the FAQ. You can crank this puppy clockwise 1 or 2 full turns to force the idle RPMs up a bit while trying to warm up the engine or do other debugging.

 

You can also do a modified version of the FAQ procedure for a stone-cold engine to ballpark the TPS and ISC/idle adjustments:

1: do this test with a stone-cold engine - no heat in it at all. If you have a source of electronic components (some Radio Shacks, Frys, etc) buy a 4700 ohm resistor (any wattage - get the cheapest resistor you can find) and hook that into the connector instead of the ECU's coolant temp sensor which is the sensor in the intake manifold by the oil dipstick. This resistance simulates a temp sensor reading on a stone-cold engine so it'll stay constant as the engine begins to warm up.

2: assume the ECU "zips" the ISC to the cold-engine/fast-idle RPMs of around 1500 RPM or so... so tune the RPMs, with that idle adjust hex-headed bolt, to get about 1500 RPMs.

3: Set the TPS voltage to something above 0.50 volts since that corresponds to normal idle RPM throttle position; I'd guess around 0.8 volts.

4: check your ignition timing while the resistor from step #1 is still connected. It should be a tad more advanced than 15 deg BTDC. Normal "warmed up" engine timing is 10 deg, on a cold engine it should be 15 deg at normal idle RPMs. At fast idle it might be a little more advanced but not much... 15 deg is a good target value while trying to get timing & idle adjustments ballpark correct. You should have a smooth running engine at this point that revs to 2500 easily and seems "happy." If not... your car has other problems that need debugging - the FAQ idle procedure isn't gonna fix it.

 

Once ballparked and the engine gets warmed up, shut the engine off, remove that resistor and re-connect the temp sensor, and repeat the FAQ idle adjust procedure to proper specs.

 

mike c.

  • Like 1
Link to comment
Share on other sites

Mike C I am in your debt! What an awesome write up and thank you very much for your time to write it up. I know I am late to the "post game show" on these starquests and there isn't much out there anymore.

 

Will let you know the outcome.

 

Thank you again!!

Link to comment
Share on other sites

  • 2 weeks later...

TPS tests Good. Fuel Pressure at 36PSI. I am testing the ISC now. When I test the MPS and connect the ohm meter what should it read? I have a 6.42volt [4 AA batteries] pack I made from batteries, wire and electrical tape for the power supply. I dont see any movement from the MPS. Could I remove the unit and see if its actually trying to move but isn't strong enough? Could it be bound? Rebuilding an option?

 

I have a sense this may be my issue or the majority of it.

 

Thanks again.

Link to comment
Share on other sites

The MPS is inside the ISC assembly. Are you saying you can't see the ISC end (points downward, has a rubber bellows on the tip of it) move up/down a few millimeters when you use the batteries? If so, then I'd remove the batteries and use the ohmmeter to look for continuity through the motor wires (where you had the battery connected). It should show some continuity, probably around 100 to 200 ohms I'd guess. If it's over 1000 ohms then the motor itself is likely kaputt.

 

Otherwise, unto the screws holding the ISC/MPS bracket assembly and remove it from the vehicle. With it off the car, look at the screws in the top of the bracket - you'll see how they hold the motor/gearbox assembly to the bracket. Remove those screws and then CAREFULLY separate the motor/gearbox piece from the bracket. There is a rubber gasket in there - that's why you need to be careful... don't break it. Also, under the small cylinder "dome" in the bracket is where the MPS lives. It has a spring made of thin metal that'll fall out as you separate the motor/gearbox from the bracket. Don't loose it. Once the motor/gearbox and bracket are separate, you'll see:

* the little black box that is the MPS. A couple teeny screws hold it down... remove them and pull it straight out. It has a skinny metal rod sticking out the other end; that's what senses the gearbox position. With the ohmmeter connected to 2 of the 3 MPS wires, slowly push this sensor rod in/out, watching for smoothly changing resistance. If you get a constant resistance reading, move one ohmmeter wire to the other MPS wire and try again. I forget the exact resistance range but it's not really critical - what matters is a smooth change in proportion to the sensor rod position. Probably varies from several hundred ohms up to 3500 to 5000 ohms I'd guess. Been a long time...

 

* the electric motor inside the black cover, ending in a drive gear.

 

* the driven gear that, when turned, makes the ISC gearbox extend/retract. This has been known to jam from old/dried out grease and from damage at the "tip" end where the ISC assembly touches the throttle mechanism. At the bottom of the gearbox/shaft there is a small and thin washer to support stuff... this washer gets mangled if excessive force is applied to the ISC gearbox or if somebody feeds the electric motor with car battery voltages.

 

* the tip, at the end of the bellows, is a teeny pushbutton (called the "nose switch" or "idle switch") that serves two functions. First, it's the piece that pushes on the throttle linkage to move the linkage. Second: it is an electrical contact, sensing when it is actually touching the throttle linkage. When the nose switch tip is touching the throttle linkage (meaning your foot isn't pushing the gas pedal down which in turn moves the throttle linkage away from the nose switch) the resistance from the 3rd wire on the motor connector to chassis ground should be zero ohms, telling the ECU "engine is at idle. Adjust the ISC to get the right RPMs, change the injector firing logic to 'idle' mode." With the ISC off the car, you can press this button by hand... it should take only a teeny force to move it. It moves about a millimeter.

 

Some folks have replaced the electric motor with generic hobby shop type motors with good luck. 3 to 6 volt motors. Years ago somebody had good luck with a Radio Shack motor of all things...

 

mike c.

Link to comment
Share on other sites

Thanks mike again. I am going to go through that ISC carefully this weekend. I also found the chart for the ECI checker tool in the FSM - fuel section 14-55/56. I have the ECI tool connected and get readings on a lot of the channels, just didn't know which indicated which sensor. I plan to see if the ECU [via the ECI checker] is getting signal from those sensors. I question the harness as well which is why I got the tool to begin with. If the ECU is getting info, I know the harness is at least good.

 

The only other issue I have is the Tachometer in the car. The electrical diagram shows it to be a direct white wire with a black stripe that goes from the negative terminal on the coil to the tach in the cluster. My issue is the tach doesn't work, and there is NO white wire with black stripe connected to the coil negative, only a blue wire with a white stripe. I found that to be for the ECU engine speed information. Have you seen any issues where there may be a deviation from the FSM on the wiring? Otherwise I feel I may have to tear into the harness to find the wire, but around there the harness looks intact and unmolested. I can post this in another thread if that would make more sense.

 

Robert.

Edited by SCillini
Link to comment
Share on other sites

The tach signal goes from the ignition coil to the ECU and from there to the tach. The wire from the ECU to the tach is a shielded skinny coax wire - basically skinny antenna wire. It is a white "core" wire surrounded by a braided mesh shield and an outer brown or black insulation layer. You'll see two such shielded wires at the ECU. One is the tach signal, the other is the signal from the oxygen sensor.

 

I think the tach to ECU wire is also a shielded wire.

 

mike c.

Link to comment
Share on other sites

Great! it would seem at some point that wire was removed from the ignition coil circuit. I was reading another thread you were commenting on "Trying to get fuel pump running before DARK" in which you stated the fuel pump is controlled by a dual coil control box. When the car starts one coil gets +12 volts from the start switch for the pump and waits for the ECU to send the next +12 volts after the ECU verified ignition pulse from the engine. If this signal is not received in time, the car surmises a wreck situation and shuts down the fuel pump to prevent possible fuel spill. The symptoms experiencing strongly suggest this shutoff scenario may be a big contributing factor.

 

Additionally, you noted in the same thread the engine's COLD base timing really should be 15 degrees BTDC instead of 10 degrees BTDC because the CTS indicates at start a cold engine and the ECU adds 5 degrees advance to maintain the 10 degrees BTDC. Given that, should I actually have needed to set the timing at 15 degrees instead of 10 degrees as the FSM states? If so, my distributor may STILL be one tooth off.

 

On the ECI checker tool, according to page 14-55, the ignition pulse is montiored on terminal 1, select switch A, check switch 2 with a variation voltage from 4v to 10v. What I am now trying to find [in association with verifying the ISC function] is what wire sends the signal for the tach from the ECU. The wiring diagram as I stated indicates a direct wire to the gauge, but from what you describe that may be incorrect and the wire actually comes from the ECU. But which one and what is its color?

 

That is the conundrum I am working on now. :) And before long mike C, I am going to owe you a real bill! lol! I haven't had this much fun since I worked on my 66 mustang as a teen.

Link to comment
Share on other sites

I got through the ISC assembly today and it all checked out within spec. I heard the MPS move in and out and verified the resistance changed smoothly when the motor moved in and out. Therefore the ISC and TPS and MAF are all eliminated as individual problems. The only thing I have to verify now is the engine impulse from the coil.

 

Mike C - I looked at the coil again and I do not see the wire as you described. What I DO have coming off of the negative post is a solid black wire which goes to a small golden box [looks like a resistor or filter of some kind but it has a mitsubishi label on it so I would presume factory] which turns into a solid white wire out of that. Then it travels to a connector where the white wire turns into a white base wire with a black stripe. This wire is like 18 gauge whereas the white wire itself is probably 16 or 14 gauge. Nothing I see is shielded or exemplifying a shielded wire is connected to the coil.

 

I will also check timing tomorrow as well.

Link to comment
Share on other sites

Well today didn't go as planned, but thats life sometimes. Here is a pic of the coil I was talking about yesterday and the absence of a shielded wire. Am I missing something here? I know the pic is large, but I wanted to make sure the wiring colors were easily visible as well as where the wires appear to be going.

http://i284.photobucket.com/albums/ll2/desktopjockey/conquest%20tsi/20141220_135426.jpg

Link to comment
Share on other sites

Thank you thank you thank you for your verification. This means the PO screwed something in the dash and possibly cut something going to the ecu for impulse signal. :/ time to stand on my head and follow wires.

 

I havent been able to find what that gold box is in the fsm. I am presuming its a resistor of some sort. Other than testing continuity anything else i can do to verify that is ok? Ideas on purpose other than as a resistor?

Link to comment
Share on other sites

OK, so I have verified to the best of my ability that the timing chain is dead on accurate, the timing is set to 15 degrees mechanically with the distributor. I verified the ECU is working, its receiving signal from theMAF, TPS and ISC and have verified mechanically [including performing the TPS/ISC reset procedure] that the ISC is working and moving as it should. Signal readings are within normal expectations according to the FSM and the ECI Checker tool as well so I know the harness is good from the mixer. I checked the impulse signal via the ECI Checker tool, and it reads 12v where the manual reports it should be reading 4 - 10 volts. When the engine is running, the impulse signal drops to 10 volts.

 

I also pulled the ECU codes from the self diagnostic port and found its reporting the O2 sensor and no other codes.

 

The PO said they had upgraded the O2 sensor to the three wire sensor, and I have been searching for the upgrade steps so I can follow what should have been done, but I cannot find it. I am also wondering if this would have anything to do with the impulse signal as well and the fuel pump controller. I plan on pulling the fuel pump controller tomorrow and testing it. Everything else being equal this thing should be running. The tach on the dash isn't working either but I've been told it runs off of the ECU impulse signal so if I fix the ECU signal this should resolve the tach problem, or no? Frustrations are running elevated for sure.

 

Anyone point me in the direction of the O2 upgrade instructions or have other ideas?

  • Like 1
Link to comment
Share on other sites

A 3-wire O2 sensor is basically a plain O2 sensor combined with a heater. O2 sensors have to be quite hot to function - it takes a mile or two of driving for exhaust heat to warm factory style 1-wire sensors enough. Heated O2 sensors warm up quicker, allowing the ECU to enter "closed loop" mode faster resulting in clean emissions faster (compared to stock style O2 sensors). Three wire sensors have two wires for the heater and one signal wire. Typically the heater wires are the same color. One wire goes to a good ground, the other to a source of switched +12volts. The oddball colored wire goes to the ECU. Until the O2 sensor warms up and outputs a minimum voltage, the StarQuest ECU runs "open loop" and intentionally runs a bit rich for safety. A StarQuest should drive & idle cleanly regardless of the O2 sensor functioning or not unless the sensor is outputting 0.7 to 1.0 volt (i.e. saying "rich mixture") which is what the ECU uses to judge "O2 sensor seems to be warmed up and working... switch to closed-loop mode when not idling." A StarQuest with a dead O2 sensor just runs rich all the time, getting lower MPG and spewing higher emissions out the tailpipe. But it should run well. Also note the StarQuest ECU goes to "open loop" whenever the engine RPMs are increasing significantly - i.e. acceleration is "open loop." And it's normal/proper/expected for the ECU to throw the "O2 sensor" error code until the O2 sensor warms up... then the ECU will clear that code. That's the only code the ECU will self-clear by the way; all other ECU error codes require a battery disconnect to clear them.

 

A StarQuest does not use the O2 sensor when idling - idle is "open loop" to compensate for the dis-similar intake manifold tube lengths. StarQuest ECUs intentionally run the air:fuel ratio a bit rich at idle to get a more stable idle; the ECU turns on the secondary air filter system (the black box on the top/front of the valve cover that makes the sucking/gurgling sound) to let fresh air into the exhaust to cancel out the over-rich idle exhaust products. Many folks remove the secondary air filter assembly... that just means their idle emissions will be higher and the cat converters may not live as long. Removing the secondary air filter doesn't affect how well the engine runs/idles.

 

The ECU does not feed the tach... the wire from the ignition coil to the ECU continues on to the tach. So generally either both or neither see the ignition pulses. Voltages at the fuel injectors depend on how much fuel the ECU needs to command. The injectors are supplied with +12volts and the ECU completes the ground to fire the injectors. The ECI checker will read 12 volts (actually battery charging voltage) if the ECU isn't using that injector - which is the case on the 87-later StarQuest secondary injector while the engine idles. As the ECU commands more fuel, the ECI checker will read less and less voltage on the injectors. What is really happening is the ECU is commanding the injectors ON and OFF. When the injector is OFF, the ECI checker will read full 12 volts. When the injector is ON, the voltage is much lower - just a few volts (thanks to the injector ballast resistors). This ON-OFF-ON-OFF pulsing happens several times per second - MUCH faster than the ECI checker can respond... so the ECI checker reports the "average" voltage. You should see the voltage change while revving the engine - on either injector.

 

Fuel pump controller?? No such animal... unless you mean the "ECI relay." That relay is turned ON by the ignition key START position and then held ON by the ECU as long as the ECU sees ignition system pulses. Without those pulses, the ECU figures the engine stalled... perhaps from an accident... so it kills power to the relay to kill power to the fuel pump. Don't wanna pump gas out potentially busted fuel lines!

 

There's nothing worse on a car than undocumented changes... especially somebody that hacks the wiring for no good reason.

 

mike c.

Link to comment
Share on other sites

I agree on the undocumented changes issue. I am certainly going through this with this car. The engine was completely rebuilt yet the PO apparently has NO CLUE on what was done to it. I am leaving out that the entire interior was replaced to black from the original maroon. Issues there are for other threads. I am beginning to think the owner before me actually bought the car with the changes already performed, couldn't get it running themselves so flipped it. I have asked on several occasions about the specs on the engine and I cannot get a straight answer. That is part of my frustration. At least so far everything I have checked and followed through on has seemed ok, I just haven't come across my real issue . . . yet. They rebuilt the instrument cluster so I am hoping from your description of the impulse signal that since the ECU gets the same signal the Tach does, I can fault the gauge [not exactly good news either]. I love how the FSM wiring diagram shows the signal going straight to the ECU from the coil with no breaks or junctions. :/

 

ECI relay is what I was talking about, not Fuel pump relay. I am concerned whats happening is the fuel is being shut off or part of the relay may be bad. The engine runs smooth for several seconds then just shuts down; it sounds like either the ignition just cuts off or the fuel just cuts off because the engine "coasts" a couple revolutions before stopping. This is what is leading me to think the control relay on startup isn't switching to the relay controlled by the ECU. I can keep it running though if I "toy" with the accelerator however the engine starts to run rough and pop [not backfire] - apparently because the injectors are pumping more fuel than the engine needs in response to the accelerator. Just thought of this after thinking it through - what about vacuum advance problem? I will test the vacuum advance to see if its working and strike it off the "unknown condition list" although the issue remains whether the advance is plugged or connected. I know I am using the correct vacuum port, of the three at the base of the mixer, I am using "D" or the one towards the front of the engine. All others are closed off except the bottom port used for the waste gate and the HVAC system.

 

Understood on the injectors, the ECI Checker does reduce voltage within the expected range while the engine runs and I verified 12 volts at each injector connector.

 

Thanks for the wake up call on the O2 sensor. If I had thought the sensor's operation through I would have remembered the three wire is just a built in heater. So many theories and unknowns with this car info gets scrambled.

 

Mike C thank you for talking me through this. If I can ever repay your assistance I will do my best.

 

Robert.

Link to comment
Share on other sites

Look for a couple unplugged wires hanging from the wiring harness bundle squeezed between the air filter canister and fender. There should be a thicker than average black+white wire, and a 2-wire connector. The black+white wire is the fuel pump test terminal, it's also the one folks jumper to the battery + post to force the pump to run. Hook a voltmeter, mechanic's test light, or even a small light bulb (one of the side marker bulbs) to this. While trying to START you should see +12 volts/the bulb should light. When you release the key, the +12volts should remain. See which happens first: loss of voltage or engine stalls. If the voltage drops while the engine is running then the ECU isn't seeing the ignition pulses. If the engine stalls and then the voltage drops, the engine isn't being killed by a missing ignition pulse signal.

 

Air leaks anywhere in the plumbing will screw up the engine - anything that makes the airflow sensor (MAS aka AFS) read incorrectly. The MAS is the #1 input to the ECU especially for idle. It takes only a puny air leak to lead to a lousy idle. Bad MAS units have been found before... and most MAS failures do NOT lead to an ECU error code. Plus, the MAS unit on 87-later cars is actually 3 different sensors... only one of which is even checked by the ECU for functionality. The ECU error code logic is pretty lame actually... it can't detect anything but total failures of sensors. If you have access to an oscilloscope that is the best way to test a MAS other than replacing it with a known good unit. Take a look at the wiring connector at the MAS itself... does it look like it has been altered in any way? Like maybe a 7th pin was chopped off to fit the StarQuest's 6-pin connector? If so, somebody has installed a MAS from a Eclipse/Laser/Talon car. They can work... but they often don't work quite right around idle on StarQuests as they "read" the airflow volume differently.

 

Look closely at the fuel injectors for faintly stamped lettering in them. Or are they aftermarket? Let's make sure your car has the right sized injectors. Folks that want to "crank up the boost" hear that they need to increase the fuel volume to match or else they'll lean out the engine and blow it up. So they make stupid mods that crank up the fuel flow ALL THE TIME instead of only when it is actually needed. Swapping the black primary injector for another secondary, or a black injector from earlier model year cars, is a common screw-up. Actually, buying a brand new primary injector from Mitsu can cause this too - Japan was sending 850 injectors as replacements for the 580ish 87-later primary injector. BAD BAD BAD. The FAQ has a note about injector flow rates and the letter codes (faintly) stamped into injectors to identify them. 87-later cars should have "L" and "M" code injectors. And the 580cc black "L" injector is the only one that needs to be plugged in to get the car to idle. If the car idles with the green "M" injector plugged in, the wiring to the two injectors is reversed.

 

Do you know if any PO tried to add an air/fuel gauge? The most common place to tap into the O2 sensor signal is at the ECU. That's a shielded wire... and it looks JUST LIKE the ignition pulse wire. So wiring there could be swapped, or the braided shield of the wire got shorted to the internal signal wire (the O2 sensor and ignition pulse wires are both shielded wires at the ECU - basically skinny coax cable like skinny cable TV wire). Look for signs of wire hacking at the ECU connectors.

 

I have seen errors in the factory service manual wiring diagrams too. For example, the air conditioning diagram shows a transistor that is supposed to control the compressor clutch... What the diagram shows is that transistor shorting the +12volt clutch wire directly to ground. I seriously doubt that's how it functions! It'd cook the transistor and blow fuses all the time. A screw-up in wire colors at the ignition coil wouldn't surprise me either. Actually, the ECU/tach shielded wire from the ignition coil "-" post doesn't actually come directly from the ignition coil... it comes from the ignitor module. The wire from the ignitor T's inside the harness someplace; one side of the "T" is the blue+white wire and the other is the "white" shielded wire that goes to the ECU and then to the tach. The input to the "T" is also a blue+white wire going to the ignitor. Has the wiring near the ignitor been hacked at all? Somebody try to install the MSD ignition system maybe? Undo the harness wire cover around the ignitor module a few inches... I don't know how far down the "T" might be as I've never had a need to find the "T" junction on my cars. Do other gauges and idiot (warning) lights on the dash work? Interior fuse #13 feeds the tach and the idiot lights. There is a black ground wire coming from the back of the instrument panel that attaches to ground #8 in the car... that's behind the instrument panel. If the dash voltmeter is reasonably accurate then the ground and fuse 13 are probably correct. The tach needs both +12volts and a good ground to function.

 

mike c.

  • Like 1
Link to comment
Share on other sites

  • 4 weeks later...

OK, Now that its above 32 degrees and the holidays are over, I had a chance to work on the car.

 

I connected a voltmeter to the single wire for the fuel pump as instructed. When I connected the lead to the positive wire, I immediately got 12.83 volts, but the fuel pump did not run. I had an assistant start the vehicle while I watched the voltmeter and I found that the fuel pressure would prime to 38psi [i have a pressure tester installed on the mixer] as it should. Interestingly though is that when the car is starting, the voltage through that port dropped to 0.45 volts and stayed that way till the engine died. Then it would return to 12.83 volts.

 

I take it by your instruction that the engine is not seeing the ignition pulses. That is contrary to what I observed at the ECU terminal with the ECI checker tool. The voltage would be 12 volts and when the engine runs, it would drop to about 8 volts before the engine would stall.

 

I checked the injectors and could not see any stampings, however the passenger side is green and the driver side is black so I imagine those are the original injectors.

 

I understand about the MAS sensor, the harness connector looks original minus the locking tab which it appears a PO broke off when they installed the Talon MAS. I have the original MAS reinstalled on the car and the ECI checker does see a signal going to the ECU. I can only verify the MAS as far as the FSM and the empirical testing allow me to see. I have no other ways to test the MAS/swap out with a known good one. At this point I can only say I am 90% sure its ok.

 

As far as air leaks, I have closed every single port on the mixer except the distributor port.

 

I have a serious feeling I am going to be pulling the dashboard to find a broken wire. :'(

 

Thanks again for all your help. Hope you had a good holiday.

Link to comment
Share on other sites

?? on those voltage readings. How did you have your meter connected? It should be meter's +/red lead to the black+white test wire, and the meter's -/black lead to a good ground or the battery "-" post. Your voltage readings make it seem like you hooked one meter lead to a +12volt reference instead of to a ground point. With the meter properly connected, you should have zero volts with the engine OFF, +12volts (approx battery voltage) while holding the key in the START position (the ECI relay should be picked "ON" with the key in START), and +12volts once the engine is running and the ECU can see ignition pulses (the ECI relay is being held ON by the ECU).

 

Zero volts while the engine is running and +12volts with the engine off? Sounds like some fuel pump wiring was altered? Open the fuel pump connector access cover (that black metal plate in the spare tire well, driver's side) and look at the wiring in that area. You should see a couple connectors clipped to the top of the fuel tank at the rear-bumper end of that access hole. Anything looked hacked in that area? Shine a flashlight forwards a bit to see the fuel pump, see if it looks like a factory one (basically a plain silver cylinder) that "fits" the mounting hardware properly. If you see spacers, home-made brackets, etc. then there is an aftermarket pump installed and it could have been installed improperly.

 

If you run a fat wire from the battery "+" post to that black+white wire test connector and DO NOT get fuel pressure or don't hear the pump running, somebody has screwed up the fuel pump wiring. The fuel pump power flow should be:

battery + post --> ECI fusible link --> ECI relay switch contacts --> T --> fuel pump --> chassis ground

The "T" is where the black+white wire/test connector taps in. So running battery voltage to that test connector should be a direct power feed to the pump. At the pump there are also a few oddball parts tied into the wiring - these are noise suppressors for AM radio. Unless they're shorted out they won't affect how the pump runs.

 

mike c.

Link to comment
Share on other sites

Okay. I did have my wires reversed on the volt meter. I have confirmed that the ECU IS in fact getting the ignition pulse and the ECI is doing its job. Key on/start, 12v; car starts, 12v remains; car dies, 12v remains for about a full second before going to 0.

 

That being said, I am wondering if the fuel pressure regulator is actually stuck closed or the return line may be somehow blocked. With the raw fuel smell it almost would suggest the car is actually flooding out. Will research testing it and working on that tomorrow if possible.

 

That tach not working is also gnawing at the back of my head, but I don't see where the gauge itself would be an issue since the recent test would suggest the ECU and ECI controller are getting the signals they need.

Link to comment
Share on other sites

This entire thread is almost exactly the same problem I'm having with my 86. The fuel pump relay keeps clicking on/off at random times. Tomorrow, I'm going to connect a bulb to the jumper port, with the hopes that I'll be able to tell which thing is dying first, the fuel pump or the ignition.
Link to comment
Share on other sites

Since you have 38psi at the injection mixer it is reasonable to assume the fuel pressure regulator and fuel return line are probably working okay. Otherwise you'd see something like 50 to 60psi - i.e. the full fuel pump pressure. The regulator "bleeds off" some fuel flow, returning it to the tank, to control pressure. If the return hose is blocked then the regulator can't bleed anything and pressure will quickly rise to max fuel pump pressure.

 

Try this test:

* engine off. Key out of the ignition.

 

* Remove the over-the-valve-cover-pipe (OVCP) feeding the throttle body.

 

* fold a paper towel and stuff it into the throttle body opening so it is underneath both fuel injectors

 

* run a beefy jumper wire from the battery + post to the black+white test terminal. I.e. run the fuel pump with the engine OFF. Verify you have 36 to 38psi fuel pressure. Connect to the battery first, then the test connector so sparks happen by the connector and not near the battery.

 

* Any fuel coming from either injector? If so, that injector is bad (leaking) or a wiring issue between the injector and ECU is grounding the injector, turning it ON. If so: remove the jumper, disconnect the battery, then unplug the ECU connectors. Re-connect the battery and jumper and check the injectors again. If fuel is NOT flowing now, then it seems you have a bad ECU. If fuel is still flowing, it's most likely a bad injector but it could be wiring. Unplug the wire clip at the injector... if fuel stops, it's a wiring harness issue. If fuel still flows then that injector is bad. Cleaning it may save it... worth trying.

 

* if neither injectors flow/dribbled gas that's a good thing. With the jumper wire still running the fuel pump, remove the little "L" shaped hose from the bottom of the fuel pressure regulator. Any gas there? If so, the fuel pressure regulator (FPR) diaphragm has a small leak - and that's what is flooding the engine. Replace the FPR - it can't be fixed. No fuel? Leave the hose off for a moment...

 

* disconnect the jumper wire. Disconnect it at the black+white test connector end first so any sparking happens away from the battery - a safety precaution. Now watch your fuel pressure gauge. It should hold pressure for a while - in theory for several days. Most StarQuests though will bleed off pressure after a few hours. If it drops instantly, there is a leak somewhere (injectors, FPR) or the check valve at the fuel pump outlet is bad. A failed check valve won't keep a StarQuest from running properly... it just means it'll take a couple seconds of cranking to start each time while you wait for fuel pressure to build back up. If the pressure holds for a while, great. Let it sit for a few hours. Then check again; if it has stayed above 30psi or so your injectors and check valve are probably better than half the StarQuests on the road today. Check the paper towel, ideally it's totally dry - no signs of any fuel dribbles. Even a few drops is bad... that means an injector that doesn't seal fully. It'll only get worse with time. A few drops per hour though won't make a StarQuest run poorly - in fact, you'd probably not notice any engine issues if the leak is slow.

 

87-earlier StarQuests had the ECU mounted with the connectors facing upwards. If the windshield ever leaked water, that water would flow down the bodywork "A" pillar alongside the windshield and would pour onto the ECU connectors... obviously messing them up. 88-later ECUs were flipped over to reduce this problem. So, with the battery disconnected, verify the ECU connectors are clean with no signs of corrosion.

 

A dying ignition switch can cause all sorts of issues. Remember, the ignition switch has one input post and multiple output posts for each position - i.e. instead of a single ACC output plus a single ON output plus a single START output as you'd expect from a simple switch, the ignition switch has several terminals any or all of which can be active in ACC, ON, and START positions. Ergo you could have 1 of the 2 ON outputs flaky while the other is perfectly normal. Put your voltmeter on the ignition coil's "+" post and to chassis ground and watch the voltage while the engine runs and stalls. If the +12volts goes away prior to the stall or at the instant of stall, the ignition system is being killed which obviously leads to a stalled engine. If the voltage lasts a second or two past the stall... then it's like not a power supply issue.

 

Does your fuel pressure readout fluctuate at all while the engine starts and runs for a however long it runs? It should be rock steady. If it lowers a couple PSI for a second or two that's enough to kill the idle. Can you reach the backside of the throttle assembly and rotate the throttle linkage by hand to rev the engine once it starts - and hold it revved? If it'll stay revved (2000 to 3500 RPM or so... no specific target other than well above idle RPMs) for a minute or two then the bug is the idle controls and not a fuel issue... I'd crank the idle speed control screw (the one on the linkage at the back of the throttle body - you can see/access it from the passenger side fender looking just to the left of the throttle cable) in two full turns to really crank up the idle RPMs. Make sure the "fixed SAS" screw (the one that looks like a carb idle screw) at the base of the throttle body, about 2 inches to the right of the throttle cable, is screwed OUT a turn or two as well - make sure it is not touching the throttle linkage at all. Cranking these screws like this should initially give you a very high idle RPM; the ECU will still try to run the ISC to lower the RPMs to normal after a second or two. This is just a test to make sure your idle control screw isn't turned too far outwards (too low RPM) killing the engine before the ECU can adjust the ISC motor to recover idle RPMs.

 

As for the dead tach: are the other dash gauges functioning? If I remember correctly, fuse #13 feeds the gauges. Do your battery/charge light and parking brake light illuminate with the key in the ON position, and does the voltmeter read close to battery voltage? If so, the fuse and the panel ground are probably okay.

 

mike c.

Link to comment
Share on other sites

okay Mike - stepping through your instruction, here are the results:

 

Fuel pressure test with jumper wire as instructed showed 38 PSI and held for a 30 second test. Pressure released to 35 PSI when the pump was disconnected and is still there. After about an hour, the pressure had released to 24 PSI.

 

Paper towel remained dry. I even double checked and ran the fuel pump again while watching in the injection mixer and zero fuel came out of either injector. Throttle plate was dry as well.

 

Pressure Regulator had no fuel in the vacuum line.

 

This is an 88 conquest built November 87. The ECU connectors are oriented on the side towards the front of the vehicle. The connectors appear in good shape, no corrosion I could see either on the connector or in the pinouts.

 

Checking the ignition coil, 12.4 volts while cranking and running, when engine stalled, I still had 12.4 volts with no drop except when cranking [obviously]. I could immediately restart the engine, and feather the throttle to keep it running. It would start popping and sputtering after a few seconds and would die as soon as I released the throttle.

 

I had already gone through the mixer, TPS and ISC/MPS with instructions from others on this site. All checked out. The fixed SAS screw is not touching the throttle at all, and adjusting the ISC screw helped the idle, but the engine still pops and sputters and dies as before.

 

As for the gauges in the vehicle, Fuel level works, oil pressure works, boost reads about -10psi while the engine is running. Voltage meter works and registers 12+ volts [as much as the gauge can be read]. ABS, batt, seat belt, brake do come on when the ignition is turned on.

 

Correct me if I am going up a bad tree, but what about a bad distributor? The wires and plugs are new as are the cap and rotor. Thinking basically, we need fuel, air and spark all at the precise timing to get a basic engine to run. This engine will run, but something may be killing the spark to the plugs or breaking down quickly resulting in a no spark condition.

 

I'd be lying if I said this wasn't starting to curl my toe nails.

Link to comment
Share on other sites

OK, so I worked on the car again today. Since the car hasn't really run enough to clean the plugs, I went ahead and installed a fresh set. As expected the installed plugs were midnight black with soot and raw fuel. The replacement did away with the popping while the engine runs as well as reduced the rich exhaust smell. Its 30 something degrees outside so I expect some richness while cold.

 

I also double checked the AFS. I found the specific settings where the ECI checker tool monitors the three signals coming from the AFS for pressure, temperature, and speed and getting to the ECU. All three were within FSM specs while the engine would run. I stepped further and borrowed a spark tester from a friend and tested the secondary system of the coil to see if possibly the coil itself was breaking down once charged. Results were that the spark continued until the engine stopped turning over.

 

Given these latest results, the car is getting air, its getting metered fuel, and its got spark as long as the engine is turning over. Below are links to pics and vids of what I am working on and did tonight.

 

http://i284.photobucket.com/albums/ll2/desktopjockey/conquest%20tsi/th_20150126_195158.jpghttp://i284.photobucket.com/albums/ll2/desktopjockey/conquest%20tsi/th_20150126_195210.jpg http://vid284.photob...0126_193106.mp4

Edited by SCillini
Link to comment
Share on other sites

At this point I wonder if the issue isn't with the basic engine & fuel system stuff... but instead problems with vacuum operated stuff. A larger vacuum leak will kill the idle. For testing in the garage/driveway, you really don't need any vacuum hoses connected on 87-later StarQuests to have the engine idle and rev gently. Unplug all of them from the intake manifold & throttle body and cap the ports. yes, even the little "L" shaped one feeding the fuel pressure regulator... that hose is only essential when boost pressure is present.

 

I'm especially concerned about three items:

 

1: fat vac hose that feeds the power brake booster and ABS assembly. I've seen brake vacuum lines rot near the ends allowing a big vacuum leak. Spraying carb cleaner around the hose and listening for a change in the way the engine runs/idles will identify leaks.

 

2: EGR valve that isn't closing properly. The only way to know for sure is to remove it from the vehicle and examine the two larger exhaust-soot encrusted ports. You'll see the "pintle valve" that controls exhaust flow between these two ports; if the valve doesn't seal properly you'll have exhaust gasses going into the intake manifold at idle... and the engine will sputter to a quick halt.

 

3: auto transmission: there is a vacuum line near the #4 cylinder intake manifold runner that runs down to the transmission's vacuum modulator. Sometimes the modulator itself fails - allowing engine vacuum to draw transmission fluid into the intake manifold. Doesn't burn too well, makes a mess inside the #4 cylinder too. That hose can have a vac leak as well obvioiusly - and being a long hose it's has plenty of places to get damaged. Stick-shift cars have a screw blocking this port in the intake manifold - make sure it's there!

 

mike c.

  • Like 1
Link to comment
Share on other sites

Another source of idle issues is the air filter canister itself. If the lid doesn't seal to the tank, or if the gasket between the lid and the airflow sensor assembly leaks, the engine will struggle to run or idle. If the lid clasps/latches don't snap tightly on your car, or if the supports have busted off, an alternative are little metal clips used on many Toyota cars of the 70's and 80's. Toyota used these clips on air conditioner and heater boxes underneath the dash. They are about three - quarters of an inch long, half inch wide, have a gentle arc to them with tight twists on the ends. They look like little leaf springs actually.

 

mike c.

Link to comment
Share on other sites

 Share

×
×
  • Create New...