I've got the throttle thing sorted out. The throttle cable had too much slack on it.

My new problem is timing and fouling of plugs. My trailing timing does not line up. leading is dead on. Engine warmed up, RPMs around 750ish. TPS set to 1V in idle position. But trailing timing is advanced just about halfway between the leading and trailing marks.

I went through my ECU pins looking at voltage. I found a couple of oddities:

1) Clutch switch on the front of the clutch pedal was dead. I replaced that and confirmed the ECU is seeing a signal from it.

2) AC relay and A/C switch both show 12V whether A/C switch is on or off

3) Neutral switch shows no change at ECU between Neutral and in gear...but IIRC I do not have a neutral switch on my JDM transmission.

4) Oxygen sensor was way off, but that might be because my engine is running really rich??

It's supposed to be Idle-below 1V, Accel -.5-1V, Decel -0-.4V.

Mine idled at 29.6mV, I wasn't driving with DMM on, but I throttled the pedal Accel-22mV, Decel-26mV

5) The other oddity was my TPS narrow at first seemed to be dead. Even though it showed 1V at the sensor and shows a clean sweep. At the ECU it showed very badly 2.64V and never really changed much as the throttle changed. Before writing off my TPS I checked it again another day and it read dead on that time. I checked it at the sensor first. Then checked it at the ECU and it had the same reading....very strange. It might not have been backprobed correctly the first time, but I thought I checked it and double checked it.

6) Full range on the TPS is definitely off by a bit. It should read .8V at idle and mine sits at .54V. Can I get some feedback on that one? On the one hand off by .26V doesn't seem bad, BUT that is a significant amount when it's 1.4 of the total spec reading.

I have an Innovate wideband on it's way to me. I'm hoping to have it installed this weekend.

When mines went bad, the engine tended to idle slightly higher than normal.


I believe the circuit pulls to ground - the switch grounds out at the A/C compressor?
Also, I believe there's a pressure sensor switch (for safety reasons) is wired in series to the circuit...which could be the culprit?


I know this affects the cold-start idle on initial cranking...


This is odd, but it should not affect anything outside of slightly worse gas mileage...
It should see at least 0.5VDC at idle and decent acceleration.


This might trigger one of those nasty E-OMP error codes which could trigger limp-home mode...


-Ted

^ I checked codes a while back and the only ones I had were for my solenoids from emissions delete and the AWS which my JDM does not have.

Although that is not entirely true. I pulled codes shortly after install and initial startup and had codes for every major sensor. After resetting the ECU they all disappeared..well other than the emissions and AWS.

I'm leaving now to reset idle, recheck TPS voltage, check timing, if time premits I'll backprobe my ECU and see if those readings are still the same.

As a general overview, I think I have some fundamental wiring issues going on. My AEM gauge will spontaneously show way over boost/vac. The last time it did it 5psi boost was the reference point during it's shutdown procedure. On startup and shutdown it will generally go to it's baseline then either dropdown and start operating, or shutdown. I didn't look when I first started up the engine but saw it while driving. When I turned the engine off the closing ceremony had it reference to 5psi.

Also my aftermarket alarm was so out of sorts that I disconnected the alarm cpu entirely. I didn't have constant power going to it at all and when I started disconnected negative wiring at the battery my alarm siren starts going off very faintly. Very strange in that the alarm should not have been getting any juice at all. When I reconnected the ground wire for the siren (connected directly to the battery sidepost) my siren starts going off full blast...again without having constant power to the alarm cpu.

Other oddities, my dome light is non-functional. The last time I checked voltage at my TPS I turned IGN to ON and did not get any ticking of the BAC for some time..then all the sudden it kicked on. Now I did have my ECU harness for TPS (IIRC the 2 harness) disconnected. Part of working today will be checking out whether having that harness disconnected replicates that same thing.

My wideband should be here tomorrow. Hoping to have it up and running this weekend. Maybe afr will shed some light on whatever is going on?

I think I forgot to address this, but...
Ignore the trailing mark for now.
I believe the trailing ignition timing is dependent on the boost sensor signal, so that's why it's not doing the proper 15-degree split.
As long as the leading is spot on...


This sounds like it's isolated to the aftermarket boost gauge itself?
If so, then it has nothing to do with your engine running problems...

This is a classic case of a ground loop circuit.
Like the above, this should not have anything to do with your engine running problems...


I doubt it.
I still have my money it's the TPS or related...


-Ted

Thanks for the feedback Ted,

Something very curious happened. I found my battery was nearly dead, and not enough to crank the car. Jumped it and was letting the engine warm up and the battery charge. After it got up to temp I starting setting the idle. It was sitting at just a nudge above 1k. When it had previously dropped down to barely idling and bouncing just above 0. I had been driving it like that for a week or so and after the engine warmed up idle would sit around 800ish. So I turned down the idle to just over 500. About this time my efan kicks on. I toggle my manual switch to have it run longer and bring the temps down further so my fan isn't kicking on and off while I'm working. I toggle the fan off after temps get down low enough and as soon as I do my idle drops from the 500rpm to just barely idling.


Also while the engine was warming up and the battery was charging my boost gauge was at 0psi rather than vacuum. It would go up from there if I toggled the throttle. When I turned the engine off closing ceremony took the boost gauge to almost 10psi.


I restart the engine, boost gauge is reading correctly at about 15in/hg but rpms are still amazingly low.


Can someone explain the difference between the two idle adjustment screws.

Fast Idle

[IMG]http://i191.photobucket.com/albums/z...605_155159.jpg[/IMG]


Not sure the name on this one, but man the pic sure captured the pollen on it (hood as been sitting up on it)

http://i191.photobucket.com/albums/z...605_155217.jpg

Fast idle is tied to thermowax and turning that screw does nothing for my idle. I am JDM with JDM BAC so I have no idle adjustment there. My only way of adjusting idle is the second pic...of which I do not know the name.

I struggled with the idle setups on our JDM Tii.

I took the TB off to adjust the fast idle. I used a heat gun to test it. I put pictures and descriptions of what I did either on this or the other forum. The FSM is wrong.

The FSM talks about how to set the butterflies but nothing mapped to what was on the TB.

I used that big screw with the sleeve to set the idle. I used the two 12mm nuts on the throttle cable to ensure I had plenty of slack. IMO the engine should start and idle with the cable removed.

Don't forget the ISC. I grounded it straight to the negative terminal as I found that the bolt I was using under the main fuses wasn't working as a ground.

^ Yeah I remember you posting about your TB and I found the thread on the other forum about rebuilding them. I'm in the process of tearing down a spare USDM S5 turbo TB. I was actually going to work on it, but then found I was out of gas for my disposable torch.

So the fast idle is for when the engine is still warming up? Adjusting my fast idle does nothing. The sleaved idle is what I should be (and have been using to set idle).

ISC? Not familiar with that abreviation. I went through my grounds, or all of them I could think of and find. I did the grounding mod at the ECU. I cleaned up all the grounds down to clean metal. I've got the engine grounded on top of the keg to the firewall. I checked the engine harness ground that is on top of the engine keg. I have common grounds for gauges and headunit at the shifter bracket. I have another common ground at the driver kick panel. I went through my grounds for my stereo's amps. IIRC I got the one at the driver shock tower. I even got the one under the trailing coil. Right now my fan and alarm siren are grounded directly to the battery. Then I have an additional battery neg terminal to body chassis ground.

When you get engine idling / running problems dependent on coolant temp, this implies the coolant temperature sensor could be the cause...
Does it check out?


You talking strictly about the aftermarket boost gauge, right?
I think the aftermarket boost gauge is defective.


Not possible - see above.


In the first picture, the idle adjust screw is at the very top, middle of the pic.
I believe it's the same pic as the screw on the left in the 2nd pic?
The screw should be toward the rear of the engine, closer to the firewall.
The screw should be pointing toward to the passenger (US-spec) fender; you need to lean over the passenger fender to adjust this screw.

Don't mess with the "Fast Idle" screw, as this just adjusts the double throttle plates relative to each other - yes, it should not affect idle (that much).

This is still a stock TB, right?
No removal of the double throttle plates?


-Ted

It did at the ECU pin anyway.
FSM specs for 2E:
idle cold: .4-1.8v
water temp 68: approx 2.4v

Mine was
idle cold: .532v
water temp 58: 2.357v


Gauge is practically brand new AEM gauge. It mostly seems dead on, but every so often it bugs out and reads over. The manual that came with it broke down the voltage it sees in relations to boost the sender sees. When it is bugged out often times turning off the engine and restarting it corrects the issue.

It's a stock JDM TB. It has double throttle plates. I just starting rebuilding a USDM TB that I'm going to put on once it's cleaned and adjusted.

Overall I think I have wiring issues. I certainly have one for the aftermarket alarm, the boost gauge is certainly buggy.

Maybe that faulty wiring extends from accessories to actually affecting engine performance? I don't know quite yet.

I have a Trust / GReddy EGT gauge that basically does the same thing.
I'm too lazy to redo and trouble-shoot the whole thing, but it works most of the time.
The aftermarket gauge has it's own dedicated sensor and wiring, right?


-Ted

None of the above is tapped into the "EM" (Engine Main) electrical harness, right?


-Ted

Sorry, Jeff. ISC is Initial Set Connector. I believe that is the standard term. It is the single green plug near the battery/main fuses you ground whenever setting the idle. We had it on and off so much when doing the swap I made a wire with a male spade and a ring terminal. But even though it was attached to a cleaned bolt holding the main fuses to the shock tower it wasnt delivering a good signal to the ECU.

The fast idle adjustment needs to be set so that when the engine (well the wax) is cold the green topped drum is aligned with one of the marks on the cam and when hot (the piston rises) the green topped drum is aligned with the other mark on the cam.

Good luck. It gets better! :driving:

Got the wideband wired up and calibrated. Other than pulling the downpipe and having a friend weld the bung and doing the wiring the engine hasn't been started since I last set the idle and TPS a day or two ago.

Idle was set at around 750-800rpm. TPS was at 1V and ranged nicely up to 4.75V. Full range is still off by about .2something-.3V but it ranges even with no deadspots. I verified at both the TPS harness as well as the ECU.

I finish the wideband install and start up the engine and it's back to just barely being above 0rpm. I let it warm up and no change. At idle the afr was 11.7ish. I drive the car around maybe a 1 mile loop get back and idling afr was at 10.8.

I used the simulated narrow output and tapped that in right at the ECU.

Here's a video of it after taking the drive. I was taking it up to about 3000rpm and the wideband would go to open air 22.4 and then come back down to 10.8.

http://www.youtube.com/watch?v=LtDPndOIeaQ&feature=youtu.be
I tried having it displayed directly in the thread but couldn't figure out how to do it.

-Ted

^ hehe how did you do that??? I know I've posted vids before but couldn't figure out how it's done?

The gauge will eventually move to my A pillar but I've got a situation where the screws holding the convertible top attachment were seized on (locktite at the factory?). I tried impacting them with a hammer on screwdriver and that did nothing. After stripping them I've started drilling them out. Short story long, I need to get my A-pillar off in order to change my single gauge holder to my double and add the wideband there.

I may go play with the idle again and wideband more...though I"m a little frustrated and may step away from it for a day.

One very curious thing the wideband was doing while I was driving was sporadically going to open air. Every so often it would go to 22.4 hold a second or two and then return. My ground may not be the best (under the shifter bracket). Gonna try regrounding in a better spot.

With the idle it seems like I set it, come back and it's not where I set it anymore.

You need to grab the YouTube unique "name" for the vid and dump it between the "YOUTUBE" brackets...
You can't just cut&paste the entire URL, i.e. "http://..." - that just rewards you with a nice black square on the screen. :P

Basically, you cut&paste the random letters and numbers after the "watch=" string in the URL.
So the entire URL you pasted was: "http://www.youtube.com/watch?v=LtDPndOIeaQ&feature=youtu.be"
All you need is this part: "LtDPndOIeaQ"
If you click <QUOTE> on my reply and stare at it, you'll figure it out. :)


-Ted

^I tried doing that with my post but on preview it showed a blank box, so I assumed it was a dead video

Getting back to the car. Can my full range TPS being off by .3 or so V be enough to cause my problems? The narrow range seems to be dead on. This coming week I"m going to talk to an electrical instructor about borrowing an osciloscope to test the TPS.

I'll at least pull the CEL codes today, see if they tell me anything.

Wish I could help you out with those questions... :(

Mines is an S4, and I haven't messed around with an S5 with the E-OMP to get enough info for you.
I vaguely remember the full-range TPS doesn't matter so long as it's linear...?


-Ted

Yeah, it's linear...no drops in the range on either full or narrow. I've heard an osciliscope is a better test and if it's as easy as borrowing the tool and hooking it up I'll test it that way.

As long as it's linear, with no drop outs its should be fine. Referring to '89 FSM, page F2-81 cites a resistance measurement spec for the TPS, full & narrow ranges. To test, disconnect the TPS connector, and measure resistance across the TPS connector terminals it references at closed & wide open throttle conditions. Do this with the engine at operating temp OR you can use a screwdriver to lift the fast idle mechanism off the thermo-wax plunger so the TB is not sitting at a high idle for the closed throttle measurements.

Full range TPS @ wide open throttle should be 3.4 - 5.1 K-ohms; at closed throttle it should be 600~900 ohms. The narrow range is 0.8~1.2K ohms closed, and 4.0~6.0K ohms WOT.

^ that is a good point, I've done both resistance and voltage tests on this TPS. I'm told that testing for voltage is more reliable because it tests the TPS under load. But I can cross-reference voltage test with resistance test and they should both be good. If one is good you'd assume the other is. If they aren't the same then something is amiss.

I've opened up my FSM to look over the checklist for rough idle. I'll go down the line and test all those.

One question related. The water thermosensor:
I'm curious if it would work to test the water thermosensor for resistance but at the ECU pin rather than the sensor itself?

Reason being, to test per FSM means removing it, draining coolant, etc. Is there any reason why I can't start with a cold engine to test resistance at the ECU and let the engine warm up while testing?

The problem with doing it that way is your DVM/ohmmeter puts some voltage across the terminals you're trying to measure a resistance across. If it's still connected to the ECU with power applied you could potentially harm the ECU. Also, unless you have a good aftermarket temp gauge, how will you know the water temp when you're taking the resistance measurement?

Assuming you do have a temp gauge, one option would be to run the car until warmed up, then jot down the water temp off the gauge and shut it down. Disconnect the battery. Then disconnect the ECU connectors, and do your resistance measurement across the two ECU harness connector pins that connect to the water temp sensor. That will only get you one test sample though, so you still won't know if the temp sensor works across its full range. IIRC, the FSM gives 3~4 temp points to measure resistance on that temp sensor.

Our RTek 2 shows the temperature reading from the ECU on the Palm. Does the 1.7 not have that capability?

If its a RTD or resistance temp device it should have table somewhere about the resistance readings. So say 1k ohm means 72 degrees. Wouldn't really matter if you got multiple readings as long as you get a good resistance reading.

Also resistance will determine your voltage. So if your TPS has a "dead spot" it won't matter if you're reading voltage or resistance. One you when off, the when on.

Yes I have an aftermarket temp gauge, a Prosport, which isn't name brand, but it' seems to be about dead on. The sender for the temp gauge is tapped into the front of the water pump housing and is taking a reading at the same spot the OEM thermosensor is.

I also have a temp gun I could point at the water pump housing, radiator, hose, etc.

I was thinking the same thing about taking multiple readings, per FSM. I could test as the engine is warming up. Turn off the engine at each temp test and continue to the next....ugh but that means taking the UIM off each time to get to the thermosensor harness.

No logging on Rtek till you get to 2

The FSM has 3 temps to take resistances at and what the corresponding resistance should be, but not really a full scale that I know of. I suppose I could take resistance readings when they are at the resistances for a specific temp (per FSM) check the temp of the coolant to see if they line up.

With the TPS, yes we are kinda saying the same thing...I was more round about. But laws of electricity say that if there is a dead spot in voltage test there will be one for resistance also. I meant something is amiss, as in it wasn't tested correctly..not probed correctly and such.

Don't need to take the UIM off... What I was suggesting in my previous post is you take your readings off of the 2 water temp sensor pins on the ECU connector, after disconnecting it from the ECU. BTW, the resistance reading you'll get this way will be a little higher than what the FSM states for the temp sensor, but it shouldn't be by much more than a few ohms. This is because you're also measuring harness cable losses to the sensor.

Forgive my ignorance, but I'm confused on how you're talking about testing resistance. IIRC the harness at the sensor is 2-pin. So to test resistance between those two pins wouldn't I need to find out which pin on the sensor harness corresponds to the ECU pin and then test resistance between that pin and the other pin on the sensor it'self?

I'm confused on what 2 pins on the ECU I'd be testing for resistance. 2E is the only water thermosensor pin on the harness/ECU. I originally was thinking I could put one DMM probe on that pin and one ground to chasis...but that's not the same as checking resistance at the sensor itself...or is it and I've simply overthought things and gotten myself confused.

For what it's worth, I was thinking the same thing about added resistance for the small load the harness will add.

My bad, I should have given you the reference to the schematic diagram on FSM page F2-8. If you follow the water thermo sensor wires back to the ECU connector on the diagram, you'll see you can measure your resistance between pins #2E and 3D. Pin 2E goes to the water thermo sensor and connects to nothing else; 3D is labeled "E2" on the ECU side, which I believe is a common +5VDC supply to a bunch of the sensors, including the water thermo.

I stepped away from the car out of frustration. Before I did that I tested the thermosensor at the ecu for both resistance and voltage. I couldn't test the lowest temp what with it being summer but the other temps checked out fine.

That same day I checked TPS and it's still good and at 1v. Took it for a drive and it seemed better idling but afr are still 10.8 at idle. Then between stops it developed worse idle and hot start issues. It did it 3 times and that's when I parked it for a week or so.

IIRC the IAT tested fine at the ECU. I'll look at my notes and also test it again, this time at the sensor.

I guess next up is to test AFM/MAF. One thing to note, I have a Cosmo AFM/MAF for a 13B. I'm assuming that could be a factor in the rich idle?

I have a spare set of coils to swap on in hopes of a magic bullet.

Is the ISC the same as the single green harness used to get CEL codes?

When checking timing I never jumpered anything. I simply got it warmed up, idling around 800ish and pionted a timing gun.

That could very well be the case, at idle and elsewhere in the power curve as the ECU uses the AFM as a primary input to manage FI. Is the Cosmo AFM a different part # than the original AFM that went with your ECU? I know AFMs are not interchangeable between S4 & S5 or T2 & NA's. I have no idea if the Cosmo AFM is interchangeable with anything else. What's the reason for using the Cosmo AFM, larger diameter opening/better flow?

I think that's what he was referring to - the single terminal green connector by the battery that you ground when checking CEL codes, adjusting idle and timing. When grounded, the ECU inhibits the BAC from messing with idle speed, which you'll need to check/adjust idle or base timing.

I ended up with the Cosmo AFM by accident and many years ago. My original turbo swap was done by a shop in Indy, AIM Tuning. While getting parts together for the swap I bought off eBay what I was told was an turbo RX7 AFM. I didn't know any different till many months later when the swap was completed. By then I couldn't go back to the seller so I used it. I've been using it since oh...about 2008 or so. No issues...well till now :P

At one time you could read the part # and that's how I verified it's a Cosmo, but now that sticker has been worn down quite a bit.

Thanks for the heads up on jumpering that, this is the first I've read about it.

Here's the update and it couldn't happen in anyone else's build thread than mine.

I test my AFM and find that the IAT sensor within it is dead to the world. I have a spare N350 so I test it as good and pop it on. It's like night and day. Idle sounds much better. After a mile or so drive afr are sitting at 11.7 or so. I'm all happy and take it out for a long drive of blissful thoughts.

And then I burn up my EGI fuse a mile from home. I tried swapping bigger fuses just to get it home and burned up all of those. I ended up towing it home. So now I have to find my short. I've always been suspicious of the electronics in this car...and now I have some confirmation on it.

Yes, the FSM refers to the connector I'm referring to as the "test connector" in chapter 4. If you search the club site for "initial set connector" you'll find lots of hits but it is referred to as "initial set coupler" as you can see in this thread.

And yes, you ground the ISC whenever you are adjusting your idle or setting your timing. Otherwise the ECU will fight with you.

Good to hear about the AFM. Bummer about the short.

Definitly have a short. Look first for it in the areas you've messed with last. Sounds like you've spent lots of time under the hood and may have recently yanked the UIM a few times. One dumb mistake I made when removing the UIM was forgetting to reconnect the harness plug back into that ASV (air supply valve?). ASV is the PITA to get to one on the backside of the UIM, close to the firewall. When unconnected or not ziptied out of the way, that part of the harness is long enough to dangle down & hit the hot DP - instant 12VDC short on the EGI circuit when it hits the DP and melts the connector. In my case, the short was permanent since the melted plastic "welded" it to my DP, so it was pretty easy to find & figure out. If yours is just dangling, it might just be making intermittent contact as you drive around.

Thanks for the heads up on the ISC and about finding my short.

Its definitely an intermittent thing with the short. Last week when I started having hot start issues I noticed that one time the whole engine sputtered in mid drive. The whole engine lost power just for a split second but then kept on driving. That and the engine compartment has felt warmer than it should...specifically around the main fuses. Probably the short making contact then not, making contact then not. Probably heating up the fuses but not burning it.

Things I've recently messed with are the leaving my OEM O2 sensor wire unconnected. I'm using the simulated output on my Innovate and have it spliced in right at the ECU. But IIRC the OEM O2 wiring is a straight feed into the ECU. I don't think the O2 wiring routes through the main fuse.

I've been testing my TPS but that has all been back-probe.

I don't have an ASV on my engine, but I do have some harnesses dangling here and there.

I'm tapping into the green 6pin harness by the battery for my efan power. Does that feed into the EGI fuse?

[IMG]http://i191.photobucket.com/albums/z...essforefan.jpg[/IMG]

Are you talking about MAIN power for your fan?

Most (effective) electric fans need at least 20A of current.
Almost nothing on the stock electrical system can support that kinda current with the exception of the starter...
I will run dedicated power wires for electrical fans.
Minimum 10-gauge wiring and as short as possible - through a relay.


-Ted