[AaronDC]

I started a thread in the Hardware Section (link below for background info). Not much response there. I now have a much more specific question, so I'll give that a shot here.

viewtopic.php?f=1&t=20486


ECU Catch Code: SFG0
Strategy: KHAI3
Car Model: 1998.5 SVT Contour


During cranking, we are getting spark at the spark plugs, but NO fuel injector pulses.

While testing and diagnosing, we have an incandescent bulb (tail light socket and pigtail) connected to the injector pigtail. That would show a flicker if the fuel injector was pulsing, correct?

We checked for continuity between the injector pigtail and the corresponding pin on the ECU plug, so there are no wire problems between the two. The positive 12V wire in the injector pigtail has positive 12V on it any time the key (ignition) is on.


The fuel pump is now connected to the Contour harness so that it is controlled by the ECU. When you turn the ignition to the on position, you don't hear the typical 2 seconds of fuel pump running. Instead, it's two short pulses. I'm not sure if that is important to diagnosing the problem or not.


What controls fuel injector timing? What controls spark timing? I expected that both were controlled by the same components.


PATS is turned off in the ECU, and we're seeing spark. So, I don't think it is the issue.


Thanks in advance!

[decipha]

are those returnless? if so u would need to adjust for that, would be where I would start

is that a duratec 3.0 L ? dohc? any go fast goodies?

[AaronDC]

The Escape 3.0L is a returnless setup in the Escape. We have retrofitted it to a return style. The Contour ECU is setup for a return style. In general, the ECU doesn't know that it has anything other than the stock 2.5L. All the 2.5L sensors are in place, and connected.

It is a Duratec 3.0L. We put the SVT Contour cams in the motor. No other performance mods.


Right now though, we have the NO performance mod. If we can't get this thing to start in the next couple days, we are very likely to miss our first race.

[Paulie]

Check to make sure throttle position is not at wot, that would shut off injectors while cranking.

[AaronDC]

To get to the injector pigtails, we have to have the manifold out of the way. Meaning that the TPS isn't even plugged in as we're testing stuff. Would the ECU see a missing TPS as WOT?

[Paulie]

To get to the injector pigtails, we have to have the manifold out of the way. Meaning that the TPS isn't even plugged in as we're testing stuff. Would the ECU see a missing TPS as WOT?

If the TPS is unplugged, it should read zero so it shouldn't be WOT. That doesn't rule out a wire being shorted in the harness.
Do you have any codes right now? Can your SCT handheld do any datalogging? Can you hook up a 12 volt LED to PCM pin 82 (PATS LED) and ground?

[decipha]

no an unplugged tps causes the ecu to use the rativ value so that shouldn't be of any concern

what are you using to tune hardware and software wise?

can you email me or post your current bin ?

[AaronDC]

If the TPS is unplugged, it should read zero so it shouldn't be WOT. That doesn't rule out a wire being shorted in the harness.
Do you have any codes right now? Can your SCT handheld do any datalogging? Can you hook up a 12 volt LED to PCM pin 82 (PATS LED) and ground?

I will check the TPS harness, just to make sure it isn't shorted.

I do get a couple of CEL codes, P1000 (readiness) and P1681 (something PATS related). The P1681 is the only reason I originally thought something wierd was going on with PATS. We have every component from the donor car, and it's key attached to the RFID sensor. I wouldn't expect a PATS problem, but the code suggests otherwise.

I can datalog with the SCT handheld, but can I learn much from that without getting the engine running?

I looked through my pin-out for the ECU, and pin 82 doesn't show up. I can confirm later if it is used or not. Is your suggestion of pin 82 specific to this ECU? Or a generic Ford thing?


Thanks for all the input.

[AaronDC]

no an unplugged tps causes the ecu to use the rativ value so that shouldn't be of any concern

what are you using to tune hardware and software wise?

can you email me or post your current bin ?

I haven't started doing any tuning yet, but plan to use BE with a pass-thru cable. I have done a few of these motor swaps into Contours, and the motor should start up on the stock tune. If I don't get my pass-thru cable in time (on order, sort of), I have an extra quarterhorse that I could use.

The current bin in the ECU is stock SFG0 with one exception. "Alternate PATS SW" has been changed from "1" to "0" in hopes of ruling out a PATS issue. Assuming that fully turns off PATS in the ECU, I'm wondering now if there is a separate PATS box that is interfering.

[gm2]

I'm working on the car also, thought I'd chime in.

Also worth noting is that we bypassed the ground circuit of the starter relay trigger. The engine will not crank with the ignition switch otherwise due to lack of ground at the starter relay. According to the wiring diagram, that ground wire is routed through the PATS system.

[Paulie]

For a 1998 Contour, PCM pin 82 is the antitheft led output. If you had an led and it was flashing rapidly, you would know you have a PATS issue. Code P1681 is PATS transceiver signal is not received. It sure seems like PATS is NOT turned off in the tune. I really think its a PATS issue.

[AaronDC]

About connecting this LED. I assume I should use a 12V LED, right? Are LED's sold for a specific voltage? I'm guessing the ECU pin will be the positive feed to the LED, and I connect the other end of the LED to a ground.

Based on how the PATS LED in my Focus works, I expect that I should see constant, slow LED flashing with the ignition off and batter connected. With the ignition on, I should see no LED flashing if everything is working correctly, right? Fast LED flashing is bad.

[Paulie]

About connecting this LED. I assume I should use a 12V LED, right? Are LED's sold for a specific voltage? I'm guessing the ECU pin will be the positive feed to the LED, and I connect the other end of the LED to a ground.

Based on how the PATS LED in my Focus works, I expect that I should see constant, slow LED flashing with the ignition off and batter connected. With the ignition on, I should see no LED flashing if everything is working correctly, right? Fast LED flashing is bad.

Yes 12 volt LED. All the above is correct. When you turn the key on, the LED should illuminate for about 2 seconds and then go out. If it starts flashing with the key on you have a PATS problem.

[AaronDC]

Thanks for the quick feedback.

Assuming I find that I have a PATS issue, do you have a suggestion for diagnosing the problem? Or maybe a suggestion for the work-around? We would prefer to NOT keep the PATS functional if we're given the option.

[AaronDC]

Well, it seems that it is a PATS issue. I'm getting the quick flashing light with the ignition on.

There's only one parameter in the strategy file that seems to be associated with PATS, and it doesn't seem to change the problem.


I'm at a loss at this point. We're starting to look at other ECU options to get it running. MegaSquirt has been considered. We've also considered swapping in a Gen IV ECU along with a mild re-wiring of the ECU to completely avoid the PATS stuff.


Any suggestions? This is really starting to affect the probability of having the car running on the day of it's first race.

[decipha]

with PATS disabled the light should blink fast

forget about pats and just set the no fuel cut sw /done

[AaronDC]

What? I like the /done part, but I'm not sure I'm following.


There is a "SW to Enable No Fuel Cutout". Stock, it's set to 0. I assume you want me to set it to 1, right?

[AaronDC]

I tried changing the no fuel cut sw this morning. No change. Still no injector pulsing.

[fordman2304]

Hey guy... the DTC P1681 is for the PATS Transceiver signal not recieved, follow this pinpoint test

E1 : clear dtc's.....does the code come back? if yes then E2

E2 : check transciever power circuit for" voltage", circuit 95 gn/bk wire color this is pin "1" harness side and ground are your readings atleast 10 volts? if no... repair the wire

Im at work but start there and i will fill you in if you need more info for testing

[Paulie]

Are you using a Quarterhorse and Binary Editor to turn off PATS? Maybe the tune is not being accepted by the PCM for some reason.

[AaronDC]

I'll try the transceiver test tonight after work. I'll post back the results.


I am tuning with BE through a QH. I hoped to be direct flashing, but that hasn't worked out yet. I think the PCM is taking it. When I press "write", BE shows the progress bar moving, and the fuel pump runs the whole time. I'm seeing all the same reactions as I do when I write to my Lightning (which runs correctly).

[decipha]

that dont mean shit, the j3 could be disabled on that ecu theres a jumper you may have to jump and probably do


with the nofuelsw set the ecu gives 2 shits about pats, it'll run regardless so you have a communication issue

i had the exact same problem on a crai8 ecu the other day, the qh worked fine but the tune wasn't being applied, i knew because it was setting the pats code when i had it disabled, flashed it and problem solved!

[fordman2304]

I just thought the guy has to start somewhere... and all the info is based soley upon the DTC and the pinpoint test follows after. As a "troubleshooter" I start with the least intrusive test as possible then analyze the data that is before me.

[AaronDC]

The SFG0 ECU doesn't seem to need jumpers soldered on. There are a few sets of pads that look like they could have jumpers soldered on, but they are at the other end of the ECU from the J3 port. All of the images I found searching showed people soldering pads adjacent to the J3 port.

I'll check out the transceiver voltage a little later tonight.

[AaronDC]

Update from this morning's work: I've cured the B1681 code. I inspected the transceiver, and it was receiving good voltage. The rest of the circuit though, was not making it to the ECU. I patched the wires in, and the code went away. Also, I'm now getting the two second fuel pump run that I should have with a key on sequence. Once I heard that, I was certain it would start.

I was WRONG. I didn't have time to put a light on an injector pigtail, but I didn't smell any fuel at the exhaust (just headers at this point), and no hint of combustion sounds.


The plug on the ECU actually didn't have the wires in place for the transceiver. I'm starting to think now that the engine harness is from a non-PATS car. Luckily, the two wires I just connected should be the only difference.


Is it possible to datalog while attempting to start the car? Would that tell me anything about what I'm fighting here?

[decipha]

i always start my datalog before cranking, thats the only way you can see whats really going on during crank and initial crank

just make sure the ignition is on and start the log

[AaronDC]

I've confirmed with the PATS output light, that PATS is not in the way anymore. I get the 2 second fuel pump cycle like it should now as well.

Still not getting it to start though.

I confirmed that I'm still getting spark. The intake manifold is back on right now, so I haven't been able to put a light on the fuel injector circuit again, yet. That generally requires a second set of hands, which also haven't been available in the last couple days. I suspect though, that I'm still not getting fuel injector pulsing because I'm not smelling gas in the exhaust manifolds and no hint of combustion sound.

I took a datalog while trying to start. It's attached to this message. I wasn't sure what was really valuable to a datalog while cranking. It was the first time I've ever datalogged with that strategy, so I only ran a few values just to make sure it worked. If there is a certain value that is more helpful, let me know.


Any ideas what to do next?

[fordman2304]

You really need to check for "Vpwr" to the injectors at least "2" of them on seperate banks.....with " key on". I suggest a voltmeter, or noid light,, the noid will load the circuit but may not flash if there is high resistance. And for shits and giggels check fuel pressure to see if theres a pressure drop of at least "5psi" while cranking.

[Paulie]

Looking at your log, battery voltage seems very low while cranking.

[AaronDC]

You really need to check for "Vpwr" to the injectors at least "2" of them on seperate banks.....with " key on". I suggest a voltmeter, or noid light,, the noid will load the circuit but may not flash if there is high resistance. And for shits and giggels check fuel pressure to see if theres a pressure drop of at least "5psi" while cranking.

We have positive voltage at the fuel injectors. One of the next steps is hook up a noid light to test at the injectors.

Thanks for the suggestions about checking voltage at the transceiver. We had voltage there, but checking that gave me the idea to check continuity back to the ECU. That was where the real problem existed. Well, the PATS problem anyways.