motor stumbles after it warms up

[sspence81]

88 1/2 Samurai that has been converted to EFI with a 89/90 ECU and wiring harness. Starts perfect and idles great. Idles down after a few minutes fine. Runs and drives great until it gets hot. Then it starts to stutter sporadically at highway speeds and sometimes (rarely) will backfire when you let off of the accelerator. It has died several times when pulling up to a stop sign and when taking off easy. To restart it takes several seconds of turning it over and when it finally starts it acts flooded. It never dies if I rev it up a bit on take off. My wife commented on the smell of fuel when we were tooling around in it with the top off. My first thought was either fuel pump or regulator. Put a fuel pressure gauge on the banjo fitting come off of the fuel filter and started the motor. Pressure stayed a constant 40 psi. Pinched off the return line for just a few seconds and the pump fluttered between 75-80 psi. No fuel leak from the regulator vacuum nipple. Allowed the motor to get hot and start to miss sporadically while setting in the driveway. Watched the fuel pressure while it was stuttering and it didn't fluctuate the needle even a little bit. So I'm confident in the fuel pump. New plugs, cap, rotor, coil. I'm I right in assuming since the problem occurs after it is up to temperature that it isn't likely anything to do with IAC, dash pot, etc? I'm almost positive that the TPS is not calibrated correctly since it has been moved. I marked it before I moved it but I only marked one side and it moves on both sides. ECU was rebuilt a few weeks ago by Speedometer Repair Guy in Winona, MN. Connected and disconnected the vacuum advance and it doesn't make a difference that I can detect. I'm 80% fabricator and 20% mechanic so any help would be appreciated. Also, FIXKICK refers to adjusting "duty cycle" but for the life of me I can't figure out what he is referring to.

---

Forgot to mention that the fuel pressure immediately drops to zero after the motor is shut off. Does not hold any residual pressure in the lines.

[fixkick]

88 1/2 Samurai that has been converted to EFI with a 89/90 ECU and wiring harness.
but what was the donor car?
Is it TBI EFI?


Starts perfect and idles great.
Idles down after a few minutes fine.
Runs and drives great until it gets hot.
Then it starts to stutter sporadically at highway speeds and sometimes (rarely) will backfire when you let off of the accelerator.
what is compression, WOT, at wide open throttle.?

It has died several times when pulling up to a stop sign and when taking off easy.
To restart it takes several seconds of turning it over and when it finally starts it acts flooded. Does WOT cranking clear the flood,? all EFI does that, its feature to cut fuel, cranked.

It never dies if I rev it up a bit on take off. My wife commented on the smell of fuel when we were tooling around in it with the top off.
Ok sounds like rich misfire. is the CEL lamp glowing key on, it must then go out running? i guess spark tips are black.?


My first thought was either fuel pump or regulator.
Put a fuel pressure gauge on the banjo fitting come off of the fuel filter and started the motor.
Pressure stayed a constant 40 psi. (good for sealevel, spec)
Pinched off the return line for just a few seconds and the pump fluttered between 75-80 psi.
No fuel leak from the regulator vacuum nipple.
Allowed the motor to get hot and start to miss sporadically while setting in the driveway.
Watched the fuel pressure while it was stuttering and it didn't fluctuate the needle even a little bit.
So I'm confident in the fuel pump. New plugs, cap, rotor, coil. (2 wire or 3 wire low voltage connector ?on distributor., wish new donor car)
I'm I right in assuming since the problem occurs after it is up to temperature that it isn't likely anything to do with IAC, dash pot, etc?
the ECU uses a map , so air leaks all cause racing with perfect mix. Jeep like. same.


I'm almost positive that the TPS is not calibrated correctly since it has been moved. I marked it before I moved it but I only marked one side and it moves on both sides. ECU was rebuilt a few weeks ago by Speedometer Repair Guy in Winona, MN.
Connected and disconnected the vacuum advance and it doesn't make a difference that I can detect.
I'm 80% fabricator and 20% mechanic so any help would be appreciated.
Also, FIXKICK refers to adjusting "duty cycle" but for the life of me I can't figure out what he is referring to.
yes that is me, the suzuki uses those exact words, as do all PWM modulation schemes..
The ECU controls idle speeds at all times, (using electric solenoid ISC) and modulates it PWM way, and 50% duty, is perfect.
50% means it can speedup 50 and slow 50, and not run out of air. at ISC. it modulates 200 times a second to do that,
my jeep uses stepper motors and are different.
So whats missing now. after the swap, EGR? CAT, other things?

---

Forgot to mention that the fuel pressure immediately drops to zero after the motor is shut off. Does not hold any residual pressure in the lines.
That is 1 of 3 things.,
1; injector leaks.
2: FPR leaks.
3: tank check ball leaks.

if you pinch return get 60psi, as most do yours is healthy, and then cut engine, and keep it pinched if it still leaks down , its not the FPR leaking , cuz line pinched, so is injector , or the tank check ball.
to prove ball ok, then just lift off TBI air horn, cut key (2nd person) if injector leaks, key off, bingo, bad injector. and not tank ball.
the key must cut injector at speed of light, well near there. 3/1000th of a second.

cheers to you and welcome!!

[sspence81]

Thanks FixKick for the response. I replaced the ECT and the problem appears to be fixed. I say appears because it was sporadic so I guess it could come back anytime. An hour of driving it around town and it didn't act up once. It is a TBI out of a 90 model samurai. The ECT did test ok with the meter but since the problem was coming and going I went ahead and replaced it anyway. So far so good. I'll let you know if it acts up anymore.

[fixkick]

the ect will lie if air pocket on top of engine, keep burp it and the ECT is under water and works.
common prob. after work

if the ECT goes dry, on tip, engine will run like a pig and suck gas, due to thinking its a cold start.

[Terebit]

I inherited this Sammi for $$ from sspence (so I think it's ok to hijack the thread). We drove it for a couple days and noticed an occasional engine stumble and infrequent backfire when off throttle. We also noticed a prevalent fuel smell and what seemed to be excessive fuel consumption (but haven't driven it enough to calculate). My wife was driving it and it stalled at a stop. She restarted and said it was missing and generally running very poorly. She tried to limp it, but it stalled out again and she had to get it towed home.

Recap: 1988 1/2 Samurai, 1.3 with 90 TBI conversion, ECU = 89 Tracker, coil/distributer is vintage 88, EGR valve & associated parts removed, NAG switch removed.

I've been trying to troubleshoot the problem: It starts and idles (a little rough) when cold, stalls once it starts to warm up. Plugs are black and exhaust is smoky (reeks of fuel). TBI looks wet inside (fuel seen streaming on throttle plate).
- No codes (not sure if the ECU to harness wiring is correct); CEL comes on steady prior to start and goes out while running. Diagnostic fuse in fuse box doesn't do anything (NAG removed). TBI wiring harness has a Diagnostic Switch Terminal behind the battery. Jumper does not produce any flash codes (no code 12). sspence said he never tried to check codes.
- Fuel pressure seems good (40 psi at filter port and 50 psi deadhead after FPR while cranking), FPR leakdown and vacuum checks (smoothly decreasing fuel pressure) seem OK, but TBI is flooding.
- Haven't done any other sensor/actuator checks yet. Not sure whether I should replace the ECU first or not.
- Not sure if the vacuum lines are routed correctly for this configuration (smog [EGR valve, VSV, dash pot, CAT] removed). Searching for a diagram or photo (mainly MAP connection).

Next step?

[fixkick]

old sammi, are one of the best cars off road. many love them, nimble wins, in the end. (high cg and weight are killers)

I inherited this Sammi for $$.
We drove it for a couple days and noticed an occasional engine stumble and infrequent backfire (to tail pipe or air cleaner) when off throttle.
We also noticed a prevalent fuel smell and what seemed to be excessive fuel consumption (but haven't driven it enough to calculate).
so its running rich, and if too rich and air leaks near,, the exhuast on all cars explode called muffler explosion, and clear name.
My wife was driving it and it stalled at a stop. (my guess would be the poor spark plugs just fouled out. (
She restarted and said it was missing and generally running very poorly.
She tried to limp it, but it stalled out again and she had to get it towed home.
its a non stock car, and as such diagnosis can be horrid. lacking all facts of the changes.

Recap: 1988 1/2 Samurai, 1.3 with 90 TBI conversion, ECU = 89 Tracker, coil/distribute is vintage 88? what?, EGR valve & associated parts removed, NAG switch removed.
I guess you mean the donor car was a 1.6L G16a ,1990 TBI. but the means, ECU , harnesses, sensors, TBI intake and all parts from G16,
did the Check engine light get deleted, some how? that is first. always. if the ECU is upset, it be wise to ask it , after all it can go to limphome or worse BACKUP mode, the ECU is a 56B30 on 5speed. is that it?


I've been trying to troubleshoot the problem:
It starts and idles (a little rough) when cold, stalls once it starts to warm up. Plugs are black and exhaust is smoky (reeks of fuel). TBI looks wet inside (fuel seen streaming on throttle plate).
- No codes (not sure if the ECU to harness wiring is correct, its not); CEL comes on steady prior to start and goes out while running. Diagnostic fuse in fuse box doesn't do anything (NAG removed). The nag has nothing to do with CEL, when you pull then NAG switch from the cluster, the CEL now owns the circuit for ever.
first off the

TBI wiring harness has a Diagnostic Switch Terminal behind the battery.
Jumper does not produce any flash codes (no code 12). IT MUST, or is wired wrong or is now STUCK in backup mode.
code 12 are MANDATORY, this is A number 1, always, 89 to 95 years.
it just a lamp and 2 wire, one wire is power,12vdc from cluster power buss, the 2nd wire goes to ECU CEL pin, a supper simple circuit,
the schematic is clear here.
sspence said he never tried to check codes.
the 19
- Fuel pressure seems good (40 psi at filter port and 50 psi deadhead after FPR while cranking), FPR leakdown and vacuum checks (smoothly decreasing fuel pressure) seem OK, but TBI is flooding.
- Haven't done any other sensor/actuator checks yet. Not sure whether I should replace the ECU first or not.
- Not sure if all the vacuum lines are routed correctly for this configuration (with EGR valve removed). Searching for a diagram or photo.

Next step?

to use any TBI system, or conversion
you MUST wire up the CEL correctly and a DIAG pin jumper port
Must or spend forever trying to guess why the ECU is unhappy. hopeless this is...

we are talking here, 4 wires.
2 for the lamp
2 for diag. one is ground.
that must work.
if the lamp is present, and power on one side, and ecu CEL pin other and the lamp is good ,. (14v @100ma or less 50mA is spec)
and if the lamp does not come on ever, the ECU is toasted.
i see yours glows, if wired right that is a good sign.
but the diag pins are not present. its just 2 wires. ground and diag.
if you jump that pin to ground and it wont flash a code, the ECU is bad. 12s or any code, even code 53. 53 can mean wrong EGR.

if flashes 12s, we can work out why it failing,
keep in mind 12s does not mean all is well , it means no shorted or open sensors and spark present, and CMP firing.
The ECU can also short out 1 of the 2 injectors lines, and it will go rich for ever. and the ECU is blind to this error.
This ecu also goes rich if the 02 fails, it can go super rich, (cured in 1996)
Id also test the map fully on using my MAP page.

keep in mind the your car has old mech, distributor, so we check timing at idle and then scream the throttle to be sure those sticky advance weights work,
see 30deg. of advance or more, gunning the throttle , if not it will waste fuel big time.
if spark is ok, then fueling is wrong and that is 100% ECU and its sensors.

hope that helps,. cheers.

[Terebit]

[quote='Terebit' pid='7204' dateline='1477502920']
...engine stumble and infrequent backfire (to tail pipe or air cleaner). Tailpipe, I think. I just heard a few pops, sounded like from the rear.

Recap: 1988 1/2 Samurai, 1.3L with '90 TBI conversion, ECU = 89 Tracker, coil/distributor is vintage 88? what? I think the distributer is from the original '88 1/2 engine, but this could have had an engine swap at some point.
I guess you mean the donor car was a 1.6L G16a ,1990 TBI. but that means, ECU , harnesses, sensors, TBI intake and all parts from G16. Did the Check engine light get deleted, some how? the ECU is a 56B30 on 5speed. is that it? I was under the impression that the engine was a 1.3L and only the intake was changed (if that's possible) The number on the ECU is 33920-60A31.

...no flash codes (no code 12). IT MUST, or is wired wrong or is now STUCK in backup mode. Is there a way to get out of Backup mode, if stuck there? New ECU only?

Replaced the O2 sensor a few days ago, no change.
I'll try to find the circuit problem to get flash codes

[fixkick]

[quote='Terebit' pid='7204' dateline='1477502920']
...engine stumble and infrequent backfire (to tail pipe or air cleaner). Tailpipe, I think. I just heard a few pops, sounded like from the rear.
all 100% normal on any engine rich, and some leaks or cracks in the exhaust, (fuel , plus air, = boom )


Recap: 1988 1/2 Samurai, 1.3L with '90 TBI conversion, ECU = 89 Tracker, coil/distributor is vintage 88? what? I think the distributor is from the original '88 1/2 engine, but this could have had an engine swap at some point.
I guess you mean the donor car was a 1.6L G16a ,1990 TBI. but that means, ECU , harnesses, sensors, TBI intake and all parts from G16a. Did the Check engine light get deleted, some how? the ECU is a 56B30 on 5speed. is that it? I was under the impression that the engine was a 1.3L and only the intake was changed (if that's possible) The number on the ECU is 33920-60A31.
60A32 IS A 1989/90 ECU, AND IS 1.6l TBI.
All sidekicks are 1.6L with EFI, (

the samurai
A new 1.3 liter four-cylinder engine with throttle-body fuel injection was introduced with 66 hp (49 kW) in September 1991
as you can see the sammi, is the only 1.3L TBI in usa.
there are no Sidekick/trackers with 1.3L TBI. it was bumped off carbs to TBI + 300 more cc, in 1989.

...no flash codes (no code 12). IT MUST, or is wired wrong or is now STUCK in backup mode. Is there a way to get out of Backup mode, if stuck there? New ECU only?


SIDEKICK g13 1.3L 1989 ONLY WAS SOLD AS CARB. + dizzy only spark.
SIDeKicK G16A 1.6l 1989/90 runs the TBI but ECU does not run spark , only dizzy runs spark for 2 yeas, and behind. 89/90 SAMmi ran 80c ECU.
Sidekick G16a 1.6L 1991 to 1995 56B TBI , and ECU now runs spark.

Replaced the O2 sensor a few days ago, no change.
I'll try to find the circuit problem to get flash codes

mixing up all those parts is not easy, the harness must match.

[fixkick]

ok, it may not be in backup mode.
most times the CEL is 100% dead (out ) in backup mode.
backup mode, means, the ECU is blown up, and now in lawnmower mode, a very very very crude, "read TPS angle and now inject off angle" mode. like that.
limphome means a major inputs is bad.
there are other things that can cause rich.
1: high fuel pressure. FPR bad mostly.
2: leaking injector
3: o2 stuck at 0v. (usually caused by cracks in the exhaust near here, the exhaust will suck in air on all cars made with cracks. (at the negative pressure event)
4: ECU bad, in many ways, usually not this. unless the caps are bad and most do go bad caps. a $5 fix.
5: no spark advance. (easy test, in neutral, timing light.?? bingo... dissy dead, fly weights jammed, seen many.
6: CAM BELT SLIPPED, all do that, ever one shortly after 60k miles, is the service point, and the cam goes retarded and huge power loss.


here is the 60A ecu , 89/90 G16a TBI, all are.
http://www.fixkick.com/INJECTORS/89/89-90schem1.jpg
see those 2 pins there,?
a15 , and a9 (DLC ) (aka. monitor port connector on early years, or a DIAG fuse on 1989s.)
a15 blue-yell is diag mode, all one does is ground this one pin, even just key on, not running (or running) and code 12s flash, endless.

this pin must work or you will never know when the ECU sees problems.

here is the 89 diag pin , hidden in the fuse box.
http://www.fixkick.com/CEL/DLC/Fuse-with-diag.jpg

the first, 89s were made in Iwata Japan, then in march , at CAMI, canada.
at the factory move i think the DIAG pins moved too.







key on, with a jumper wire ground the ECU diagnostic pin

[fixkick]

i dont think you can mix intakes.
The engine clearly has G16 casting marks in the casting

here is the 91, block, TBI G16a
see the top left there.
11am here.
see that.


no mixing up G16 with G13
the serial number is ( see that bell boss at 9am the top of that is stamped, the engine serial.
G13yxxxxxx
G16yxxxxxx
y = year, in usa.
xxxxxx is serial, its clear as day what engine block is there, in TWO places.
head codes are vastly more complex.