Login Register

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
fuel issue
i have a 96 geo tracker 16 v 2 wheel drive, new fuel pump, fuel pressure regulator new plugs wires and distributor and bug, issue is when trying to start will hit hard then nothing...when i clamp off fuel line to the FPR it will start and run fine till clamp is removed then starves for fuel then dies....can anyone help
how many miles on engine?
how old is fuel? in tank?
did it sit for a year with old fuel, after it last ran perfectly? when was it last a daily driver? (say a full tank of fresh fuel ran through it at least every month)
sure it lean.
doing what you just did "FPR", takes fuel pressure to 2 times normal , in fact that is called the fuel pump shunt test, 60psi is what that hits. (pressure)
what is outdoor temp? parked. when this is done
what is ECT reading, parked ( unplug it (release lock( and measure ohms, with any meter) use resistance meter scale,

the starting richness is based off ECU reading this sensor.
or plug in any scan tool made, they all do 96s and newer in usa or 98s and newer canada.
OBD2 port is above right knee, D shell shape connector. and is 16pins for OBD2 tools.
plug in scan tool , does the ECT read parked ambient temps, got DTC errors. cranking?

more questions.
is spark timed near TDC cranked, if not he cam belt slipped. (timing light to #1 spark plug, easy test) even cranking, timing lights work even cranking...
is compression above, 150psi? (spec, is 180f hot)

if all above is ok
is fuel pressure at 35.6–42.7 psi (key on 3 times, will hold here) this is factory spec, and takes only good pump and good FPR and good non rusty wires to pump to get this.

here is the ECT ohms, if you have no Scan tool.


if out door temps are in the 40F range, overnight (my wild guess) then the ECT reads about 5000 ohms.
the chart above sets these rules
and injection rates, off them. cold.
if say the sensor read 300 ohms cold ,it would be very lean starts, for sure....

note my etc chart even shows ECT volts key on, running or not, (my chart is the only one that does these, calcs)

are the air filter and fuel filter clean?
finish the 60k mile interval tuneup>?

one other possible is melted and clogged, cat, preventing starts but id need to see engine power, after you did start it,
i live in missouri and the temp is about 35 degrees the fuel in the tank is probably an issue its been sitting for about 1 yr before that it had overheated water pump went out warped head..had head milled .009 was removed. new seals, gaskets, after that i had an issue with timing got it fixed with the help of your links...it ran great for about 2 days then went to crap again...then i changed fuel pump and fpr and this is where i am now Last month i took the geo to a guy that works on geo and suzuki he told me he checked everything timing fuel pressure and all he said it all checked out so i will start with bad gas
1yr can be ok,,on fuel... i think we should start over.... from scratch.
bad gas can clog those 10-micron tiny screens in the 4 injectors. (needs ultrasound reverse flow methods to clean)
what we do is if all sensors read right,
and fuel pressure is spec.
then by process of elimination, the injectors must be clogged.
did you take a sample of gas., and smell it and look water to separate out. colors is light amber in USA.

the one thing we cant do is test injectors for full flow.
we can check the pulse widths on a scope,. and see that the ECU is pulsing them prefect and if they are, then they are clogged.

You never told me door count.
the 2 door loves to rust the 2 pump wires behind the left tail light lens. or center rear bumper, same wires pink power , blk ground.
g402 screw there rusty.?
what if the pink wire was rusty, there, in that connector seen on same page.
and only 8volts to the pump.?? oops, that wont work., it needs FULL VOLTAGE.
that is why having a fuel pressure gauge, tells you, pressure good or not, and can change at next bump in road, then what, then pump has notw got full 12vdc to it.
and pressure falls.
your mechanic i bet, didnt see it failing did he,.??? if did, why didnt he fix it? (engine good, spark good, fueling wrong, find why)
so never found it.? he gave up?
to fix cars, diagnose them failing only.

first is fuel pressure
no injector runs with 20psi.
pressure first , aways.
the ECU can not know its wrong, ever. so... if wrong, it COMPUTES WRONG. (under injects)
it assumes its right and if is the then flows are correct, and you get full power. (injectors not clogged)

harbor freight sells the gauge for $20 (with free suzuki hose_
its one of the most important tools to own, saving you many long hours of labor and a bath in gas,,, best $20 ever spent, each time if discover, pressure is good. move on.

what if the pulses were small on injectors
the MAF is dirty.
if you unplug, it the MAF missing trips, failsafe it the fueling goes rich is a pig,
ever clean the maf? (from 1foot distances with maf cleaner)>

is the ECU flashing code 12s on demand>?

key on, CEL on, is it?
then insert diag jumper
key on, flash 12s?
this is first, if cel is dead so is ECU, that is if the nag switch is not present as it is on all Fed cars. usa.
if it has the nag, and will not flash 12s remove the switch, (or slide it one time)

key CEL lamp.
and code 12 flashing or if not starting crank for 5full seconds, still code 12 or code 42 or 41 or?
next is ECT ohms at 34f or what ever, F temp is there. 5000 ohms cold is good ect, at that temp. this causes proper injection to start any engine, at that 35f temp (coolant) fuel vapor pressures, are why,...
cold fuel is hard to vaporize. so runs richer there. all the way to -40F , richer step by step.

do the MAF tests next.
if the maf is weak, then engine goes lean
the voltages are on my maf page.
key on
The voltage observed should be 1.0–1.6 volts.
idle is (exact specification is 1.7–2.0 volts) at idle

check for loss of spark anytime engine will not start
check timing too, if the timing belt idler is loose, not set with its index tang it goes loose and spark goes way out of time, with cam.
ok , it wont start so does test fuel work no, IT MUST. if not say so.
my guess i will. and sounds perfect for 3seconds, repeat 3 times for good feelings effects.
so the FPR pinch test works.
what is next.?
fuel pressure, (repeat any time engine fails to run and test fuel test passes)

my guess is engine is ok, and CAT not melted and EGR valve is closed.
a not running engine is very easy to diagnose. after all its failing now..!!!


keep in mind a dry rail means dead pump.
but a dribbling rail dont mean a damn thing.
only spec pressure runs engine and makes full power. only.

my motor cycle runs with 0 pressure, gravity tank,, no EFI will ever do that.

the trouble with old cars, tests pass, and fail 1hr later.
must be dogmatic in your tests.
to find these.
it is a 2 door, when i replaced the fuel pump i cleaned all wiring and connection near and around tank. The tank had a lot of rust in tank a lot. after putting in new pump pump purred like a kitten..i primed fuel lines. don't have a pressure tester YET but it wanted to start then i clamped the fpr hose and it ran fine till clamp was removed
that rust, goes right past a filter and lands in those 10 micron injector screens
the fuel filter can be about 40microns, so it lets through, 30 to 10, packing the injectors. like clock work,
rust is totally bad in any fuel tank.

take one injector out, send it to , witchhunter
see report, before and after
if its ok, then, good
if not all 4 are bad.
$22 to find out.


getting rust out of screens is not easy.
on injector clean can do that,
would take, acid .
so they back flush them .
the rust is like tiny rocks there. stacked up. in a pile.

ever get the cel to flash 12s?
this is like step one on all cars,
flash codes
or scanned DTC errors.
takes 5min work? a paper clip


if the lamp stick on
its mr, Nag


best done on dead engines, cranking, so the ECU can SEE the errors live and report them live.
i am going to change all injectors and probably the fuel filter i dont have a shop all work is going to be outside and we are now under a winter weather advisory till mon i will be getting me a fuel pressure tester just to make sure. im not really a mechanic more of a parts changer so the tools timing light multi meter sockets are pretty basic but i will get it done then i will get back to you...Thanks
is that filter 20years old?>

the filter packed, and using gauge on its filter, test port. (6mm screw)
the test port is on the filters, input banjo bolt, on end...
so when the filter packs, the test port goes to 60psi
and the fuel rail can be like 20 or less. due to packing full of rust.
the rail needs a $10 fitting to measure there. in my pump page, summitracing sells it.
if the filter packs this overloads the pump, full time.

your pinch test shows the engine is running lean,.
we have only this to go on.
only tests prove what his happening,

all ways posted , to do DIY rail fitting, and the 2267 adaptor shines


adapts metric, to NPT, usa pipe fittings, so any gauge fits.


so the shop pulled the tank, replaced pump, saw rust and then kept the old filter there?
no way.

the pump tests are very easy.
on the 16v there are many,,, not just pressure regulation

test 1 to 7 are covered here

my guess, ?
test 4 fails like a dog....

your pinch test is test 6 what we call the shut test ..60PSI and that is the max the pump can do.)

i then do the PFR tests. at least 3 ways. key on, idling and gun throttle. test.
or use my vacuum hand tool, to do many vacuum tests..

fpr test
Static: 36-43 PSI, keyon, not started (say 40) (zero vacuum)
idle: 29.8–37.0 psi (say 34) caused by 20 inches of vacuum)
and gunned to Wide open throttle , the pressure tries to hit 40 psi, (caused by vacuum near 0)

the injector balance tests are way more hard, and tedious.

here is the filter test port being tested. its very easy, test,(depressurize first)

the guy i took it to said he changed the filter.... while doing the " shunt" it runs awesome but once i take the vise grip off it goes to crap...starts hesitating can not hit the throttle or it tries to die
the fuel gauge is just $20 at harbor freight and has the suzuki hose.
why not check fuel pressure first.?
if engines good, and sparks good , then this is a fueling issues.
those start with good fuel, then good pressure.
1: pressure, above 30psi not 60 or it floods
2: 1996 16v why not use a scan tool see if there are big problems with sensors next (like maf dead, or the like) DTC errors, and all sensor can be watched for valid ranges.
3: if all that checks out, by process of elimination, then your 10 micro screens are packed. (deep in side the injectors) 10 micros is super tight screens.

here is duck 1, (as in knocking over duck 1)


no EFI will run right with bad fuel pressure.
it's not going to happen.

here im doing it.
[Image: TBI-conn-1w.jpg]

btw, why cant your mech, find this.?
all ASE shops can do this in like 30min work (after all we know spark and engines ok , it runs, so....its Fueling bad)
pressure good
scans good (no bad inputs to ECU).
injectors clogged..

MAF , no scan tools.

off my MAF page, step 9
Next move the probe to the output pin. Gray-black wire, the voltage key on, no start is 1.0–1.6 volts , if not MAF may be bad.
got that? about 1.3V?
start engine pinched (shunted)
step 10. (same page)
Start car, this pin must be 1.7–2.0 volts. at 800 RPM, if not MAF is bad, or RPM is wrong. of rpm is higher so will be the voltage, due to airflow is presented as voltage, ( 1v key on , 4v wide open throttle up a hill)

this engine EFI , injects based on MAF outputs (air flow)
so if its stuck low, there will be pathetic fuel flow.
got DTC errors for MAF? not scanning a modern car, is a horrid waste of time, if its screaming, DEAD MAF. etc.

Forum Jump:

Users browsing this thread: 1 Guest(s)