Sentra 97 AFM or fuel pump? (or maybe something else)
Me
runs fine under low/normal load, but cuts out up hills etc. No CEL. No
black smoke - it just dies.
Have replaced filter as first step.
Used parts supplier has a bin full of used pumps for about US$60 each,
and doesn't sell many or hear of many failures, so he suggests that it's
the AFM rather than the pump or relay etc. He could be right, but it
doesn't seem like the AFM to me, as it's running very well/smoothly
under low load, not hesitating etc on acceleration etc and only starts
cutting out when it's been under load for 10 seconds or so.
I also checked MAF signal voltage, and that's looking okay (I think), at
warm idle, it's steady at around 1.4v, signal voltage increases and
falls smoothly, it's steady at about 2.5v @ 3,000 rpm. But, I also hear
that the AFM unit failures are related to internal dry solder joints, so
it might test okay but fail on road test.
On the other hand, the pump also seems to be operating as normal, can
hear it when the ignition turned on, and it's cutting out after about 5
seconds. (not sure if that initial 5 second run is just based on a timer
circuit though - the fact that it cuts out "as normal" might be
meaningless wrt fuel pressure)
I don't have a fuel pressure gauge, but the fuel hoses feel much softer
at idle than I'd have expected if they were at about 34psi (FSM spec).
Any suggestions appreciated for "next step"?
Should I go out and buy a fuel pressure gauge, or perhaps set up some
long leads for multimeter probe to see if I get some strange readings
from the AFM when it's on a test-drive?
Me
> I also checked MAF signal voltage, and that's looking okay (I think), at
> warm idle, it's steady at around 1.4v, signal voltage increases and
> falls smoothly, it's steady at about 2.5v @ 3,000 rpm.
Found spec in FSM (GA16DE):
1.0 - 1.7v at idle (in "N", no load)
1.8 - 2.4v @ 2,500rpm (in "N" no load)
Rechecked, unless it's an intermittent "dry solder" problem with AFM the
AFM is within spec.
cse...@mts.net
>On the other hand, the pump also seems to be operating as normal, can
>hear it when the ignition turned on, and it's cutting out after about 5
>seconds. (not sure if that initial 5 second run is just based on a timer
>circuit though - the fact that it cuts out "as normal" might be
>meaningless wrt fuel pressure)
>I don't have a fuel pressure gauge, but the fuel hoses feel much softer
>at idle than I'd have expected if they were at about 34psi (FSM spec).
>Any suggestions appreciated for "next step"?
>Should I go out and buy a fuel pressure gauge, or perhaps set up some
>long leads for multimeter probe to see if I get some strange readings
>from the AFM when it's on a test-drive?
>
shelf. When the car was running the lite would be on - when it
stalled the lite stayed on, meaning I was getting power to the pump. I
would jump out and squeeze the fuel return hose while cranking the
engine from under the hood. No fuel pressure pulse detected. Fuel
pump faulty. Replaced pump - problem solved.
On my kids car the same sort of symtoms. I wired a test lite as
before. Except no power to pump when it stalled. Located the fuel
pump relay and when it went into a stall, I jumped out, felt the relay
and it was way hot. Replaced the relay, problem solved.
Both situations were intermitant without any set times of failure.
And if I let the car cool down for 5-15 minutes it would clear up
until next time, Maybe 5 min or an hour.
I would also chech the Crank and Cam sensors. These should show up as
a "no crank signal detected" or "No crank reference". You might want
to get an OBDII reader and see if your storing any codes (The idiot
lite on the dash doesn't always come on for all trouble codes).
P
--- news://freenews.netfront.net/ - complaints: ne...@netfront.net ---
willshak
Don't dismiss a clogged cat converter.
--
Bill
In Hamptonburgh, NY
In the original Orange County. Est. 1683
To email, remove the double zeroes after @
robbie
My older Sentra had a bad problem that felt like fuel starvation. I
diagnosed FOREVER until it quit running entirely. Then i found out the
coil was bad. It had evidently been "going bad" for quite a while, but
I had checked it and it had a spark and was within specs when cold but
evidently did not work properly under load.Never a CEL in all that
time... If you have a junk yard nearby it would be worth your time to
swap it out. Instantly cured hesitation, a crappy idle, lower power-
the 1.6 runs like a sewing machine again.
willshak
I wasn't suggesting that the cat was the problem, just suggesting
another thing to check.
I had a similar problem with my 86 Pulsar. It didn't happen all at once
but just got worse over time.
It turned out to be a clogged cat. Replaced it and all was well.
Me
Robbie, I saw your other thread, and assumed, as I was with mine, that
AFM was the most likely culprit as it's a common fault, but it checks
out okay, the signal voltages are even, within spec, - and the typical
MAF failure symptoms of uneven idle / hesitation aren't there.
Bill, I've got a spare cat/exhaust manifold assembly here, but I doubt
that's it, as when it's not missing/cutting out, it's running very well
under load/high revs. (OTOH if it's coil/ignition related and I don't
get this sorted properly, I guess I could kill/overheat the cat)
Plugs are fairly new, leads etc seem good, and anyway when it cuts out
under load it's a very hard abrupt cutout on all 4 cylinders, not a
typical symptom of one or two bad leads or plugs.
I now think the fuel pump etc is okay, I removed back seat and cover,
and can feel the fuel hose pressurise and pulse when engine is revved. I
don't have a gauge to put inline to check pressure, but it's probably
okay. If I pull the fuel pump fuse under load, cutout from fuel
starvation isn't as sudden/abrupt as what's happening, it slows,
sputters and dies, rather than just completely cutting out "hard".
It is also only happening when engine reaches normal operating temp -
when it's cold/warming up, it's running perfectly. Danmed nuisance, as
it takes a couple of "laps" up and down around local hills for a test run.
This car (japanese domestic market model) doesn't have an OBDII
connector, it's not throwing a CEL and I can't check for (pending) codes.
It's making more sense to me now that it could be the coil,
breakthrough/arcing only when engine is warm/under load, as the ECU
changes ignition signal based on load and engine temp. The distributor
was leaking oil from internal seal, fixed some 20,000km ago and clean as
a whistle now, but it has had engine oil sitting in it, so that might
have eventually "got to" the coil (coil is inside distributor).
cse...@mts.net
>This car (japanese domestic market model) doesn't have an OBDII
>connector, it's not throwing a CEL and I can't check for (pending) codes.
No OBDII connector?? Seems to me that all cars {Asian at least} were
OBDII compliant starting in 1996. Prior to that they were propriatary
OBD using various methods for pulling codes.
Me
>> This car (japanese domestic market model) doesn't have an OBDII
>> connector, it's not throwing a CEL and I can't check for (pending) codes.
>
>
> No OBDII connector?? Seems to me that all cars {Asian at least} were
> OBDII compliant starting in 1996. Prior to that they were propriatary
> OBD using various methods for pulling codes.
>
> P
>
markets) I think OBDII became standard by about 2003 for most Japanese
manufacturers. This has a proprietary OBD connector (mounted in the
fusebox). There are some alternatives to OBDII readers such as:
http://www.blazt.biz/products/cable.php
(for some reason I'm getting a malware warning from Google Chrome for
parts of that site - I think it's an error, but be careful)
Me
>>> My older Sentra had a bad problem that felt like fuel starvation. I
>>> diagnosed FOREVER until it quit running entirely. Then i found out the
>>> coil was bad. It had evidently been "going bad" for quite a while, but
>>> I had checked it and it had a spark and was within specs when cold but
>>> evidently did not work properly under load.Never a CEL in all that
>>> time... If you have a junk yard nearby it would be worth your time to
>>> swap it out. Instantly cured hesitation, a crappy idle, lower power-
>>> the 1.6 runs like a sewing machine again.
>>
Bingo!
Bought a used distributor from a wrecker's yard this morning (pulled it
out myself from a low km wreck, and took the cover off to check it was
clean inside), swapped them over, set timing, and problem fixed.
(Mine wasn't hesitating on acceleration, or idling rough, which is
typical symptom of MAF going bad, but it was a fuel-starvation like miss
under load, only when warm, with no CEL).
Cheers / Me
Me
Bugger it - I should never have posted the above - premature smugness on
my part. Sure enough, later today on a slightly longer run, it started
showing exactly the same symptoms - so bad I was lucky to get home, as
it completely died a few times.
I have since cut the cover-plate from the MAF, removed the PCB,
resoldered all components / connections on the PCB, cleaned (carefully
with small artist brush and CRC carb cleaner) the hot wire and
temperature sensor resistor, reassembled, glued cover-plate back down
with RTV, and have taken the car on several truly long runs - problem
seems to finally be solved.
So dammit, it seems that it was the MAF / AFM after all.
I can return the used distributor, but it was quite cheap, from quite
low mileage car, and when I set timing with inductive light, at idle
speed the new distributor seemed to be much steadier/less "jumpy" than
the old one was, so although probably not the cause of the problems, the
old distributor was probably not very good / marginal anyway.
robbie
good catch on the MAF. dry solder joints suck! i can tell you are a
diy maniac, so glad you got it cured on your own without dealing with
a mechanic. I have heard so many people tell stories about even
dealerships throwing parts at a problem without a true diagnosis.
apologies to the GOOD mechanics on this group! BTW I bought a Consult
device (it is serial based - old tech) that plugs into the non-obdII
cpu port. it came with ConZult freeware version which doesnt do much
but will pull codes and let you see the timing. it cost me about $29
on ebay. Just search for Nissan Consult. No i am NOT the seller, just
a satisfied buyer. With the right software you could leave it
installed and access on the fly adjustment to fuel etc but needless to
say I wasn't really interested in it for that reason.
Me
Thanks - I see that the "consult" cables and software is also available
here, USB version, about US$80 or so.
There's a design issue with these MAFs, on GA15DE and some other Nissan
motors, the MAF is combined in the throttle body, so it gets hot. I
think on some GA16DE, the MAF is on the intake close to but separate
from the throttle body, and probably gets less vibration and less heat,
but they can still be a problem. It's running perfectly now.