I have a 1984 Chevy Celebrity station wagon with a 2.8L GM V-6 with a
carburetor on it. The engine was rebuilt about 25,000 miles ago and passed
about 10,000 miles ago. I also replaced the EGR valve at that time and
cleaned out the EGR
passages (in order to get it passed)
A few weeks ago I bought a used OTC 4000E scantool with the idea to use
on both the 84 Chevy and a 95 T&C that I own. When I got the scantool I
decided to make a
baseline of both vehicles.
The 95 Chrysler came out fine, lots of fun with the scantool testing out
it's features, lots of fun
statistics and info from the engine and transmission off the CCD adapter.
When I plugged it into the Chevy, however, while I got lots of fun
statistics and such out of
the engine there, the one thing that caught my eye was that the scantool
insists the engine is
runing rich. It even has a red light on the scanner that illuminates marked
Anyway, before I go digging, is there anything obvious that anyone
remembers about these
particular vehicles that I should check first? My gut feel is it's in the
carb somewhere, as
the carb was not replaced when the engine was rebuilt.
Other than that the Chevy runs fine, although it has a lot worse fuel
mileage than I expected
on a 2.8L, that is why I think that the scantool is right.
What is the MAP sensor voltage, key on engine off, and with
What is the MC dwell reading at: idle, fast idle (2200 or so),
during light cruise at 40-45 MPH?
What is the O2 sensor doing?
Does anything change if you disconnect and plug the canister
purge hoses (plug the carb side)
What happens if you ground the MC dwell lead (green single bare
connector underhood) mainly to O2 sensor and RPM?
TPS at curb idle?
Both choke vacuum breaks working?
What color are the sparkplugs?
The early E2SE carbs had problems with the rubber tip of the MC
solenoid pintle breaking off and then not having adaquate seal
when energized to go lean, there is a test in the service manual
for checking this with a mity-vac and a piece of W/W hose and
some jumpers. The A/X body carb chassis in general were sensative
to having the charcoal canister get saturated with fuel from
people trying to pack the gas tank when filling up. The main
metering system in this carb runs the M/C solenoid in parallel
with an adjustable main jet, someone may have monkeyed with it.
The idle mixture -is- adjustable and should be adjusted to around
35* varying slightly + - 5*. Carb gaskets can not leak at all,
mainly the base gasket and the throttle body to main body
gaskets, vacuum leaks here really wreak havoc. IIRC the base
gasket was actually an electrical EFE heater grid, they liked to
melt, many were replaced with a plain gasket, check the service
manual as to what belongs there, the only way to cover up a
failed/missing EFE grid is to over compensate with a richer
mixture. Carb is somewhat sensative to float level, easily
measured with a special gauge that drops in where the mesh screen
is for the bowl vent on the carb top, or can be measured the
conventional way once the carb top is off. Gauge should still be
available thru Thexton, Lisle or one of the other specialty
You have to use an AC Delco PCV valve for proper PCV metering,
aftermarket PCVs only screw things up.
Leaking needle and seat.
Cracked fuel bowl or other monkey handed mechanical malfunction.
Although often claimed to be the left half of a Quadrajet, these
carbs really were crap.
Oh yeah, ignore the little red light on the scan tool, the carb
when set up correctly is biased slightly rich. Everything is
working fine when the MC dwell is varying between 25 and 40
degrees under the conditions I described above.
MAP sensor voltage is .52 with key on and engine off, going to 3.3-3.7 volts
with key on
and engine idling.
3 degrees at idle, about 20 degrees at fast idle, about 29-35 degrees
This is the odd one. At idle at about 625-650 RPM the O2 sensor voltage
stay under .20 volts, but it fluctuates throughout that range. But at
anything above idle,
ie: cruising fast idle, whatever, the O2
voltages are constantly all over the map. The scanner appears to get an
second, and the O2 sensor can swing from .08 volts in 1-2 seconds, to .78
volts in the next
second, then to .45 volts in the next second, then to .12 volts in another
second then back
to .52 volts in another second. And it was still doing this even after a 45
minute freeway drive.
The rest of these tests will have to wait until the weekend, unfortunately.
But, is the O2 sensor
reading normal? I'm kind of thinking that it isn't. Certainly the exhaust
stream couldn't be that
I have the service manual and I'll look this one up.
Now that one I know about as the shop that dropped in the rebuilt engine
discussed that with me and he said the same thing. With my engine, when
separated the carb that heater/gasket was not in good shape, I guess the
gasket must stick to the manifold and carb. But, the heater part of it
did work. So when they put it back together the mechanic did reuse the
heater/gasket and added form-a-gasket sealer and such to make sure it was
well. He did warn me that any attempt to separate it again could
destroy the heater/gasket.
When I cleaned out the EGR passages to get it through emissions, I did have
lift the carb. (the passages were completely plugged and I actually had to
a hammer and long wood drill bit with a sharp tip to clear the impacted
from the passages, I had a shop vac going in the manifold throat to catch
pieces) Fortunately though the sealer seemed to have the action of instead
the gasket being permanently glued to the manifold and carb, it actually
quite easily without damage. I did make sure when I put it back that the
fully seated. It also passed emissions then so I am pretty sure that gasket
Carb is somewhat sensative to float level, easily
Okay, that's normal for your system which actually uses a
differential pressure sensor instead of a MAP sensor.
Three degrees is too low, the computer is seeing a lean condition
and commanding the MC solenoid to richen things up.
If you can correct it to be 35 degrees by backing the idle
mixture screw out, things should be okay. If not, start looking
for vacuum leaks with the carb to manifold base gasket being the
about 20 degrees at fast idle,
Close enough to correct given the idle dwell.
Damn near perfect, a tick low but again, could be influenced by
whatever is driving the dwell low at idle.
Expected given the MC dwell readings, shows that the system is
trying to make a correction, 3 degrees (6 actually) is the
absolute limit on a rich command.
Good, that's what supposed to be happening.
Your ECM is a very slow baud rate unit, what is displayed on the
scanner is apt to lag behind anything you might observe live on a
DVOM or labscope.
The GM CCC systems do not try to target a supposed -ideal- O2
sensor voltage, things swing around a bit just like they do in an
EFI system. If the system goes lean periodically, the cat-con is
assured a supply of oxygen, if the system goes rich periodically,
the cat-con is assured enough HC to keep the fires lit.
So, yes, the exhaust stream will vary WRT O2 sensor voltage, when
it doesn't, that's when the spend money light gets turned on.
The problem with re-using the gasket and gooping it is that you
can still suffer vacuum leaks thru the laminations of the gasket,
but before you do anything else, try adjusting the idle mixture
richer (out) to obtain a MC dwell of around 35 degrees slightly
Shame on the shop that installed the rebuilt engine for not
cleaning the manifold properly, it's in both of your interests
that chunks of carbon do not break loose and jam a valve open,
plus they should have verified the EGR operation to insure that
the ring lands and bearings in your new engine didn't get the
shit pounded out of them from the inoperative EGR.
I'd do a little sniffing with propane around the base of the carb
to see if in fact there is a vacuum leak there.
But honest to god, aside from the low MC dwell at idle, this
thing is pretty close to spot on.
Motorsforum.com is a website by car enthusiasts for car enthusiasts. It is not affiliated with any of the car or spare part manufacturers or car dealers discussed here.
All logos and trade names are the property of their respective owners.