87 Dakota stuck in open loop?

My 87 Dak is stuck in open loop and performance is poor. Stalling at stops, diminished power,
hesitation on acceleration, etc. All these probs vanish when I disconnect the fuel mixture
solenoid in the Holley 6280 feedback carb and the damn thing performs better than new! seriously

When connected, the fuel mixture solenoid in carb is running stuck at 20% duty cycle when engine
rpm is high enough for computer to engage it. This is what tells me that the system is stuck in
open loop

Here is what I have ruled out as the problem:


1. engine temperature sensor - checks out within range of factory specs. I also checked it by
installing in its place a resistor of 720 ohms to simulate a warmed up engine. Closed loop
operation did not take place.
2. wiring of engine temperature sensor to computer - no shorts or opens
3. O2 sensor - checks out within range of factory specs


I am in the procees of sifting through Alldata's info on this vehicle to see what sensors
trigger closed loop operation and check them out one by one. There are not many on this vehicle.

I am beginning to suspect that the computer may be unable to function in closed loop mode and is
limping by in open loop. Is this possible?

Thanks for any and all help.

Jack
California

 

The "lean burn" computer on is not capable of keeping up with this odd fire
engine with some miles on it. There was a reason dodge went to throttle body
fuel injection the next year and multi port soon after. I spent many years
tiring to keep mine running well. My local dodge dealer drilled out the idle
adjustment plugs to get more adjustment and that helped, but what really
made the difference was eliminating the computer and getting a mopar
performance distributor and a accell super coil. That made it run great til
I sold it at 175,000 miles.

KS

 

Hi Kevin,

I feel that I am getting tantalizingly close to understanding what is going on with this setup.

I confirmed that the mixture control solenoid in the carb is working properly. I removed the
worn rubber grommet on the end of the plunger and replaced it with a suitable piece of plastic.

I then disconnected the battery to drain the computer memory and hooked up the duty cycle meter
again and started it up. This time, the readings were different. The dwell slowly increased as
the engine warmed up and fluctuated with rpm change like it is supposed to. Readings swung about
the 50% mark like they are supposed to. This hadn't happened before. Then, after about ten
minutes, it stuck at 20%. The dwell reading always went directly to 20% after warmup and stuck
there. This seems to be pointing to the fact that someting is throwing things out of whack as
the engine reaches a certrain temperature. The only sensors that I know of that affect fuel
mixture are the engine temperature switch which tells the computer that the engine is warm
enough to go to closed loop operatrion, the O2 sensor and switch portion of the switch/relief
valve which sends air fromm the air pump into the exhaust manifold when the engine is cold. This
fresh air causes the O2 sensor to read a lean mix and compensates by enriching the air fuel mix
for better running while cold.

I just replaced the O2 sensor ($30) and I am waiting for the engine to cool down, have
disconnected the computer and I am waiting for the engine to cool down to see if the new O2
sensor has any impact on the problem.

I clamped shut the hose that sends air to the exhaust manifold yesterday, but that did nothing
to change things. That was before I got the idea to drain the computer memory. If the new O2
sensor does not make a difference, I will drain the computer memory and clamp the air hose to
the exhaust manifold again. If the problem turns out to be a faulty computer, I have a line on a
"remanufactured" one from Standard Motor Products through rockauto.com for $99 plus core.

But first I want to exhaust all the other potential causes.

My ace in the hole will have to be something along the line of what you did. The odometer just
flipped over to 100,000 miles the other day and the damn thing is in GREAT condition otherwise.
It just needs to pass California's smog test, which just got more stringent since the last time.

Jack

 

Jack,

I'm not 100% sure but I don't think the mixture control solenoid and the
computer will do much for idle emissions if thats all they test.

Kevin

 

Kevin,

This year California made the test more rigorous. Vehicle is put on a dynamometer and tested at
25 and 40 mph in addition to idle. Otherwise, I think you are right. At any rpm, as long as
there is no load and no movement, the system reverts to open loop and feedback mechanism is
bypassed.

I've made various measurements under various conditions and the results are interesting.

First of all, after complete warmup, with the feedback solenoid connected or unconnected, the
new O2 sensor reads at the upper end of its voltage range, .95 volts, when the throttle is open
even a little, which is indicative of low oxygen content in the exhaust gases which points to a
stuck rich condition. When the throttle is closed, the O2 sensor reading is at the low end of
it's range, almost no voltage at all, indicative of high oxygen content which points to a lean mix.

The funny part is, when the truck is moving under load and the O2 sensor reading is indicating a
rich mix, the duty cycle of the feedback solenoid reads only 20% indicating that the computer is
responding to a lean condition and is feeding more gas into the mix. This is where I get a
drivability problem in that the truck stalls when coming to a stop or even just slowing to round
a corner or driving in stop and go traffic. If I set the idle higher to compensate, the engine
will sometimes race at idle going up to 1500 to 2000 rpm. BUT, with the feedback solenoid
disconnected and FULL gas flow through its jet, it drives with no problems what so ever. Idle
can be adjusted for consistent rpm at a reasonable rate and the engine never dies.

The only way that the computer can change the air/fuel mix is through the feedback solenoid. The
only other major thing that the computer does is to advance the spark when needed using an
intake manifold vacuum sensor in the form of a diaphragm.

The warmup phase is another story altogether.

Jack