Okay I'm usually pretty good about figuring these things out, but my truck has me stumped. 1988 Mitsu SPX, Federal Emissions, manual transmission.

Carb has been rebuilt 3 times now by the same shop. It runs about the same(terribly). Here's what has been replaced: 02 sensor, fuel pump, fuel filter, engine, catalytic convertor, carburetor.

If I probe the TPS I show .6 volts at idle, and 4.6 at full throttle and a smooth transition in between. If I apply vacuum to the EGR valve, the engine dies. The hot air inlet and the choke both work fine.

At full operating temperature the O2 sensor voltage stays at .8 volts - it's definitely running rich. If I unplug the coolant temperature sensor(the 2 prong one) with the engine running I hear a click from a relay, but see no change in the way the engine is running.

The carb shop tried to "fix" my problem by grounding the yellow wire of the MCS to the frame. It passed emissions with this bypass hooked up. It also made the thing totally run like crap.

Having removed that and readjusted the idle, if I touch that wire now to the ground the idle immediately goes to crap, and the o2 sensor voltage starts to fall. So I have concluded that grounding the MCS leans out the mixture, based on what the o2 sensor does. So it almost looks like the computer is either bad(I could be the first one to have that happen reading the other messages....) or there is another sensor somewhere in the feedback system that is keeping the computer from having control.

My 4 cyl dwell meter hooked up to the MCS wires at idle stays at 45 degrees. It says at 45 degees without anything hooked up as well, leading me to believe there is NO signal being sent to the MCS. Can anyone tell me if that means the ECM is commanding the carb to be full-rich? No signal at operating temp either means the computer is brain dead or something is convincing the computer that things need to stay rich(another sensor somewhere that is bad?).

Anyone have any idea as to what the heck might be wrong?

Chris

If you need an ECU I have some, both Cali and Federal... <img src="/forums/images/graemlins/cheers.gif" alt="" />

manual shows .25v at closed throttle. Adjust TPS. I have an FSM .pdf for a carb 87 mitsu. Want a copy?

I suspect either a bad feedback solenoid valve (meters about half the fuel thru the main jet well to control mixture via ecu), or bad fuel cut solenoid valve.

I can't find any error code generation in a carb ecu - you have to use a checker or vom to check harness connector voltages.

Is there 3 wires on the coil - terminal? One to igniter in the dizzy, one to the tach, and one tach signal to ecu. If the last one is missing, the ecu gets no rpm signal.

If you plug the canister to carb hose, does the idle improve? Bad bowl vent valve, not in rebuild kits.

Yes, please <img src="/forums/images/graemlins/wink.gif" alt="" />


Okay, the carb shop said that they replaced the MCS. They did do something with it because the wires were cut and spliced with crimp-on butt connections, which were total crap. I opened up the harness and soldered/shrink wrapped the connections.



That's what I've concluded as well..there is no CEL, and where the early FI would let you read out some pulses with a voltmeter, the carb model apparently does not have that feature.



I don't know, but I will check tomorrow.



Same here, I will check this. I assume you are talking about the rubber hose that says "To Carb" at the canister right?

Here's what we found out tonight. I used a jumper wire to short the two pins of the 'temperature switch'. The one with the pins that look like |-. No change. So I pulled one of the lines from the vacuum tee on the intake(the one below the carb) to introduce additional air that wasn't being metered by the carb..and the oxygen sensor voltage immediately began to fall.

If I open the throttle slightly the voltage across the MCS starts to fluctuate, as though it is trying to find a good dwell range based on the o2. But the o2 sensor reading never changes, and after about 5-6 seconds of the thing fluctuating the voltage goes constant again. It looks like the computer is trying to go into closed loop and eventually it dumps out when it doesn't see the o2 sensor swing.

Chris

PM me an email addy and I'll zap it over. 5meg.

The MCV is NOT the primary fuel controller. You need to check the FBSV, which manages fuel per the ecu. Only m/t trucks have the MCV, and I haven't figured out yet what it does from the manual. FBSV should line up with the main (dside) barrel and the wires come out the float cover.

You have to disassemble the connector to remove the FBSV from the cover. Manual shows how.

Get a can of carb cleaner and start spraying the hoses, carb base and intake to head flange. If there is a vacuum leak the idle will smooth out. I found out on mine when I put my hand over the top of the carb and slowly close it off and the RPMs came up (I must have a massive vacuum leak! <img src="/forums/images/graemlins/zombie.gif" alt="" /> ). Anyway, great way to find vacuum leaks.