4 replies to this topic

[andatche]

I'm having trouble solving an irregular misfire at idle on my SPI at the moment and wondered if anyone had any ideas. Some information...

When starting the car from cold, for the first 20 seconds or so it seems to idle fine every time (I'm guessing during this initial period the ECU is using some fixed, preset fueling/ignition map?). After this it begins misfiring irregularly but quite frequently, sometimes several times in a row but usually a few seconds apart. Each time it misfires it spits a small amount of fuel out of the exhaust but no blue/white smoke or anything. If I depress the accelerator very quickly then the engine will bog down/stall completely or backfire through the inlet manifold. Once the engine has reached operating temperature the problem seems to be a bit better. It still misfires occasionally but doesn't spit fuel out of the exhaust and no longer bogs down/stalls under rapid depression of the accelerator.

The engine is a freshly rebuilt & run in 1275 with new piston rings, new valve guides, valve stem seals, oil pump, water pump etc, Kent 274 cam & duplex timing gear, standard forged rockers, standard airbox with a K&N element. Compression is good on all cylinders, it isn't using any oil or water and under normal driving conditions doesn't seem to have any issues, though is perhaps running a bit rich as under hard acceleration or engine load there is a smell of petrol inside the car.

So far I have checked/changed the following with no success:

• Valve Clearances (Checked and rechecked after about 100 miles, set to a "loose" 12 thou)
• New spark plugs, HT leads, Dizzy cap and rotor arm (After some mileage the plugs don't seem to show signs of rich or lean running, neither did the old plugs)
• Changed the coil for a known good one from another SPI
• Replaced all the vac hoses and cleaned out the fuel trap
• Removed the ECU and cleaned the MAP port with some cotton wool on a radiator overnight
• Checked the lambda heater relay by swapping it with the electric fan relay (working)
• Checked the lambda readings on a code reader (Swings between 200mv-500mv when warm, swings less when cold. Getting a little slow perhaps - will replace)
• Checked the other sensors (Air temp sensor is faulty but unplugging it completely doesn't make any difference. CTS seems fine)
• Replaced the stepper motor (Was having erratic idle issues before, since fixed the problem)

I shall replace the air temp sensor and lamda sensor on pay day (I guess these explain the rich running but would they have a severe enough effect to cause the misfire? Especially as the lambda value is almost ignored during cold start?). I also have a new fuel filter which I shall fit at the weekend. Meanwhile, does any one have any other suggestions?

Cheers.

[zorgster]

sounds like the problem I have with our 1996 Sidewalk... spi...

Took it to Fast Cars in Croydon where he hooked it up to a diagnostics... found faults in map sensor, coolant temp sensor and throttle potentiometer. When cleared and he finally experienced the misfiring and loss of power again, only the CTS fault returned...

I took it home and drove with one or two problems along the way... i took off all the vacuum pipes and found a 8mm crack in the elbow that attaches the red pipe to the back of the injection housing... i've taped it up and ordered new ones... tho the tape (good electrical tape) hasn't resolved the issue..

i drove last night to pick up someone from the station, got there ok .. when leaving as soon as i pulled out and put into 2nd gear touching the gas pedal killed the engine... eventually i could start her up and had a choppy ride home...

i've got a new CTS to put on too.. if i can reach it .. apparently if the signal from the CTS is erratic, it can mess up the ECU.. especially if the values it is emitting are within the acceptable values that the ECU responds to... so one moment you get a good mixture the next it's too rich...

could also be the wiring between the sensor and the ECU...

i also found a slight kink (but not yet kinking (not when cold)) in the vacuum pipe... and wonder whether when the engine got hot... and the pipes warmed up .. perhaps the pipes soften and bend at the kink thus cutting off the vacuum... ?

I'm having trouble solving an irregular misfire at idle on my SPI at the moment and wondered if anyone had any ideas. Some information...

[andatche]

Thanks for the input Zorgster,

I have a Crypton ACT code reader and according to the live data on that I'm getting a good MAP reading, the CTS reading seems accurate and TPS readings seem reasonable too. The only fault logged was the inlet air temp sensor which I know is faulty. Unplugging this (should make the ECU use its default "limp-home" value - 30C iirc) doesn't seem to make the problem any better or worse though so I suspect it's unrelated. The only thing I have noticed is that when the misfires occur the ignition advance seems to briefly jump to ~27 deg, then settles back to ~15, I'm not sure if this is the cause of the problem or an effect of the ECU trying to correct whatever is causing it.

I have replaced all the vac pipes with new ones and there are definitely no kinks or cracks/leaks on any of them.

[zorgster]

Another Sidewalk owner (that bit is not relevant) ... suggested that one thing overlooked in the diagnostics is debris in the petrol tank... a piece of paper or fabric floating around that occasionally gets sucked over the exit pipe ... then releases itself to let fuel out ...

i'm not suggesting it's the cause ... just passing it on... i'm not quite sure how to eliminate that as a cause of fuel starvation ...

what about the fuel pump relay?

Edited by zorgster, 20 August 2009 - 04:38 PM.

[andatche]

The fuel pump doesn't seem to make any change in pitch or switch off briefly when the misfire is occurring so I assume the relay is OK. I shall double check the tank and make sure it doesn't contain any debris. I also have a new fuel filter which I will use to replace the existing one at the weekend.

I have a feeling the problem is fueling/mixture/sensor related as if it was electrical I wouldn't expect it to improve when the engine warms up and the ECU switches to open loop. One slightly worrying thing, when I started the car this morning there was a small amount of white smoke from the exhaust, though it was raining last night so it could just be moisture in the exhaust evaporating. By the time I'd driven a mile (I stopped to check) there was no sign of smoke at all and the car doesn't appear to be using water.