flat spot on acceleration?

It could be a malfunctioning oxygen sensor, when they die the ECU thinks the engine is running lean and richens theair/fuel mixture--bogging is a sign of an overrich mix.

When O2 sensors age they often have to be hotter than the 650*F minimum operating temperature of a new one to produce a useful and accurate signal, this could explain why it's only a cold or cool engine problem.

jetwash-cleaned

Hi,

many thanks for the suggestion, we did try disconnecting the sensor to see if it made any difference...but there was no change...

could it still be the sensor?

If there was no change when you disconnected the O2 sensor then there's even a better chance that it is the problem. Unlike many sensors that use electricity to measure something, O2 sensors actually generate electricity in proportion to amount of oxygen in the exhaust gasses, as compared to the amount of oxygen in the atmosphere (O2 sensors are vented to admit air to the "other" side of the sensor element). This is approximately 0.45V or so at the optimal 14.7:1 air/fuel ratio (this is not technically correct, but it's close enough for this discussion) ...

If the voltage is above 0.45V the mixture is judged to be rich and the ECU reduces the amount of fuel being injected, if the voltage is under 0.45V the mix is judged to be lean and the ECU increases the amount of fuel. The sensor itself consists of a cone shaped platinum plated, porous zirconia, ceramic element that is exposed to exhaust gas on the outside of the cone and atmospheric air on the inside of the cone (through the earlier mentioned vent). The difference in oxygen content between the exhaust gas and atmospheric air triggers a galvanic reaction that generates a voltage--when O2 in the exhaust is low (indicating an overly rich mixture) the difference between O2 in the exhaust and plain 'ol air is large and the generated voltage is large, the voltage drops as the amount of O2 in the exhaust increases (indicating a lean mix).

Obviously when you disconnect the sensor the output is 0.0V, making the ECU think the mix is lean, which makes it try to increase the amount of fuel being injected, which makes the engine run far too rich, which makes it bog when you try to accelerate (which inherently feeds more gas to the engine anyway).

As O2 sensors age the platinum becomes contaminated or corroded and the output drops, so your observing no difference when you disconnected it actually supports the diagnosis of a failed O2 sensor.

-cliff-

In which case the engine management system should flag a fault code.

Except that the eng> many thanks for the suggestion, we did try disconnecting the sensor to see

Disconnecting the sensor will have no effect on cold running. It will make the motor run rich when hot and it will cause the computer to log a fault with the oxygen sensor.

Unlikely.

You need to clear the computer to get rid of the spurious oxygen sensor fault you've induced, then go for a drive, then query the computer to see what faults it logs. Instructions on how to do all this will be in the workshop manual, thought doubtless someone has put them on the net somewhere too.

Thanks for the advice guys!

We have tried querying the ECU as instructed by the manual but no lights appear anywhere...(see my new post - just where is the engine check light!)

Surely now we have disconnected the o2 sensor we should now be getting a fault code, like you say, even if we didn't get a fault code before...so we're just wondering if we've got a dead bulb, or a disconnected wire to the check engine light, but until we know whereabouts this light is...we can't check the darn thing!!!

gosh, can you tell it's been a long day...

Jo.

The ECU on the early Miata (I have a '90) seems to be a brain dead sort of fella'--the O2 sensor was broken off by the PO's fumbled attempt to change it out when I got mine, there was no fault code logged.

You are of course correct that the O2 sensor is ignored when the engine is cold (and also above 4K RPM), however the original poster also indicated that the problem occurred when the engine was warm started after 1/2 hour or so--I'd still take a close look at the O2 sensor.