Warm misfire

[Eddypunch]

When the engine is cold and warming up it's fine. Soon as it's been at temp for a couple of minutes it's starts to misfire at like 2500rpm.
Thanks

[Brianmoooore]

It would really help if we knew what engine you have!

[Eddypunch]

Ahh sorry it's a m20 2.0 auto

[Brianmoooore]

Age? Massive changes to the M20B20's management and external bits throughout its production run.

[Eddypunch]

1989 it's an aswell auto

[Brianmoooore]

Look up "testing blue temp. sensor" in the wiki, and check its resistance with the engine stone cold and at full running temperature AT THE TERMINALS OF THE ECU PLUG.
If the hot running figure is too high, check out the condition of the wires under the rubber covers of the 7 pin round plug and socket hidden under the inlet manifold.

[Carmo13]

A failing ignition coil usually only starts to play up when hot.
May be worth testing this.

[Brianmoooore]

Failing crank position sensor is also temperature sensitive, but the first thing to establish is the health of the signal reaching the ECU from the coolant temp. sensor, as in my last post.

[Eddypunch]

Put a new blue temp sensor in and it was good for 2 days, now it's started doing it again, just not as bad.

[Brdjo]

What spark plugs are you using? When did you last change them? Is engine run on LPG?

[Eddypunch]

I put some in last week think there ngk.. no just petrol

[Eddypunch]

I'm going test the blue temp sensor at the ecu later

[Brianmoooore]

Put a new blue temp sensor in and it was good for 2 days, now it's started doing it again, just not as bad.

DO NOT THROW PARTS AT THE CAR. Only replace anything based on the results of testing.

[Eddypunch]

ah ok yeah il test this one and let you know what it's saying

[Eddypunch]

Not very good with a mutlimeter, but here goes. When the engine is cold I'm getting 1.090 and when it's hot I'm getting .246 and at the sensor it's 936 when warm

[Brianmoooore]

Hot reading is about correct (slightly high, but this could be down to the delay between stopping the engine and measurement).
Cold reading is very low, and, if the engine was stone cold, is a cause for concern, but wouldn't give the symptoms you're reporting. Was the new temp. sensor genuine OEM or aftermarket?
Since you went ahead and changed the blue sensor, you will have already disturbed the wiring it was important not to disturb before testing, so you need to take the plug and socket under the inlet manifold apart and check it physically, including under the rubber boots on the back of them.

[arrisbmw]

interesting what you say, why is it important to test BEFORE you disturb the wiring.
I was thinking maybe i could disconnect the plug under the inlet manifold 1st, to check my ohm readings cold and hot of the blue temp sensor. therefore not disturbing the ecu plug. I need to work out what pins in the connector , is it a screw type like the x20 or simple pull push ? I remember you previously stated its a weak point in the wiring and can be liable to corrosion. thanks

[Brianmoooore]

ECU lives tucked up out of the way in a warm, dry cabin - an idea that some French manufacturers could do well to adopt, so there's unlikely to be any 'poor connection' issues there.
The blue temp sensor and connector under the manifold live in a much more hostile environment, so are much more likely to encounter problems, especially when the latter has a known design fault. It was subject to a BMW TIS, but this came at a time when the majority of E30s had left the dealer servicing network, so the most remain unmodified.
The plug and socket are a miniature version of the C101 engine loom connector and the diagnostic socket.

[arrisbmw]

O.K thanks .So does that mean its a screw type ?, i,m not familiar with the c101 connector.

[Eddypunch]

Tight ok thanks.. It's seems to be what it likes.. one day it's fine, the next it's missing when warm again. Starting to annoy me. I don't know what to do next, maybe get a m50 dropped init!!

[paultv]

Just a thought and easy to verify:

I had something like this, but missing at 3000 to 4000 then clear again, turned out to be a burnt out suppression resistor in no5 plug lead - spark normally jumping the resistor, but randomly not always and worse when hot.

Measure resistance of each plug cable from dizzy to plug cap - should be 2K if I remember correctly, just check they are all the same resistance anyway to rule this out.

Paul