NA6 Rough Startup idle

[Tumbles]

I can't really explain this too well in words, so I have linked a video because it's multimedia. (Video sound is terrible, and I might actually get a better video tomorrow.)

Basically, when the car starts up, it starts up rough and doesn't sound like a normal startup. Also takes a few seconds of cranking for the engine to fire up.

Reasonably new to cars overall, so I wouldn't know how to diagnose what it is.

Any ideas on what's going wrong would be appreciated.

Video: http://s132.photobucket.com/user/Tumble-W33d/media/20131022_160544.mp4.html

[Tumbles]

No-one has any idea?

I've read threads on a rear coolant sensor failing, but I'd prefer to know that that is what is going wrong before I spend money on fixing it.

[Okibi]

Start with the obvious plugs n leads.

The fuel filter might be worth replacing if it's long in the tooth, will be worse if you often drive it on empty.

[Tumbles]

I bought the car just over a month ago.

Oil+oil filter done.
Fuel filter replaced already. (It had a hole in it from rust - real funny story including me getting petrol in my eye.)
Spark plugs are in the 5 heat range, they're all good according to my inspection, but I will be changing to 6 heat range soon.
Leads were voltmeter tested and they're in great condition.
I also replaced the Rocker Cover Gasket at Nitro Dann's workshop and there's no leaks there - although I doubt this would have any issue with startup anyway.
I compression tested the engine (fully warm - oil pressure stable after 20 minutes of driving around) and got 170 PSI in the first 3 cylinders and 180 PSI in the 4th. I think that's normal though.

Anything else I should check?

[gslender]

Looked at the video.

The car was cold and she started and barely idled near 900. The first thing that comes to mind is the two types of fast idle systems.

One is connected and controlled by coolant and is on the intake manifold. Take it off and test in cold water (it should be open) and then try hot near boiling water (it should be closed).

The other is harder to get to and is under the TB. I think it is probably this one. It is electric and is controlled by the ecu. It also has coolant going through it, and an air gasket that commonly perish.

Start by checking the connector is clean and secure.

It will require removal of throttle body and should be checked by connecting a 9v battery to one of the connectors. It should rotate around really fast. Spray with WD40 and cleanup and replace that rubber gasket.

G

[manga_blue]

Here's the NA6 idling bible for you:
http://www.miata.net/garage/isc.html

Also, testing spark plugs leads at 1 or 2 volts with a multimeter does not give you any real indication of whether or not they are failing at 20,000V when they're in use. You still have symptoms of lead problems.

[Tumbles]

Looked at the video.

The car was cold and she started and barely idled near 900. The first thing that comes to mind is the two types of fast idle systems.

One is connected and controlled by coolant and is on the intake manifold. Take it off and test in cold water (it should be open) and then try hot near boiling water (it should be closed).

The other is harder to get to and is under the TB. I think it is probably this one. It is electric and is controlled by the ecu. It also has coolant going through it, and an air gasket that commonly perish.

Start by checking the connector is clean and secure.

It will require removal of throttle body and should be checked by connecting a 9v battery to one of the connectors. It should rotate around really fast. Spray with WD40 and cleanup and replace that rubber gasket.

G

This seems to make sense to me. I'll try this out next week, since I'm flat out this week.

Thanks.

Here's the NA6 idling bible for you:
http://www.miata.net/garage/isc.html

Also, testing spark plugs leads at 1 or 2 volts with a multimeter does not give you any real indication of whether or not they are failing at 20,000V when they're in use. You still have symptoms of lead problems.

Idling bible seems really complicated. I'm going to need to spend some time with it. Thanks for the link!

How should I test them? Also, I don't get any misfires or anything whilst driving? Should that not happen if plugs are gone?

[gslender]

I've seen rare cases where cold leads / plugs would misbehave but act fine when hot and dríven hard.

I would though call that rare and think the low idle is really abnormal even with a bad lead it should be hunting and trying to idle at 1100 when cold. I think you've got a idle valve stuck.

A bad coolant sensor at the rear of the engine (under the coil pack) is another potential, but easily sorted by removing the connector and seeing the engine run rough when hot. The default ecu behaviour when sensor disconnected is to assume a cold engine. If the engine behaves the same when sensor disconnected then the sensor might be at fault.

G

[toppertee]

I had idle problem with mine, I ended up finding that I had a couple broken bolts/studs on the inlet manifold .
Not saying this is the problem.

[Tumbles]

I had idle problem with mine, I ended up finding that I had a couple broken bolts/studs on the inlet manifold .
Not saying this is the problem.

Everything looks quite unbroken thankfully haha. I will be quite paranoid however, and check it out once daylight comes about again.

I've seen rare cases where cold leads / plugs would misbehave but act fine when hot and dríven hard.

I would though call that rare and think the low idle is really abnormal even with a bad lead it should be hunting and trying to idle at 1100 when cold. I think you've got a idle valve stuck.

A bad coolant sensor at the rear of the engine (under the coil pack) is another potential, but easily sorted by removing the connector and seeing the engine run rough when hot. The default ecu behaviour when sensor disconnected is to assume a cold engine. If the engine behaves the same when sensor disconnected then the sensor might be at fault.

G

Okay. I'm having trouble understanding the bad coolant sensor bit of your post, so the following is to clarify your meaning. Sorry for being slow.

Default ECU behaviour when the sensor is disconnected is to assume engine is cold, and thus runs the engine rich. (Does not affect idle at all? Does not do anything apart from running engine rich.)
If I disconnect a good sensor, when the engine is hot, the ECU will think the engine is cold and run it overly rich?
If I disconnect a bad sensor, when the engine is cold, the ECU will think the engine is cold (solving the problem temporarily), but will result in rich running when engine is hot?
If I disconnect a bad sensor, when the engine is hot, what happens?

[Okibi]

For the slow start it wouldn't hurt to check your battery connections are good and your engine earth is too.

Have you tried with another battery in case the CCAs on your one are right down?

[Pokmx5]

You could possibly have an air leak somewhere... sounds like idle missfire.

[gslender]

A bad coolant sensor at the rear of the engine (under the coil pack) is another potential, but easily sorted by removing the connector and seeing the engine run rough when hot. The default ecu behaviour when sensor disconnected is to assume a cold engine. If the engine behaves the same when sensor disconnected then the sensor might be at fault.

G

Okay. I'm having trouble understanding the bad coolant sensor bit of your post, so the following is to clarify your meaning. Sorry for being slow.

Default ECU behaviour when the sensor is disconnected is to assume engine is cold, and thus runs the engine rich. (Does not affect idle at all? Does not do anything apart from running engine rich.)
If I disconnect a good sensor, when the engine is hot, the ECU will think the engine is cold and run it overly rich?
If I disconnect a bad sensor, when the engine is cold, the ECU will think the engine is cold (solving the problem temporarily), but will result in rich running when engine is hot?
If I disconnect a bad sensor, when the engine is hot, what happens?

I'm suggesting that if the sensor is really broken and not working it might be telling the ECU the engine is like -20 deg when in fact it is more like 15 degree (when cold and first starting) this would make it run really rich - too rich idle tends to make it run lower. A lean idle will run faster.

I'd still think you have an idle valve not opening or working correctly - that fast idle valve on the top of the intake manifold could be the problem.

G

[Tumbles]

Bumping an old thread because issue persists.

Fuel pump replaced with a 2nd hand one. NGK Plugs and leads installed.

Car has a long crank time and sometimes doesn't fire. I suspect fuel is the issue.

Fuel pressure regulator and fuel relay under dash are my next items to rule out. I've bought a rear coolant sensor but haven't installed it yet, since I'll do it at the same time.

Fuel tank is clean and new fuel filter is less than 2000kms old.

Anything else that I should check over?

[93_Clubman]

Anything else that I should check over?

Check Fuel Inj relay in the main fuse/relay box in the engine bay.

Also check for the 'blackwire' issue the inside the top of the fuel tank gauge/pump assembly: viewtopic.php?f=55&t=46425