I always set mine up with the headlights on.

In addition, it makes sense to clean the ground point in the engine bay (under the brake booster in LHD cars, no idea how to describe the location on RHD cars) and the point where the engine ground strap is attached to the chassis. (12/99)

I have encountered the same thing, with idle speed dropping with the headlights on, but only when I'm drawing to a halt (i.e. on the brakes). It's noticeable because, apart from hearing and feeling it, I can see the headlamp beam wobbling on the back of the car in front. As soon as I come off the brake the idle speed comes up again.

You need to check your light switch monitor function (a facility of the diagnostic connector)

How would I go about that? (BTW I am sort of familiar with the diagnostic connector. I have used it to check for fault codes.) (12/99)

Check switch monitor functions just as you do for ECU fault codes, but use the MEN terminal rather than FEN. Your LED (or other test indicator) should be on with the lights on (or with the brake lights on). There are quite a few other functions as well - full details in my manual (advert!) These signals tell the ECU that the engine is getting a bit more load so that it can compensate. (12/99)

I know this might sound a bit stupid, but I think on my car at least, it's related to the speed of the heater fan. I was experimenting today, cos mine too started to drop revs and lights dim when stopping. This also happened when I gave it a quick rev when stationary. Only seems to happen in cold weather (car not cold though) and mine did cut out a few times earlier in the year too.
Conclusions believe it or not:
When fan not on or at position 1, revs drop right down and lights dim when stopping or after revving.
When fan speed 2 or more, revs don't go below idle and lights don't dim.
Also when car stationary and at idle and fan speed is reduced from 2 to 1, revs drop on their own for a few secs.

What's going on?! Any ideas? Give it a try and post your conclusions to the list so I know I'm not going mad! (12/99)

Sorry to bang on about my diagnostic connector interrogation manual again, folks, but those of you who have invested will know the answer to this!

Basically, there are a number of situations which change load on the engine, either mechanical or electrical, when you do something like switch the lights on or engage a gear or press the clutch, etc. So there are a number of switches which tell the ECU what you are doing, and this helps the ECU compensate by enrichening the mixture or changing timing or increasing idle speed, etc.

One of these switches is connected to the blower knob (and incidentally another to the A/C switch if fitted), but it only tells the ECU that you're on blower speed 2,3 or 4. Go on blower speed 1, and the ECU doesn't know, but when the engine is cold, things are a bit stiff, and the mixture is critically balanced so the extra load on the electrics is enough to cause a noticeable drop in idle speed compared with having the blower on "off". So this is quite normal.

If the engine slows down significantly on speed 2, etc, the ECU switch sensor may have failed. You can check the operation of all these switches via the MEN terminal in the diagnostic connector box - full details in the manual!!! (NB see the donutz site!) (12/99)

I think you will find that you have a problem with the idle airflow controller or the actual valve. Shows up in winter particularly. Can't remember the acronyms off hand - bloody old timers disease The controller is under the throttle body and the valve is on the manifold. Pull the plug off the controller when at normal op temp, the revs should rise to somewhere between 1200 and 1500 if all is OK.

The other thing that is very important is that the computer idle revs and the throttle butterfly position match exactly at idle when the engine is at normal op temp. If not the airflow when cold could be to little - rich cut or too much - lean cut. Pull the plug off the throttle body at idle, if the revs rise or fall the butterfly positon is not correct, adjust the idle screw under the little rubber cap and replace the plug. Repeat undtil there is no detectable difference plug on or off. If one is to err, err on the side of faster rather than slower; only a tad mind. (12/99)

I have this problem also. When the engine's cold (first 10 mins of running usually), when I come to a stop the revs drop to 600RPM or so.

I checked this on the miata.net page, and apparently there is a TSB that mentions this problem on the Mk2. The TSB says its something to do with different metal compounds somewhere in the engine (Sorry, I can't remember all the details specifically now) and they heat up at different rates which causes the problem. Something to do with the valves etc.

Its never caused my car to stall so I've never really bothered. Also, my location means I have plenty of open roads to drive down in the morning before I hit any traffic, so the car is usually warmed up before I hit the stop-start-stop jams in the morning.

Have a look on miata.net in the Technical Service Bulletin areas and you'll find the TSB for this problem. Print it out, and take it to your dealer. They should fix it for you free of charge I guess? (12/99)

I was talking to this bloke down the pub last night..........

...........and he turned out to be an electrician type person, who had just read a book on engine management systems - mmmm interesting.

Anyway he said the EMS continually monitors the electrical load on the whole car, and adjusts the fuel/air etc accordingly. Having the fan on more would increase the amount of fuel used, therefore less likely to drop revs/stall. He also thought that there must be a fault somewhere in the system for the revs to drop / stall. He suggested wiggling the fuses, getting the wiring to the heater fan checked out, but when he heard it happened to other MX5's (anyone else found this happening?), he said it is more likely to be an inherent fault and you wouldn't be able to do anything about it. Looks like we'll have to drive about with the fan on number 2 then. (12/99)

I found this earlier today, as I too have this problem, though it has never (quite) stalled the car. Has anyone on the list ever managed to get a UK dealer to check their cars against this service bulletin? It looks like it was soured from Mazda North America, so would (probably) have been distributed around the world including UK dealers?

I also noticed the report about the loom around the injectors? (http://www.miata.net/garage/tsb/wiring_recall.html). Has this ever been found on a UK spec MK2? (12/99)

Check your base idle.
Turn off all electrical equipment.
Open the diagnostic connector and you'll see a diagram inside the cover.
Connect the contact labelled "TEN" to the one labelled "Gnd" using a few inches of wires.
Start the engine and check the idle speed.
Adjust the screw in a metal cylinder sticking out of the top of the throttle body to get the idle to around 850-900 rpm.
Sometimes this screw has a rubber bung over the top of it. I find that I need about 1100rpm as a base, but I do have the extra load of a supercharger.
Note that you cannot properly adjust the base idle without connecting Ten->Gnd.
Unless like me your Throttle Position Sensor needs adjusting! (2/00)

For the past month or so my car has been suffering from the idle drop problem.

Come to junction, brake, slow down, stop, revs drop, cough splutter, car idles like normal. I've tried all the base idle settings etc but without success. I have noticed however that every time the car gets a service the problem goes away for quite a while. I've assumed it's a tuning problem because of this.

Earlier this week, I happened to check my oil. I hadn't checked it for a while and knew it was probably due a top-up. The dipstick didn't even register any oil so I filled it, got about 1.4L in it, DOH!

Should check a little more often but it's rained every weekend I haven't been busy recently. Since the oil topup the idle problem has completely gone. I can only assume the problem is oil or at least oil pressure related. With a nice maximum (actually slightly above) level of oil the problem has gone away completely.

Check your dipstick folks. (12/00)

I appear to have cured my idle droop by setting the base idle back to 900 rpm. I don't have a calibrated rev counter so used the dash one, combined with listening to the engine note as I connect and disconnect the jumper across the GND and TEN connections in the diags connector. Basically I guessed that the ECU would set the idle to the correct rpm and that if I matched that rpm when the ECU was in diags mode I'd be close enough.

I ran the car up to normal temp, and then with it sat ticking over on the drive, connected gnd and ten. The revs rose quite a lot and the dash rev counter showed nearly 1500 rpm. I knocked this back to a point I thought was about right and the droop seems to be cured. I guess the ECU was over compensating when knocking the revs down to the correct idle speed causing the droop? Car was serviced in the summer and I have only changed the oil since. It is due one now but the idle droop has only been there a month or so - is it possible for base idle to creep up by itself or not? (1/01)

Basically I guessed that the ECU would set the idle to the correct rpm and that if I matched that rpm when the ECU was in diags mode I'd be close enough.

That's how I used to do it and it works well. With a supercharger I found you need to set the base idle higher due to the very slight drag of the blower.

I ran the car up to normal temp

*Very* important and the part often missed. The engine really does need to be up to full heat and have been running for a fair time.

It is due one now but the idle droop has only been there a month or so - is it possible for base idle to creep up by itself or not?

Yes, I used to find it needed a tweak every so often. I also found that Mr. Mazda often set it wrong at services, and I suspect this was due to setting it without the engine full warmed. I always used to set the idle again myself after it had been for a service, but now I do my own service so I always know it's right. :-) (1/01)

If you've checked everything else (base idle, timing, plugs, leads, fuel injectors, sensors (ie no fault codes), oxygen (lambda) sensor) the next resort is to strip down and clean the Idle Speed Control valve.

It sits under the throttle body. Bit of a fiddle to get out, but just spraying cleaner into it doesn't work. Strip it down and clean everything with carburettor cleaner. That will restore the plunger which actually controls the air going into the engine to proper operation - it's probably well clogged up. Then give it a good blast (driving wise) (1/01)

Just had my 1997 Monza (1600cc) in for its MOT. Didn't realise how comprehensive the emmission tests are.

The mechanic hooked my car to his SUN Engine Tester and everything was fine apart my the tickover speed which was 827 rpm as opposed to the limit of 800. He was helpful and attempted to adjust the throttle stop under the damper but to no effect. The machine refused to issue a pass printout.

He then carried out the rest of the test without problems. I noticed however that he had keyed into the SUN machine the type of car as Mazda 323. He said this didn't matter as the engines were the same but would try the engine tests again, This time he keyed in MX5 and low and behold the SUN tester printed a pass report.

He did say that the tickover was varying though and the SUN tester must have caught the right moment. I must admit that the needle on the car's rev counter sits just to the high side of 800. I notice in the manual that a faulty ISC can cause tickover problems but since until my visit to the MOT station I hadn't noticed anything at I will just leave the thing alone. Any advice from someone who understands the idle system would be appreciated though. (6/01)