Idle Dip

[Fuzzlet]

My SE has started to get the traditional idle dip. Its a 2004 model with 43k kms on it. After doing a search I've found a few possible problems/solutions:
-Cleaning the throttle body and/or intake manifold
-Cleaning the EGR valve etc
-Cleaning the idle control equipment etc
-Re-adjusting the idle control
-Faulty O2 sensor (however the car is only 4.5 years old so I'm hoping it wouldnt be this)

I'm gonna have a look around the car in the coming days to track this problem down, but was wondering if anybody has had this problem caused by a different problem to what is listed above?

[Alex]

Happens to me sometimes, normally it's when the temp gauge in centre but the engine isn't really "warmed up" yet, seems to go away after 1-2 enthusiastic runs from a standstill (say 3/4 throttle from 0-speed limit shifting at around 4-5k)
I think it could be due to the O2 sensor giving false readings because the inbuilt heating element is past it's due by date but that's just my opinion.
Mine isn't too bad (most of the time not noticeable) so I haven't done anything about it.

Others have also had some success with unpluging the battery for an extended period of time (say 30+ mins) to "restart" the ECU, but I haven't tried this.

[blackster]

Before jumping into the most sinister causes, how about looking for any vacuum leaks.

[doctor_bob]

Sorry for the ancient thread bump, but my car seems to be starting to do this as of the last few days, pretty much as described by Fuzzlet (not an SE though; NB8A). I was ignoring it but it dipped kinda low tonight; touched the 500 rpm mark and hovered around 600 for a second or two before returning to normal; gave the throttle a bump after the first few times it did it until it went away...

I'm not very experienced at troubleshooting these types of things or working on engines, most of my car work has been confined to regular maintenance type stuff. I've found guides I think I'll be able to handle for re-adjusting the idle control and cleaning the EGR valve.

Not sure how to approach the other items on that list, or if there's anything else to check. Some advice would be very welcome.

So, let's assume what looks like the worst case there; how much roughly does a new oxygen sensor cost? Is this the sort of thing someone with limited experience could perform?

[93_Clubman]

So, let's assume what looks like the worst case there; how much roughly does a new oxygen sensor cost? Is this the sort of thing someone with limited experience could perform?

Aftermarket replacement O2 will cost about $100 - OEM will probably cost 3-4 times that.
The appropriate universal Bosch 4 wire O2 should do the job, & from memory Okibi & Tbro posted something on the best one to use & he also has an NB8A:

viewtopic.php?f=29&t=25609&hilit=sensor&start=30

viewtopic.php?f=29&t=25609&hilit=sensor&start=15

viewtopic.php?f=76&t=37547

Easy to replace if it's a straight swap of 4 wire O2 sensor for 4 wire O2 sensor. From memory early NA8 was 3 wire, so replacing it with a 4 wire involved minor work.

[hks_kansei]

Just remember that you need to use a special kind of sensor safe anti-sieze on the threads. (I think some brands give you a small sachet in the box)

Apparently some kinds can make fumes and cause incorrect sensor readings.

[johnnyreble]

Yes I had this problem on my NA6,
After a lot of research I discovered through a friend in USA that it is actually a lag problem.... The TPS switches off fuel to the engine at closed throttle position ( decel switch, this is purely for emissions ) when the engine reaches idle speed the ECU switched fuel back on and there can be a lag between hitting idle speed and the fuel being switched back on. and this causes the dip.

However on the NA6 it's a simple fix .. Disconect the red wire from the ... TPS ( decel switch ) the result is a smother idle with no dip at all, On the NB I don't know which wire is the decel switch, however if you increase the idle speed to 1000 to 1100 rpm using the throttle stop screw ( not trying to fool ECU via gnd 10 ) you should get rid of the dip and this will also confirm that it is in fact lag in the system. Why this lag occurs is still a mystery

I hope this is of some use to you

Johnny

[hks_kansei]

WOuldn't disabling the fuel cut on closed throttle cause worse fuel consumption?

Not necesarily at idle, but whenever you get off the throttle (ie: coasting in gear) you don't want fuel being injected into the engine to be wasted, or possibly just cause it to run stupidly rich and ruin the cat.

[doctor_bob]

Thanks for the replies. Probably not going to have much of a chance to work on it until the weekend (plus it was behaving quite a bit better today...)

if you increase the idle speed to 1000 to 1100 rpm using the throttle stop screw ( not trying to fool ECU via gnd 10 ) you should get rid of the dip and this will also confirm that it is in fact lag in the system.

That confused me a bit; so are you saying I should avoid following this guide when adjusting the idle? What is the alternative method? Is it the same as in those steps; as in do everything there minus bridging GND & TEN?

Forgive my ignorance

[johnnyreble]

Yes forget gnd 10 and adjust it with the throttle stop screw not the idle screw ...try it you can always put it back, and no it doesn't increase fuel consumption, well not enought that you would notice, Mine has been running like that for some time.

Johnny

[93_Clubman]

Bob, as you probably have it's worth noting that JR is specifying this as a fix for an 1989-1993 NA6, as did Rob Spargo in his Know Your Car article.

If you haven't already, it would also be worth having a look at how other NB8 owners have solved idle dip: search.php?keywords=nb8a+idle+dip&terms=all&author=&sc=1&sf=all&sk=t&sd=d&sr=posts&st=0&ch=300&t=0&submit=Search

search.php?keywords=nb8b+idle+dip&terms=all&author=&sc=1&sf=all&sk=t&sd=d&sr=posts&st=0&ch=300&t=0&submit=Search

[doctor_bob]

From everything I've read... I think it's probably the oxygen sensor based on my car's symptoms. I saw a link here the other day on how to test if it was functioning within normal parameters with a multimeter....So much for time to work on it this weekend though; family stuff with fathers day chewed just about every bit of time I have left after housekeeping

[project.r.racing]

sounds more like a IAC issue if it is just a slight rpm drop for 1-2 seconds. normal when they get old and dirty. oppon decelleration or when the aircon compressor kicks in etc.

[doctor_bob]

Thanks for the response project.r.racing

Some quick googling suggests this could be rectified by cleaning it; what are your thoughts on this vs replacement if it turns out to be the culprit?

Also can you suggest a reliable test? There was something I found - although I think it was for an NA - suggesting to disconnect the connector and test the resistance between the terminals, which should read between 10.7 & 12.3 ohms. Another one (*sounded a bit dodgy) suggested detaching the IAC valve while keeping the electronics plugged in, and then covering the hole with the engine running; removing one's hand from the hole should result in the valve closing, covering the hole should result in it opening, according to the writeup.

It goes away completely after a few "spirited takeoffs" as with Alex's old response, also once fully warm it will not happen again until it fully cools down (ie turning the car off then back on while still warm will not result in idle dip). The car will idle smoothly until I start driving it; if I turn it on and idle until warm the revs are steady until I drive, stop, and put it out of gear; now with this in mind if I get it fully warm before stopping the first time it will not happen at all.

Hasn't been as bad as it was that one time it dipped to 500RPM the night I posted. Barely noticable now. *scratches head*

[doctor_bob]

Begin ramble:

My father and I had a bit of a look at it today...

So firstly some updates to my story, the car was due for a regular service and being a bit timepoor I took it to a friendly mechanic. When I got it done the guy threw some injector cleaner in the petrol tank (didn't ask for it actually, not sure what type or if it could be harmful to mx5s...?). Lots of black sh!t was ejected from the exhaust. After running on the injector cleaner for about half an hour it started to develop a rather disconcerting (set of) problem(s); when applying the throttle it would give normal response to about 10% but then bog down and felt like no extra power was added until squeezing it to a bit past half way open at which point ALL OF THE POWER would come at once (very scary driving it home that day as it was raining, almost lost the tail on a few roundabouts as the power kicked down unexpectedly ). Whenever this was present it would also idle extremely rough and dip quite low, almost to the point of stalling (but not quite, didn't actually completely stop at any point).

After this presented itself initially it would be fairly consistent in being able to be reproduced: When turning the car on it would start to happen once the oil started to come to temperature - not before - and remain for an indeterminate amount of time (quickest it went away was about 5 minutes, longest was over an hour). Both the bad idle and the bogging throttle would stop happening "all of a sudden" and then the car would run perfectly normally until the engine is stopped and turned on again. Interesting thing is it didn't matter how long the car was off for - even stopping it for a few seconds and kicking it over again and it would re-appear straight away (unless long enough to cool down, in which case it would return with the temperature of the engine...)

I drove it for a few days to see if deteriorated or improved, and whether running the tank of injector cleaner through completely and refilling would help. Nothing changed at all.

Anyway I think we've narrowed the issue down to a faulty EGR valve, but I'd be interested in some opinions;

It was the first thing we were going to look at (being the easiest) and when we pulled it apart the first thing we noticed was that the piston in the valve was sticking in a few places and that it did not move through straight/flush as there was a bit of play in it. When soaking it in carby cleaner there was a bit of leakage through the piston to the outside of the unit. Needless to say there was quite a bit of gunk in there, we didn't actually get it all out. I'm planning on taking it off and soaking it overnight to loosen it more. Perhaps there was a bit of grime that was filling some of the small gap creating play and when the injector cleaner came through it was blasted out, causing the piston to have too much play.

When firing it up after reattaching the idle was perfectly normal and then drove perfectly for the first 5 minutes, then it happened again for a few minutes, then went away again. Second time it was there but went away very quickly (under two minutes) and there is an overall improvement in the idle... so what we did might have helped, assuming this is the culprit. The last thing dad sprayed in there was some WD40 (I asked if using this was wise, he seemed to think it was ok & knows a sh!tload more than me so I didn't argue), which may have lubricated it for the first few minutes until the residue all burnt off... the piston felt smoother when we put it back together anyway. I have some theories why it might go away after a while too, regarding polishing the surfaces from the solenoid hammering away at it constantly...

So the reasoning behind thinking this is to blame; This was the first time it had started at temperature and NOT presented the problem since first encountering it. The bogging down, if I had to describe it, feels like like something is trying to open but sticking then suddenly coming loose, and in a tactile sense the piston in the EGR valve seems to stick in the same way. Of course I can't test this theory without replacing it... having said that, if there's play in the piston and it's sticking this is something I should replace anyway, right...? So I'm thinking one way or other I'm probably up for a replacement... Some advice would be awesome

Sorry for the long rant. I'm not mechanical people, I'm computer people. But I'm trying dammit