y33 idle jerk / hesitation discussion (lets fix this!!)

[miata007]

I actually had my battery checked one time at Kragen. The guy said the battery was bad and needed new one. It's been over 16 months, and is still starting the car on one try every morning. Could this cause the stumbling problem?

Some updates on my 97Q. Rather some occasion stumbling, it is happening more frequently now. At the end of a road trip (20min or more) after all warmed up, the car starts the stumbling continuously (very often)when I am at a stop/light. Not sure what to do now. It's at 170,000mi. I am hoping it will last for another 30k before I donate the car.

007

[UM97Q4.1]

Miata, me and you are on the same page here, same year, same mileage and exact same things going on. Mine has definetely gotten worse over the past couple weeks, and once it gets warmed up it is absolute crap.

Bean, how is the alt. holding out. I wish it was that easy for me but as I stated I just had the alt. replaced. Could it be something that is killing the alt.?

Bullit, how far are u from Marietta? Might be time to see Q45tech.

[96Qowner]

Ok, we have several possible issues:

1.) Dirty intake path.

2.) Intermittent MAF performance.

3.) Unreliable waveform from battery/alternator.

4.) List more if you think of them.

It's very difficult to solve for 3 or 4 variables. I suggested hooking up real-time Consult software and checking the log when symptoms appear. Brian suggested cleaning the intake path. Dennis suggested reading the waveform. The MAF could be swapped out with a known good one. To my knowledge, no one has done any of those - correct me if I'm wrong.

If the intake path is dirty and produces the symptoms AND the alternator or battery is bad, and also produces the symptoms, then sometimes a cleaning will solve it and sometimes a new battery will solve it and sometimes a new alternator will solve it. But if you don't fix ALL of the possible problems, you still get the symptoms. Seems fairly obvious, and since erratically trying things hasn't worked, I suggest doing some of the things that haven't already been done.

Actually, a straight question to Dennis might be useful. Dennis - how does T3 deal with this symptom? Haven't they experienced it?

(BTW, my name's Richard.)

[Q45tech]

We diagnose systems record reams of data and meditate on the dozens of voltages.

Many times we keep the cars a few days have techs drive them back and forth to home [20-30 miles each way] with Consult II attached.

We have replacement parts that we swap and eventually zero in on the things that don't throw codes. [coils, ecu, maf, etc etc.]

Most codes require multiple misfires WITHIN a number of minutes or they are ignored.

I sure if you pay for 8-16 hours of diagnosis we can find any problem, then of course the parts replacemnt cost.

[maxnix]

There can be multiple causes with the same symptom. If you don't have a correct new as OEM base to start with, diagnosis may be well close to impossible as there may be more than one mode of failure occuring simultaneously, and even different modes simultaneously in different cars with the same or similar stumbling symptom.

The title of this thread presumes there is only one cause for the stumbling symptom, which is patently untrue as revealed by the previous posts on this topic.

[Mopar]

is it possible that the hesitation comes from the injectors? I really suspect them

[bullittandy]

There can be multiple causes with the same symptom. If you don't have a correct new as OEM base to start with, diagnosis may be well close to impossible as there may be more than one mode of failure occuring simultaneously, and even different modes simultaneously in different cars with the same or similar stumbling symptom.

The title of this thread presumes there is only one cause for the stumbling symptom, which is patently untrue as revealed by the previous posts on this topic.

I didn't clean my intake and the last time I felt a stumble was about 30K and that was for about 2 weeks.

What I don't understand is why the experts won't say "The solution is unclear." Saying that the solution is to replace every worn part so that the engine is OEM new is about the dumbest and least helpful thing I've heard.

[UM97Q4.1]

I am no mechanic by any means, but I am willing to bet money that it is the MAF. If I had the money, time, and knowledge to do this myself I would. In your mind if my car has a problem it is obviously due to neglect on my part. I love this car, have had it for 7 years. This problem has been going on at least the past 5 years.

My friend was a manager at a shop (nationwide chain) and I would say that in 05 he had this car more than I did trying to figure this out. I have already spent a ton of money on this car and had many different methods applied to remedy this. Basically anything he recommended I would say go for it.

[Q45tech]

I think what most owners and shops without a Consult II miss is that you can set up a series of 12 voltages to monitor and have the system active awaiting your push of a button to record.................the system records 3 seconds of PREVIOUS data PRIOR to your pushing the button.

So assuming you have coordination when the stumble occurs you push the bottom and you have the voltages associated with the event.........nothing out of the ordinary you monitor 12 different ones eventually if you understand the ecu workings totally in your head you'll find the culprit.

Same with spark in cylinder you can record all the primary pulses and SEE how the spark firing gets reflected [superimposed] to the recovering input.

Many dealerships don't have all the tools they need to 100% diagnose much less independents.

I've had to engineer and develop weird sensors or go outside the industry to find ebay substitutes designed for something else and adapt for vehicle use.

http://www.picotech.com/Pico_T...g.pdf

http://www.picotech.com/auto/e....html

The problem is spending $10,000 for ocassional use equipment will never be cost effective and an almost impossible sell to management in small shops just as a $6500 Consult II useful for just one brand is tough for a generalist non specialty shop.

[bullittandy]

With all due respect, how does this previous post help anybody with a 97+ Q45 solve this weird stumbling problem?

What's that expression? Can't see the forest for the trees.

[96Qowner]

Heheh, Andy ... with all due respect, you're merely dismissing the advice.

Let me say for the experts - "The Solution is Unclear".

I did some thinking about this on a long drive yesterday. Let's start calling it a "syndrome". It's becoming obvious that there isn't ONE cause for these misfires. If anything, I would say it's the fault of the Y33's ECU programming. For one reason or another, the VH41 will misfire. Sometimes, it's at idle, sometimes it's on a long highway upgrade in 4th gear. Sometimes it's accelerating mildly in city traffic.

I haven't seen any evidence that it's fuel-delivery related, so we look at electronics - sensors, connectors, etc. We look at the air-delivery path. Reread what Q45tech said:

We diagnose systems record reams of data and meditate on the dozens of voltages >... < when the stumble occurs you push the bottom and you have the voltages associated with the event.........nothing out of the ordinary you monitor 12 different ones eventually if you understand the ecu workings totally in your head you'll find the culprit > ... <we swap and eventually zero in on the things that don't throw codes. [coils, ecu, maf, etc etc.]

Isolating ONE component requires bringing each one up to spec, either by swapping or cleaning, etc. Unfortunately, if MORE than one is suspect, they ALL have to be right to solve the problem - kind of like how replacing your struts won't fix everything wrong with the suspension.

I have another suggestion: Texasoil swears by a dielectric connector treatment called Stabilant-22:

Stabilant-22 ... Miracle cure!!

Someone (or everyone) should clean their MAF and treat the connector and all other connectors they can get at with Stabilant-22. If there's any improvement, that will be significant. It's cheap.

Someone should get a BG intake cleaning - same idea - improvement says that intake condition is a factor. It's not expensive.

Etc.

Here's an interesting thread that I found by taking Brian's advice and searching old threads (hint, hint):

Wife got tired of stumble, so off came intake for a kerosene cleaning, and same for EGR and IAC valve. Used only carb cleaner on lower runners. Seems to have cured the stumble. Will let you know later. Dremel helps a lot, but inside of penum is really rough cast with lots of nodules on the surface. Cleaned as best I could.

'97 Q45, stumbling idle

[maxnix]

If anything, I would say it's the fault of the Y33's ECU programming.

Here's an interesting thread that I found by taking Brian's advice and searching old threads (hint, hint):

'97 Q45, stumbling idle

I certainly don't follow that line of reasoning. The ECU doesn't change the data on the maps it follows. I still believe it is component faliure or degradation, singly or in combination, which causes these symptoms. The task is to determine which components, which is made easier by refurbishing the degraded ones before attempting diagnosis.

Thanks for reminding everyone about the original thread that renders this one so superfluous.

Perhaps with some judicious editing they could be combined for future searches?

[96Qowner]

I certainly don't follow that line of reasoning. The ECU doesn't change the data on the maps it follows.

I'm just thinking it's too sensitive, and/or doesn't deal with erratic sensor input very well. It seems to me that the stumble is being created/allowed by the ECU itself - needs a tweak in the map.

Brian, near the end of that thread you said your stumble returned. Did you solve it a second time?

[Q45tech]

Maps are passe since OBd2. Equation coefficients are filled in and solved then added unknowns in 5 other equations to arrive at injector open time.

About 25 equations must be processed and solved each 1/100 of a second.

MAF and O2 voltages can trim equations by +-20% EACH.

Unfortunately without removing and flow test injectors [not at high but idle pressure] one doesn't know the variances in flow.

[maxnix]

I'm just thinking it's too sensitive, and/or doesn't deal with erratic sensor input very well. It seems to me that the stumble is being created/allowed by the ECU itself - needs a tweak in the map.

Brian, near the end of that thread you said your stumble returned. Did you solve it a second time?

Yes, by replacing a faulty coilpack. I'll go back and edit.

[maxnix]

Maps are passe since OBd2. Equation coefficients are filled in and solved then added unknowns in 5 other equations to arrive at injector open time.

That explains the dearth of OBD II ECU upgrades, since the hard coded equations them selves would have to be modified.

I always thought OBD II just relied on more sensor inputs.

[beancan]

The stumble came back... With a vengance. It was doing it bad enough I thought for SURE the check engine light would come on.. But it didn't.. Then it quit missfiring.. (#$(@$(# car.. I have an extra coil pack.. Guess ill play musical coil.. Its a pain..

[bullittandy]

I didn't clean my intake and haven't felt a stumble for a long time.

I also replaced my spark plugs about 5K miles ago which means that I had to remove the coils which I would assume would have killed any coil on its way out-but who knows.

I did just replace a coil on my wife's V70 and it stumbled at idle but drove decent at speed. This is a 5 cylinder which is not smooth to begin with so a dead cylinder was very noticable compared to the Q which is likely much smoother with a dead cylinder-even more so when the cylinder is intermittently dead.

[Q45tech]

OBD2 equations are encrypted, so the only way reflashers get the code is by bribing someone inside or an ex employee thus things cost significantly more.

The G35/Z has promoted a whole industry in replacement aftermarket programmable ecu for non street emission cars.

True OBD2 requires back up sensors in case primary failed much like the 3 sensors in 90-96 CAS assembly. Or the use of a MAP with a MAF.It takes 2 -16 bit ecu, one just for diagnosis part and 1 to do calculations.OBD2 in fail safe can run on either ecu when MIL is on as a slow backup just to keep engine running.

http://140.134.132.124:8080/ds...8.pdf

http://www.arb.ca.gov/msprog/m...3.pdfh ... me/qqq.pdf

[maxnix]

http://140.134.132.124:8080/ds...8.pdf

http://www.arb.ca.gov/msprog/m...3.pdfh ... me/qqq.pdf

Thank you for the great links, Dennis. I will read them all!

And I like FirstLook. Too bad it doesn't weem to interface readily with a laptop.

[miata007]

Trust me guys, I have tried using the search feature and you do not get every bits of info related. That's why we need pointer from people who has seen it and are nice enough to share.

007

[Q45tech]

It is very simple to solve ONCE you find out the exact cylinder that is misfiring by syncronizing the exhaust pulses to a reference and adding in the time delay from exhaust port to exhaust pipe [measurement point].

One only has 4 pulses per rpm with a V8 and thus at 660 rpm idle = 165 pulses per minute or 2.75 per second.

If you can feel a misfire, you and electronics can hear it at exhaust pipe ---------the shape [acostic waveform] of the sound will be different.

All you have to do is syncronize the display so you can deterimine which of the 8 in a row sound pulse is #1 cylinder and then using firing order [1,8,7,3,6,5,4,2] then determine which one is misfiring.

Think about the time delay where the exhaust valve opens ~~130 degrees after the spark plug fires and stays open for 240 degrees after that.Hint each peak of exhaust pulse sound is exact 90 degrees of rotation from every subsequent one [assumming rpm are steady]

How long in time does the exhaust pulse take to get from exhaust valve thru header, thru cats , thru muffler to end of exhaust pipe?

Duh you use the Consult to turn of individual cylinders to calibrate which is which.

All we are doing is creating a different system than OBD2 uses [O2 sensor] to measure misfires................the O2 reacts weirdly to a misfire the exhaust sound changes with a misfire.

You can even look at the primary waveform [coil input] as a misfire will reflect back from the spark plug thru secondary to primary coil and show wierdness.

Hundreds of patents on detecting misfire, using Google " misfire detection" I get 150,000 hits.

A power balance test can give you insight since weaker vs stronger cylinders might give you a place to start looking. Do an in gear [lower than 1200 rpm] brake on loaded [AC on/off] sequence power balance test.

[beancan]

I put in another can of BG44k because the car was doing it bad.. bad enough to shake.. It quit all together and hasnt even HINTED at it.. Stupid car, Im about ready to enter it into a drift event (aka, carbon removal)

I dont get it

[miata007]

Q45tech,Thanks for the detailed write-up. Unfortunately, I don't have the brain power to comprehend and tools to test what you suggest. I wish your shop is near me and I would pay and let you figure it out. I don't trust the guys at my dealership.

007

[96Qowner]

Just was reading this thread:

1999 Q45 Coil Pack Problem ?

It would be a good idea for each of you to see if you can produce the problem long enough to throw a code - get a reading off Consult or other software. For instance, in gear, A/C on, motor hot, etc.

Gotta locate an offending cylinder, somehow.

[Q45tech]

I can't wait until members start buying used cars with 2 coils per cylinder and 2 different injectors per cylinder with partial direct injection [Lexus today, Nissan's future]. 5 separate CAS sensors. The new 5.0 liter engine will be a nightmare.

If these 10 year old designs are giving you fits just wait.

The dealers are having problems finding electronic inclined technicans who can do the work for so little pay.

[maxnix]

Gotta locate an offending cylinder, somehow.

Bingo! Any good scanner will tell which cylinder is misfiring, but it won't tell you why.

Which is why you have to eliminate all degradation from operation variables - i.e., return the intake path to as new condition, plugs, filters, etc.
Modified by maxnix at 9:19 AM 12/21/2007

[bullittandy]

Bingo! Any good scanner will tell which cylinder is misfiring, but it won't tell you why.

Which is why you have to elimanate all degradation from operation variables - i.e., return the intake path to as new condition, plugs, filters, etc.

I don't think OBD-II scanners can find a misfire without a CEL-am I wrong?

Let's be realistic, cleaning the intake is a waste of time. Replacing plugs, and cleaning the throttle body and replacing air and fuel filter is smart.

[bdijanni]

Bean, Whats going on at your house?? Figure this out yet? I was thinking about some kind of jumping bad coil signal or something complex like that. Maybe thats how people are fixing their problem, they buy 8 new coils. Q45tech said that they usually fix the problem by replacing all 8 coils and replacing the fuel rails. What do you guys think? No one ever said anything when I asked does anyone with this exact problem have an aftermarket ignition system???

[texasoil]

I have been trying to resolve an intermittent miss,skip,stall with an active customers car for several months now. Changed TPS first after I would repeatedly get stumble/missfire/miss at light throttle application (holding brakes) Confirmed intermittent TPD voltage with scope. Went away for 3 weeks--cane back-different kind of miss--on moderate acceleration-seemed like fuel starvation, changed filter (it was partially plugged) but engine seemed down on wide open power. Jumper fuel pump to high speed-lots of racket-replaced obviously on last legs fuel pump, checked FP controler connectors and had lots of wide open power now, no miss. Came back in 3 weeks-another miss-different again smooth as silk, perfect power etc for days, then shi## hits the fan for a few seconds, even dies at light. Restart and PERFECT! WTF? Banged on MAF, connector etc. to no effect. However, substituted known good (off my car) went away several weeks-'defective MAF worked fin on my car-go figure.Canme back,another miss-intermittent, but read bad when there. Maybe 4 days just fine, then SHi#$.

Checked engine out on 'brake-stand'-noticed lots of lifting-engine mounts shot-then found throttle position/loading where engine would lift up, and stumble down, recover back up-and stumble down,etc.etc for as long as I kept it loaded-varying throttle position around no effect on pattern so TPS OK--HAS to be a wire/ connection being moved and broken,re-making!. I had 'wiggled' all the connectors and MAF connections, but it was OBVIOUSLY a connection issue as nothing else was changing, only the amount of engine lifting up! Removed cleaned and used Stabilant 22 on left side (driver side) ignitor , missfire went away never to return. ( I went ahead and did all the engine connectors anyway-all of them, AND battery terminals and main lead connectors. Been 3 weeks now. Owner had been using cheap no-name gas, I had also used BG-44 in tank, it helped some.

Summary--a plethora of factors can be involved in 'misfires' this '95 low mileage (61K) had been on diet of cheap gas the last 6 months, fouled up intake valves and fuel injectors, AND TPS was faulty AND fuel pump worn out AND bad connection on ignitor AND bad connection on power leads at battery+ low voltage = maybe fire, maybe not. Maybe trigger injector, maybe not.