22 replies to this topic

[Diller]

This past weekend at Road Atlanta I had a new problem that seemed to get worse over the weekend. The problem develops when the engine is hot and starts around 5K RPM. The problem is intermittent but showed consistency in turns 2, 4, 5, and 11. The problem would also pop up from time to time between 7 and 10A. This is my video from Sunday's race that shows the issue beginning during the first lap. Keep an eye on the tach in 11 near the end of the video that shows the worst.

 

 

I changed the coil pack, wires, and plugs before the race with known good parts (stole them from a friend running DEs) but that was not the problem as we put my parts on his car and he didn't see any of the same symptoms. My best guesses right now are fuel pump or cam sensor.

 

Another symptom is that I am getting a lot of noise into my tach wire that goes to my traqmate and autometer gauge that makes the tach jump to 10K during the engine miss. The tach wire passes near the coil pack so that may not be directly related.

 

Any help would be greatly appreciated.

[Ron Alan]

I was going to say welcome to the 99 world of misfires but you are in a 94? Anyways...Tom Hamptons post near the end of this thread is very interesting. My 95 has never missed but the tach signal jumps all over the place...

 

http://mazdaracers.c...he-missfirepop/

[Diller]

Yep, the car is a 1994. The plan is to swap the fuel pump, cam sensor, and check over the entire wiring harness for any broken/damaged wires. I just want to cover my bases so I can bring a box-o-parts to the next race or dyno session.

 

Another thought was a clutch slipping but I don't know if it would show up like this, especially without any clutch smell. 

[Tom Hampton]

Where did you tap in for your tach signal?

[Diller]

Where did you tap in for your tach signal?

 

 

The harness that goes to the test port on the drivers side of the engine bay near the air filter. It is then split again in the cabin between the traqmate and autometer tach.

[Tom Hampton]

Do you still have the stock tach? If so did you notice if it did the same thing?

Routing the tach wire past the coil pack is not a great idea. There is a lot of opportunity for emi coupling through there. Single wires like that are very susceptible. If the emi couples into your new tach wire it could couple back into the rest of the circuitry,including the ecu.

I doubt that's the main issue, but it's something to fix.

[Diller]

I agree going near the coil pack is not idea. I was hoping it was far enough away to limit the noise. The new tach sits in front of the OEM gauge cluster so I don't have a good view on it but I did not notice the jumping. I was planning on moving the tap to the wire near the ECU or shielding the wire where it is.

 

The jumping tach has not been noticed until the engine miss happened. I agree it needs to be fixed but I don't think it is a cause to the problem, just a symptom.

[davew]

I very much doubt that is fuel related. It is something in the ignition system.

 

First, what did you change on the car last, before the problem started.

Second, disconnect all the extra devices. Traqmate, gauges etc.

Third, carefully inspect all wires to coil pack, cam sensor along with plug wires and connections inside the coil pack. If any black residue, change both coil pack and wires.

Forth, tap on AFM with a 3/4" wrench (or similar sized solid object) with engine running. Don't hurt it but, give it enough of hit and see if the engine miss occurs when the AFM is hit.

Fifth, inspect wire connectors at ECU

sixth, ccheck all engine grounds, especially that hard to find one on the intake manifold. If you find a bad ground, be very suspicious of the coil pack and ecu going bad shortly.

 

dave

[Diller]

There was nothing changed between my last race at Atlanta Motorsports Park and this race at Road Atlanta. The only difference besides the track is temperature. AMP was around 35F and RA was closer to 80F. Coil pack, plug wires, and plugs were changed and the problem did not follow those parts to another car.

 

This car is a 1.8 so I don't have an AFM. I will be sure to check all the engine grounds as I may be pulling the motor to do some other maintenance items. 

[KW78]

The biggest clue I see here is that when the car misfires, the tach actually increases 2-3k RPM.  

 

I have had both a 1.8 car and a 99 car do this.  You might check your tach trace on the TM to see if it shows up there, one of these did.  

 

From memory, the ignitor function (which is on the fender on a 1.6) was moved into the computer on the 94-95, but on the 96-97 (and up) moved into the coil pack.  I didn't pick up what year your car is.

 

I would duplicate the problem again, then hit the connector at the ecm with DEOXIT and work forward under the dash to any and all other connectors.  Try it again.  If problem remains, swap in another ecm. 

 

Sometimes the act of swapping parts just exercises the connectors and solves the problem.  I see that far more often than actual bad parts, so if you get it fixed, put the old parts back in as a ABA verification of a bad part. 

 

Then, if you haven't already, MOVE THE COMPUTER from the rules idiocy of leaving it in the rain, heat and vibration.  I put them in the glove box and coil up the harness that reaches the rear.

 

THIS CLASS NEEDS TO BE ALLOWED OPEN WIRE HARNESSES, IT IS AN ALL YEARS PROBLEM!!!  IT IS ALSO TIME FOR A SPEC ALTERNATE CHEAP ECM, LIKE MEGASQUIRT!

 

My .02

Kyle

[Jamz14]

I completely agree with Kyle that we should be allowed a simplistic aftermarket wiring harness that is easily disconnected and replaced. Not sure about the ECU change he mentioned, but willing to consider it. I think however that when you have a tunable ECU like megasquirt, it will benefit more those that can spend a lot of money on the dyno and hinders those that aren't able to go to the dyno all the time. I think it will tend to grow the gap between the tip of the spear and the shaft.

 

Maybe I am wrong about that though.

[davew]

As a dyno owner, I would love to see programable ECU's.

 

3 hours dyno time times 40 cars ;-))

 

Not to mention the wire harness install and the consistent revenue from retunes.

 

Great idea for Dave, Bad idea for the class.

[SaulSpeedwell]

I very much doubt that is fuel related. It is something in the ignition system.

 

 

Agreed.  Fuel generally can't change that fast and would not be reflected in the tachometer at a frequency that high. 

 

Run extra grounds.  Body to injector rail, battery to body, body to PPF, check the kill switch/alternator/starter connections, make sure the ECU "grounds" are good, and consider bypassing the trick tach and Traqmate the next time this is an issue. 

 

Monitor battery voltage as well.  This can be as cheap as the $11 Wal-Mart analog "voltmeter" duct-taped to the dash and tapped into the cigarette lighter.  Analog is best here - when the miss occurs, glance at the meter - if it is flickering all over the place, be suspicious of the battery/battery wiring/killswitch/alternator/starter "stuff".

 

-Mark (3/7ths Electrical Engineer) 

[Diller]

Thanks for the suggestion guys. Good to have some things to try before the next track weekend. 

 

The car is a 1994 so I guess the ignitor is in the ECU. Would a bad signal from the cam sensor cause this? I have played with Megasquirts in the past and if they get a bad tach signal from the cam/crank sensor, the ignition is a crap shoot.

 

Any chance that the traqmate can log vehicle voltage? I have one open spot in the anolog I/O box that I could use if it is as easy as plugging in a fused power wire.

 

I am going to do some parts swapping just to rule out a few. Plus, if the problem still exists, I have some spares for the future.

[Adax]

 

Any chance that the traqmate can log vehicle voltage? I have one open spot in the anolog I/O box that I could use if it is as easy as plugging in a fused power wire.

 

 

 

Yes. Just connect it.

[fotostars]

Yes. Just connect it.

 

Right, bring it to an analog input and it will be logged...

[Diller]

Perfect. I'll hook that up and see if I can see any variations.

[ppridday]

You find anything more on this?  Last weekend I was out at my first track day of the year in my 90 (1.6) and part way through the day I developed a misfire/loss of power that was intermittent.  Sometimes lasting 8-10 seconds and then clearing up for 10-20 seconds or so and then recurring.  This was at all rev ranges as well.  It would also do it as I was cooling down to pit exit.  Pulled the plugs and they looked OK.  Would they remain clean on an intermittent issue like this?  I would assume my best option would be to replace plugs regardless and wires with it.  Next I would assume coil pack and possibly pull injectors??

 

From what I have gleaned on here it doesn't sound like a cam position sensor??  Correct me if I'm wrong here...   I wish I had some video but some dickwad stole my Gopros just before the weekend...

 

Thanks in advance for any info..

 

Paul

[Kyle Freiheit]

Check the ground on the back of the fuel rail that is for the coil packs. I had one that was finger tight that did the same thing. Hope that helps.

[Steve Scheifler]

Is it a miss, rough like one or more cylinders are not hitting at all, or a smother loss of power from all 4? Only under hard throttle or also light throttle? All RPM? Can you reproduce it at all in neutral? I assume you do not have a wideband installed?