Repeat ignition coil failure, 3rd time replacing in 7 months
 

First a little history on my rx8's long journey to this point

Purchased 2005 s1 rx8 in November 2012, 6 speed mt of course with 84k~ miles

Noticed not too long after purchasing the car i was getting an average of about 9mpg which I'm sorry even for a rotary is just shit mileage, i got some help from some forum members in my post here https://www.rx8club.com/series-i-tro...-243359/page2/ which eventually led me to my current situation of getting about 16mpg city/highway makes no difference.

So my rx8 which just hit 95k miles gets about 16mpg average and repeatedly fails ignition coils within 3-5 months of having them installed, does anyone have any idea why these coils would keep going bad so quickly? I'm not FI currently and I have asked mazda repeatedly to reflash my ecu to make sure it was within factory specs, I have an appointment scheduled again with my local mazda dealer to replace the failed coils tomorrow again (parts warranty pays for replacement) but having to keep taking my car to mazda and still getting consitently worse gas mileage than other rx8 owners I'm starting to wonder if there is a bigger problem here and I've run out of ideas on what it could be, I normally consider my self pretty knowledgeable when it comes to cars but this one .... I just can't keep her running perfectly for more than 3 months at a time.

Quick summary of work done to my rx8 since i purchased it about a year ago

Engine replaced (reman installed by mazda)
Starter replaces 2x (installed by mazda)
Coils replaced 3x (installed by mazda)
Plugs replaced 2x (installed by mazda)
Plug Wires replaced 2x (installed by mazda)
Clutch replaced (stage 2 installed by mazda upgrade wasn't necessary but i saved on labor during engine replacement)
Flywheel replaced (11lb~ installed by mazda see above)
Thermostat replaced (engine temps low in old diagnostics)
Repeatedly cleaning maf sensor
Mutliple cleanings of k&n air filter (rechargable)
Cat-delete pipe installed (thought maybe cat was clogged)

I've checked for vacuum leaks several times and keep up on my oil changes but this car is like a bad omen it seems and that really sucks because i want to love it like i did the day i got it again. :Eyecrazy:

Question: How are you diagnosing coil failure? What diagnosis method are you using?

I'll await your answer before posting my thoughts on what I think is happening.

Generally it starts out the car feels a little sluggish but overall is running fine, eventually it gets worse to the point where I have trouble keeping up with traffic..pushing my rx8 to 6 grand just to keep up with a corolla, Eventually the computer will throw a code, had the code run and it gives "Misfire on cylinder #1" which is the same code it's thrown the last couple replacements, I'll have to get the exact code number but it's been consistent everytime.

That is P0301, no need to go get it. P302 is the rear rotor, and P0300 is "random".

I suspected that you were just using the misfire codes, and your coils are actually completely fine. There are things you have not addressed which CAN cause misfires:

- E-shaft sensor could be fouled (needs cleaning)
- E-shaft sensor profile could be off (needs reseting)
- Fuel trims could be off (needs reseting)
- Vacuum leak could exist that you haven't detected yet (you can confirm or disprove this with ODB2 data at idle)
- Engine compression could be failing (need a compression test at your mileage anyway, you are nearing or just past the end of warranty coverage on the engine)
- The catalytic converter could be failing (clogs from misfires, but can then cause misfires in return even if the original problem is solved)
- Your front O2 sensor could be failing (can be damaged by misfires and/or a clogged cat, and then cause misfires in return


All of those can cause misfires, and you haven't examined them yet.

You are correct on some of these points however I would hope that mazda's diagnostic tools would at least find if the fuel trim was off or there were vacumm leaks present.

- Engine compression could be failing (engine is a newer reman with less than 10k on the clock, last compression test was above factory specs I'll have them test it again tomorrow)

- Catalytic converter I replaced with a mid-pipe, updated my post to reflect that

-I've reset the eshaft sensor by doing the brake pedal method a couple times in the past

so that leaves us with

- E-shaft sensor could be fouled (needs cleaning)
- Your front O2 sensor could be failing

assuming Mazda's diagnostic reports are worth the paper theyre printed on.

Here is data from the last thread (late febuary,2013)

Compression test results:
Rotor 1 - 8.8 , 8.8, 8.9 @ 288rpm
Rotor 2 - 9.2 , 9.1 , 9.0 @ 290rpm

Diagnostics results:

Response Lean to Rich - Min: 0.292 Max: 255.992, Value: 1.949
Response Rich to Lean - Min: 0261 Max: 255.992, Value: 1.425
H02SB1S2 Monitor - (Empty on chart)
Low Sensor Voltage for switch time calculation - Min/Max: 0.299V, Value: 0.299V
High sensor voltage for switch time calculation - Min/Max: 0.400V, Value: 0.400V
Rich to lean sensor switch time - Min: 0s , Max: 0.080s, Value: 0.032s
Catalyst monitor bank 1 - (Empty on chart)
Rear to front switch ratio - Min: 0 , Max: 7.999 , Value: 0.960
EVAP monitor (large) - (Empty on chart)
Evamp Monitor Large Leak Check - Min: 0.618mA , Max: 255.996mA, Value: 2.089mA
Evap Monitor 0.040 inch - (Empty on chart)
Evamp Monitor 0.040 inch leak check - Min: 2.085mA, Max: 255.996mA, Value: 2.089mA
Purge Flow monitor - (Empty on chart)
Purge Flow Monitor blocked evap system line check - Min: 0mA, Max: 18.648mA, Value: 18.621mA
secondary air monitor bank 1 - (Empty on chart)
secondary air functional check - Min: 0s, Max: 4.992s, Value: 0s
misfire cylinder 1 data (rotors #1) - (See Below)
ewma misfire counts for last 10 driving cycles - Min: 0, Max 65535, Value: 5
misfire counts for last/current driving cycles - Min: 0, Max 65535, Value 9
misfire cylinder 2 data (rotor #2) - (See Below)
ewma misfire counts for last 10 driving cycles - Min: 0, Max: 65535, Value: 7
misfire counts for last/current driving cycles - Min: 0, Max: 65535, Value: 30
Thermostat Monitor - (Empty on chart)
Engine coolant temp - Min: 71c, Max: 6513c, Value: 74c

I'll try to get more information tomorrow when i have mazda "diagnose" the problem again, also keep in mind this isn't me stating the coils are failing but mazda after they run through their diagnostic they find bad coils, replace them, wait 3 months, repeat.

Actually no. If they don't specifically test for it, they won't discover a vacuum leak. And while they might be able to pull up the fuel trim values, those values won't mean anything since they will vary from car to car, sensor to sensor, and drivestyle to drivestyle.

Ok, compression is probably fine then, however, to me this points further at an e-shaft sensor profile problem. Techs often don't reset it when doing engine replacements.

ok, that's crossed off

Never mind my prior comment then, this resets fuel trims and the e-shaft sensor profile correctly

Those are good numbers, no concerns about compression scores for sure.

The only thing of note on that is the misfire counts, you can see it's happening to both rotors.

The O2 sensor is probably fine, but those tests won't always reveal a failing one.


So the list has shrunken to an e-shaft sensor problem (likely just fouled if so, it's rare that they actually fail, though possible), a vacuum leak, or a possible front O2 sensor failure.

This is all stuff that you can test yourself with a $25 OBD2 bluetooth adapter and a $5 android smartphone app. Likely far cheaper, and certainly far less time wasted on your end.

Just a thought in case you are inclined.

I actually just ordered an adapter, it's something I meant to do months ago but put on hold when my nexus 7 dash project was put on hold. Thank you for your help, any tips on cleaning the e-shaft sensor my self? I'll talk to mazda about it tomorrow and try to get them to test the sensor for failure.

The internals of the sensor is basically just a magnet, the exterior is just entirely coated in plastic to keep the elements out. Random bits of even faintly metallic material can attach to the sensor and start interferring with the magnetic signature. This does means that you can just wipe it off with a rag. You can use spray cleaners too, like MAF cleaner or carb cleaner or WD40, though considering how close it is to the accessory belts, you should avoid getting any lubricant on those.

Awesome, I'll give it a shot, will update tomorrow after i get back from the mazda dealership.

check ur alternator and fuel pump

S1 alternator is barely enough juice even when it's brand new, this is also one of the reasons why they updated to 110 amp in S2. and lots of people overlooked this part, remember, it's one of the things that keeps ur whole car running. and you can't test it correctly until it's hot (after completely warmed up)

Fuel pump when it's about to die, will give you random misfire/lean issue. check that.