Misfire Under Heavy Load Only

KJMS3

Greenie Member
Hello everyone, I'm new here so I'll give a small intro here before I get into it. My name's Keelen, I'm 22 years old and I've had my 2008 Mazdaspeed 3 for a little over a year and a half now. Bought it with around 196000 miles on it bone stock, now with almost 212000 miles, and it was treating me incredibly well until about 6 months ago when this misfire began, and since it's my only running car right now, it's been daily driven since the day I bought it, including now. I like to think I'm fairly well versed in the world of cars and modifying for someone my age, but this car has had me completely stumped, which brings me here, as I've tried everything I can think of.

Everything on the car is from Corksport unless I state otherwise here: 2.5" short ram intake, test pipe, 80mm cat-back, cleaned and flow tested OEM injectors from Graveyard Performance, injector seals and O rings, HPFP, E85 safe HPFP seals, 50/50 VTA BPV, CST4, 3.5 BAR MAP sensor, oil catch can, Tial EWG, 3 port boost controller, polyurethane motor and trans mounts (the set of 3 that CS sells), NGK 6510 plugs gapped at .026". I believe that's everything, and all mods were installed by me at home following CS instructions most of the time for torque specs and tips specific to whatever job I'm doing. I've been tuned by Erik at Dramatune since my first major mod (intake) and have had revisions with every mod from the beginning, and before CST4 I was on an E30 mix. I do also have a catless downpipe on backorder from CS right now. Would also like to add that the EVAP system has been gone through and mostly replaced, since the car threw a CEL for the EVAP system on my way home from buying it and I had to fix it before smog.

Now for background on this misfire. It initially came about when I was K04, stock boost controller, stock MAP sensor, stock HPFP seals (only seals! first mod I did to the car was a HPFP when I got it). Car started misfiring BAD one day so I threw a new set of plugs in it which then caused the "only under heavy load" condition, which I will get into. Sent an email and datalog to Erik to get a second opinion, he recommended new coil packs and a compression test, so I bought a set a new OEM coil packs and sent him my compression test results (which were all good), still misfiring! And nothing in the logs pointing in any direction. So I then moved on to the fuel system and got the E85 HPFP seals from CS since I already had the injector seals and O rings from them and the injectors from Graveyard, still misfiring! Since I had seen oil in the intercooler, my next thought was that my stock K04 had finally given up on me (or was on its way to), so I saved up and got the CST4 with EWG like I had been planning on doing after fixing the misfire anyways, so I figured it's worth a shot! Got my base map from Erik and still misfiring! One thing I found odd, however, is that with the bigger turbo, the car allowed more boost to be made before it started misfiring and gets better MPG when compared to my K04 setup. K04 misfired around 8psi, now with the CST4 close to 12psi. So at that point I began thinking that maybe there's a short somewhere in the wastegate circuit, since right when I start to hear that sweet EWG crack open is when it starts to misfire, so last weekend I put an new ECU in the car, and still misfiring just the same! And now with an ABS code (P2022 if I recall correctly) from the new ECU. Cleaned the MAF today and got another compression test just to double check as well. Compression test number as of 4/14/2026 are as follows:
Cyl 1 - 205
Cyl 2 - 195
Cyl 3 - 205
Cyl 4 - 205
Compression test was done with coolant temp at 135 F. The last test around august 2025 (when I sent the results to Erik) was done hot and yielded similar results. The car drives COMPLETELY normally with no CEL codes until I try to do a small pull and pass that threshold of when the misfire starts (now about 12-13psi boost). When I sit on the misfire it does throw a CEL for most commonly a random misfire P0300, but I have seen P0301, P0303, and P0304 as well since this started. Sometimes it gets some KR while misfiring as well, not always but maybe about half the time, the most I've seen was a little over 3. Also should add that it doesn't misfire at the same amount of boost every time. Sometimes it allows more, sometimes less, so the 12-13psi is just around where it usually does. I've tried experimenting with seeing how the car reacts with higher and lower outside temps, intake temps, and coolant temps, but none seem to make a difference.

That's all the info I can remember to include right now off the top of my head, but if I remember something else I'll be sure to add it. Any questions feel free to ask and I'll reply as fast as I can. I included a datalog, a vid of what it sounds like, and a pic of the plugs which were installed about a month ago when I did the turbo job here as well, the datalog was taken at the same time as the video. In the picture the left plug is cyl 1 and right is cyl 4.

Any help/guidance is much appreciated!!

Edit: Looks like the vid didn't upload since it's a .mov file from my phone, here's a link that should open to it Misfire Vid.MOV
 

Attachments

Last edited:
hate to tell you this but its one of two things, the tune sucks and is causing the misfire or the tune sucks and has blown up your motor

dont use dramatune
 
hate to tell you this but its one of two things, the tune sucks and is causing the misfire or the tune sucks and has blown up your motor

dont use dramatune
Was on the same tune from there for 7 months before the misfire started happening, if it was the tune wouldn’t it have started right when I got the tune?
 
where are you located? did it start when the weather changed this spring? what was the weather when you got the tune finished.

the plug on the right looks pretty rough but i cant tell if the electrode is melted or if its just not a great picture, can you post a closer pic of the plugs
 
I’m in central Cali, and no it started around August 2025. The tune that was on the car when it first started was competed around January or February 2025 if I recall correctly, but the car was just fine through the worst of summer here in June and July before the miss started, and it didn’t get any better when the weather returned to how it was when the tune was finished. As for the plugs the pic I got wasn’t the best but they were all intact, double checked my plug gap while they were out yesterday and all were still good at .026”
 
The problem with tunes that are "just good enough" is exactly that. Any deviation from conditions it was tuned in is a recipe for disaster.

And your... "Tuner..." has a history of dumpstered engines; in fact, there's a whole ass thread on here about how many engine scatterings he's responsible for and his responses to those issues. I guarantee you he will blame you for something or brush it off as failed parts.

Anything to make a quick buck and shuck responsibility.
 
Edit:
Looked over the logs.

You're at half pedal and the throttle is over 70% (basically WOT) and the shit is struggling to make boost at an RPM and timing (3 deg at 4k+?) it should have no issue with. Do yourself a favor, pop the hood while parked, start it and pull the valve cover vent hose (cap it with your thumb on the hose side) and see if the valve cover breather port is trying to suck in or blow out; that will be your answer.

From the look of the log though, I'm thinking ringland on 3 or 4 has gone to valhalla.

Also, stay the fuck away from Graveyard; hes the hard parts version of Dramatune.

BRB gonna add this thread to the confirmed kills list.
 
Dammit, I was really hoping it wasn’t gonna be looking like anything major. I’ll do that test with the valve cover breather tomorrow after work and report back with what I can only hope will be good findings. If not, guess I’m selling it for a better home since I don’t have time for that mess rn :(
 
Oh it definitely won’t, I love speed3s and I’d love to take the time to use this as an opportunity to build the motor and do something more wild with it, but I just don’t have the time or funds rn with refreshing my turbo Miata. Plus its my daily so I can’t have it down for more than a weekend
 
@Enki I pulled that hose off while it was running like you said. No excessive blowby or vacuum. Seemed to me like the PCV system was working completely normally. Held a piece of paper up near it and it started vibrating from the PCV valve doing its job. I would think this points to normal?
 
Well considering the vacuum hose goes from the intake (after the throttle body, thus, making a LARGE amount of vacuum against that hose) to the block...You should be seeing strong vacuum against the port, not a vibration in airflow.

Think about why there would be a vibration in the airflow if vacuum is being pulled against it from all cylinders.
 
Maybe I got the wrong hose then bc you’re saying it’s on the intake after throttle body which could only be the manifold. The one I pulled off is on top of the valve cover, towards the driver side of the car, and from factory would go to the intake on the turbo inlet. Since I have a catch can tho, mine routes to the catch can before making its way to the turbo inlet
 
Even with the other side of the PCV system blocked off? The barb on the intake manifold is blocked off (as per CS catch can install instructions) and the PCV valve hose (that’d normally go into the intake manifold) is tee’d into that valve cover hose before they both go into the catch can. I don’t mean to sound doubtful I’m just trying to understand.
 
So you're vent to intake then? That would mean you have crankcase pressure at idle, which isn't exactly great. Also, the hole for anything to pass through flow wise off the valve cover is quite small, and since the PCV plate up front should be doing most of the heavy lifting regarding evac, feeling pressure at the valve cover still ain't great.
 
Yes, it all vents to the intake which is why I thought the pressure vibration was normal. So is it possible I’m lucky enough to not have had drama blow my shit up? Like would it be worth it to try getting a tune from someone actually worth a shit or would I just be throwing more money away at this point like I’ve been for the last 7 months trying to fix this?
 
After reading all of this... I am lost, not because I don't understand the problem. But the testing and PCV checks..

A PCV system is regulated by the valve coming off the manifold going to the block. A good working valve puts a slight vacuum on the system. To measure it, generally you have to use a gauge that reads inches of water and it's done via the dipstick tube. Sounds like you've plugged the valve cover and you're now routing it differently which I'd need to see pictures to understand.

Pulling the cap on these cars is not a good test. The chain is below the cap and throws oil and air and such out. It's not like other cars where you can twist the cap and watch is jiggle in place to see whats going on.

I've tested a lot of PCV systems over the years on a large variety of cars.. It's a really simple system, even in the European world with diaphragms and added complexity they still work the same.

If I am reading the original story correctly, the misfire was present before all the mods... It was never diagnosed at that point... Compression numbers were good back then.. You threw new plugs and coils at it, and when it didn't fix it.. You threw a bunch of mods at the car.. You made the mistake thinking some oil on the inside of the intake around the turbo was part of the problem..

The misfire sounds like a hard miss that is most likely a ignition issue. Just cause you've thrown new parts on the car doesn't mean they are good. I'd have enough money to buy build a pretty nice motor from all of the bad out of the box parts I've seen over the years. So you need a known good set of coils to try. Or you need someone who can see what's going on electrically with all the signals to see if a coil is dropping out or not.

As for the motor health. Leak down test, not just a compression test. It'll tell you if a ring land has gone bad..

Something I haven't heard about happening on this platform, but I have seen on other cars. A partially broken valve spring causing valve issues under heavier loads. This is tricky with out tearing out the cams and pulling the lifters to have a closer look.

Last note, just because you got a set of injectors flow tested, I wouldn't completely trust them. I got a bad set from graveyard a few years ago when building my car. I've never asked if they were using proper equipment to test at the high pressures GDI injectors run at... but I do know most "flow testing" and cleaning is done at normal port injector pressures... It's hard to find a proper place that tests GDI injectors at the proper pressures.
 
His PCV system is set to vent to intake (pre-turbo) so the port on the intake manifold is capped. This means any crankcase gasses would not immediately outflow intake gasses and can disperse more evenly and let the car run reasonably normally until the blowby gets extreme (I've seen how it acts both VTA and stock config and the difference is night and day).

Nobody said he should pull his oil cap. I did that on mine because I have a genpu valve cover with a baffle there, that largely prevents the oil from coming out. What you see moving the paper towel is blowby, not chain air.

Also Graveyard is the hard parts version of Drama, from what I've seen and discussed with people.
 
Back
Top