2004 to 2020 Mazda 3 Forum and Mazdaspeed 3 Forums banner
1 - 20 of 35 Posts

·
Registered
Joined
·
14 Posts
Discussion Starter · #1 ·
Hi all,

New member here with 2015 Mazda 3 touring, manual transmission and 160k (km) on it.

So the issue came couple months ago, with car jerking at low speeds and in general feeling very sluggish. Almost like there was a delay between the gas pedal and delivering power. No CEL or any codes, which is the biggest riddle of it all.

Took it to the shop where they did the following:
  • throttle body cleaning
  • new spark plugs, all 4
  • 2 new ignition coils (misfiring on cyl 1 and 4, although no code). This solved lack of power but not the sluggishness and jerkiness
  • new MAF sensor, as the old one was showing lower voltage than per requirements.

Whenever the car was serviced and battery disconnected, the car after would perform like new for couple of days. Total night and day difference.
Same goes when MAF was replaced (although battery was not disconnected then). I thought replacing the MAF solved the issue.

But to my disappointment, couple days later, jerkiness and poor performance would come back. Every time. The worst is at low rpm and between 1st and 2nd gear.

At this point I am at loss. Only thing that comes to my mind is bad MAF wiring, as the issue is very, very inconsistent. Like there was a spell casted on the car, for lack of better word.

Any suggestions are welcomed. Thanks
 

·
Registered
Joined
·
14 Posts
Discussion Starter · #3 ·
How's your clutch pedal engagement? Pressure good? Any change in the bite point?
CK
Hi CK, appreciate your response.

No issues with clutch pedal or pressure. I got this confirmed at the dealership with at my regular maintenance. Also changing gears is very smooth in terms of clutch engagement.

What boggles my mind is that after disconnecting the battery, or replacing the maf sensor, car works like brand new for couple of days. That's why my initial thought was it's something maf related, or simply once the sensor goes through re-learning process it keeps reverting back to the old issue and start jerking again.
 

·
Registered
Joined
·
8 Posts
Have you tried changing the air filter? If you found the MAF had an effect its clearly something to do with airflow of some kind. Had the same issue a few weeks back, a lot of jerking, bucking, insane amounts of hesitation, even used 98 fuel. In the end was just the air filter went bad.
 

·
Registered
2018 Mazda 3 GT
Joined
·
3,415 Posts
Fuel injectors clogged? A/F ratio changes between 0-25 miles after ECU reset, which can occur after battery disconnect.

Not a fan of these, but maybe try some fuel additive engine cleaner?

The MAP sensor needs cleaning too. MAF sensor cleaner works well. It's in the front of the engine and has a bolt or screw securing it in.. i forget which.
CK
 

·
Registered
Joined
·
14 Posts
Discussion Starter · #6 ·
Have you tried changing the air filter? If you found the MAF had an effect its clearly something to do with airflow of some kind. Had the same issue a few weeks back, a lot of jerking, bucking, insane amounts of hesitation, even used 98 fuel. In the end was just the air filter went bad.
Air filter has been recently changed, couple months ago, with OEM one. So that can't be the issue. But I was thinking along the same lines, something is probably wrong with airflow, as it gets temporarily cleaned with MAF replacement and battery disconnect.
 

·
Registered
2018 Mazda 3 GT
Joined
·
3,415 Posts
Have you tried changing the air filter? If you found the MAF had an effect its clearly something to do with airflow of some kind. Had the same issue a few weeks back, a lot of jerking, bucking, insane amounts of hesitation, even used 98 fuel. In the end was just the air filter went bad.
I sure hope a mechanic would check that before replacing the MAF sensor.. :unsure:
CK
 

·
Registered
Joined
·
14 Posts
Discussion Starter · #8 ·
Fuel injectors clogged? A/F ratio changes between 0-25 miles after ECU reset, which can occur after battery disconnect.

Not a fan of these, but maybe try some fuel additive engine cleaner?

The MAP sensor needs cleaning too. MAF sensor cleaner works well. It's in the front of the engine and has a bolt or screw securing it in.. i forget which.
CK
This very good point. Fuel injector service was my next step to discuss with the shop, but I was thinking if there is a way to try and pinpoint what an exact issue might be based on other peoples experience. But very good observation.

I didn't know the fact that A/F ratio gets changed after 25 miles after ECU reset. That might explain why car worked so much better after battery was disconnected.

But that wouldn't explain why car worked fine after MAF replacement, as the battery was not disconnected during that time. It could be though that new sensor simply needed some time to learn the parameters.

No idea honestly, never experienced something like this. Either the problem is constantly there or not. This way, it's so hard to pinpoint as it's very inconsistent or it reappears couple days after the service was done.
 

·
Registered
2018 Mazda 3 GT
Joined
·
3,415 Posts
Exhaust is another consideration.. increased (something clogged) or decreased (something leaking) back end pressure can cause your issue.

I read that spraying soapy water (dish soap and water in a spray bottle) around the exhaust manifold and piping will reveal leaks if a vacuum is hooked up to the tail pipe and set on reverse (clean the vacuum first). If bubbles appear, you found a leak.
CK

 

·
Registered
2020 CX-5 GT 2.5L 2011 3i 2.0L
Joined
·
123 Posts
Try disconnecting the upstream O2 sensor, then start it up and see how it runs that way. The CEL will come on immediately with a O2 circuit code set, but no concern there. With the O2 disconnected, the engine management system will stay in open loop and use the default value table.

If it seems to be running ok with the sensor disconnected, stick with that until you're well past the point in time when performance normally goes bad. And if it does continue to run ok well beyond that point, then the problem lies somewhere in the closed loop processing, and the first thing you might consider is the upstream O2 sensor.

You could also diagnose by examining the live OBD data stream (with the sensor connected), but the above test is a very simple and easy thing to try, and will give a quick answer, one way or the other.
 

·
Registered
Joined
·
8 Posts
Hi all,

New member here with 2015 Mazda 3 touring, manual transmission and 160k (km) on it.

So the issue came couple months ago, with car jerking at low speeds and in general feeling very sluggish. Almost like there was a delay between the gas pedal and delivering power. No CEL or any codes, which is the biggest riddle of it all.

Took it to the shop where they did the following:
  • throttle body cleaning
  • new spark plugs, all 4
  • 2 new ignition coils (misfiring on cyl 1 and 4, although no code). This solved lack of power but not the sluggishness and jerkiness
  • new MAF sensor, as the old one was showing lower voltage than per requirements.

Whenever the car was serviced and battery disconnected, the car after would perform like new for couple of days. Total night and day difference.
Same goes when MAF was replaced (although battery was not disconnected then). I thought replacing the MAF solved the issue.

couple days later, jerkiness and poor performance would come back.

Every time. The worst is at low rpm and between 1st and 2nd gear.

At this point I am at loss. Only thing that comes to my mind is bad MAF wiring, as the issue is very, very inconsistent. Like there was a spell casted on the car, for lack of better word.
1
- ECU mapping, potential issue with VVT but i still wouldn't take airflow out of the issue as its a major part of slugishness.

2
- what rpm? 560? 750? 1k-2k 3.2k? A normal slugish response would naturally occour between 1.2-2.6/2.7k as the VVT needs to re-adjust

3
- maybe just check for the correct grounds or if there is too many.

Honestly id need more info to understand whats happening, for all i know it could even be that the VVT part of the engine is flooded with oil and can't function, the other thing could be a vaccume leak somewhere in the intake or even in the cooling system. All depends on what info iv got to work with. At the very least, remove your coolant cap, and put it back on using pressure, pushing down on it as you twist. (DO NOT DO THIS WHEN ENGINE IS WARM OR HAS BEEN RUNNING!)
 

·
Registered
Joined
·
1,529 Posts
Battery could be hovering just below 2.4V too. Disconnecting it for a couple of days might be just enough to let it get back up near 2.4, but then after hooking it up for a couple of days with the usual parasitic drain, if it's dying and goes below 2.4, can lead to low RPM bucking, among other things. Is it the original battery? I'd get a meter and check it first thing in the morning to rule out a dying battery.
 

·
Registered
2013 Mazda 3 i Hatchback
Joined
·
384 Posts
So, it sounds like the car runs fine when using the default fuel mapping, but after the car "learns" based on various sensors, it no longer runs correctly. If I had a scanner hooked up to it that could show live data, I'd be interested in seeing (when the car is not running correctly) the long and short term fuel trims, the MAF readings, the O2 readings, and if the car has completed all of the normal drive cycles.

I suspect that the fuel trims will be out of whack (it should be close to zero normally, anything above 12-15% will throw a CEL). If so, it gives you some things to investigate. A host of issues can cause the fuel trims to be out of spec, ranging from sensors reporting bad data (MAF, MAP, IAT, ECT, and O2), injectors not operating properly, poor fuel pressure, intake/vacuum leaks, or perhaps extreme carbon build-up on the intake valves (only likely if you only do very short trips).

Without the scanner, you'll just be guessing, but hitting it with some injector cleaner won't hurt. As others have suggested, preventing it from using it's closed-loop corrections might be a good test to run. Next time the issue returns, try unplugging either an O2 sensor or the MAF sensor. This will give you a CEL, but it will force the car to use the default fuel maps. Now, I'm lazy, so I'd unplug the MAF as it's way easier to get to. The car won't be as fuel efficient, you might be down a few HP, and you'll be looking at that CEL, but if it runs like that and the jerking goes away, it will confirm that it's definitely the adaptive fuel mapping that's causing the issue. You won't want to run that way for too long, just enough time to confirm that the jerking doesn't return.
 

·
Registered
Joined
·
22 Posts
Have you tried just using MAF cleaner spray? I had a check engine light on the last Mazda that was cleared and never came back after I did this. I read something about aftermarket intakes sometimes causing build up on my sensor. I'd definitely try a few bucks into that spray before having a mechanic look at replacement.

Nevermind, you said you replaced it already. You can try to hook up Torque with a cheap BT ODB2 transmitter, reset the ECU and watch your datalogs for a few days. My guess is that since it doesn't reoccur immediately, something's not being read correctly. Since it doesn't happen until the car learns it's cycles, you may be able to see a very noticeable difference in your logging between, say, day 2 (no issues) and day 3 (issue), I'd definitely start with your airflow and vacuum, other stats are kinda hard to determine.
 

·
Registered
Joined
·
14 Posts
Discussion Starter · #16 ·
So, it sounds like the car runs fine when using the default fuel mapping, but after the car "learns" based on various sensors, it no longer runs correctly. If I had a scanner hooked up to it that could show live data, I'd be interested in seeing (when the car is not running correctly) the long and short term fuel trims, the MAF readings, the O2 readings, and if the car has completed all of the normal drive cycles.

I suspect that the fuel trims will be out of whack (it should be close to zero normally, anything above 12-15% will throw a CEL). If so, it gives you some things to investigate. A host of issues can cause the fuel trims to be out of spec, ranging from sensors reporting bad data (MAF, MAP, IAT, ECT, and O2), injectors not operating properly, poor fuel pressure, intake/vacuum leaks, or perhaps extreme carbon build-up on the intake valves (only likely if you only do very short trips).

Without the scanner, you'll just be guessing, but hitting it with some injector cleaner won't hurt. As others have suggested, preventing it from using it's closed-loop corrections might be a good test to run. Next time the issue returns, try unplugging either an O2 sensor or the MAF sensor. This will give you a CEL, but it will force the car to use the default fuel maps. Now, I'm lazy, so I'd unplug the MAF as it's way easier to get to. The car won't be as fuel efficient, you might be down a few HP, and you'll be looking at that CEL, but if it runs like that and the jerking goes away, it will confirm that it's definitely the adaptive fuel mapping that's causing the issue. You won't want to run that way for too long, just enough time to confirm that the jerking doesn't return.
"it sounds like the car runs fine when using the default fuel mapping, but after the car "learns" based on various sensors, it no longer runs correctly" - hit the nail on the head here!

Unfortunately, without the CEL it's very difficult to identify where the issue is and where to begin with. The problem is always there, as it only works properly with default cycle for couple of days after battery is disconnected. Jerking is worst when driving at low rpms in 1st/2nd gear, shifting between gears (1st/2nd/3rd), and going from high speed (highway) to low speed (traffic jam). I would describe it, in my own simple words, as the car has a hard time finding its own balance. Also, from time to time I would get a rough idle, oscillating between 500 and 700 rpm.

I'll try injector cleaner, that's a cheap thing to check. I recently disconnected the MAF sensor, but the idle went to 1500rpm and car sounded very, very rough so I gave up on that test.

I am thinking getting it into the shop again, but it's difficult to find a mechanic who is willing to spend time and troubleshoot the issue without CELs. When I took it to the Mazda dealership, and it was a reputable one, they dismissed it as faulty spark plugs and dirty TB.
 

·
Registered
Joined
·
14 Posts
Discussion Starter · #17 ·
Battery could be hovering just below 2.4V too. Disconnecting it for a couple of days might be just enough to let it get back up near 2.4, but then after hooking it up for a couple of days with the usual parasitic drain, if it's dying and goes below 2.4, can lead to low RPM bucking, among other things. Is it the original battery? I'd get a meter and check it first thing in the morning to rule out a dying battery.
Hi there, thanks for chiming in. The battery is new, not even a year old, so that shouldn't be an issue. I could check though, at least that's an easy one...
 

·
Registered
Joined
·
14 Posts
Discussion Starter · #18 ·
Have you tried just using MAF cleaner spray? I had a check engine light on the last Mazda that was cleared and never came back after I did this. I read something about aftermarket intakes sometimes causing build up on my sensor. I'd definitely try a few bucks into that spray before having a mechanic look at replacement.

Nevermind, you said you replaced it already. You can try to hook up Torque with a cheap BT ODB2 transmitter, reset the ECU and watch your datalogs for a few days. My guess is that since it doesn't reoccur immediately, something's not being read correctly. Since it doesn't happen until the car learns it's cycles, you may be able to see a very noticeable difference in your logging between, say, day 2 (no issues) and day 3 (issue), I'd definitely start with your airflow and vacuum, other stats are kinda hard to determine.
Thanks. How would I know what the correct data is? Is there a reference table to refer to? It's a very logical step, but might be above my level of expertise...
 

·
Registered
Joined
·
14 Posts
Discussion Starter · #19 ·
"it sounds like the car runs fine when using the default fuel mapping, but after the car "learns" based on various sensors, it no longer runs correctly" - hit the nail on the head here!

Unfortunately, without the CEL it's very difficult to identify where the issue is and where to begin with. The problem is always there, as it only works properly with default cycle for couple of days after battery is disconnected. Jerking is worst when driving at low rpms in 1st/2nd gear, shifting between gears (1st/2nd/3rd), and going from high speed (highway) to low speed (traffic jam). I would describe it, in my own simple words, as the car has a hard time finding its own balance. Also, from time to time I would get a rough idle, oscillating between 500 and 700 rpm.

I'll try injector cleaner, that's a cheap thing to check. I recently disconnected the MAF sensor, but the idle went to 1500rpm and car sounded very, very rough so I gave up on that test.

I am thinking getting it into the shop again, but it's difficult to find a mechanic who is willing to spend time and troubleshoot the issue without CELs. When I took it to the Mazda dealership, and it was a reputable one, they dismissed it as faulty spark plugs and dirty TB.
And forgot to add - absolutely the worst performance is when AC is on, specifically when shifting in lower gears.
 

·
Registered
Joined
·
8 Posts
That bucking is the clutch and transmission. Its from dropping the clutch when its trying to revmatch, my advise would be know the gears a bit better, for mine its 6k 1st gear to 3.1k 2nd then 6k 2nd to 3.1k 3rd, then 6k to 4th is 4k

If that doesn't help try to just hold a tiny bit of throttle before lifting the clutch or when decelerating hold the clutch right before bite point so it revmatches, then downshift, hold the clutch, shift, ect. You will literally feel the car click into gear once the clutch and transmission speed matches in the correct gear.

Also theres almost no way to stop bucking in 1st when downshifting from second unless you hold the throttle slightly to keep it at 4k the let off the clutch. The way i do it wears on the clutch more but i find it safter in an emergency then rev matching the traditional way.

forgot to mention gearing... 0-50km 6k you want to downshift into 1st from 2nd at 40km thats around 2.4-2.7k in 2nd if i want to downshift into 2nd from 3rd, i have most of the rev range, anywhere from 60-80km 4k-6k 4th to 3rd, anywhere from 60-100km in 4th thats 2.7-4k down to 3rd thats 2.3k-5.7k
 
1 - 20 of 35 Posts
Top