known problems of Mazda skyactiv 2.5 hard re-starting?

Joined
Feb 28, 2015
Messages
1,012
Location
MD
I’m not on any Mazda forums so I'm just sticking it out there to see if anyone has any knowledge of this issue, or access to Identifix and is curious enough to look it up.

I have a 2017 Mazda 6 2.5 skyactiv direct injected.

I’m sure it’s due for valve cleaning, although Mazdas supposedly don’t gunk up the valves even with gdi due to some magic they work in the heads and air velocity and temps or whatever. Regardless, valve cleaning has not been done.

115k on odometer. 5k religious oil changes since new (I’m original owner). It doesn’t use any oil. Clean the throttle body and replace the engine air filter around every 15k. Run Costco top tier almost every fill up, I’d say 90% of the time. Plugs replaced with the same exact as oem at 100k just for maintenance (I wasn’t aware they call for them at 75 or 80k or whatever it is). Old plugs looked fine, nothing alarming at all. It was doing this hard start at times then too I remember, and I had hoped it was because I ran over the plug interval. They are not from Mazda (they are from RockAuto).

I drive it pretty easily, an easy 25-30 mile commute. But I make sure at least once every couple days I pull it hard up a long hill or wind it out getting on the highway where it can stay wot for awhile.

For the last couple years, after it’s been on a drive, shut down for a short period of time, anywhere from minutes to an hour or three, it sometimes starts funny. Like it doesn’t crank the normal length of time (shorter) it fires and stumbles a bit, idles low, then clears up. Never had a check engine light. I’ve never scanned it to see if anything is pending and never watched any kind of data. It’s just finally getting bad enough now that I’m going to look into it (maybe). Yesterday after it sat about 3 hours it started like this again but actually stalled. Second start was perfectly fine.

I suspect a leaking injector, but I’m not overly familiar with direct injection and not familiar with Mazda at all as far as their issues.

I suppose it could be dirty valves somehow absorbing the fuel, but I feel like that’s a little more far fetched.

Any experience or knowledge of any common issues with these engines? And this hard re-starting issue? Or if someone is curious and nice enough to explore Identifix, I’d imagine someone has posted about this on one of the same engines throughout the Mazda models.
 
Last edited by a moderator:
My old RAV4 had something similar at 350k miles that was due to the charcoal EVAP canister. You probably don't have enough miles for that issue to appear.
I also never top off gas which is usually the main culprit of a charcoal canister clogging or becoming over saturated with fuel.
 
A leaky DI injector is easy enough to diag, On a hot shut down....Rail pressure will increase if the injectors are holding.
I’ve been out of a real shop since 2013. In a government shop that is ill equipped.

Is rail pressure a bit of data that I can pull up with a capable scan tool?
 
Thanks.

Having never diagnosed anything on direct injection…you say rail pressure will increase (if no leak present).

Is that due to heat soak?

Will an autel have the spec with the data point? I’m sure there’s a period of time it should hold at least xx psi as well, which would be beyond whatever number autel may list as the spec. Is there a generic time you look for fuel rail psi to hold?

Are gasoline direct injectors prone to fail? They’re similar to diesel injectors aren’t they?
 
So I’ve had a scanner on it off and on a few days now.

No faults or pending codes.

No data is off.

Of course I have not experienced the starting issue since I’ve had the scanner.

Fuel rail pressure rises after shut down as you indicated it should.

The car had had impeccable maintenance. I’m wondering if I should remove the intake to clean the valves, but I want to understand why dirty valves would cause a stumble on hot starts on a GDI engine. I understand how it does on a port injected engine, I just don’t when it comes to GDI. I’ve read many people recommend valve cleaning far sooner than the 115k I’m at, and list a slew of reasons (misfires, poor fuel economy, loss of power etc) which all make sense, again, for port injection. How do these issue arise with direct injection? Wouldn’t dirty valves just choke the engine of air?

I did pull up a TSB about hard starts due to oil pump and or oil pump pickup issues, but I’m not seeing any oil pressure issues on any start ups either.
 
The first step in diagnosing a problem is to scan the car for codes. The "Service Engine Soon" light is on only when an emissions-related code is present. I recommend OBDFusion app: https://www.obdsoftware.net/
I’ve had a scanner on it for the past 5 days or so. No codes. No pending faults. No data is off at all when I monitor fuel pressures, fuel trims, oil pressure, MAF, cam crank signals, etc.
 
I wanna say we fought with one of these last year and a MAF fixed it.
Do you know where it’s reading intake air temps from? Because that’s the only bit of data that I feel like is slightly off.

Example: car sat 8.5 hours at work, from 35-44degrees outside. On startup to go home (44 degrees outside) one IAT reads 60 degrees. I didn’t realize at that point it has 2 IAT inputs. Cruising down highway on the way home I scrolled and realized there are 2 IAT, just called 1 and 2. 1 is always close to outside air temps, within 5 degrees I’d say. 2 is always about 10-12 degrees higher than 1, putting it well off outside air temps.

I can understand these values being off as they’re underhood and get affected by underhood temps and or heat soak. But on first start when all should be pretty cooled off, I didn’t expect to see 60 degrees. Although I did not have my starting issue either.

I’ve been hinted at checking purge valve, which is a sensible direction as if it’s dumping / stuck, it throw A/F off. Unsure how I’ll ever duplicate or verify that at this point though. I know I can vacuum check them, maybe rattle it around a little to see if it holds vacuum or it suddenly loses it, but that’s not necessarily a replication of what’s going on…it’d be more that it’s sticking open.
 
I’ve had a scanner on it for the past 5 days or so. No codes. No pending faults. No data is off at all when I monitor fuel pressures, fuel trims, oil pressure, MAF, cam crank signals, etc.
That is weird. I suggest you get OBDFusion and the appropriate interface module so you can log various data streams. Those data can be downloaded into excel or google sheets for analysis.
 
Back
Top Bottom