HORRIBLE cold start!

[sphinxrider]

Hi guys, I don't post much, but since my Mirage ES 5spd is a January 2018 purchase there's not too much to say for me.

Specs:
2018 Mirage ES 5spd hatch, 19k miles.

Oil is quite new, less than 500 miles on it.

Issue:

I drove my Mirage about 15 miles on the 7th, at around 8pm, mostly highway, made a pitstop at a grocery store and the finished my trip.
The temperature dropped to 28f/-2.2c, and it snowed a little bit.

the car sat parked until today, the 8th, until I started the car around 2pm, temperature around 38f/3.3c.

Total sitting time, 18 hours (car has sat longer than that in the past before with no issues).

1st start attempt:
I turned the key and the car half halfheartedly started and immediately quit after I let go of the key, on it's own (it didn't even idle, just died immediately).

2nd start attempt :
I turned the key and the car cranked for a solid 20 seconds with no indication that it would start, it just cranked and cranked.
I applied a little bit of throttle and the car started very slowly, barely running, I pushed the throttle until it smoothed out, and I cautiously let go and the engine almost immediately died again.

As it almost stuttered to a stop I pushed the throttle up to 2,000 RPM and kept it there for about a minute, when I let go of the throttle the car idled at 1,500 RPM (for reference my car has never idled higher than 1,100-1,200 RPM when cold started, EVER)

I immediately started driving not trusting the car and got on the highway and drove about another 15 miles back, for the rest of the day the car ran and idled normally with no issues. I did notice that it seemed a little down on power, though that may have been in my head.

Has anyone else had cold start issues like this?

I have been running my car almost exclusively on 93 octane since I got it, and the recently I filled up 3/4s of the tank with regular because I was a bit short on cash. I figured these cars are rated for regular gas/87 octane that it wouldn't hurt anything but now I am suspicious, I filled up at a place I don't normally get fuel, though I am not convinced it was necessarily the fuel, I only add that info to try give as complete a picture as I can.

I've driven the car and started it in much colder weather than this, and never noticed a performance drop, so I am baffled by this.

Are there any common issues that could cause this?

Again, 19k miles on the clock.
I am hoping against hope this doesn't happen again, but I'd be lying if I said this didn't shake my faith a bit in the reliability of this car.

Thanks guys!

[Loren]

I've had the car catch a bad start once or twice. It does something like almost start... and then not. I think what happens is that it's getting flooded by a false start. Then you hit it again and it re-primes, and hits it with another rich "cold start" cranking pulse and floods it some more. Once you recover from it and get running, it's fine. Owners manual usually states to crank the engine with full throttle to enable flood-clear mode. I don't know if the Mirage manual says that or not, but it's pretty typical for most modern cars.

It's remotely possible that running the car on 93 octane exclusively has taught the ECU that it can run ignition timing advanced toward the high end of its range. Ignition timing "trim" would be maxed out because it's never getting into a situation where it's sensing knock and retarding the timing for that.

If that's happened, and then you run some 87 through it, especially in "demanding" situations like cold weather, maaaaybe it will take it a bit of driving for the ECU to relearn. I don't think that has much, if anything to do with your random start flooding issue. But, it could be a factor.

Once the ECU has learned "okay 87 octane is my new norm" and the ignition timing trim tables are all adjusted to that, you can absolutely run 87 octane without issue.

[Mark]

This is what my 2017 owner's manual states on this topic -

If the engine still will not start, the engine could be flooded with too much gasoline.

While depressing the brake pedal (CVT) or the clutch pedal (manual transaxle), push the accelerator pedal all the way down and hold it there, then crank the engine for 5 to 6 seconds.

Return the ignition switch to the “OFF” position and release the accelerator pedal.

Wait a few seconds, and then crank the engine again for 5 to 6 seconds while depressing the brake pedal (CVT) or the clutch pedal (manual transaxle), but do not push the accelerator pedal.

Release the ignition key if the engine starts.

If the engine fails to start, repeat these procedures. If the engine still will not start, contact your local Mitsubishi Motors dealer or a repair facility of your choice for assistance.

There's an extra section for CVT owners -

Startability of CVT vehicle with ambient temperature of -4 °F (-20 °C) or lower

When the ambient temperature is -4 °F (-20°C) or lower, it may not be possible to start
from a standstill even with the selector lever 5 in the “D” (DRIVE) or “R” (REVERSE) position.
This phenomenon occurs because the trans-axle has not warmed up sufficiently; it does not indicate a problem. If this occurs, place the selector lever in the “P” (PARK) position and let the engine idle for at least 10 minutes.

The transaxle will warm up, and you will be able to start normally.
Do not leave the vehicle during warm-up operation.

[benwerks]

throw a bottle of techron into the fuel tank the next two or so times when you refuel.

[Cobrajet]

Yeah, this sometimes happens with these cars. Mine did almost the same thing a couple weeks ago, but additionally threw a CEL shortly after it began to run smoothly. The CEL self-resolved after about a dozen ignition cycles.

https://mirageforum.com/forum/showth...esterday-P0340

[Eggman]

Has anyone else had cold start issues like this?

Yep, it happened to me once. I did the Clear Flood thing with the accelerator - it ran rough, then smoothed out and all is well since.

Somewhat related, there are some TSBs regarding cold weather operation that cover freezing relays. I don't think this is the case with you though, but who knows.


TSB: Rich Air/Fuel Ratio at Cold Ambient Temperatures
TSB: Error Codes with Cold Weather Operation
TSB: Extreme Cold Ambient Temperature - Relays Fail to Operate

[sphinxrider]

Hey guys thought I'd post my findings.

First, thank you for posting this info, I really appreciate it. It's good to know I'm not alone on this one.
I have done more reading on the forum and found others having this issue as one of you mentioned.
The car hasn't done it since then so hopefully the rest of this winter will be trouble free.

I have driven the car until the last bar in the fuel gauge was blinking at me and refilled with 93 octane as per my usual norm.


There is something I want to comment on as far as 87 vs 93, the car feels like it has significantly better throttle response on 93 over 87, I found very noticeable dead spots in the power band on this car running on 87.
On 93 it feels much smoother, better power delivery (every bit helps), fuel economy is similar though I seem to pick up a couple mpgs on 93 (not much mind you, but enough to notice).

Just my impressions.

[Loren]

Makes sense. 93 octane allows the ECU to adjust to more advanced timing, which will improve throttle response and torque. Better FE goes with that.

Most people who have tested that say that the difference in FE isn't enough to make up for the extra expense of premium fuel. But, if you like all of the other benefits, why not?

[sphinxrider]

Makes sense. 93 octane allows the ECU to adjust to more advanced timing, which will improve throttle response and torque. Better FE goes with that.

Most people who have tested that say that the difference in FE isn't enough to make up for the extra expense of premium fuel. But, if you like all of the other benefits, why not?

Honestly, being stick and only having 78hp having better response is worth it to me, I think I only spend about $5-7 bucks more per fill up, and I don't drive that much.

I dont think I'd notice with a CVT but with a stick I usually shift between 2-3k rpm, maybe slightly higher, and I noticed a BIG dead spot in power around 2800-3000rpm.

This is anecdotal but one good example of how I noticed the power difference on fuels;
I was driving downhill (slight downhill on a 55mph highway) and put my foot down in 4th gear @2800rpm, NOTHING happened, the car just sat there until it decided to give me some acceleration.
I tested this a couple times to see if it would do it again, and it did repeatedly.

When I put 93 in the tank I tried the same test, and on an uphill and the car didn't hesitate, it gave me power under the same 4th gear 2800 rpm conditions but on an uphill.

I live in a hilly area so I'll be sticking with 93 lol

[Pimp_Daddy_Patty]

Unless Mitsubishi is doing something different with their newest ECUs, I haven't come across an ECU that will advance ignition timing beyond what it's originally calibrated to do.

At times of higher ambient temperatures,and prolonged loads that builds higher combustion chamber temps, the ECU will sense knock and apply an amount of knock retard to maintain safe combustion. The result of this is slightly lower power output until the knock threshold returns to it's normal range. In a case such as this, the higher octane fuel would maintain a higher knock threshold which the ECU would not have to pull any ignition timing.

Higher octane fuel typically burns slower, and in an engine only calibrated for lower octane fuel, would actually cause peak cylinder pressure to happen a bit later after top dead center. This in turn can lower power output by a small amount. From my experience in adding turbo kits to NA engines, I would typically advance timing in the vacuum range of the ignition map to maintain part throttle power and fuel consumption. The difference isn't much, but I still managed 40mpg on cars that never attained such fuel economy numbers in factory, NA form.