So I took the car out of storage and it started fine, idled for ~30 minutes fine, then suddenly died after driving about 100 feet. Thought I had just let the revs drop too much (it sometimes dies/ rough idles with the AC/defrost on if the revs go too low, I think that’s a vacuum leak). Restarted as the car was still rolling and kept driving, then it died again despite keeping the revs around 3000 as I braked for a stop. Car shut off, I parked it, and the RPM gauge went sort of nuts with the key in the ignition but engine off.

Background: Car has been sitting for about 3 weeks, it has snowed a ton lately. Tried to keep the snow off until the last week or so. Very humid inside when I opened it up (has started to warm up recently and melt). Also found mice food in the engine bay, but no apparent wiring damage.

Seems like a wiring issue? All the fuses I checked are fine (just replaced them all), but it feels like what happens when you pull the EGR fuse and crank it. Battery is new, starter and alternator were replaced last year. Car drove fine a few weeks ago Anyone have any ideas? I’m guessing wiring damage / a short somewhere, I just can’t see anything wrong.

here’s what it sounds like: https://files.catbox.moe/oar2tv.mov

 

This really sounds like an electrical or sensor issue, not a fuel or battery problem. The big red flag is the RPM gauge acting erratic with the key on and engine off — that almost always points to a bad ground, moisture intrusion, or damaged wiring. Given it sat for weeks in snow and you found mouse food, rodent damage to a sensor wire (especially crank or cam position sensor, EGR, or engine grounds) is very likely even if it’s not obvious. Moisture can also cause corrosion in grounds, fuse boxes, or ECU connectors and lead to sudden stalling and random gauge behavior. A vacuum leak may explain rough idle with AC on, but it wouldn’t cause the car to die at higher RPM. I’d start by checking and cleaning all engine grounds and closely inspecting sensor wiring looms for hidden damage.

 

Ok, thanks. Guess I’m getting it towed to the shop, damnit. Just noticed the PF Master is showing “No Connection” to the ECU now as well

 

My car has been stalling out as a result of corrosion at the connection to the alternator.

Your alternator drive belt is ok?

I suppose you have checked to see if a passenger has stomped on the wire connecting the PFC to the ECU.

My speedometer tics and does not move at temperatures close to freezing but starts working again once the interior warms up.

You could have condensation causing a short in the current weather where it normally does not.

Also consider that you may have had condensation in the gas tank and now have diluted fuel.

 

Yeah, unfortunately that pretty much confirms it’s an electrical/communication issue. If the PF Master can’t connect to the ECU, that usually points to a loss of power or ground to the ECU, a blown ECU-related relay, or damaged CAN / signal wiring — rodent damage or moisture would fit perfectly with what you’re seeing. At that point towing it in is the smart move before chasing your tail or risking more damage. Hopefully it ends up being a chewed wire or corroded ground and not the ECU itself. Good luck, man — that’s a frustrating one.

 

No passengers in a long time (everyone is too big to fit in these buckets, lol). Condensation definitely seems like it could be a factor. I'm hoping it's that and it'll fix itself and not mice or bad corrosion. Belts were replaced last year as well with the alternator. Car just got dropped off at the mechanic so I guess I'll find out more next week after they've had time to diagnose it. I didn't check the relays yet, just the two main fuse boxes (the third, longer fuse box was relocated to in front of the FMIC and is below an air dam so it's kind of hard to access but there doesn't seem to be any water in there either).

 

Someone here had a relocated main relay box, mounted tipped to one side. Rain etc. getting into the box caused all sorts of problems.


The water can migrate into the front harness which is huge and also can cause all sorts of issues.

Also check front wheel wheels to see if the liners have been broken through and the front harness exposed.

 

Thanks. When I inspected the front relay box it didn't seem wet at all but I wasn't able to open it up at the time. The local mazda dealer just spent about a week looking for problems but reported back that they didn't see any broken/corroded wires or anything and they weren't able to find a cause. The front wheel liners are brand new so I doubt it's that, didn't see any problems with them when I checked earlier but who knows if water got inside there. I'll probably have to tow it to a rotary specialist now and have them try to figure it out. I wonder if it's a bad ECU or hopefully something dumb like a bad relay (going to try replacing those).

 

Waiting on the new relays to arrive then I guess I'll take it to a specialist. The engine cranks and starts now, but dies within 30s each time.

Things that were checked and ruled out:

• fuel pump
• bad gasoline
• sensors checked & relays checked (appear to be working, but the mechanic mentioned one of the two main green/yellow relays in the main fusebox was acting strange at first, so I'm going to try replacing those)
• abnormal amount of smoke when it started (looked white to me, I was only able to see it run for a few seconds when I visited and just saw whisps, but he told me it was smoking a lot)
• wiring checked in engine bay, appears fine

 

The green and yellow relays are very important. They are also hard to remove and can get crushed in the attempt.

 

Long shot but consider you may have gunk or other debris in your gas tank that is clogging up the sock.

When the car sits the debris drifts away from the sock, but when you run the car again it clogs the sock up again.

How clean is you air filter?

Have the mice set up camp in there?

I once had a pinched o-ring in the fuel line connection to the fuel pump. This contributed to my car running lean and, ultimately, contributed to detonation.

 

Just replaced the green & yellow relays and tried, no luck. Then I tried the deflood procedure with the EGI fuse removed for about a minute and tried again, still no luck. It fired up for about a second then died, that's the best I got even when I tried fluttering the gas at one point.

 

There are redundant circuits tp the fuel pump. You probably need to remove both the green and yellow relays. Also the fuel pump relay in the main relay box (at the front) and the fuel pump fuse inside below the dash.

Another way to is to disconnect the electrical connector at the fuel pump itself. That makes sure no power is finding its way to the pump.

By now your plugs may be hopelessly flooded you should consider replacing them., even if they look new.

 

If it is unusually cold outside (you can see your breath) you will see a lot of water vapor when starting the car up. It will fade away as the motor gets hotter.

 

The white smoke on start up however, may indicate that coolant is seeping into the combustion chamber when the car is standing, indicating failing coolant seals.

Is there any coolant showing up in your oil.? Are there exhaust bubbles in your coolant. when you peer into the upper filler neck?

Perhaps try a cooling system pressure test.

When was the last compression test done?

 

Yeah, didn't touch the relays under the dash yet. They checked all over for mouse damage and didn't find any. The air filters were replaced about a year ago and are fine. I was hoping the extra smoking when they started it meant it was flooded. Maybe I didn't deflood it long enough, who knows. Going to pass it off to a mechanic who knows rotaries and just bite the bullet on this one.

 

Just realized I hadn't uploaded a video. Here's what it sounds like.

 

It's alive!! The spark plugs were bad

I guess maybe due to moisture or something, I dunno. They were replaced about a year ago (maybe 5,000 miles on them at most).

 

5000 miles is about right. Change the oil then as well. Consider changing the O2 sensor more frequently than you think.