Cranking issue, starter gets hung up/kicks back
 

1990 miata
94 1.8 w/ jrsc
mspnp 90-93
460cc

Read a ton of threads about starting problems, but only seen this mentioned once. The car is and also has been hard to start in the cold, i know that was related to my cranking fuel settings...but its a track car only so i wasnt that concerned. However an issue has come up that is more of my concern, the car will "kick back" when trying to start at times. I can't just hold my starter button down to get the car to crank for longer then 2-3 seconds.

It's almost like a distributor type car with the timing too far advanced. It's actually done is so violently once it kicked the timing belt off a tooth. I've played with the cranking advance and it doesn't seem to help. I actually just replaced the starter today because mine died (housing cracked, more then likely from the kick back issue).

My trigger angle is set at 65 +22.5, cranking advance was set at 18, but ive since changed that to 8. A friend of mine thinks it could be literally a fueling issue, as in the injectors are pushing too much fuel during cranking and hydrolocking the motor. However it usually takes me 4-5 tries to get the car to fire off if under 40 degrees, so if it was that flooded i don't think it would start at all. At higher temps the car starts first or 2nd try consistently.

I'm sure someone has gone through this issue, so please any help is appreciated.

Have you verified the cranking ignition timing with a light? The only thing that comes to mind, is that some coils with integrated ignitors will trigger early if you use too much dwell.

Backfires could also indicate your cranking PWs are too lean.

I have not, however with how little i can crank before it typically does it i find that it would be hard to get a reading. As for the coil, its using the 1.6 stuff still, so its seperate.

Doesnt act like a backfire, no noise...just stop cranking, gets hung up.

If anyone needs any more info from my msq file let me know. Dwell settings and what not are still set to what they were on the base 90-93 mspnp file.

I'm going to play with some more cranking settings tomorrow for the fuel. Also may try unplugging the injectors and cranking the motor to rule out flooding possibilities

ecu support full throttle flood clear?

Mine did that...then I got an optima battery and everything was fine.

Looking forward to an answer.

I've seen this too on a few occassions. Cold mornings, first start of the day. Temperatures have risen here in Texas, so I wasn't able to get a satisfactory solution due to inability to consistently duplicate. Also, haven't worked on my warmup settings all that much -- have been focused on warm idle, getting my larger injectors working and the turbo install.

A few rules of thumb for cranking:
1. Run as little advance as possible. 0° works just fine. Unfortunately, I have a MS-I (MSPNP) and a separate cranking advance is not an option for me.
2. Cranking in the cold takes a LOT of fuel. And, the fuel will puddle. No getting around that unless you're direct injection.

You are NOT hydrolocking with fuel. That's just dumb. Wouldn't waste my time chasing that. Have you compared your injector volume to your squish volume?

Let us know how you do. I think this problem is pretty common.

Needs new starter?

I encountered a situation with a weak starter where the starter would not have enough torque to spin the engine quickly enough against the compression in the cylinder. The spark would dutifully fire before TDC but the starter, not moving things quickly enough, would not get the piston over TDC before the cylinder pressure went way up from combustion. This caused a kickback against the starter. Battery was swapped before finally replacing the starter, which cured the issue.

Your results may vary.

The contact at the ignition switch gets dirty and eventually doesn't pass enough current to the starter. iirc. I currently have this issue and traced it to the ignition switch. I'm too lazy and busy to open it up and clean the contacts. i replaced the starter when my trans was out just in case (didn't fix it), and my batter is fairly new.

Next time starting seems weak.... Jiggle the key and lift firmly on the key as you are twisting (use common sense, don't break the f***er off!)

it could be the starter or battery, but my guess is the ignition switch.

just post your msq. im sure its just your settings.

no tps

Have a new battery in the car already

Im not sure what you mean by squish volume haha

I just put a new starter in last night, havnt had more then once chance to try to start it cold.

I suffered from this and installed a starter button

It's on my laptop ill get it up here later after i try a thing or two tonight.

but just to confirm, you only have this issue in cold?

well that's when its noticeable because the car takes longer to start. When its warm it fires off usually first try.

I actually just went out and tried cranking it cold and it didn't kick back at all while i messed with settings. So could've just been a faulty starter. I'll update as i find out more.

I've only experienced it on first start on a cold morning. I have a new battery and new starter. Unsure about ignition switch condition. Also, need to look at the battery cables (pretty sure the ground cable is a bit frayed). Slow starter speed is a real possibility. I will say that it is really disconcerting when it happens.

Squish volume is the volume left in the cylinder when the piston is at TDC. Our engines are non-interference (thank goodness). There is still quite a bit of air volume in the cylinder at TDC -- much more than a fuel injector can fill -- although I'm having trouble finding a reference for exactly what that volume is right now.

Well the car still does it, but only after 2-3 failed attempts of starting. I've been adjusting the cold start settings...if i get those right then this is no longer an issue haha.

Track car right?

Any progress on this? I've been experiencing the same ever since my turbo install. Not as violently as what you've described, but it's unnerving.

I'm still having this problem too (MS1 V3.0 on a 91), the issue with mine is noise on the CAS signal which causes the MS to think the engine is revving beyond cranking speed (I've seen 3000rpm on datalogs! - at atmpospheric pressure), this means the MS chucks in a load of fuel and a load of timing.

I have a filter 0.1uF CAP on the MS to try and filter any noise, I have a known good CAS which I will fit soon to see if my CAS is marginal.

If you've double verified that your timing is dead-on, it sounds like you don't have enough cranking puslewidth at lower coolant temps.

Might want to investigate how the CAS is wired to the MegaSquirt. Moving the wiring away from things like coils and using shielded wire is helpful. MS2 and MS3 have some software noise filtering capabilities that can also help with oddities like that.