stock idle AFR should be about 12-12.5 with no airpump. its a little stupidly complex, but with the restrictive stock exhaust, there is a lot of exhaust gasses that stay in the chambers, which means Mazda needs a rich mixture to not have the engine misfire, as misfire = high emissions.

if you don't mind a couple of misfires, you can go leaner, and if you have a freer flowing exhaust you can go leaner still.

if its lean the idle will get "weak" or it'll randomly stall.

i'm under the impression that when the start pin is activated on the ECU, it just fuels by coolant temp, and RPM. so a non flooded intermittent no start can be the coolant temp sensor.

 

Wow. Pics worth 1000 words here. Basically appears that your AFM is worn out with a dead spot when closed. Sending PM.

While the coolant temp sensor is the main determinant of crank fueling, I don't think the AFM is totally ignored by the ECU. See diagnostic tables within Section F2 of the FSM for evidence of this.

 

here is the NF01 boost sensor graph. unit at the bottom is mm/hg

 

It's 3" exhaust all the way back, and from the wideband readings, I was seeing around 12.1 AFR's all the way through the rpm band while just free-reving the engine with the fuel maps set to stock.

If i pulled up to 35% fuel on the low throttle setting at 800rpms using SAFC the idle would get weak, but when the engine randomly stalled yesterday, I was actually holding the throttle around 3000-3500 rpms and watching the wideband and the revs just dropped and the engine stalled. Then it would not start again until I pulled the MAF connector.

We have some rain rolling through right now, but if I get a chance to check the temp sensor today, I will do that and report back with the information for peace of mind.

Thanks for the graph too, more information on parts being used on my car never hurts.

 

Got an N370 MAF from another member on the boards this week. Hooked it up and the car started right up, idled fine and drove perfectly. The wideband is going back on this weekend to retune for the stock injectors.

Thanks to everyone for the help, having some others to bounce ideas off of most definitely kept me on the right track. Much rotary love.

I'm probably going to try and find another NF01 AFM for later down the road, when I do go standalone and hopefully get a better understand of its differences in resolution from that of a N370 AFM.

Thanks again everyone

 

cool! glad you figured it out. i'll scan in the page for the NF01 and N390 (13B cosmo) AFM's, in theory the cosmo AFM should support larger injectors and read more air flow (hp), but you'd have to figure out what size injectors

 

Terrific news Ian! I'm happy you solved your problem.

Forget about another AFM, it's archaic technology. All modern ECU's use 'speed/density' to determine fuel demand and ignition timing. That means no AFM, they rely solely on the pressure and temperature sensors for thermodynamic inputs to calculate airflow.

 

J9, that was my thoughts, but I know I'd have to use a different EMS to support larger injectors. SAFC advances timing too much if I have to decrease the fuel maps to get the proper AFR's. I'd still appreciate the information if you would be kind enough to PM those pages to me.

Monty, I think you're right. Standalone is something I most definitely need to look into, it would make future tuning a lot easier. In the nearer future, preventative mods are going to take place. Starting with an aftermarket FPR and s4 secondary rail to support the fpr.