Hi all,
I have a turbo NB Miata. It's a ton of fun but I think I'm in over my head trying to iron out some of the remaining wrinkles.

My target is ~8 PSI boost. I am happy to spend more time/money to prioritize reliability.
I have hardly any experience with TunerStudio but I am comfortable with math.
I swapped out the ECU, confirmed ignition timing, got a shop to install the turbo + exhaust, and then got the car tuned by a tuner. It generally runs well but for some reason is now rich overall.

I would appreciate any advice you might have.

Here's the build:
Vehicle: 2004 NB (~85k kms), 1.8L
ECU: MS3 Pro PnP
Injectors: Flow Force 640cc Injectors
Turbo: Flyin' Miata custom spec turbo system for NB chassis (GT2560R turbocharger) - also with coolant reroute
Blow-off/Bypass Valve: Recirculated bypass valve
Downpipe: FM 3" downpipe, NB chassis
Catalytic Converter: FM 3" catalytic converter for 3" downpipe, NB chassis
Muffler: MagnaFlow 4 X 9in. Oval Straight-Through Performance Exhaust Muffler 11229
Clutch: Supermiata Sport Clutch 94-05 Miata 1.8
Gas: 94 octane (located in Canada)

Electronic Boost Controller: none
Fuel Pressure Regulator: stock
Fuel Pump: stock


Here are some specific problems:
1. Rich Idle
It idles somewhat rich. If I lean out the idle squares on the VE table, I find that the idle tends to oscillate.
I went over the data for my injectors and I'm not sure if I should be changing the injector size (in Engine and Sequential settings) from 640cc to 752cc based on this:
https://goflowforce.com/pages/instru...3LoniVCA9qQxhM
I tried changing it and the car wouldn't start. I'm guessing I'd need a full re-tune.

2. Occasional Stalls
These typically occur on inclines or if there's any load on the electrical system (e.g. A/C). Do I just need to adjust the PWM Idle Voltage compensation? Is there an objective way to do this?

3. Foul Smelling Exhaust
Despite using the FM catalytic converter, the exhaust is quite foul smelling. I am guessing that this is related to the rich idle resulting in unburden hydrocarbon products, but not sure. Would this still be expected even if the tune was improved?

4. Runs Learner over Time
The longer the drive is, the more the car runs lean. Is this due to heat soak? Is there anything beyond what's in the the Flyin' Miata kit that I should be doing in the engine bay?

5. Do I need to install/upgrade other equipment to help address the above problems?
The tuner recommended an aftermarket fuel pressure regulator, but based on what I've read on this forum it seems like that is not necessary when pairing these fuel injectors with the Megasquirt 3. Is that necessary?
I also looked at upgrading the fuel pump. It's stock but also >20 years old. If replacing it makes sense, with my goals it seems like a Deatschwerks DW200 would be reasonable.
I also already have an electronic boost controller that I plan to install.

I attached a log with a drive. I wasn't able to go into boost much, but I think the main issues are outside boost anyways.
I've included some pictures of the engine bay and undercarriage in case I missed anything that could be important.

Thanks for any suggestions!

 

Attach your tune too. We can make some minor changes that'll help with all your issues.

You're right on #1, changing injector size will require a retune. You could scale by a percentage, but I wouldn't trust that to be enough. But questions 2-4 require looking at the tune.

Edit: holy **** you weren't kidding, 11:0 idle. Yeah that'll stink. Post the tune and we can help more. Idle advance isn't doing anything to help, you should be able to idle at 14.7 easily.

 

Thanks for the reply. I thought I had saved the tune with the log but it looks like it didn't work. I'm 90% sure the attached tune is the one corresponding to the log.

Would it be fine to just leave the fuel injectors set to 640 cc? They are technically 640 cc but the FlowForce site makes it seem like they're more on NB Miatas with the higher fuel pressure. Does that affect anything if it's already tuned assuming that they are 640 cc instead of the 752 cc adjusted for NB fuel pressure?

 

Just my $0.02
I ran my stock NB fuel pump/rail/etc... with a set of FF640's for a good while without issues. I liked those injectors quite a bit and would absolutely expect them to idle fine at 14.7 on pump gas. I have similar issues to what you describe on my 1050cc injectors which I assume is them being non-linear at those pulse widths.

Interesting too that Overrun fuel cut is off and boost control is enabled. I also noticed that launch control is enabled, which I fully support, but wasn't sure if you were aware of that. At a glance the ignition table seems decent, maybe a bit relaxed in cruise (as is the AFR table) but nothing major there.

 

Try the attached while logging, it's a bunch of changes to see if it'll actually control idle afrs, might need more changes. The rest of the tune looked ok, but definitely some unrefined parts.

 

Thanks for the helpful replies.

Awesome - I'll keep it stock then unless it should just be replaced due to age.

Would I be correct to assume I'd do all of the fine tuning of these tables after the idle is fixed?

Ok great, thanks. Would it be best to fix the tune first and then turn on EGO?

Thanks. I already own an EBC so I'll install that but otherwise I'll leave it as is. I don't plan to run E85 or go much beyond ~8-10psi so I'll plan to swap in the DW100 when the other one gets too old. Would you ever swap the fuel pump prior to it dying? Is there a risk of engine damage if the fuel pump fails under boost?

Should I turn on overrun fuel cutoff and disable boost control? It was not a deliberate choice by me but maybe the tuner chose those options. I already have an EFI boost control solenoid kit but haven't installed it yet. If I install it, then would boost control make sense?
I have no idea why launch control is on. I don't think the clutch is wired into the ECU so I doubt it's doing anything. I'm not particularly interested in launch control at this point.

Amazing. Thank you so much. I'll give it a shot.

 

I tried out the tune that you sent. In terms of RPMs, it idles great - even with the A/C turned on. It'll be a huge relief to not have to look at the RPMs every time I come to a stop. I really appreciate the changes you made for drivability.

The start was a bit jerky but I've had that before too. After the start, the AFRs at at idle were initially rich but that's probably WUE or ASE, right?
I've attached a log of a brief drive. I tried turning on the A/C part way through.

 

Try the attached, just warm idle, you should be able to idle steadily at the 14.2 target, but the little idle I did see didn't seem happy, oscillating between 13.6-14.6, but it never had a chance to steady itself.

 

Again, just my opinions/theory. Curly has way more real world experience with this stuff.

 

Thanks for all the feedback.

I tried the latest tune today and attached a couple of logs. The 6:57 AM one has a stall that occurred when I was coming to a sudden stop at the bottom of a hill. If I'm reading this correctly, it looks like the MAP spiked causing the stall. I've had this issue before and found out that it was the clutch compressing the MAP vacuum line to the ECU. Looking under the dash I don't see anything obvious but I guess it could have happened.

Thanks again for your help.

 

Hi.
I am not familiar with the MS3 pro, but I do have some comments.

I notice that you have an injector dead time that loods to be set for 1.27mS at what I assume is 13v. The web says that this brand and size should be set to 1.15mS.
If this is off, then this is not going to cause your problem, but it won't help you dial in your idle.

Looking at your log, I don't understand a few things.
-at time stamp 243.584 and 252.387 your AFR mysteriously drops to 25:1. I can't see anything else in the log that would reflect that the MS3 is trying to make this happen.
maybe I am missing something that someone else can comment on.

-time stamp 255.176 the ignition timing gets bumped down to 5.3 degrees. This appears to be caused by idle rpm timing correction being "on" and having large amounts of timing correction in there.
maybe Curly can comment.

-at time stamp 254.476 the rpm's are decreasing and the map is rising. This is where the engine stalls. The fuel map is particularly bumpy 900-2000rpm at 60-90 map, but they are somewhat bumpy everywhere
So why is the AFR stable as hell through this time period. (and the engine cannot stay running) This make no sense to me.

What part of Canada are you in?

 

It's a really rough tune just to see if it'll idle at leaner than 11:0, haven't looked at the rest of the tune yet. The dead time could very well be the reason it's doesn't seem to be stable much leaner than 14:0, but that's still loads better. I did activate aggressive idle timing correction to stabilize RPM, haven't had a chance to look at the latest log though. Oreo's doing some good time stamps to look at though!

 

Sorry for the big delay in replying. Got swamped at work.

I'm basing it off of the FlowForce instructions here: https://goflowforce.com/pages/instructions
I'm using the FF640 table with the 4.0 BAR column (NB Miata)




I'm not sure why this is happening. Since stalls and very low RPMS (without a complete stall) happen to occur on hills when the clutch is already in, wouldn't that point to a mechanical rather than software issue? I was thinking that perhaps the fuel pump struggles on inclines. Could that be a possibility?
Alternatively, perhaps the rate of RPM decline is too rapid for the software to compensate. The lightweight flywheel probably does not help. I would expect that to occur when rapidly stopping at flat locations as well, which doesn't seem to be the case.

Edmonton

Thanks again for the tune.
I have attached several logs, some with stalls that almost always occur when coming to a stop with the clutch in on an incline.

For 2026-05-27_18.29.47, at 128 sec the car stalled. This was coming to a rapid slowdown on a hill.
For 2026-05-27_07.00.27, at 320 sec the car stalled. This was again on an incline. I see that the MAP dropped low just before the stall and then rose quickly. I put it into 4th gear since I still had some speed to restart the engine while moving. I'm guessing that is not a good thing to do but I was in traffic.
For 2026-05-29_19.01.54, at 101 sec the car stalled. This was also on a hill. The MAP oscillates just before the stall. I can't figure out why it happened.