Edit - See post #12 and #17 for relevant info on the cam trigger and a ~4000 RPM misfire it was causing.

Feb 21, 2024 - Currently I've found that custom FW with a small overflow fix is necessary to use the 36-2 wheel without issue. Others have had success with other cam trigger profiles, but they did not work for me.
Leaving the info below as I think some of it is still relevant.

I did double check my timing and updated the trigger angle to 208 to be spot on, but at 210 it was still very close to perfect.

End Edit

Hi All,
I was setting up the Flyin' Miata 36-2 trigger wheel earlier today and was surprised that I couldn't find any Speeduino specific posts about this, so I figured I'd make one.
Keep in mind that I only set this up today, so I don't have a ton of data backing this up.
Regardless...
First off, this is on a 2000 NB with a BP4W using the stock sensor and location. My ECU is a SpeedyEFI PNP, but this should work for any Speeduino based boards.
FM has instructions on how to setup the trigger wheel for the Hydra, as seen below. NOTE, THESE ARE NOT ALL CORRECT FOR SPEEDUINO

Replaced this with some poor writing on top because it was showing up in google searches and I wanted to make it obvious that these were not for Speeduino.



These are largely correct, but the trigger angle is incorrect. I'm not sure how the trigger angle works in Hydra, but in Speeduino it is degrees ATDC that the first tooth (the first tooth after the 2 missing teeth) passes the sensor.
I originally set my trigger angle to 80 and my car would not start, and was backfiring like the Air/Fuel mixture was off.

Conveniently, FM has a photo in their instructions which shows the teeth relative to the sensor.

Photo from Flyin Miata's instructions

Each of the tooth/gap pairs on the wheel should be 10 degrees, given that there are 36 teeth and 36 gaps (if you imagine the missing teeth are there).
Counting from the gap to the sensor, it becomes pretty obvious that there are not 8 teeth between the sensor and the gap at TDC. I counted 21, which led me to a value of 210 degrees.


Marked up for clarity

The rest of the settings are largely like FMs, although for Speeduino it should be Missing Tooth, 36 primary, 2 missing, Crank speed, as shown below.




With the settings shown above, I was able to fire the car first crank. I verified the timing with a light and it looked spot on. It was the first startup for this engine, so things were a bit hectic, and I will likely go back and double check everything later, however at this point I feel confident in this setup.

Feel free to correct or comment if you see something incorrect. I'm far from a pro, just figured I could point someone in the right direction and save them a bit of headache.

 

Did you make any other changes to the trigger setup? I've tried your identical settings but my car was running extremely rich (9-10AFR) and struggling to idle. I swapped the trigger wheel to the OEM again to see if anything had changed but it fired right up and idled perfectly.

Did you try changing the gap and if you did it have a stark difference in how it ran? I had it gapped to the recommended credit card thickness which is about 0.030".

 

Oddly enough, I'm actually chasing the same issue right now. My car doesn't want to idle above ~12.0 AFR. I just rebuilt my engine and changed injectors so I assumed it was from another change I had made. I did take the car out on the road and it was driving very smoothly through the rev range

I believe I gapped mine to about the same. I used the largest feeler gauge on our set but I can't remember the value of the top of my head.

I am not seeing sync loss while the car is running.
​​​

 

I did just find this on the speeduino forums. This was posted a couple weeks back. Seems like a switch to running semi sequential injection might be necessary

 

Yea my hunch is its definitely timing related, something like too much fuel for when spark is happening. My plugs were completely fouled from maybe 10mins of testing at idle but didn't really want to push it due to how poor it was running. Did you also try fiddling around with semi sequential injection to see if it made any difference? Wish we had a setting like on Megasquirt where they just have Miata 36-2 too would make it so much simpler.

 

I have not played with fueling settings yet. Like I said, I changed a lot (compression ratio, trigger wheel, injectors, valve job) when I rebuilt my engine, so there's a lot I'm working out right now. My engine also started making some noise that might be a bearing failure, so I'm focusing on that.

I am planning to do some diagnosis on that noise today. I am going to research more on the fueling options and will see if those have any benefits today when I fire the car up. I'll update here with my findings.

 

I have two potential changes that might help. I've seen other threads where people running MS suggest switching to a semi-sequential injection setup. That is option 1.
The other, based on another forum and anecdotal evidence from a friend running a 36-2 wheel is to use the "4-1 cam" for the secondary trigger type. I saw this mentioned on the Speeduino forum as well, although it sounds like it might affect VVT.

Research is ongoing. I've got a question posted in the Speeduino discord, so hopefully that yields some results. I'll update as appropriate.

 

I'm going to do a bit of a brain dump here. Hopefully it will help someone else along the way.

What I've gathered is that the Miata's cam sensor pattern is not supported by Speeduino, however the combination of the stock crank sensor and stock cam sensor is supported in the various Miata trigger pattern options.
What this means is that when moving to a 36-2 wheel, Speeduino based ECUs (such as the SpeedyEFI PNP's) don't have a reference point for the orientation of your cam.

People suggested moving to semi-sequential fueling which I didn't understand at first. However, as I thought about this and realized what was going on I started developing an idea of why that was. This explanation might be wrong, but it makes sense to me.
If there is no reference for the cam position, then the ECU has no idea which set of intake cams are open. Imagine setting the crank to TDC with your valve cover on and trying to tell which set of valves is open.
That's why semi-sequential is suggested when no cam signal is available, because there are multiple possibilities, so you need to spray both to cover your bases. This is a downgrade from sequential, but not a catastrophic.

I plan to try semi-sequential today and expect that will work better. Opposing lobes on the cam look like they're 1&4, 2&3 for a 2000 Miata, same as firing order, so that will be the injector pairing I'll try.

If anyone more experienced has corrections, confirmation or critique please chime in. Again, I'm slowly wrapping my head around this and want to make sure I'm not missing something.

As I posted above, there is supposedly some code that has been developed to support the Miata's cam pattern on Speeduino, but it isn't tested/integrated yet (as of Jan 2024). This is a common trigger upgrade, so I assumed it was already supported, but I didn't do my research and now I'm paying for it.

When I was running the single tooth cam setup, I was seeing full sync in my datalogs, but I'm not sure how that's possible if what I've said above is correct. I've also heard anecdotal evidence from a few people that the 4-1 cam signal works, but that didn't work when I tried it.

 

so, i'm on this same boat however, I used your settings and it worked right up.
i did set the signal filter to OFF as weak was giving me sputters.

hopefully they release an update to this. i am staying with the FM with this settings until then.

 

I just went out and tried a variety of settings.

Semi-Sequential did not work for me. I was talking to HJPro earlier today and that's the setting that his car liked.
Sequential with the 4-1 cam pattern did not work for me. That settings is working for a friend that I autoX.
Sequential with the 1 tooth cam seems to be working on my car. I'm getting full sync and no sync loss.

I really don't understand how or why that is the way things are working. As far as I know, all of the cars listed above are running 1.8 BP4W Engines with SpeedyEFI boards.
I've got my car idling well enough at this point on open loop. It's hunting around a few hundred RPMs, but not wildly. I've got other things to investigate (IE knocking sound ), so I'm going to leave this be for now.

It seems like most people are finding some combination of settings that works for them, so I guess for now the strategy is to try the combos and see what works. Or, if you're smart, don't swap your trigger wheel until this is sorted.

 

Too late for not changing. Although it did take me 10 mins. I’m lazy and stubborn. Car drives and idles fine.

I have another issue I need to sort out and it is starting. But it was having that issue even before the wheel swap.

im gonna message you if that’s ok.

 

Posted this elsewhere, but wanted to leave it here as well since it's relevant. I might change back to the 4-1 cam pattern and see if that works with my revised fuel table.

I fixed the misfiring this weekend. I'm going to do my best to explain it here although I'm still trying to figure out exactly what was going on. TLDR Fueling issue related to my trigger setup on Speeduino

Speeduino does have support for a missing tooth trigger wheel (36-1, 36-2, etc...). However there is currently no support for the Miata's cam trigger. In theory this means you need to run semi-sequential fuel, however I had my car setup with a different trigger pattern and the car seemed to be running fine. I tried a few other settings including semi-sequential and a different cam pattern, but the car didn't want to run. I DID have the car fire up, but it was dying as cranking enrichment tapered off. I believe I could have changed my fuel table and corrected this, but at the time it didn't seem like an improvement over what I had.

I found some custom firmware that was released a few months back with support for the Miata cam profile. I compiled and installed that yesterday. I had a similar experience where the car was dying after cranking. Convinced I now had a valid trigger setup, I started adding fuel. I had to nearly double my idle fuel and retune my entire VE table, adding ~15-30 points across the map. Maybe more in some places. I would not be surprised if the trigger setup is added to the next firmware.

With that change I'm now able to rev the car happily into the 5K range, although I'm waiting to put a few more miles on the engine before I want to get to redline. There is no trace of the hesitation I was seeing before.

I believe that the cam signal is never referenced once the car is running, so I'm not sure that the cam profile actually fixed the issue. Like I said, I think one of the other configurations I used would have worked as well if I retuned my fuel table. I was not seeing sync loss before, I was hitting my AFR targets, and I was setup to run sequential fuel.

My idle is still unhappy above ~12 AFR, but with some new confidence in my trigger pattern I'm going to see if I can correct that via tuning. I also added some fuel injector cleaner to my last tank of gas, so I'm going to see if that does anything.

 

More info related to the post above. When I switched to the FW that supports the cam pattern, I had forgotten to plug my cam sensor back in after doing some debug and capturing some trigger logs. So I was running at half sync and semi-sequential fuel. That seems to be what fixed the misfire. Presumably I was running sequential fuel and it worked up to ~4K RPM.

When I realized that I had left my cam sensor unplugged, I corrected that and immediately saw full sync. I had noted that the car was really jerky on light throttle, such as cruising. The idle had to be super rich and was still erratic. Getting back to sequential fuel fixed those issues. My car is now very smooth on light throttle and my idle is significantly improved (although I think some voltage correction might be causing an issue, but that's separate).

Long story short, if you're going to run the 36-2 wheel I would recommend getting the FW that supports it. I'm guessing that this cam pattern will be in the next Speeduino FW release, but I'm not sure. It sounds like some people have their cars running without this FW/cam trigger pattern, so YMMV.

 

and what firmware is that?

wonder why they removed it.

 

It isn't an official FW, it's code developed by someone who has written some other cam trigger patterns. It hasn't been removed from any FW, it hasn't ever been added because it was only developed a couple months ago and I' m not sure how much testing has been done on it.

The FW is located at the link below. Use at your own risk. You'll need to compile it, but there are instructions on that one the Speeduino docs. I did have to change the naming of the header file for the time.h library when I was compiling it. I believe it's based on the firmware from ~Oct 2023.
https://github.com/mike501/speeduino...h%2B1Secondary

 

Awesome.
yes, I am no stranger to Visual Studio

 

Just to be thorough, the custom firmware fixed most of my issues. In the latest speeduino FW release there is an overflow bug that causes injector pulsewidth to drop almost to 0 at 914 RPM. That was what was causing my stumble at idle after switching to the custom FW. There is a fix in the master branch of the Speeduino FW.

I let the creator of the FW know that I've been using it without issues. Sounds like he's going to try to get it moved into the master branch so it will become available once it is reviewed/pulled in.

In summary, as of Feb 2024, my car is running flawlessly using the 36-2 wheel with the custom firmware + overflow fix.

Relevant links :
Cam Trigger Pattern FW : https://github.com/mike501/speeduino...h%2B1Secondary
914 RPM Overflow Fix Commit : https://github.com/noisymime/speedui...5f5201db0e3a69


Settings I'm using currently

 

mind sharing your compiled FW?
i wanna give it another go and see if I have the same luck.

 

Not sure how to go about that. I was looking into it to share with someone else though.

Export compiled build or something like that in the Arduino IDE? I assume you're on a 2560 based board?

 

you could zip up the source files and share them on email?