Miata Turbo Forum - Boost cars, acquire cats.

Miata Turbo Forum - Boost cars, acquire cats. (https://www.miataturbo.net/)
-   MEGAsquirt (https://www.miataturbo.net/megasquirt-18/)
-   -   Engine dies when I hit rev limit, comes back with a key cycle (https://www.miataturbo.net/megasquirt-18/engine-dies-when-i-hit-rev-limit-comes-back-key-cycle-82748/)

Mech5700 01-20-2015 11:28 AM

Engine dies when I hit rev limit, comes back with a key cycle
 
4 Attachment(s)
Can't seem to figre this one out... started happening recently. Hit the rev limit, engine dies out. If I'm in gear and coasting down, cycle the key and it comes back to life. When i work the throttle as the rpms are dropping, afr responds like it's still getting fuel. Seems like I'm losing spark. Lowered rev limit to 2k and it doesn't do it. Historically I lose sync when I hit rev limit, but never has caused any major issue. This happens with spark cut or fuel cut as well. No scary noises or anything mechanical going on. Runs like a raped ape till rev limit. A couple things have changed before this started happening, not sure if related or not.

-wired for seq ignition
-had a tuner messing with it on the dyno

Log and tune attached.

Halp :(

Braineack 01-20-2015 12:27 PM

It really shouldn't be losing sync when you hit the rev limiter.

can you take a composite log of this happening to see what the cam/crank sensor inputs look like when you hit the rev limiter and get out of sync.

Mech5700 01-20-2015 01:20 PM

It always has. Along with a voltage spike and battery light... we talked about that via email but never came to a conclusion.

Braineack 01-20-2015 03:05 PM

yeah can you do a composite log of it? gotta go to the diagnostics tabs, check the save to file button.

get it to cut out, and stop the log. then email/upload.

Never seen or heard of this happening so I wanna see what the inputs are doing when you lose sync.

Mech5700 01-20-2015 03:12 PM

Ok. I'll get a log on the way home tonight. It gets a code 31 and 32 I believe btw.

Mech5700 01-20-2015 04:09 PM

2 Attachment(s)
Was able to do a quick log at lunch. Hit rev limit in 1st (didn't die), shifted and hit it in 2nd, hit rev limit and it died. quick key cycle and it did not come back to life. turned off for a second and cranked back up in neutral and it started back up while coasting.

Braineack 01-20-2015 05:49 PM

that log doesn't show any loss of sync. That log is literally the signals your cam/crank sensors to sending to the MS. It's not logging one missed event.

but your datalogs do.

that's really interesting...

Mech5700 01-20-2015 06:46 PM

So what do you suggest we do? Can there be possibly something funky with the megasquirt itself?

Mech5700 01-20-2015 08:35 PM

6 Attachment(s)
Lowered rev limit to 7k, with the hysteresis at 100, spark cut only, and got much much better response. No tach flopping, only had batt light flash maybe once (when it lost sync and voltage spiked), and didn't stall out. Tried 7100rpm and it was hit or miss. Would sometimes stall, sometimes not. Switched back to 7k and it was good again. First log attached shows the one time it lost sync, the second log shows 3 consecutive redline shifts without any problem.

Mech5700 01-20-2015 08:58 PM

might be a long shot, but I found this while searching sync loss 31 and 32.

Megasquirt Support Forum (MSEXTRA) ? sync loss #31 ms3x (View topic)

Not exactly sure of the science behind it, but it kinda sounds like I could experiencing something similar? Will run fine up to 7k, dislikes 7100, and absolutely hates 7200?

hrk 01-21-2015 05:22 AM

Subscibed, mine does it at 7600 rpm. Have logs and info gathered and MS3X being looked at DIYAutotune if they can reproduce it on their bench.

Braineack 01-21-2015 08:44 AM

I might have you pull the unit and open it up and have you readjust two pots on the inside to help clean up the crank/cam signals to see what effect that has on it. I may have them set for a 90-97 install, where the 99+ need different settings.

only thing I can think of right now.

I'll email you detailed instructions on what to do.


the developers released code that should prevent the voltage spikes on any sync-loss event. I can walk you through upgrading to this firmware as well if that doesn't solve the issue--you really shouldn't be seeing sync-loss at all.

DNMakinson 01-21-2015 09:12 AM

Scott, please post your preferred FW update procedure here (rather than off-line) so we all can learn. Thanks.

Mech5700 01-21-2015 10:23 AM

Ok, mech5700@yahoo.com

Instructions on pot adjustments and how to do a firmware update, pleeeeeeeease

I'll copy over what scott sends me if he doesn't post it up, Dave.

Braineack 01-22-2015 10:57 AM

2 Attachment(s)
setting crank/cam pots.

Remove the MS.

Remove the two studs holding the Serial port to the case.

Remove the four screws on the dual DB37 side of the case.

Slide the PCB out of the case.


Lift the wide ribbon out of the way to expose the adjustment pot you need to turn.

https://www.miataturbo.net/attachmen...ine=1421942233

It's R11, the top one closest to ribbon connector labeled "adjustment for cam input".

Using a small percision screwdriver, turn it 6 turns CCW.

Now turn it back 3.5 turns CW.

remove the two studs holding the green expander board to the case cover.

gently set it out of the way to expose the crank input adjustment pots.

https://www.miataturbo.net/attachmen...ine=1421942233

shown above it's the two blue pots.

turn the top most pot (middle of board) 12 turns CCW. (same units do not have this pot, ignore this step if you have a resistor installed there instead)

turn the bottom most pot -R56 (edge of board) 12 full turns CCW.

now turn it CW 10.5 turns.

reinstall the expander back into the case cover.

carefully slide the pcb into the case (third slot up)

carefully position the expander board into the case (fourth from top)

replace the 4 screws.

replace the studs on the serial port.

reinstall the MS (dont forget the MAP sensor).

test it out.

Mech5700 01-22-2015 11:18 AM

Got it.


And for the firmware update?

Braineack 01-23-2015 08:35 AM

It's a simple process:

D/L the latest firmware. Pre 1.4 beta 10: MS3 pre-1.4 beta 10 firmware released - Megasquirt EFI

1. Save your current MSQ file.
2. DISCONNECT coils and/or Ignitor. The flashing process can damage coils if they are left connected during the process.
3. Turn car to key ON and connect laptop to MS.
4. run the ms3-loader.exe (or whatever it's called) from the firmware directory.
5. follow the prompts, you can go through just clicking enter for the options, it will find your MS com and everything.
6. let the loader run, takes about 2-3min.
7. Once it's complete open TS and create a new project. Detect your firmware and let TS find and d/l the new firmware. [I like to create a new project for any new firmware I run]
8. Once connected, load the msq you saved. Look at the error log page after loading, spot check the settings to make sure everything ported over correctly. Some firmwares make changes that add options or change how things work so you need to make sure nothing is going to be wrong. Typically new values have some crazy high value in there that are obvious.
9. double check your AFR, AIT, CLT calibrations. They shouldn't have changed, but doesn't hurt to make sure.
10. turn off MS. Reconnect coils.
11. Start car.


Every once in a blue moon the flash might fail, where you'll have to open the case, slide out the board and install a BOOT jumper on the ms3 daughterboard to attempt another reflash.


I'd try the adjustments first and if that solves the issue I'd wait until 1.4 is a final release before running the new firmware. Beta firmwares can have random bugs.

Mech5700 01-23-2015 11:38 AM

I don't care what they say about you, you're a swell fella.

Thank you. Will report back within the next few days.

Braineack 01-23-2015 12:21 PM


Originally Posted by Mech5700 (Post 1199579)
I don't care what they say about you, you're a swell fella.

lies.

831Mazderp 06-05-2017 10:12 PM

Ive been experiencing the same issue with the limiter at 7k with spark cut; what became of this?






Originally Posted by Mech5700 (Post 1198431)
It always has. Along with a voltage spike and battery light... we talked about that via email but never came to a conclusion.



All times are GMT -4. The time now is 12:38 AM.


© 2024 MH Sub I, LLC dba Internet Brands