MS2/eextra 3.57 Cuts out at 4000 rpm

[Goosenhaben]

Greeting all.
This little Civic of mine's been giving me issues that I haven't been able to sort out. Even after reading the manual and surfing the archives for answers.
What is happening is that the engine will cut out at 4000 rpm. I've messed around with the noise filters and for a time got it up to 5000 rpm.
In fact it runs and drives very well, until you hit 4k rpm.
The distributor is new, injectors professionally cleaned, engine's been rebuilt. All parts are off the shelf stock. The vehicle is a '91 with '99 intake, injectors and distributor. I custom built the wiring harness to accomidate.
I'm using MS2/extra with 3.56. Tuner studio MS v3.0.22 (MS2/Extra release 3.3.3)

2017-08-06_13.15.28.msq

I'm an ASE master L1 certified mechanic with about 20 years experience so I'm no stranger to electronicly controled devices on vehicles.
But I haven't, until now, started working on the 'front door' as it were. My knowledge and experience has helped with other problems but his one has stopped me.
From what I can figure out is I am losing spark control. It just stops. Like you've hit the rev limiter.
Now I have messed around with the noise filtering and it has just made it worse. If I adjust the noise filter lower that where it is, the ceiling drops to like 2-3K rpm. If I enable tach period rejection, it doesn't change the 4K limit and the engine becomes hard to start and stalls at idle.
So here are some screen shots I ve made. Hopefully they will help. The first is at idle. Yellow is attached to JP1, green is attached to J1.

jp1 yel jp grn at idle.jpg

And here we are at 4000 rpm. The yellow trace is moved to coil control out of the MS. In my case, to the ignitor in the distributor

grn at JP yel at coil out.jpg

For grinns I had traced to where on the board the coil control comes out. I'm not tech savvy enough to be able to tell you which pin it is but it's on the board in the chip holder. And I lose the signal there so I can rule out anything with the transistor I put in to run the ignitor.
And I'll stop here. I feel like I'm rambling.
Thanks for reading and your time.

[jsmcortina]

Please post a datalog that shows the cut out.

James

[Goosenhaben]

Not sure which log would be the best one so I made a few for you to sort through. All done during the cutout.

2017-08-13_11.25.23.csv

2017-08-13_11.24.15.csv

2017-08-13_11.22.45.csv

[Goosenhaben]

Three more. Although I think two are test files so I could find them. They were done at idle.

2017-08-13_11.14.40.csv

2017-08-13_11.03.37.csv

2017-08-13_11.00.41.csv

For grinns I had reflashed the unit to 3.4.2 to no effect.

[rickb794]

Wrong log type, need MSL logs not CSV.

[Goosenhaben]

Alrighty, here it is.

2017-08-13_17.07.02.msl

[kjones6039]

I, for one, would to see the MSQ that made that log. You've got some weird stuff going on with your RPM!

Can you post the tune for that log?

Ken

[Goosenhaben]

MSQ is in the first post. I took the log before flashing.

[jsmcortina]

Is the spark output still working at the cutout?
(PS. MSQ is in first post.)

James

[Goosenhaben]

No.
Also in first post.
Second image the yellow trace is out to ignitor. Green trace is crank signal after conditioning.
I noticed that after looking at the MSL that injector pulse also goes away. So it looks like the unit stops doing its job.
Not trying to throw anyone off, or if it's related, but another issue I've noticed is that when ambiant temperature gets into the ninties it gets extremely hard to start, if it'll start. Once it is runnung, I can keep it running and seems OK. But will most likely stall at idle. Still have the 4K cutout. And tooth and sync errors. Below 90, no issues save for the 4K thing.

[kjones6039]

MSQ is in the first post. I took the log before flashing.

Sorry! Missed it.

Ken

[rickb794]

First I would replace the 3 or 4 year old firmware that is 3.3.3.
I would start a new project and a new tune starting with the default msq.
You can go back to the old tune to verify setup.
This will insure nothing from the old MSQ gets brought over to bother the new firmware.
Exporting/importing the tables should not be a problem.

Although the issue seems to be within the MS i have other questions about your hardware.

What is your injector's measured impedance?
What are you using for rpm input trigger? Not familiar with Civic Distributor.
What are you using for ignition output? Dumb coil or smart coil?
Have you honored the grounding recommendations 3.2 of the manual? Pins 15-19 to the motor? Sensors grounded to the MS not to the car?
Are you using resistor plugs and not using low resistance plug wires?

[Goosenhaben]

First I would replace the 3 or 4 year old firmware that is 3.3.3.
I would start a new project and a new tune starting with the default msq.
You can go back to the old tune to verify setup.
This will insure nothing from the old MSQ gets brought over to bother the new firmware.
Exporting/importing the tables should not be a problem.
I have already reflashed to the new firmware but didn't do a new project. I could reflash again and do a new one.

Although the issue seems to be within the MS i have other questions about your hardware.
I've been thinking the same thing.

What is your injector's measured impedance?
12 ohms

What are you using for rpm input trigger? Not familiar with Civic Distributor.
Use VR sensors. Honda uses 3 in the distributor; a 'crank' sensor with 24 teeth, a cam sensor with 4, then a TDC on #1 cyl. I'm using the 4 pole one but
I've set the 'crank' sensor to be 24-1 if necessary.

What are you using for ignition output? Dumb coil or smart coil?
Yes? The coil itsself is simply powered but grounded through the ignitor in the distributor. I've added a transistor into the unit to run the ignitor as per
an article I've read.
https://www.diyautotune.com/support/tec ... our-honda/

Have you honored the grounding recommendations 3.2 of the manual? Pins 15-19 to the motor? Sensors grounded to the MS not to the car?
Yes.

Are you using resistor plugs and not using low resistance plug wires?

Plugs are resistor and wires are standard.

[Goosenhaben]

rickb794. . . that did it.
I now have a proper revlimit of 6500 rpm as programmed.
Now I'll have to wait to see if the high ambiant temp is an issue.
So I did reflash the unit to the latest firmware. Then I reentered all parameters manually. Even the tables. I didn't wan't ANYthing to possibly crossover.
Result : Success!

I also haven't turned on the noise reduction.

As to why I took so long? Well, after all the frustration, learning, hours spent on research and repeated failure didn't inspire me to get it done. That and my inherent lazyness. All I kept thinking about was what was I going to do when this didn't work.

And here's the MSQ and the log to prove it!

RX Civic Set.msq

(118.81 KiB) Downloaded 31 times

2017-10-01_18.54.57.msl

(641.32 KiB) Downloaded 25 times

Thank you all for your help.