Can BT/TS drop cause an ecu reset?

[Dan_Austin]

I have about six months with BG code 3.76 on a sequencer
and recently added a bluetooth dongle to enable using
shadowdash.

It may well be a coincidence, but since then I have had
several stalls, all just off idle, all while TunerStudio 2.0.6
was running. I happened to note that the last couple
occured at about the same time the BT connection dropped
disconnecting TunerStudio. I have been increasing
InterWriteDelay to deal with the BT drops, 30ms was not
enough, so the next test will be at 35ms.

Not every drop that resulted in TS going offline, so I am
not convinced is is related, but the timing of when it
started happening and then seeing TS go offline moments
before a stall prompted me to ask.

[grippo]

Dan, I have made changes quite a while ago to incorporate better serial comms in order to accomodate Bluetooth tuning. I was waiting for a final change in TS before releasing this code. This change has been made and there is an alpha release of TS to accomodate it. I just need to make some changes to my code and ini file to be compatible. So I expect to release it in the next few days and you can test it then. I will let you know here, and if it works for you I will release it to everyone.

[Dan_Austin]

That is excellent news. I'll get the TS alpha and prepare for the
release. Will the update require new software on the router side?

[grippo]

That's a good question. I don't think it will require a change on the router side, since the router only deals with CAN, but the serial pass-thru mode where TS talks to the router through the main processor/ router CAN link has been giving me fits all day.

[grippo]

Ok attached is the latest code version for Microsquirt and Sequencer. Code changes relative 3.780 code are:

Added capability to change rev limit rpms when N2O signal changes.
Added nested interrupts to allow operation with Android phones/ Bluetooth
Added crc32 envelope for serial error notification.
Restored Priming Pulse max to 65 ms.
Fixed number overflow problem in warmup mode.
Fixed bug in non-wheel mode.

Let us know how it works out for you and I will release it on the standard site. The ini files are the same except the SEQ one has a command set to indicate the code is in a Sequencer as opposed to a Microsquirt. No changes were required to the Router code or ini file.

[grippo]

Forgot to mention that this code needs TS Beta version 2.1.17 which can be obtained from the EFIAnalytics link.

[Dan_Austin]

Excellent.

I have not had to do an update yet, so to confirm-

I need the S19, ini and MS-II Download Utility V2.00
Launch the downloader and configure and connect the serial cable
Ground the purple bootloader wire
Power up the ECU
Select the new S19
Power off the ECU and disconnect the bootloader lead
Power on the ECU
Launch TS and load the new INI in th project properties.

Is it recommended to pull the fuses for the coils and injectors
during the update or is that excessively cautious?

[grippo]

That is the procedure. You will know the code loaded properly when the Downloader says "verification succeeded". It may give an error message after that, just ignore it. After the ini is loaded and everything comes up, you want to load your msq, which should be made with the old code just before you download the new code. You may get some warnings because your new ini has some extra inputs in it.

As far as pulling fuses, that is a good idea. When I was satisfied the inputs looked good and it got sync in cranking, then I would connect ignition, check that the coils don't get hot just sitting there without cranking, then crank just a bit and check for spark. Then connect injectors and try to start. This is not a major build change so I don't expect problems, but since you are the first, it pays to be cautious.

[Dan_Austin]

The update process was painless. The only error on merging the
MSQ and new ini was about the ipada array size not matching.
One item that concerned me at first is the ECU was cycling the
fuel pump while TS generated a difference report before allowing
me to push the config. I pulled that fuse as well and started
checking all the settings.

The config looked OK, so the coil fuse was re-installed and the
coils did not heat up, on cranking sync was quick and spark followed
as expected. I re-installed the FP and injector fuses and she fired
right up. Idle was erratic, and I noticed that my AFR was not
moving. I reset the AFR calibration table to LC-1 and it came to
life and reported very lean conditions. I had been running the
car before the update, so it was already up to temp, so I ran VEAL
and it fattened my idle up nicely and the idle smoothed right out.
All other sensor readings were as expected.

I have not had a chance to drive it yet, but I did give it awhile at
idle with TS connected with bluetooth and did not observe disconnects.
I will take the car out for a test and re-tune tomorrow, but so far
it looks good.

[Dan_Austin]

Today's tuning session went well. I managed about an hour
with not BT drops, using VEAL for a good bit of that. I had
a couple issues with hot restart, but I cannot say that is new
as my normal driving does not often involve a hot restart.

With the hot restart I noticed very lean conditions that lasted
for 30 to 60 seconds. I would suspect heat soak on the IAT, but
that did not happen that I can recall on 3.77, but it may well
have.

I've attached my msq to have a second set of eyes look at it.
The engine is a 2.9L turbocharged straight six. I have moderately
large injectors at 475cc/min and Delco D585 coils. I am using
a EGR solenoid vavle for my IAC that defaults to open.

Thanks for making the update available.