Bugs and Feature reqests

[Alec]

So are there any plans for user-configurable units within the Bluetooth app? That's really the only thing holding me back from picking up a Bluetooth module.

[Jadzwin]

Do you mean metric / imperial ?

[Alec]

Do you mean metric / imperial ?

Yea, imperial units.

[Jadzwin]

Will be added in next release.

[Alec]

Will be added in next release.

Will that also be the case for the EMU client? Viewing imperial units in the EMUlog would be nice as well.

[Jadzwin]

For EMU client it will be much more complicated.
I think it is possible for logger and gauges (not for tables and parameters)

[clem]

Hi jakub!
My electronic servo-Wastegate actuator works great! But now I got a bit deper in the stuff of servo PWM signals and want to inform you: The frequency is not that important, it determines just how often the servo checks its position and makes correctures (every time a pulse occures, the servo checks that)
The most important thing is, that the pulsewidth is very constant! Most servos use 900us to 2100us of pulse width (take note of the unit!!)
When I connected my servo to emu i recognized that the pulse wiggles inconsitently for for several us (20 to 50us, measured with oscilloscope) and that makes the servo permanently work to "correct" its position even when EMU sends constant Duty Cycle. this results in approx. 5 Degrees of servo rotation (with 100% servo rotation of 280 deg)
Is there any way you could improve that? I do not recognize it in the MAP Log very much (maybe 2 to 3 kPa "oscillation"), but it would be nice for the servo not to work when it is not necessary.
But I assume that is the boarder of possibility, as injection time resolution is at 16 us?

Thanks!!

[fieldy]

Another inquiry from me: Would it be very compicated to implement in the software, to disable the rpm based fuel cut during the time were afterstart enrichment and/or afterstart idle dc increase is active ? Would make startup on E85 easier under all circumstances.

Thanks

[Jadzwin]

Do you mean overrun fuel cut ?
I can add an option to increase these RPM (cut and resume) when afterstart is active.

[fieldy]

Exactly, this would be awesome!

Thanks

[clem]

Hi Jakub. As I read my post several times now: the resolution of the pulsewidth is not the problem, but the change in width when it should be constant is not optimal.
Have you had time to take a look at that?
thanks

[RHD]

Hi Jakub. As I read my post several times now: the resolution of the pulsewidth is not the problem, but the change in width when it should be constant is not optimal.
Have you had time to take a look at that?
thanks

have you installed a flyback diod on the solenoid? if not this may help

[clem]

Well as I do not run a solenoid (inductive load) I did not install a diod. that is not necessary in my case as I just run an 880ohm resistor to pull up the line to the "high state" voltage, emu just switches that ohm - load to ground ("low state"), so no diod is needed. correct me when I'm wrong!!

may a different resistor value make a change ? in oscilloscope the square wave looks pretty clean though...

[Mobilelpg]

I sent the info that was asked for by email but have received no replies?

I have been doing some looking over the datalogs from everytime the software freezes and have found that the freezing is not related to RPM or Boost so much but is in fact everytime it logs a Knock event!
Without fail everytime the datalog shows a knock activated fuel increase and ignition retard, the software freezes.
I am using a Phormula KS4 for knock readings and that shows everything as fine (I will be logging next time with the KS Pro det cans so will have a better idea if it is genuinely hearing knock).
I am however wondering if there is a glitch that is causing the freeze and a knock event, or if it is a knock event that is causing the software to freeze.

I can upload another datalog which clearly shows numerous freezes and each time there is a knock event at that exact moment if anyone would like to look over it.

Edward.

Please contact me at info(at)ecumaster.com to solve your problem.

Hi,

I have renewed all of the earths, including fitting an earthing kit and adding extra earths but the issue is still the same.
I had a strange issue today where whilst driving with the laptop connected and logging, my VE 2 table got completely messed up, my wideband went to 10 AFR and the engine wouldn't run.
Thankfully I was running on LPG and the table switches automatically to VE 1 when I switch back to petrol, so I drove back on petrol thinking there was a fault in my LPG system.
Only when I go back and started checking the logs did I find that the VE 2 table had messed up completely and the VE in the datalog was showing over 5000%.

Unfortunately this forum says I can not upload the datalog as it is over 2mb even compressed but I have uploaded the Emu file and you can see how messed up the VE 2 table is (only a few minutes earlier the table was fine)
Any opinions would be appreciated.

Edward.

Thanks for the reply.
The laptop is always running off the battery.
I will look into the earth points involved.

Edward.

[PSI-Motor]

Please send me your map and log at info@psimotor.se


Regards Mattias

[Jadzwin]

We are wroking on your issue. We found a potential bug when communication is broken during transfering data to device (data corruption may occur and I think this issue corrupted your data). In version 1.116 there is improved communication protocol, however tests shows some issues and we need to fix it in next version.

You can disable knock sensing and check if the problem still exist.

I sent the info that was asked for by email but have received no replies?

I have been doing some looking over the datalogs from everytime the software freezes and have found that the freezing is not related to RPM or Boost so much but is in fact everytime it logs a Knock event!
Without fail everytime the datalog shows a knock activated fuel increase and ignition retard, the software freezes.
I am using a Phormula KS4 for knock readings and that shows everything as fine (I will be logging next time with the KS Pro det cans so will have a better idea if it is genuinely hearing knock).
I am however wondering if there is a glitch that is causing the freeze and a knock event, or if it is a knock event that is causing the software to freeze.

I can upload another datalog which clearly shows numerous freezes and each time there is a knock event at that exact moment if anyone would like to look over it.

Edward.

[Jadzwin]

The problem is that timer resolution and timer priority used for Boost PWM control doesn't allow better control.
For PWM solenoids it doesn't matter.

FB diode is required only for inductive load.

Hi Jakub. As I read my post several times now: the resolution of the pulsewidth is not the problem, but the change in width when it should be constant is not optimal.
Have you had time to take a look at that?
thanks

[clem]

Ok thanks anyway. I already assumed that.

[Mobilelpg]

Map and log file sent.

Regards,
Edward.

Please send me your map and log at info@psimotor.se


Regards Mattias

[Mobilelpg]

Hi,

Thanks for the update.
I intend to disable knock sensing next time I take the car out and I will update here the results.

Edward.

We are wroking on your issue. We found a potential bug when communication is broken during transfering data to device (data corruption may occur and I think this issue corrupted your data). In version 1.116 there is improved communication protocol, however tests shows some issues and we need to fix it in next version.

You can disable knock sensing and check if the problem still exist.

I sent the info that was asked for by email but have received no replies?

I have been doing some looking over the datalogs from everytime the software freezes and have found that the freezing is not related to RPM or Boost so much but is in fact everytime it logs a Knock event!
Without fail everytime the datalog shows a knock activated fuel increase and ignition retard, the software freezes.
I am using a Phormula KS4 for knock readings and that shows everything as fine (I will be logging next time with the KS Pro det cans so will have a better idea if it is genuinely hearing knock).
I am however wondering if there is a glitch that is causing the freeze and a knock event, or if it is a knock event that is causing the software to freeze.

I can upload another datalog which clearly shows numerous freezes and each time there is a knock event at that exact moment if anyone would like to look over it.

Edward.