[WIP] BMW Z3

[andybpowell]

Can you please post the Zombieverter settings .json you currently are running?

As the GS450/IS300h code should be scaling the PotNom to 3500 it seems to not be scaling it alot for some reason.

will do

[andybpowell]

Can you please post the Zombieverter settings .json you currently are running?

As the GS450/IS300h code should be scaling the PotNom to 3500 it seems to not be scaling it alot for some reason.

params (1).json

(1.78 KiB) Downloaded 25 times

[andybpowell]

I got the data logger to work though it stopped after 24 seconds but the torque spikes at each end are curious

log (3).csv

(4.86 KiB) Downloaded 18 times

[Bigpie]

Did your log cover a torque cut?

[andybpowell]

Did your log cover a torque cut?

Not sure I've had to pop out for a while I'll try it again a bit later

[tom91]

Just a few observations:

You limit throttle two ways:
"bmslimhigh": 50
"throtmax": 50

when you do another plot can you add Potnom as this is the calculated throttle command vs your throttle positon reading to see if they stay linear.

[andybpowell]

Just a few observations:

You limit throttle two ways:
"bmslimhigh": 50
"throtmax": 50

when you do another plot can you add Potnom as this is the calculated throttle command vs your throttle positon reading to see if they stay linear.

but there is no bms enabled ?

[andybpowell]

weirdest thing, I've just done 3 runs with the data logger and everytime it cuts the data logger stops

log (6).csv

(972 Bytes) Downloaded 21 times

log (5).csv

(972 Bytes) Downloaded 18 times

log (4).csv

(2.71 KiB) Downloaded 19 times

[rstevens81]

you haven't got a bad earth or something essentially cutting the 12v?

[Bigpie]

Go back to the graphs if that's more reliable?

[tom91]

How are you driving during the logs, your cut outs do not seem to be "consistant" with any over signals.

Can you do it with the Pot1 and Potnom so we can see what happens there.

The Torque cuts, do you loose wifi connection too when it cuts out?

[andybpowell]

you haven't got a bad earth or something essentially cutting the 12v?

pretty sure 12v is ok because if I stop straight away and stop and restart the log it starts again straight away so no loss to zombie also inverter and battery voltage match and remain stable so not losing power to inverter

[andybpowell]

Go back to the graphs if that's more reliable?

yes I will go back to graphs as they seem unaffected can't test anymore today as the school is kicking out (damn pesky kids)

[andybpowell]

How are you driving during the logs, your cut outs do not seem to be "consistant" with any over signals.

Can you do it with the Pot1 and Potnom so we can see what happens there.

The Torque cuts, do you loose wifi connection too when it cuts out?

no don't lose wifi, I will have to go back to the graphs as plot is less temperamental

[Zechy]

As far as I can tell the bodge in lastest release is only in the prius inverter for some reason. Having looked at several posts here and on discord it appears that the temperatures are 8 bit signed value ( maybe it was just assumed to be 16 bit ) . The MTH value when read gets placed into an array of unsigned values that then gets converted to float. I think this Is why Damien was seeing 255 at negative temperatures ( this is also another clue that values are 8 bit and not 16 ). The unsigned value needs converting to integer before converting to float either like this, for 300H, temp_inv_water=int8_t(mth_data[20]); or by declaring temp_inv_water as int8_t in GS450.h.

Thanks royhen99. temp_inv_water=int8_t(mth_data[20]); worked for me. Quick run around the block. Smooth as can be!

[andybpowell]

Thanks royhen99. temp_inv_water=int8_t(mth_data[20]); worked for me. Quick run around the block. Smooth as can be!

Since this power cut was stopping me progressing with the car I decided to drag my arse into the 21st century and learn how to compile a binary for the zombieverter it took quite a while to install the required software but I got there eventually and made the change as per the quoted post and the problem has now gone the only problem I now have is that I compiled it from the 2.04A release but I'm running a modified version of this that jamie sent me to make my brake booster work so now goes but doesn't stop never mind plenty of things to get on with at least I would have if it wasn't pissing down

[Bigpie]

https://github.com/damienmaguire/Stm32-vcu/pull/76 Has the fix, apply that and you're golden.

Also https://github.com/crasbe/ZombieBuild is the easiest way to build

Code: Select all

temp_inv_water=int8_t(mth_data[20]);

What about when it's below 0?

[andybpowell]

https://github.com/damienmaguire/Stm32-vcu/pull/76 Has the fix, apply that and you're golden.

Also https://github.com/crasbe/ZombieBuild is the easiest way to build

Code: Select all

temp_inv_water=int8_t(mth_data[20]);

What about when it's below 0?

no idea about below 0, I was just seeing if it worked I haven't even looked at the temps it's throwing out If the rain stops I'll get out and try it and log some values see what happens

[Bigpie]

int8_t is unsigned integer. Did it fix the cutting out though?

[royhen99]

uint8_t is unsigned integer, int8_t is signed integer ( -128 to 127 ).

[Bigpie]

uint8_t is unsigned integer, int8_t is signed integer ( -128 to 127 ).

HAHA I totally misread that

[andybpowell]

int8_t is unsigned integer. Did it fix the cutting out though?

yes running fine

[andybpowell]

https://github.com/damienmaguire/Stm32-vcu/pull/76 Has the fix, apply that and you're golden.

Also https://github.com/crasbe/ZombieBuild is the easiest way to build

Code: Select all

temp_inv_water=int8_t(mth_data[20]);

What about when it's below 0?

Thanks for giving me the benefit of the doubt but your confidence in me was misplaced all I got for my time was a screen full of errors it seems that editing that one line of code was my limit I shall stick to licking windows and eating Crayons
I've got to drop the l210 anyway to lock the motor to gearbox coupling to stop that bloody rattle

[Bigpie]

BIN PM'd

[andybpowell]

Thanks to Jamie coming through with an updated Bin making up for my shortcomings we have some more data from the l210 with is300h inverter
I'm pretty sure I'm no longer getting the harsh cut but the car is juddering a lot due to my bad choice of mounting the entire motor to gearbox coupling, but current still max's out at 120 Amps

I'm going to pull out the L210 to fix the coupling to lock the input shaft solid hopefully at the end of the week so if anyone wants me to do any runs before that to get specific data let me know hopefully the l210 wil be out and back in quickly but sods law say's something will slow me down or get in the way you know the way it is.