[Timus]

Quick checklist:
- vvt need to be disabled
- pin assigned only to this function
- wiring, controlled pin will go to ground when wmi is enabled(verify with multimeter if there is voltage between +12v and wmi output pin when wmi is enabled)

[shelbyspeed]

I just recently got this bench tested and driving an LED. One thing I noticed had to be active before it would drive the duty cycle was "tank empty," had to be ON and I jumped it out. I was then able to drive the circuit. This may have been my jumbled mess of testing equipment but once I turned it on and jumped the pin out the duty cycle drove consistently and predictable.

[AndyPonting]

Quick checklist:
- vvt need to be disabled
- pin assigned only to this function
- wiring, controlled pin will go to ground when wmi is enabled(verify with multimeter if there is voltage between +12v and wmi output pin when wmi is enabled)

Thanks Timus. I'm sure I've got all these covered but will retrace my steps and make sure the pin isn't assigned elsewhere.

Edit: silly question, does the function require the engine running? Of course I can set minimum rpm and map to 0 and trigger with TPS for testing? Only the car is loud and my neighbours are... Well they've not complained yet!

[AndyPonting]

I just recently got this bench tested and driving an LED. One thing I noticed had to be active before it would drive the duty cycle was "tank empty," had to be ON and I jumped it out. I was then able to drive the circuit. This may have been my jumbled mess of testing equipment but once I turned it on and jumped the pin out the duty cycle drove consistently and predictable.

Even with this function disabled? Or am I reading right that this *must* be enabled for wmi to function. I can change the polarity from inverted to normal to show "tank empty" off.

Andy

[Jaap Struyk]

Morning!

Where in the code can I change IAT to CLT? Is that currentStatus.IAT?

Want to cool the engine with water when CLT is higher then xx degrees, WMI is controlling a second injector setup parallel to the fuel (methanol) injectors.

Greetings Japie

[LPG2CV]

take a look at 'void wmiControl'

it may not be as simple as swapping iat for clt, as clt may be on a different config page

[Timus]

Sorry for delay.
File you need to modify is auxiliaries.ino.
Around line 688 in current commit on github.

You need to change this line:

Code: Select all

if( (currentStatus.TPS >= configPage10.wmiTPS) && (currentStatus.RPMdiv100 >= configPage10.wmiRPM) && ( (currentStatus.MAP / 2) >= configPage10.wmiMAP) && ( (currentStatus.IAT + CALIBRATION_TEMPERATURE_OFFSET) >= configPage10.wmiIAT) )

To something like this:

Code: Select all

if( (currentStatus.TPS >= configPage10.wmiTPS) && (currentStatus.RPMdiv100 >= configPage10.wmiRPM) && ( (currentStatus.MAP / 2) >= configPage10.wmiMAP) && ( (currentStatus.coolant + CALIBRATION_TEMPERATURE_OFFSET) >= configPage10.wmiIAT) )

[Jaap Struyk]

Morning!

Thanks for the response!
Didn't make the change however due to running into other WMI related issues do, one of the major things is that I can't get wmi pulsewidth to show up in the log to confirm either it (is/should be) working or not.

I have an injector driver board hooked up directly to the arduino pin controlling an extra per cylinder set of injectors (EV14 2200cc) some moths (with March 2021 firmware) ago I tested this setup with fuel (methanol) instead of water on the system monitoring my lambda values to find out at what value in the control table the injectors start to give fuel and that was around 20 at 134Hz so 19 is injector dead time, 255/100*85 as max 85% duty cycle value and added WMIpw to datalogging in TS showing me the table entries on replay.

Currently I have my water tank filled up and the water injectors are under water pressure, when testing at idle even with a table value of 150 the engine keeps running (with that amount at idle it should die instantly), at 25 it looks like the engine is running less smooth and EGT's are dropping a bit, unfortunately I can't get WMIpw to show up in the log anymore to confirm the software is functioning as it should, in TS it's value format is activeInactive and I can't change that. (however in the log it keeps steady at inactive)

Update: on reboot with table values above 30 the engine won't start until I lower them and reboot suggesting table values can't be changed real-time, when restarting the engine exhaust is pretty foggy suggesting left over water is burned.

Update2: WMIpw is coming to live when I change the tank empty bit described by shelbyspeed a bit further in this thread.

Is there a way I can visualize and log what the WMI output pin is doing?

[Timus]

unfortunately I can't get WMIpw to show up in the log anymore to confirm the software is functioning as it should, in TS it's value format is activeInactive and I can't change that. (however in the log it keeps steady at inactive)

That's weird, originally it was a percentage and was working OK, I see that there were some modification to TS config and status memory layout and maybe it's broken now.

Update: on reboot with table values above 30 the engine won't start until I lower them and reboot suggesting table values can't be changed real-time, when restarting the engine exhaust is pretty foggy suggesting left over water is burned.

WMI table is relaying on same functionality as all other tables and therefore should act exactly the same including real time modification. WMI table is shared with VVT one and maybe there was some changes to it lately, but without checking it I cannot tell.

Is there a way I can visualize and log what the WMI output pin is doing?

It should work just fine.

Overall, I thing the best way is to try to use older versions of firmware and check on which version it's stopped working correctly, sadly this january is very busy month for me and I will not be able to look into it. If you can't wait you can try to debug it yourself or wait to february when I will be able to look deeply into code.

[pazi88]

Quick look in the speeduino FW, and looks like wmi duty was never in the list of log items. Only gauge in TS. And it seems that something is fishy with the wmi tank empty input settings. I can make fix for that in next FW.