[help needed] Looking for a solution for the Subaru EG33

Your chance to introduce yourself and your vehicle
User avatar
Dron_Gus
contributor
contributor
Posts: 451
Joined: Wed Nov 13, 2013 1:11 pm
Location: S-Pb
Github Username: dron0gus

Re: Looking for a solution for the Subaru EG33

Post by Dron_Gus »

image.png
image.png (29.28 KiB) Viewed 19377 times
This is how IAC is connected to ECU on EJ20G. I don't think think they invented something new for EG33. Actually this is the same.
sleepingAwake
Posts: 113
Joined: Fri Jan 13, 2017 4:15 am
Location: CT

Re: Looking for a solution for the Subaru EG33

Post by sleepingAwake »

Dron_Gus wrote:
Thu Jul 09, 2020 9:23 am
image.png
This is how IAC is connected to ECU on EJ20G. I don't think think they invented something new for EG33. Actually this is the same.
Dron_Gus wrote:
Thu Jul 09, 2020 9:21 am
I connected center pin to +12 and drive open and close ends using PWM. I was able to rotate IAC from fully closed to fully opened by changing balance of duty cycles. This is what implemented now for two-wire IAC.
I relayed what you've said to the ECU. Didn't seem to help.
sleepingAwake
Posts: 113
Joined: Fri Jan 13, 2017 4:15 am
Location: CT

Re: Looking for a solution for the Subaru EG33

Post by sleepingAwake »

Verified IAC functions properly when connected to a standalone PWM controller.
Nothing with RUSefi.
User avatar
Dron_Gus
contributor
contributor
Posts: 451
Joined: Wed Nov 13, 2013 1:11 pm
Location: S-Pb
Github Username: dron0gus

Re: Looking for a solution for the Subaru EG33

Post by Dron_Gus »

I will try new FW this weekend. But I can test it only on my own HW. So I'll check that nothing is broken in SW only.
sleepingAwake
Posts: 113
Joined: Fri Jan 13, 2017 4:15 am
Location: CT

Re: Looking for a solution for the Subaru EG33

Post by sleepingAwake »

Dron_Gus wrote:
Thu Jul 09, 2020 11:31 am
I will try new FW this weekend. But I can test it only on my own HW. So I'll check that nothing is broken in SW only.
A screenshot of the idle settings in TS would be helpful if possible.
User avatar
Dron_Gus
contributor
contributor
Posts: 451
Joined: Wed Nov 13, 2013 1:11 pm
Location: S-Pb
Github Username: dron0gus

Re: Looking for a solution for the Subaru EG33

Post by Dron_Gus »

sleepingAwake wrote:
Thu Jul 09, 2020 11:37 am
A screenshot of the idle settings in TS would be helpful if possible.
Ok. I'll do. Previously I did all configuration by code/defaults modification. I run TS for few times only. Time to learn new tricks.
sleepingAwake
Posts: 113
Joined: Fri Jan 13, 2017 4:15 am
Location: CT

Re: Looking for a solution for the Subaru EG33

Post by sleepingAwake »

Dron_Gus wrote:
Thu Jul 09, 2020 12:14 pm
sleepingAwake wrote:
Thu Jul 09, 2020 11:37 am
A screenshot of the idle settings in TS would be helpful if possible.
Ok. I'll do. Previously I did all configuration by code/defaults modification. I run TS for few times only. Time to learn new tricks.
I'm also using mostly defaults. Of course I have to make some changes, like pin assignment and selecting double solenoid.
But I haven't played around with the other Idle settings in TS. Just trying to establish basic hardware function at this point.

Regarding voltage should I be getting 3.3volts on either pin? With the PWM set at 50% I'd expect 6v connected to a 12volt source. Looking at the schematic for Frankenso 0.4 the low side drivers output 5volts. So maybe 3.3volts is what I should expect and its not enough to drive the IAC?
User avatar
Dron_Gus
contributor
contributor
Posts: 451
Joined: Wed Nov 13, 2013 1:11 pm
Location: S-Pb
Github Username: dron0gus

Re: Looking for a solution for the Subaru EG33

Post by Dron_Gus »

sleepingAwake wrote:
Thu Jul 09, 2020 12:18 pm
Regarding voltage should I be getting 3.3volts on either pin? With the PWM set at 50% I'd expect 6v connected to a 12volt source. Looking at the schematic for Frankenso 0.4 the low side drivers output 5volts. So maybe 3.3volts is what I should expect and its not enough to drive the IAC?
Low side output can output only 0V. :) High level voltage depends on where other side of load is connected.
These outputs are capable to drive these coils.

LEDs (D414, D419) is on?
Are you sure that you have +12V on central pin?
What voltage do you see on these two outputs while MCU in reset? You should see 12V (or 0V in some cases) but not something in the middle.
sleepingAwake
Posts: 113
Joined: Fri Jan 13, 2017 4:15 am
Location: CT

Re: Looking for a solution for the Subaru EG33

Post by sleepingAwake »

Dron_Gus wrote:
Thu Jul 09, 2020 12:53 pm
sleepingAwake wrote:
Thu Jul 09, 2020 12:18 pm
Regarding voltage should I be getting 3.3volts on either pin? With the PWM set at 50% I'd expect 6v connected to a 12volt source. Looking at the schematic for Frankenso 0.4 the low side drivers output 5volts. So maybe 3.3volts is what I should expect and its not enough to drive the IAC?
Low side output can output only 0V. :) High level voltage depends on where other side of load is connected.
These outputs are capable to drive these coils.

LEDs (D414, D419) is on?
Are you sure that you have +12V on central pin?
What voltage do you see on these two outputs while MCU in reset? You should see 12V (or 0V in some cases) but not something in the middle.
D414 and D419 are not lit up.
The homemade oscillscope I was using isnt working correctly. Please disregard what I said about 3.3volts going to IAC pins.
I just rechecked with a proper meter and I'm getting 12v+ on either pin. Holding down reset button and still getting 12v+.

You would think I could move the valve. Makes me think its a SW or config issue now.
sleepingAwake
Posts: 113
Joined: Fri Jan 13, 2017 4:15 am
Location: CT

Re: Looking for a solution for the Subaru EG33

Post by sleepingAwake »

Is there any special settings or code that I needed to implement to get the IAC working?

Really seems like a SW issue now.
I have continuity from the IAC to the gates on the drivers and from the sense pins on the drivers to the pins on the STM32.
Solid 12v+ on the two coils at the IAC.
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: Looking for a solution for the Subaru EG33

Post by AndreyB »

Do we need flyback diodes here?
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
sleepingAwake
Posts: 113
Joined: Fri Jan 13, 2017 4:15 am
Location: CT

Re: Looking for a solution for the Subaru EG33

Post by sleepingAwake »

User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: Looking for a solution for the Subaru EG33

Post by AndreyB »

Yes, no one uses this this on a real vehicle so far. I believe @Dron_Gus has tested this on a bench with fix own hardware not Frankenso.
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
sleepingAwake
Posts: 113
Joined: Fri Jan 13, 2017 4:15 am
Location: CT

Re: Looking for a solution for the Subaru EG33

Post by sleepingAwake »

AndreyB wrote:
Fri Jul 10, 2020 1:45 pm
Yes, no one uses this this on a real vehicle so far. I believe @Dron_Gus has tested this on a bench with fix own hardware not Frankenso.
I'm just bench testing at the moment as well. I just happen to have a spare IAC to test with.

For testing and manual % movement in TS was it just as simple as setting pin assignment and selecting Double Solenoid?
sleepingAwake
Posts: 113
Joined: Fri Jan 13, 2017 4:15 am
Location: CT

Re: Looking for a solution for the Subaru EG33

Post by sleepingAwake »

sleepingAwake wrote:
Fri Jul 10, 2020 12:50 pm
Is there any special settings or code that I needed to implement to get the IAC working?

Really seems like a SW issue now.
I have continuity from the IAC to the gates on the drivers and from the sense pins on the drivers to the pins on the STM32.
Solid 12v+ on the two coils at the IAC.
Do I need to implement settings in FSIO to get a PWM output? Or is this baked into the code once I choose double solenoid and set pins?

Verified my meter could pick up a PWM signal with a working standalone PWM motor control and I dont see a PWM signal at the IAC just 12v+.
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: Looking for a solution for the Subaru EG33

Post by AndreyB »

Pretty much zero connection with FSIO.

Do you have a logic analyzer? can you get a $9 logic analyzer on eBay?
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: Looking for a solution for the Subaru EG33

Post by AndreyB »

Can you please provide any data? Can you record a video and post it on youtube of how you adjust manual idle position and how does the valve respond? what kind of power supply are you using?
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
sleepingAwake
Posts: 113
Joined: Fri Jan 13, 2017 4:15 am
Location: CT

Re: Looking for a solution for the Subaru EG33

Post by sleepingAwake »

AndreyB wrote:
Sat Jul 11, 2020 5:34 pm
Can you please provide any data? Can you record a video and post it on youtube of how you adjust manual idle position and how does the valve respond? what kind of power supply are you using?
I could post a video.

I'm using a 3s lipo to provide 12volts.

When I adjust the manual IAC slider in TS and hit burn the signal never changes and the valve never moves. Solid 12v+ and no PWM signal observed.

Only way I get it too move is with the separate PWM motor controller I have. I also observe a PWM signal using that spare controller. So I know the valve works and my meter is able to detect the PWM signal.
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: Looking for a solution for the Subaru EG33

Post by AndreyB »

Please spell out what is 3s lipo. There is a chance that googling what is "3s lipo" is too much effort not just for me and you are making it harder to help you by adding those extra steps. I could be very wrong here.

Do we need flyback diodes here? Single-solenoid idle requires a fly back diode.

Actually can you move your IAC at least in one if you run it as single solenoid? We know that single solenoid code works for sure. If single solenoid logic would move it one way that would tell us at least something.

Unfortunately the community is too small to provide a better level of support :(
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
sleepingAwake
Posts: 113
Joined: Fri Jan 13, 2017 4:15 am
Location: CT

Re: Looking for a solution for the Subaru EG33

Post by sleepingAwake »

AndreyB wrote:
Sat Jul 11, 2020 5:43 pm
Please spell out what is 3s lipo. There is a chance that googling what is "3s lipo" is too much effort not just for me and you are making it harder to help you by adding those extra steps. I could be very wrong here.

Do we need flyback diodes here? Single-solenoid idle requires a fly back diode.

Actually can you move your IAC at least in one if you run it as single solenoid? We know that single solenoid code works for sure. If single solenoid logic would move it one way that would tell us at least something.

Unfortunately the community is too small to provide a better level of support :(
3s lipo is 3 lithium cells in series to create an 11.1v pack. Fully charged its around 12.2volts. Can output 2-4 amps without issue.

IDK if we need flyback diodes. I'll try what you suggested with single solenoid right now.
sleepingAwake
Posts: 113
Joined: Fri Jan 13, 2017 4:15 am
Location: CT

Re: Looking for a solution for the Subaru EG33

Post by sleepingAwake »

AndreyB wrote:
Sat Jul 11, 2020 5:43 pm
Please spell out what is 3s lipo. There is a chance that googling what is "3s lipo" is too much effort not just for me and you are making it harder to help you by adding those extra steps. I could be very wrong here.

Do we need flyback diodes here? Single-solenoid idle requires a fly back diode.

Actually can you move your IAC at least in one if you run it as single solenoid? We know that single solenoid code works for sure. If single solenoid logic would move it one way that would tell us at least something.

Unfortunately the community is too small to provide a better level of support :(
Just tried moving the IAC in Single Solenoid mode. Nothing. Not seeing a PWN signal at the IAC. Just 12v+
I just purchased a Logic Analyzer as well.
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: Looking for a solution for the Subaru EG33

Post by AndreyB »

sleepingAwake wrote:
Sat Jul 11, 2020 6:02 pm
Not seeing a PWN signal at the IAC. Just 12v+
Please reiterate how exactly do you check for PWM. We _know_ that single solenoid works, we _know_ that single solenoid valve required the flyback diode.

I am not very electrical so that's all help I can provide :(
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: Looking for a solution for the Subaru EG33

Post by AndreyB »

Pictures, videos, napkin drawing of what wires go where.
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
sleepingAwake
Posts: 113
Joined: Fri Jan 13, 2017 4:15 am
Location: CT

Re: Looking for a solution for the Subaru EG33

Post by sleepingAwake »

AndreyB wrote:
Sat Jul 11, 2020 6:06 pm
Pictures, videos, napkin drawing of what wires go where.


Not sure whats up with the 64v gauge reading TPS in TS. Happened yesterday after updating erasing STM32 and reuploading latest firmware. It's not really on my radar right now anyway.
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: Looking for a solution for the Subaru EG33

Post by AndreyB »

Please note the new red note. Please enable self stimulation in "controller->ECU stimulator"

image.png
image.png (144.35 KiB) Viewed 19290 times
image.png
image.png (123.53 KiB) Viewed 19290 times
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
sleepingAwake
Posts: 113
Joined: Fri Jan 13, 2017 4:15 am
Location: CT

Re: Looking for a solution for the Subaru EG33

Post by sleepingAwake »

Alright that was it I needed to enable self simulation.

The valve does show signs off life but goes to a fully closed position irregardless of manual IAC position changes.
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: Looking for a solution for the Subaru EG33

Post by AndreyB »

sleepingAwake wrote:
Sat Jul 11, 2020 7:04 pm
Alright that was it I needed to enable self simulation.

The valve does show signs off life but goes to a fully closed position irregardless of manual IAC position changes.
Does your meter show different duty cycle or voltage according to your input?

Flyback diode time if you ask me.
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
sleepingAwake
Posts: 113
Joined: Fri Jan 13, 2017 4:15 am
Location: CT

Re: Looking for a solution for the Subaru EG33

Post by sleepingAwake »

AndreyB wrote:
Sat Jul 11, 2020 7:13 pm
sleepingAwake wrote:
Sat Jul 11, 2020 7:04 pm
Alright that was it I needed to enable self simulation.

The valve does show signs off life but goes to a fully closed position irregardless of manual IAC position changes.
Does your meter show different duty cycle or voltage according to your input?

Flyback diode time if you ask me.
I do see a different voltage/duty cycle with my meter now. I installed the flyback diodes this is the behavior with them in.
stefanst
contributor
contributor
Posts: 703
Joined: Wed Feb 17, 2016 12:24 am
Location: USA 08530

Re: Looking for a solution for the Subaru EG33

Post by stefanst »

Just for diagnostics, if you flip the two inputs, does the valve go to full open instead of full closed?
sleepingAwake
Posts: 113
Joined: Fri Jan 13, 2017 4:15 am
Location: CT

Re: Looking for a solution for the Subaru EG33

Post by sleepingAwake »

stefanst wrote:
Sat Jul 11, 2020 11:58 pm
Just for diagnostics, if you flip the two inputs, does the valve go to full open instead of full closed?
No it does not. Could it be because of the low coil resistance? 9 ohms to center tap.

I would think this valve works like a seesaw. More current on one side versus the other moves the valve open to close.
But as I move the slider the voltage observed on either coil is always matched. It changes from 448mv to 6.02v but each side measures the same. I thought I would observe low voltage on one side versus the other. An imbalance. To move the valve.

Updating this post with a screenshot from Pulseview showing the signal at PE6 and PC13 (my iac pins).
Attachments
PWM from IAC.jpg
PWM from IAC.jpg (231.66 KiB) Viewed 19232 times
Post Reply