Autowindow on a BRZ/FRS (exploring possibilities)

[Acree]

From what I've seen, typically LIN will use the same pins as typical UART on microcontrollers. If it comes down to it, it wouldn't be too difficult just to program an 8 or 10 pin entry level microcontroller to do what you need.

Do you have any experience in CANBUS systems or master/slave hardware for LIN?

-Acree

[scottmcphee]

There's lot of choices for MCU's with LIN, CAN, and UART hardware ports/lines on-chip. Attaching hardware to the bus is not the difficult part, it's the proprietary protocols Toyota/Subaru use on these lines - that's the hard part. I don't have this level of tooling (time?) to reverse engineer the protocols - but if you have ideas like to hear them!

If we have choices and can stay off the CAN and LIN bus, that'd be the way to go.

The UART is a bit different story. It's dedicated and point to point only between the doors, and probably carries very few messages, I'm willing to tackle that one. Can likely sniff, capture, and replay a packet to get passenger window to go up/down on command.

[scottmcphee]

I suspect the passenger door "power window regulator switch assembly" (F6), is either powered(active) all the time, or powered but wake-up on UART Rx. It also seems that without being told something by the driver's door over UART, the passenger side doesn't know the state of ACC power in the car.

I suspect the driver's door p.w.r.s.a (E8) has its power/active state controlled by LIN. Similarly, without some message over LIN, the driver's door doesn't know the state of ACC power in the car. In other words, letting E8 know that "it's OK" to respect (allow) a window switch to operate the window.

I'm really curious to see if the white wire test I'm asking for has different results on each door, for this reason.

As you can see, dropping an "ACC ON" message onto CAN, probably cascades that message using different physical network technologies all the way through to the passenger door.

But I bet we can induce a "ACC ON" message to be generated by the car, simply by tweaking a line somewhere momentarily.

[Acree]

According to the manual, they are both powered all the time. I would assume the same exact thing you are saying, that the driver's side is on wake-up over LIN which in turn sends wake-up over UART to passenger side.

From the small exposure I've had to it, LIN is standardized just like UART. There are different ways to compose a message but none of it is going to be proprietary like the CANBUS. It seems to me the easiest thing would be to piggy back on that LIN line like what you said, and reproduce the correct message. It would take an afternoon with an o-scope, but it wouldn't be very difficult.

-Acree

[chadstyle]

Originally Posted by Acree

According to the manual, they are both powered all the time. I would assume the same exact thing you are saying, that the driver's side is on wake-up over LIN which in turn sends wake-up over UART to passenger side.

From the small exposure I've had to it, LIN is standardized just like UART. There are different ways to compose a message but none of it is going to be proprietary like the CANBUS. It seems to me the easiest thing would be to piggy back on that LIN line like what you said, and reproduce the correct message. It would take an afternoon with an o-scope, but it wouldn't be very difficult.

-Acree

This makes sense..and if LIN is standardized unlike CAN BUS then this may be the path of least resistance. If this is the case then the install would be simplified as well since only the drivers door would need to be tied into. Unfortunately I don't have an oscilloscope to test the wires...

[scottmcphee]

I've not yet played with LIN, if it's like you say, then yay!

An oscope may not be as useful as a small logic analyzer, like ant8 or bus pirate.

It's not guaranteed the drivers side passes all LIN traffic to passenger.
It is guaranteed window up/down commands traverse UART, destined for passenger door.
It's a question what actually originates the passenger commands, probably the driver window module.
I fear there is no LIN command to control drivers window up/down.

All of this conspires to suggest the device needs to install inside or run wires to at least the drivers door. And if it installs inside the drivers door I know that I have to run at least one wire to ECU, for an input/output function needed for autowindow.

[Sportsguy83]

Originally Posted by scottmcphee

I've not yet played with LIN, if it's like you say, then yay!
.....

Scott, where in Canada are you? Maybe we can go in the Canada Sub-Forum and get a Canadian user willing to help and you can actually get a test car?

[scottmcphee]

All this LIN, CAN, yack yack yack is great fun... But I want to re-focus on my current ask, and is something anybody can do with a simple volt meter:

Okay next person to go into either door, check out some wires on the power window regulator E5 on left door or F1 on right door. Say which door because they might act different.

What is voltage observed on some wires running between the window switch and window motor, on these pins on the window motor connector:
Pin 5 white wire, when car acc power is active, and again when acc is cold.
Pin 1 white wire/black stripe, when acc power is on, window doing nothing, then again while window rolling up using switch on door. And does voltage change/drop when window reaches top edge? Continue to observe voltage and roll window down.
Pin 3 green wire, same tests as pin1.

Use a good chassis ground point wire for the negative lead on meter.

Me having access to a car at this stage is not a real problem, it's me having time to go into the car and do things with a person volunteering their car is the issue. By fanning out work at this stage is the only way I can make progress, it's sort of a group discovery thing required here. My best is to help lead this expedition - team will make things happen. This is how I developed most of the autowindow features for VAG cars - never seeing some features on cars in person, but building electronics that integrate well and are lasting to this day, over a decade later.


Thanks!

[Sportsguy83]

Originally Posted by scottmcphee

Okay next person to go into either door, check out some wires on the power window regulator E5 on left door or F1 on right door. Say which door because they might act different.

What is voltage observed on some wires running between the window switch and window motor, on these pins on the window motor connector:
Pin 5 white wire, when car acc power is active, and again when acc is cold.
Pin 1 white wire/black stripe, when acc power is on, window doing nothing, then again while window rolling up using switch on door. And does voltage change/drop when window reaches top edge? Continue to observe voltage and roll window down.
Pin 3 green wire, same tests as pin1.

Use a good chassis ground point wire for the negative lead on meter.

Understood.

Hopefully @thejevans has time Thursday night :happy0180:

[thejevans]

Yeah, there's snow everywhere right now. I'm gonna have to pass for now.

[colHolm]

I might have time this weekend...

[trueno86power]

Can't help you sorry mate! ( just seen you mention me :P ) My car is stored for winter!

[scottmcphee]

Originally Posted by trueno86power

Can't help you sorry mate! ( just seen you mention me :P ) My car is stored for winter!

A Winter Princess!

..wish I had parked my car for the winter, got into a front ender accident on ice yesterday. Not fun.

[trueno86power]

Oohh crap no. But I wish I parked my car in the summer to. Got stuck in 2 hails storm last summer... lol

Anyway back to topic!