BlueSaab v6.1

[Karlis]

BlueSaab v6.1 is finally here.

It comes in two flavors: with and without SID text control.

If your car does not have SPA (Saab Park Assist) system installed, you can use BlueSaab firmware version that has the SID text control enabled. However note, that if you do so, Night Panel function and any warning messages won't be displayed on SID as long as BlueSaab is engaged.

If your car has SPA (Saab Park Assist) system installed you should use BlueSaab firmware version that does not have SID text control enabled. DO NOT, I repeat, DO NOT install BlueSaab firmware version with SID text control if your car has SPA! We are not responsible for any issues you may encounter or any damage done to your car or you if you do so. You have been warned.

Attached .zip file contains:

    - BOM (it also shows which parts are mandatory and which parts you will be ok without)
    - Text file containing links to PCB and CDC connector
    - BlueSaab v6.1 firmware binaries
    - BlueSaab flasher tool for macOS
    - MS Windows users should use Flasher-STM32 GUI tool (link in links.txt file)

CDC connector is made by TYCO-AMP-TE CONNECTIVITY; part #827229-1. One place we we used to purchase these from "Connector People" are selling them now for a price of an arm and a leg. So look for alternative sources...

That's about it.

BlueSaab_v6.zip

[Seth]

Excellent work!

[homerisback]

Many thanks Karlis ! Can't wait to feed my v6 modules with it

[EDIT @ 15:30 CET]
Just back from my car where I tested my bunch of v6.1 modules : worked like a charm with SID scrolling text (didn't try without text) ! Amazing job

It gave me a fresh look on the issue I had since a few months : no more scrolling text with my v4.2 which I had factory resetted thinking about a faulty RN52... nothing to do with Bluetooth chip : my phone is faulty and doesn't send track tags to BlueSaab anymore and take days to respond to SEEK+/- or NXT.
With my old iPhone4 everything works flawlessly

[j_szczygiel]

Congratulaitons! :) Can't wait to set it up. However, there is no CDC connector type included in the text file. Could you post it here and update the ZIP package please?

[stm32]

Finally got my first 6.1 assembled and now doing some testing on it prior to hooking it up on the car (9-5 2008 with Denso satnav).

Anyway, few questions. I got the software loaded no issues and everything seems to be working fine on the STM32 side. However, I'm failing to get any power from STM32 to the POWER EN pin on the RN52. Is there some kind of wakeup thing i.e. there needs to be traffic on CAN before the STM32 pulls the POWER EN pin high on the RN52. I do get power on the actual power pin just not to the pin that wakes up RN52.

Second question, did you manage to get the RN52 auto setup done for the STM32 version? Or is it still a requirement to configure the RN52.

Anyway, awesome work with bluesaab and especially happy with the route you've taken with the STM32!

[Karlis]

PWREN pin on RN52 should be always getting 3.3V as it's always asserted high from STM32 (i.e. it never changes state). Having said this, RN52 is configured to go to sleep after a period of inactivity (no active connections for 10 minutes). It then takes a CAN message coming from IHU to wake up RN52 and make it reconnect to the last known device on PDL.

RN52 gets configured on every reset of BlueSaab. There's an initialization routine that takes care of all necessary RN52 settings, so no manual configuration needed. This also the case for "legacy" systems starting with versions 5.x.

[stm32]

Ok, cool! Thanks for the information Karlis. I'll do some double checking to ensure I haven't fried anything during soldering.

[j_szczygiel]

Do you guys have or share an online BlueSaab BOM i.e. on Mouser/Farnell/TME or at least Google Sheets? That would be much easier to order all required parts at once than manually digging in a particular store and searching all the parts one by one...

[Seth]

It helps to search the forum...

http://bluesaab-forum.2349123.n4.nabble.com/V6-1-6-1-1-PCB-tp1360p1410.html

[Karlis]

The BOM can also be found in the attached .zip file of the original post in this topic.

[stm32]

Bluesaab v6.1 fully assembled and working straight out of the box on 2008 Saab 9-5 with Denso satnav and SPA. Great work Karlis and others!

Only question I have is that the IHU buttons 1-6, random etc doesn't seem to be working. Increasing the gain etc.. I assume there is something different in the Denso unit? Steering wheel buttons are working fine!

[Karlis]

Congrats!

I haven't figured out how Denso button events are being handled on the bus as I don't have access to such a system to do traffic analysis. I would assume the events are similar but are being broadcasted with different CAN node IDs (?).

As for RN52 volume gain - I disabled that part of the code as it's basically useless. RN52 volume gain should always be at max setting - 15. That allows relatively "smooth/close" volume levels between Radio and CDC modes on IHU.

[stm32]

Ok cool! I'll capture the bus events and share them with you as soon as I get some spare time again to tinker with the car.

[stm32]

Ok some new "features" on bluesaab using the v6.1 with the 2006+ Denso.

At somewhat random intervals IHU loses the CDX (bluesaab), Bluetooth turns off and the unit just blinks the green light with the power led on. After this happens you have to change the source to let's say radio and when you press CDX again in IHU Bluetooth turns on, CDX is visible in IHU and all works just fine.

Judging by the lights it seems like bluesaab just goes to sleep. Cycling the source and bluesaab waking up after that, also supports this hypothesis. Could there be a CAN frame that is not present or different to the non Denso cars?

I'll mod the original changer I took from the car so I can capture the CAN traffic with the changer in place. Hopefully we can then resolve the issues.

[Karlis]

In v6.1 firmware, RN52 Bluetooth module is put to sleep after 10 minutes of inactivity (no active connection to audio source - phone, tablet, pc, etc.). It takes a IHU control frame to be sent out for RN52 to wake up and connect to last known device on PDL (Paired Devices List).

[stm32]

Yup, I'm aware of that but all of these "sleeps" have happened while there has been an active Bluetooth connection. Is bluesaab doing something if there is the spindown frame sent by IHU, or something like that?

Anyway I got the original CDX now modified and will record the traffic tomorrow. Any wishes on which events I should capture? Will get at least btns 1-6, random and scan from IHU screen.

I'll also try to see if I can capture a traffic that happens when Bluesaab goes to sleep in the middle of an active connection. Provided that is the reason.

[Karlis]

There are three types of "request" frames sent out by IHU to see what CDC is doing:

*) Power-up ("Hey, CDC - time to power-up!")
*) Active ("Hey, CDC - you still there?")
*) Spin-down ("Hey, CDC - time to go to sleep!")

To which CDC (BlueSaab) is supposed to reply with a group of four frames.

"Requests" from IHU have CAN ID 0x6A1, CDC "replies" have CAN ID 0x6A2.

In my experience, I *think* I have never seen anything but "Active" requests coming from IHU when BlueSaab is connected to the car. That might be because BlueSaab sends out a status frame 0x3C0 that already says something like "Forget about power-up, I'm active". Need to check my code again...

[stm32]

Here's the captured data from IHU-CDX CAN frames. I tried to press all buttons always multiple times for each capture to ensure it is easier to spot them from the CAN traffic. I only had 4 audio CDs so had to do it twice, first capture is discs 1-4 and second is discs 3-6. Second capture also has the random and scan (disc/track) buttons.

ibus-ihu-cdx-btns_1-4.csv

ibus-ihu-cdx-btns_3-6-rnd-scan.csv

Karlis if you would like some specific events captured or there are individual CAN frames that are interesting I can go back to the car and try to figure them out.

[Karlis]

Thanks.

From the looks of it there's not much of a magic happening there. The same frames with a bit of different bytes flying around. Specifically 3C8 (CDC status frame). Looks like the magazine was not present (removed) at some point during the capture. Hence the last byte of the frame changes from D0 to C0. Other than that it looks like a normal "conversation street" on the bus.

[stm32]

Cool! Oddly enough after removing bluesaab for capturing the CDX frames and reinserting it, I haven't been able to reproduce the straight to sleep from active bluetooth bug and the unit has worked flawlessly including multiple starts and 6hrs of driving. So looking good! It could be as simple as bad CAN connection or something if the connector wasnt fully seated or something.

Anyway thanks for the help Karlis!

Hopefully the captures will be of help for any future work! And thanks for awesome work on the unit!