the circuit should show some voltage - whatever it is - between 21-23, and 21-22.

The test for the switches is to measure the resistance, each function has a specific value you look for - and yes, this would be a different voltage if you were measuring volts.

The important thing to note is that there is a difference between a set of 2015-2017 steering wheel switches, and a set of 2018-2019 steering wheel switches. Not sure how that decoder maps between a subaru legacy/outback with the earlier vs the later switches.

That is a selection on their website - it asks which years, and correctly lists 2015-2017 or 2018-2019.

You can see the difference in the values below..

2018/2019 switches:

2015-2017 switches:

 

Man, so much for plug and play
I've been going back and forth with Joying once per day, not much progress. Usually some sort of simple question or job aid for something I've already done. Sent them another pic of my original dash to verify my parts again today.

@Rymar @walker, The CAN bus decoder has a 2 pin plug, one labeled 2015-2017 and the other 2018-2021. I'm using the later. This inputs to a 2 pin labeled AMP CTRL on the unit. The job aid attached with that then asks for speaker setup, but its cut and pasted. I'm not sure if those are related to the "satellite switch".

I'll play around with the voltages on 21-23 and 21-22 later. My post 9 on the thread at least indicates those pins exist on the harness. Interesting observation, there is a CAN bus test menu where it scrolls what looks like a mac address on the screen when a button is pressed. I only get feedback for "A" buttons. Nothing from "B". It looks like Pin 22 might have an issue being read.

Thanks for the diagrams and knowledge.

 

One more thing to add, I don't have anything in PIN i87-10 on the harness and PIN i87-9 is my stray wire still. Joying hasn't even tried to answer me on that and ignores the question I've asked 3 times on.

 

here's an interesting thread... the OP stated that he had to change the resistance value from 10K to 1K to get his steering wheel controls to work. I believe his car is a 2017 WRX... so that would be the 2015-2017 outback radio, the Denso units, not the 2018-2019 subaru harman versions... but there was a setting to adjust it.

said it's accessed through a "secret menu" - I'm guessing the same 8888 menu that you were in and posted about earlier



the cars that have the same headunits as a 2019 outback:

2019-23MY Impreza​

2019-23MY Crosstrek (Including Hybrid Model)​

2019-24MY Forester​

2019MY Legacy/ Outback​

2019-21MY WRX​

2019-21MY WRX STI​

2019-21MY BRZ- Navigation -equipped ONLY​

2019-22MY Ascent​

cars that have the same headunit as a 2018 outback:

2017-18MY Impreza​

2018MY Crosstrek​

2018MY Legacy and Outback​

2018MY BRZ (navigation only)​

​

might help with searches - things like "2019 forester +joying headunit" or something like that. Figure maybe the forester or crosstrek people might have also used that headunit, and have the same controls on the wheel. Maybe it's a similar solution with idoing, but no idea.

 

Since the CAN bus is a differential bus, 2 signals are required: a '+' and a '-' version of the same signal. I think this means you will need to connect to both i87-9 and i87-10 for this to work.

Also, here is the system block diagram from the Gen5 FSM. Do you have the separate power amp? It's not clear where the CAN bus connection is. What is the CAN bus used for by the Joying unit?

Steering wheel switch sensing shouldn't be related to CAN bus communication at all, as the headunit is reading analog voltages at i87-21 and i87-22. That's why it's important to look at the voltages with the unit operating; resistances just verify that the switches and roll connector are working, which is pretty likely to be true.

 

The CANBUS connection for i87 9&10 is the top left in that block diagram - that would be the HS-CAN (high speed CAN).

I (re)connected these, and also left everything on null vehicle setting wise as also indicated in the same email. Now when programming the keys, it is as if there is a button being automatically pressed repeatedly. It immediately binds with whatever selection I make (volume up), and then continues to repeat the input maxing out volume in this case. To clarify, this may not actually be the volume up key, it just happened to be the mapping I chose. I have no idea what input is being repeated. It then times out any input at all after about a minute of this behavior, and I have to pull power to try again. I can no longer select any input at that point as it shuts something down. I'm pulling the battery connection because its easier than getting behind the radio. I can't find a power cycle on this device!

with all the moving parts in this install - I'd first verify that the satellite switches are functioning as Subaru intended. The 2018-2019 page I posted earlier has the resistance values - you could measure at the radio connection, they might be just a bit off, but should be really close. Pay attention to the value - at rest, see if it's good (no buttons pressed). You should test and record the values of all the button functions.

You definitely want to verify there isn't some high impedance problem with all the things done so far - something introduced that is unrelated to the radio. Check it off the list as verified. Testing them all will help if joying comes back and says your satellite switches must be the problem... you can say "nope, I checked them against the values in the FSM, they are tested and working as expected, under the conditions in the service manual". Then you can even give them the values you wrote down. Prevents some unnecessary back and forth.

Then - move on to the two connections going to the satellite switches - at the harness going to the new headunit - there you look for the reference voltage. If the radio is going bananas because of it - joying might ask you to verify what the voltage reads. I'm guessing there, don't have any idea what voltage it should be, could be, and under what conditions. You'd be measuring something and then guessing if it's ok or not. You could measure and then ask joying I suppose, and maybe that would prompt them to give you more guidance on that. Another thing to verify checked off the list, once done.

 

The reddit poster Lightspeed810 said this about changing the pullup resistor value, in a Joying 9" unit:
"I did have problems with the steering wheel controls. I had to go into the factory settings and change the resistance from 10k to 1k. You can get into the settings using 167 as the password (took me incredibly long to find that password online)."

Given the values of resistance shown in the tables, 1K looks like a reasonable value.

Also, I think the CAN bus is used by the factory headunit to monitor these parameters, by reading them from other CAN-connected computers in the car, such as the ECU and BIU. The Joying unit probably can read most of these as well. Did you try the CAN bus test, as shown in one of the service menus?

 

Also, I think the CAN bus is used by the factory headunit to monitor these parameters, by reading them from other CAN-connected computers in the car, such as the ECU and BIU.

Except the switches are terminated at the headunit, and not other CAN connected controllers - so the other controllers would have to get their data indirectly from the headunit, where those wires are physically connected.

 

Are there any other controllers which need the "Satellite Switch" status, aside from the headunit itself? The cruise control buttons on the steering wheel are connected separately. And all 8 (or 9) buttons are only related to headunit operation (see post #40).
I think it's just the headunit which needs things like reverse status (for backup camera enable), and vehicle speed (for lockouts), and illumination level.

 

You guys are really pushing my skillset here. Thanks for pushing me through this. I've really ever only used resistance to check for connections chasing household electrical.
That being said, I'm a bit confused where this 12 pin harness is.
I measured resistance and voltage at pins 21 and 22, as well as 21 and 23 on i87. I'm worried I've broke something due to my results. I had absolutely no voltage fluctuations. I cannot find a specific fuse that might impact this glancing at the manual.
As for resistance on these same pins, I could not get a single reading on the B side with circuit on.
The A side, had my meter on 2000 ohms, pins 21-23 i87:
vol+ 1000
vol- ranged from 600 - 850, and it just flashed on the screen, i could not hold the button without it returning to complete resistance. sorry for terminology!
seek left 330
seek right 001
what is list and return?
Is my meter terrible? Operator error? This sounds like an issue. Should I connect the old head unit up and see what works? Am I measuring at the wrong harness?

The Joying unit does have some sort of pull up resistor setting in one of the menus. The secret menu is in the main settings > factory and has a password of 8888 on my unit. I don't recall if that setting was in there or the steering wheel control mapping section under the "common" settings. I think I'll hook that up and check the resistance values to see if it is somewhere around 1k.
Thanks again

 

The Joying settings are not what I thought. There is a pull up resistance setting of low med high auto. None made a difference. I can only ever program 2 buttons before it seems to have issue, they are on the A side. Then there is a mv tolerance setting, default is 100mv. This isn't key to key, just general steering wheel control setting.

 

You should measure resistance with the "i87" connector disconnected from the Joying unit, and voltage with i87 connected and the Joying unit powered up. Are you sure that i87-23 is connected to ground through their harness? You should be able to just measure the resistance from this pin (on their connector) to ground, with power off.

If it's not too difficult, I would plug back in the OEM unit and check to make sure the steering wheel switches are all working.

 

I'm confused on my readings, but here they are:

with the Joying unit disconnected, I confirmed the above resistance readings from previous. Still nothing measurable on B side.....Is my 2000 ohms setting too low on the meter, or wouldn't that just cause the value to max out?

The voltage on the same connector, but from the joying harness side of i87 with power on
23-21 3.2v
22-21 3.2v
21 complete resistance to G
21-G 3.2v
23 & 22 0 resistance to G
22-23 0v

22 doesn't seem right, which is also the B side. Is this pulling to ground from the Joying unit so I never get a resistance reading?

I hooked back up the stock radio and all buttons work. Any more suggestions?

 

That is confusing, for the voltage settings.

what should happen - you should have a reference voltage (whatever the value) and it should be the same when measured from:

21-23
22-23
23 can/should have continuity (no resistance) to a chassis / electrical ground (depends on what the radio is doing)

IF all three wires were doing what the subaru controls want to see, I would say it looks like 22 and 23 were reversed… BUT measuring 22-23 would show voltage, even if reversed. And 21-23 wouldn’t work right, it would show 0v, since both would be the same voltage…

This looks like 22 and 23 are both grounds, both have continuity to ground. I’d test voltage from 22-GND and 23-GND as a confirmation step, but there is enough in your measurements I think to guess.

This would explain why side A works - because it’s getting voltage when connected, and side B is getting no voltage. One voltage source, two grounds.


Now - it could be that this unit is running the voltage the other way through the circuit - that there is one voltage source and two grounds - it’s just measuring the voltage drop after it goes through the controls, instead of before it goes through the controls.

If that is the case, for that to work - wire 23 would need to be a voltage source, and both 21 and 22 would need to be grounds.

Another scenario - voltage flows the opposite way for this unit… and is measured at a different point in the circuit (going in vs coming out)… so if you reverse the idea - where 21 and 22 are grounds, and 23 is a voltage source - you would get voltage on both sides

if 21-23 are currently working, for the A side, and the unit is reading on the ground side (23), and is being supplied by 21 for power… you would need the B side to be connected to 21-22.

what subaru radio wants to see:

power in A side : 21
power out A side : 23
power in B side : 22
power out B side : 23

IF it’s the case that this unit is measuring after the circuit, then it would want to see:

Power in A side : 23
power out A side : 21
power in B side : 23
power out B side : 22

it’s just reversed direction on the switch wiring diagram.


Maybe this graphic will make it make sense - this is what I think is happening, without seeing any wiring stuff from joying:

SW1 is the A side, SW2 is the B side

Based on one side working, and the measurements - I think this is it.

What would need to be done is to arrange the pins, like above, for the Joying row. Power coming in - looks like it is currently pin 21. Pin 21-23 work for SW1. So power in, pin 21, would move to position 23, pin 23 (ground) would move to position 21. Pin 22 stays in position 22.

I’d take a picture before and after as a reminder if you start swapping pins around. Or tag the wires with something (like a piece of tape).

Worst case is it doesn’t work And you can move the pins 21/23 back.

 

More thoughts. If the stock head unit works with the buttons, why am I still not getting a reading on the B side for resistance buttons (21-22 i87). Also, remember that time I accidentally gave 12v to the mic....what could I have broke. Seeing that it works with the stock head unit, I'm doubtful anything broke. I do believe I turned the key to power on without starting, and no radio connected. I tested my connection and then realized my error.

 

Some findings on the harness, I'm putting this here for my records as well. So harness i87, black 20 pin connector (B20) that goes directly to head unit, and white 16 pin connector (W16) that goes to canbus decoder.

B20 to i87
2 connects to 21
12 connects to 22
15....and some of 16, connect to 23, is this accurate?

B20
15-16 559 ohms resistance

B20 to W16
16 connects to 16
15 connects to 8

W16
16 and 8 do not connect

W16 decoder receiving side
16 and 8 558 ohms resistance, matches as expected from testing B20 to i87, but again...is this supposed to? This should be ground from i87-23.

 

That being said, I'm a bit confused where this 12 pin harness is.

that is if you disassmble the steering wheel enough to get to the switch side of the roll connector - it's the first plug in the harness from the switches on the wheel... like if you replace them, you'd unplug them there... since you're at the radio end of things, you're testing there instead.

in the wiring diagram - the connector right at the steering wheel for those switches, the connection pin numbers are shown... at the audio side, the pins in the connector are shown as well. I circled both.

at the wheel - 2 becomes 23 (your ground). 6 becomes 21 (your switch A side, or SW1+), 12 becomes 22 (switch B side, or SW2+).
NOTE: These are subaru harness connections, nothing to do with the joying stuff..

A/SW1+ : 6 / 21
B/SW2+ : 12 / 22
GND : 2 / 23

I measured resistance and voltage at pins 21 and 22, as well as 21 and 23 on i87. I'm worried I've broke something due to my results. I had absolutely no voltage fluctuations. I cannot find a specific fuse that might impact this glancing at the manual.
As for resistance on these same pins, I could not get a single reading on the B side with circuit on.
The A side, had my meter on 2000 ohms, pins 21-23 i87:
vol+ 1000
vol- ranged from 600 - 850, and it just flashed on the screen, i could not hold the button without it returning to complete resistance. sorry for terminology!
seek left 330
seek right 001
what is list and return?
Is my meter terrible? Operator error? This sounds like an issue. Should I connect the old head unit up and see what works? Am I measuring at the wrong harness?

use the other picture - the one with list and return is a 2015-2017, and they had different controls for the radio. They also have different resistance reading. You want the one for the 2018/2019.

setting your meter to a 2000 ohm limit will cause the at rest test to fail - it's 100 kiloohms. 100,000 is going to be above your limit of 2000 in that setting. Likewise, volume down is going to be 3210 ohms (going by the diagram), and your meter will flash funny as it tries to read the resistance before it goes past the limit you've set (at 2000).

working fine, just be aware of any manual limits you set - you might have to jump up to the next higher value for a limit if you need that little extra bit to read the value.

The Joying settings are not what I thought. There is a pull up resistance setting of low med high auto. None made a difference. I can only ever program 2 buttons before it seems to have issue, they are on the A side. Then there is a mv tolerance setting, default is 100mv. This isn't key to key, just general steering wheel control setting.

the pull-up resistance setting is to get the unit in the ballpark range of where it should be - the mv tolerance setting is to fine tune the sensitivity of the reading - so if there are resistance values that are close enough together, or maybe really far apart, it might not distinguish between those buttons. That would be a fine tune if it gets confused between say the two seek directions, or the seek left and volume up.

maybe try the button programming in order (lowest to highest, highest to lowest) for resistance values. Does it always let you do two and fail in either order? with the chart, i'd start with each button, in turn, to program. Go from either highest resistance to lowest. Both A and B side switches have similar resistance values. See if there is anything in common.

A side is radio stuff, with these values
seek > : 1 Ω
seek < : 330 Ω
vol + : 1000 Ω
vol - : 3210 Ω

the B side is the phone stuff, with the same values

source : 1 Ω
off hook : 330 Ω
on hook : 1000 Ω
talk : 3210 Ω

 

@walker, thanks for pin/diagram lesson.

Wow, so my side A resistance readings were spot on it appears. I still can't get a side B reading for some reason. I'm going to test again when I take back out the factory unit and I'll try the mapping order too.

I've only gotten 2 buttons mapped before it stops me from mapping more. Even then, they never really work afterwards. I've done random orders and its mostly the same result. The only unique thing is when I do volume up mapping, it just continuously pulls the volume up like I'm inputting it continuously. Also, my "B20-12" pin was a little loose in the adapter. This goes to i87-22. Fingers crossed it does something.

 

Well, I still couldn't get a reading from side B....but the steering wheel controls are now working! I think the pin was a little loose that I mentioned previous. I'm not sure why that would had interfered with side A mapping. Yay, victory. I'm going to refocus on what else I need to get working on Monday.

Thanks to all that participated thus far!

I'm still hoping to get the rear camera to interface with the steering wheel. I need to get a mic hooked up, not really sure if it is possible to work with stock mic by just adding 5v. Tablet built in mic works for voice but not good enough for a call. There is also a mic it came with that I might replace stock one with. Still looking to play with HDMI out, GPS, screen auto dim with car settings, and 4g (I got a Tello SIM in the mail today for this).

@Ben78 , if you're still around, are you able to open/push the "car info" or "car set" app (name varies)? I get an error message still. Joying indicated that this should bring up a menu giving me options from the CAN bus. What do you have in there if it does open? Do you have your backup camera working with steering wheel? How about diming? I currently set it by time but want it to sync with car headlights.

@walker , any ideas if the stock mic audio input is routed through this already?

@Rymar , I'm worried I need i87-9/10 for the active reverse steering lines and auto diming. Do you know anything more on that CAN bus test? I have that stray pin 9 labeled amp, nothing in pin 10 at all connecting to joying. I do not have any sort of amp nor am I aware of anything needed.

 

Carinfo just gives a message saying "There is no original car information for this model"
My reverse camera works, as does my steering wheel buttons. I can't start the reverse camera with a steering wheel button - is that a thing?
Dimming works, in the settings there is a hardware menu with a toggle switch to see the headlight status. But I drive with my headlights on always anyway, so I mapped a steering wheel button to let me raise the brightness if I need to.

 

Well, I still couldn't get a reading from side B....but the steering wheel controls are now working! I think the pin was a little loose that I mentioned previous. I'm not sure why that would had interfered with side A mapping. Yay, victory. I'm going to refocus on what else I need to get working on Monday.

The voltage differences are going to be very small - a loose connection will cause bad readings if the resistance values are off on that side of the connection.

I need to get a mic hooked up, not really sure if it is possible to work with stock mic by just adding 5v. Tablet built in mic works for voice but not good enough for a call.

@walker , any ideas if the stock mic audio input is routed through this already?

For a 2019 - the mic shows two sets of connections, one set of three to the DCM and one set of three to the headunit. For cars without a DCM (like you've done now) it shows all 6 going back to the headunit.

reports on the forum, including the subaru official bypass box, say you have to connect power back to the microphone, and where it's connected corresponds to cars that do not have telematics hooked up - so that means the wiring effectively becomes the wiring for the non telematics cars. 2015-2017 non telematics cars only had three wires going to the headunit, or through the DCM. Only one set.

The function of the wires is duplicated - two each, for the three required connections. And all 6 end up in the stock radio at some point (either directly or through the DCM).

Now, if the redundant set actually worked, you wouldn't have to wire up the separate power with pin #4 on i87. Normally that is blank on the 2019., the DCM feeds that line into a different pin to the radio with a different function (MUTE) - pin 24. My guess is some sort of toggle for not only power to the mic, but also for the DCM to mute the stereo (and probably disabling the second mic connection).

I'd try just the 2015-2017 connections - those are correct for the MIC1 connections on a 2019 with no telematics.

So... hopefully less confusing - Connector i87:

pin# 4 - 5v power
pin# 5 - MIC SIGNAL (MIC+)
pin# 19 - MIC GROUND (MIC-)

this is good for 2015-2019, without telematics, or if you've bypassed the DCM.


The second set of mic connections - for reference - on a 2019, these exist, but don't in a 2015-2017. Some reports on the forum show that the mic to the radio stops working if you don't connect pin#4 above. That means the original radio really likes to see the MIC1 connections, and I'm not sure what it's doing with the second set. These wires do exist in the overhead console. Since you are using something aftermarket, it's probably not interesting to the radio, and you might get away with using the one above.

OR, maybe you could use this second set if the ones above don't work for whatever reason. Try the MIC1 connections first, if not, here are the pinouts for the second mic connection.

connector i87

MIC1
pin# 4 - 5v power
pin# 5 - MIC SIGNAL (MIC+)
pin# 19 - MIC GROUND (MIC-)

MIC2
pin# 6 - 5v power
pin# 13 - MIC SIGNAL (MIC+)
pin# 14 - MIC GROUND (MIC-)

a 2015-2017 car will list these secondary connections as follows (they’re not used in those years):

pin# 6 -MIC DET (5V)
pin# 13 - NC (no connection)
pin# 14 - SHIELD-GND (AUX)

The mic in those cars only has three wires coming out of it (I scoured some ebay photos of both overhead consoles to confirm existence of pins and the connector).

if all that fails, run the provided mic up there and figure out how to secure it in place.

 

OK, I may have some of this wrong, but here goes about the CAN bus.

You don't have an extermal power amp; but where there is an external power amp, for example in the H-K systems, the OEM headunit sends commands over a serial link. Is this another, private CAN bus? It might also be another type of serial bus. It is definitely a differential bus (2 signals, + and -) like CAN.
There are also 2 'interrupt' signals from the power amp back to the OEM headunit, perhaps for error conditions like overtemp or short circuits on the speaker outputs.

Anyway, the wiring diagram shows these connections between the headunit and power amp.

i262 is the power amp connector, which is wired to the OEM headunit. First the serial control signals:

and the interrupt signals:

and here is the wiring diagram:

So, one possibility is the instructions for the CAN bus connection with the Joying unit are actually to the power amp (which you don't have). This could explain why the 'AMP' signal is involved.

So, back to your setup.

The main CAN connection is on U87-9 and U87-10.

Are you sure that the Joying unit doesn't have another pair of CAN bus pins, perhaps on a built in connector instead of the external adapter? It is very common for cars to require a CAN bus connection to read reverse gear status, for example. The CAN bus test I mentioned was shown in one of the menu images you posted. But it won't work without a connection on both U87-9 and U87-10 of the positive and negative versions of the signal.

I would also expect that you will need to enter a vehicle type, in order for the Joying unit to know which controllers to query, and what CAN message IDs to use, to get some of the status info it needs. Once you have established a good electrical connection on the proper CAN bus signals, the error during vehicle selection might go away.

 

Thank you - this post has been incredibly helpful! I've almost got my Joying 9" install complete for a 2017 Outback Limited with the Harmon Kardon system.

I have one last issue - the rear speakers don't work. I left the Starlink system connected so the front speakers work fine, but I've got nothing from the back. I tried all software/interface settings I could find. Joying support responded with a customer's feedback, "Discovered the problem, the Subaru Starlink box bypassed the front speakers. My initial concern about the extra plugs was right, I have to plug them in for the front speakers to work (Reddit help)." I responded to them "thanks, but it's the back speakers, not the front."

Did any of you have issues with the back/rear speakers?

 

Anyone tried this Gps antenna adapter? Europeans, DAB adapter? SBU-SMA Subaru GPS/DAB factory antenna conversion cable

 

So at this point what is your opinion of this option. Probably too early, but how does the unit handle hot interior temps in summer time? What is your summary of good and bad changes from subaru original unit? How long did it take? It sounds like a long painful install. In hind sight, would you do it again?

 

does the radio have a real SCAN function or is it just SEEK to the next station and stop ? also, does it receive HD FM stations ? thanks

 

Hi, Just ran through the same. Quick review on that post:
https://www.subaruoutback.org/posts/6583325/