Alfa Romeo Forum banner

Integrating original Mic into new HU - Investigation & Implementation

2K views 12 replies 5 participants last post by  Marmo 
#1 ·
Hi all,

Like many of you, I've purchased a new double din HU to replace the OEM radio currently installed in my 159.

To make the system look a lot more professional, I would rather use the existing Mic that is used within the B&M system, rather than installing the one that came with the HU and routing that through.

Most threads on here I've seen state it can't be done. At least not easily!

So I've decided to start an investigation to try and see if it can be done and then give a detailed installation guide.

The plan so far:

1. Remove Mic from central cluster and analysis the Mic output using an oscilloscope (I'm an Electronic Engineer by trade).

2. Reposition the Mic to the right side of the cluster, hopefully improving sound quality.

3. Establish the pins for the Mic input signal to the B&M unit.

4. Splice into said Mic signal wires and analysis signal (again with an oscilloscope)

5. Assuming all is good, design a cable to tap into the Mic signal and input into the new HU.

I won't be covering installation of the HU unless asked. There are plenty of threads on this topic.

Looking forward to posting my results!
 
See less See more
#2 ·
Hi

I've put the mic that came with my HU on the drivers side of the cluster a little below the level of the seat. You can't see it and admittedly I don't use it, but as it is hidden the system looks just as professional as it was before.

Bear in mind that if you parallel your mic between two audio inputs then you are loading it more than either system expects and you may lose signal strength in both systems possibly meaning that you may have to shout to get it to work with both the HU and B&M. You might not, your scope will show you the loading effect.

I suspect that the threads that say it can't be done are looking to somehow extract the mic signal via the CAN bus. That presupposes that the B&M system actually puts out an audio stream on the bus and that the HU has the software to detect and use it.

So much assumption is made regarding the CAN bus. It's not appreciated that the CAN bus only passes data - it doesn't actually 'do' anything.

Good luck!
James
 
#3 ·
The CAN bus is nowhere near fast enough to put bitstream audio through. Also it would be a waste of bandwith and audio quality to even try.

I am very curious where the signal gets translated from analog to "digital". If there is just an analog signal arriving at the B&M unit, it shouldn't be an issue to re-route it.
You would however also need to measure the impedance of the microphone and see if your new headunit has the appropriate amplifier.

Maybe some B&M pinouts will help.
Did you check E-Learn for wiring info on the B&M and/or radio nav routing?
I can check when I'm home tonight. Sounds like an interesting undertaking and I think it can be done!
 
#4 ·
Not quite true....it also decides where and whether to send that data...for instance on some cars if the seat belts are not engaged it wont send the signal to release the electronic handbrake...and lots of other similar linked actions. It also changes the signal to and from a CAN bus data signal to a CAN bus controller signal. Receiving: it converts the data stream from CANbus levels to levels that the CAN controller uses. Transmitting: it converts the data stream from the CAN controller to CANbus levels. So if you are expecting you mic data to pass through the canbus unchanged (if it even goes through it....I have my doubts) I dont think it will be. Unlike you two Im most certainly only a keen amateur....so I happy to be proven wrong!!!!
 
#9 · (Edited)
The CAN Bus doesn't decide anything.

Imagine Churchill, Roosevelt and the three Armed forces Chiefs in WWII, they all had to pass information to each other and receive information from others in order to make decisions and order things to happen. There was also a postman. While the war was executed by all the famous people they needed the postman or it wouldn't have happened. Now Postie had no idea what the others were saying in the letters but he had rules, he would go to a standard post box where people had put the letters, collect the standard envelopes with an address on and deliver it to another standard post box.

The CAN bus is a postman, you give it information via a standard interface and it sends it, interested parties pick it up by a standard interface and use it. (OK that's more like Ethernet, CAN is a little different but lets not get too deep) The interfaces and the bus have to be standard because companies in the future might want to connect new equipment to the bus and they need to be sure that they will be accepted by the postman (because he is a simpleton and only knows dat rules is rules.... seriously computers are thick, but fast)

What you state about levels etc is true - but that part is standard and all kit will do the same, importantly it does not include any decision making. In your seat belt situation it is actually Winston Churchill who wont let Roosevelt release the handbrake unless the RAF says it's OK, postie just delivers the letters between them. Unlike simple electrical circuits where everything happens instantly the software in the control units can monitor other equipment over time and only make a decision if it 'thinks' it is safe (for example). Even then that control unit could decide to flash a light and wait for the driver to press a steering wheel key before releasing the handbrake.

This is why systems with software in are so hard to fathom - you can't tell what they might do just from observing them or having a wiring diagram, you have to be told what they will do. What they do is exactly what the programmer wanted them to do, so you have to ask him. So if there is no documentation you may as well cast some goat entrails....

Incidentally, it seems that audio can be sent over CAN, just google it. I suspect that the audio would be compressed to avoid hogging the bandwidth. Again this reflects the fact that CAN is just a postie and doesn't give a monkeys what you are sending. (Apologies to Postmen - you are essential!)

I can't find any block diagrams of the CAN system myself or I could be more useful!

Cheers
James
 
#5 ·
So looking at the E Learn schematics, the Mic is simply two wires going straight to the B&M module, going through D020. No CAMBus involvement.

So initially I took the Mic out of the lighting cluster and captured the signal. The first picture shows the signal captured when tapping the Mic.
This was to make sure that it is simply a standard cheap passive Mic, requiring no bais voltage to power a preamp or anything like that.

The signal shows a standard Mic signal peaking at around 5mV.

Perfect!

I'm now going to splice into the pins 8 and 24 of the B&M connector and confirm nothing funny is happening before being inputted into the module.

In regards to over loading, it shouldn't be an issue if it is simply a cheap standard passive Mic.
The input impedance of the HU amplifiers is so high you can normally split them three or four times before any noticeable change. So I'm tempted just to see how it performs. I can always just cut the connections to the B&M unit if it is quite or I measure a noticeable attenuation in the signal.

Speaking of impedance, there has been no effort in the wiring to provide any kind of impedance matching at the Mic end and the B&M end. This is all pointing in the direction that the Mic isn't anything special (think the Mic that's integrated into phone headphones).

I'll post the results captured at the B&M end soon! So far it's all looking quite simple.
 

Attachments

#6 ·
That great news!
If the mic signal is passive, I doubt the B&M will throw any errors if you fully disconnect the mic.
e,g, unplug it from the B&M or take the pins out of the connector. That way you don't have to "share" the signal. If you keep it plugged in to the B&M you might get into a whole lot of trouble and noise due to ground loops.
I doubt there are any useful B&M voice commands if you are going to use the new HU for media, navigation and voice calls.
 
#8 ·
So I've now removed the B&M connector from the outside shell so I can trace the wires from pins 8 and 24 back (picture 1).

They lead to a coax cable that has been spliced to fit in the connector, reducing the alot of the EMI shielding effects provided by the coax (but that's just me being anal!)

I used these tap wire splice connectors that cut into wire and provide an additional connection to that signal (picture 2).

When I probed the signals, there was a lot more high frequency noise than before so it was difficult to detect any change in the signal amplitude.
Unfortunately, this scope does not have any real time filtering I could used but it does allow FFT so I can analysis the frequency response in (sort of) real time.

Picture 3 shows the signal along with the frequency response.

To prove the Mic was still behaving as normal I tapped the light cluster to look for any spikes at the low end of the frequency response.

Thankfully, it was showing a change when ever I tapped the light cluster or made a loud noise (with some delay) so I think I'm just going to go for and make a cable.

Wish me luck!
 

Attachments

#10 ·
GOOD NEWS!

I made a cable using some coax laying around from work (I think its RG159 from memory), a 2.5mm subminiature Jack, and a couple peices of 20 AWG wire soldered to the shield and signal of the coax.

Wire Cable Technology Finger Electronic device


I then connected this to the screen and signal for the Mic and it works a treat.
Even while plugged into the B&M unit.

Initially it was only connected using the tap splice crimps but they kept coming loose so I soldered it in place in the end.

Wire Electrical wiring Cable Auto part Wire stripper


When I get home I'll write up a full guide to doing this mod if anyone wants it.
 
#11 ·
GOOD NEWS!

I made a cable using some coax laying around from work (I think its RG159 from memory), a 2.5mm subminiature Jack, and a couple peices of 20 AWG wire soldered to the shield and signal of the coax.

View attachment 887835

I then connected this to the screen and signal for the Mic and it works a treat.
Even while plugged into the B&M unit.

Initially it was only connected using the tap splice crimps but they kept coming loose so I soldered it in place in the end.

View attachment 887839

When I get home I'll write up a full guide to doing this mod if anyone wants it.
Thats great!

I would definately be interested in a write-up if/when I upgrade my headunit.
AFAIK maps won't get any newer than 2015 so it's coming to it more and more every year..
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top