[Madskp]

Did some more testing today measuring voltages as I turned on and off the system.

The BL3500 is recieving the 5V on its powerlink PIN 4 when a source is selected from the W1. Also the data signal on pin 6 seems to change in voltage when turning on and off.

I can also see a change in data signal voltage when changing volume.

For good measure I tried to apply 5V to pin 4 and 7 on the powerlink connector without anything else connected to the BL3500, but nothing happened.

So it might be a combination of the 5V on pin 4 and the datasignal on pin 6 that has to be present to turn the BL3500 on via powerlink, and that data signal must be different from whats comming from a BL Active, MCL2AV or any powerlink equipped BS, BM, BC

One odd thing I noticed when measuring voltages was that pin 1 on the BL W1 Powerlink connector was 12V which seems high, but the BL3500 MKII doesn’t seem to have anything connected to pin 1 so might not mean anything in relation to the BL3500

[Madskp]

Madskp wrote:

One odd thing I noticed when measuring voltages was that pin 1 on the BL W1 Powerlink connector was 12V which seems high, but the BL3500 MKII doesn’t seem to have anything connected to pin 1 so might not mean anything in relation to the BL3500

Just to be sure that PIN 1 in the powerlink connector is not used for turning the BL3500 MKII on I decided to disassemble the BL3500 so I could get full access to the plug PCB. I found out that most of the parts have to removed from the aluminum enclosure in one piece

From there I could unscrew the plug PCB.

There is no sign of connections from pin 1 of the DIN connector on the PCB. To verify further I also tried to do continuity test from pin 1 to all other connections, test points and vias on the PCB, and found no place with connction. So I’m about 99.9% sure that Pin 1 has nothing to do with activating the BL3500 MKII via Powerlink.

 

[Madskp]

Some aditionel testing with Bl3500 MKII and BL W1.

The same setup as before with the BL3500 and BL W1 connected via powerlink. In addition I also connected the BL3500 to Masterlink and set ML to on via the submenu Menu, 0, 9, GO to see the Behaviour.

In this case I cannot activate the BL3500 via the BL W1 powerlink conenction

[Madskp]

Madskp wrote:

So it might be a combination of the 5V on pin 4 and the datasignal on pin 6 that has to be present to turn the BL3500 on via powerlink, and that data signal must be different from whats comming from a BL Active, MCL2AV or any powerlink equipped BS, BM, BC

And just to be sure that both the Powerlink On signal on Pin 4 and the data signal on pin 6 in the powerlink connector is needed at the same time to activate the BL3500 MKII from the BL W1 I tried to make an adapter cable with only pin 4 and the 2 ground pins (2 and 7), and after that a cable with only pin 6 and the 2 ground pins.

I was not able to activate the BL 3500 with any of these cable configurations which supports my earlier statement that signals on both pin 4 and 6 are needed to activate the BL3500 MKII

[Madskp]

After testing with the BL W1 I am now at the point where I have no further open ideas to test for the topic of this thread. Therefor any suggestions for further testing is very welcome..

Of course if I get any Ideas myself I will continue with them

[matador]

Hi Mads,

I must confess I’m a little bit lost here, All those pin to pin voltage do confuses me.
On question thought: did you manage to get the BL3500, working as it should with the WL1?
That mean having it wirelessly connected to a master link master and having the WL1 starting the 3500?

[Madskp]

matador wrote:

On question thought: did you manage to get the BL3500, working as it should with the WL1? That mean having it wirelessly connected to a master link master and having the WL1 starting the 3500?

I you refer to what is shown in the Bl3500 user manual I haven’t tried that setup as I gues it will work and is a little outside the scope of this thread.

But to clarify the setup I tried I have modified the sketch from the user manual:

The Powerlink cable has to be fully wire for it to function as it needs both the normal powerlink on signal on pin 4, and the data signal that is on pin 6.

This setup works, but I can’t replicate it with powerlink outputs from BL active, MCL2AV or BS Ouverture, so the data signal from the W1 must be the key for this to work.

[B3OHACK3R]

Madskp wrote:

After testing with the BL W1 I am now at the point where I have no further open ideas to test for the topic of this thread. Therefor any suggestions for further testing is very welcome..

Hi!

now where you have found out that both the trigger signal and some communication on the data pin is necessary just a hint from my side.

The next obvious step would be analyzing the data stream W1 sends on the PL Socket.
You can use a logic analyzer for it or a dev board like Arduino, Raspberry Pi, etc. Just watch our re the voltage level.
Since PL, DL, MCL and IR more or less all share the same data protocol my idea is treating it like an electrical IR signal (without the carrier). Use an IR library for recording the raw data stream that turns on the BL3500. Then just play it back again and the BL3500 should turn on. Make sure to turn off the carrier frequency when sending. If it does not work straight away use an oscilloscope for debugging.

No guaranty – to be done on your own risk 😉

[Madskp]

Hello and thanks for the input.
I don’t have any experience with logic analyzers but might try to dip my to into it.

This project might actually have a tool I can use (bullet number 3):

https://github.com/jensjoachim/B_and_O_Projects

It might take some time before I get to it as I have some other projects going at the moment and not much spare time.

thanks again for the input

[B3OHACK3R]

Yes, saw that before and also the second one you linked in the other thread.

Both are a bit improvised and at least on a RP2040 MCU it didn’t work straight out of the box.

If you want to dig into it using an MCU I can recommend starting here. It supports IR and datalink so should be usable for PL data as well I guess.

https://github.com/Arduino-IRremote/Arduino-IRremote/blob/master/src/ir_BangOlufsen.hpp#L4

In the end you don’t have to understand or reverse engineer the data protocol at all. Just record and replay what comes out of the W1 PL data pin. So any software that can record and send raw IR data will work. For the sending part just make sure to switch of any carrier frequency.

[Madskp]

B3OHACK3R wrote:

If you want to dig into it using an MCU I can recommend starting here. It supports IR and datalink so should be usable for PL data as well I guess. https://github.com/Arduino-IRremote/Arduino-IRremote/blob/master/src/ir_BangOlufsen.hpp#L4 In the end you don’t have to understand or reverse engineer the data protocol at all. Just record and replay what comes out of the W1 PL data pin. So any software that can record and send raw IR data will work. For the sending part just make sure to switch of any carrier frequency.

Thank you, I will try to dive in to that when I have some time on my hands. Always great to have a starting point to learn from.

[Madskp]

Now I have a logic analyse and have tried connecting it up to the datalink pin 6 in the powerlink connection between the BW1 and the BL3500 MK2:

When activating the BL3500 MK2 via the BW1 with an ATAPE command I get this

Followed by this

Not that I am able to make sense of this command. But I have insted tried to compare it with the datasignal on pin 6 in a powerlink cable from a Beolink Active conencted to the BL3500 MK2

I can see that there are some differences in the structure of the first two parts of the signal, but also the part from the second picture from the BW1 data signal is not at all in the signal from the Beolink active, so clearly a different datasignal from the BW1 compared to a Beolink active

 

When presing standby with the BW1 connected to the BL3500MK2 I do not register any data at all. Probably the Bl3500 turn off because the powerlink on signal on pin 4 disappereas.

 

 

 

 

 

[B3OHACK3R]

Nice and now we know what’s the difference between the BW1 and a normal PL output.

I see following in your BW1 capture (second screenshot equals the previous message):
0011 0001 1110 0111 1111 0000 0000 1001
0011 0011 0100 1100 1011 0000 0000 0101 0000 0000 0001

which equals those hex values:
31 E7 F0 09
33 4C B0 05 00 1

 

The Active kit sends the following:
0011 1011 1100 0000 0100 1000 0110 0100 0001 1000 1
0011 1011 1101 0001 0100 0000 0000 0000 0000 0000 1

which equals those hex values:
77 80 90 C8 31
77 A2 80 00 01

 

At least that “33 4C” data block I think I have seen before on AAL/MCL communication.

[B3OHACK3R]

Ah, forget the last 1 bit on each message. It’s the stop bit (should be a little longer than a normal 1 bit)

[Madskp]

B3OHACK3R wrote:

Nice and now we know what’s the difference between the BW1 and a normal PL output. I see following in your BW1 capture (second screenshot equals the previous message): 0011 0001 1110 0111 1111 0000 0000 1001 0011 0011 0100 1100 1011 0000 0000 0101 0000 0000 0001 which equals those hex values: 31 E7 F0 09 33 4C B0 05 00 1 The Active kit sends the following: 0011 1011 1100 0000 0100 1000 0110 0100 0001 1000 1 0011 1011 1101 0001 0100 0000 0000 0000 0000 0000 1 which equals those hex values: 77 80 90 C8 31 77 A2 80 00 01 At least that “33 4C” data block I think I have seen before on AAL/MCL communication.

Would it be helpfull if I export the data to some binary files and upload them in a post?

[B3OHACK3R]

Madskp wrote:

Would it be helpfull if I export the data to some binary files and upload them in a post?

I have my doubts that the saleae software has a built-in decoder for such pulse/space length encoded protocols. It is not really commonly used.

[Madskp]

B3OHACK3R wrote:

Madskp wrote:

Would it be helpfull if I export the data to some binary files and upload them in a post?

I have my doubts that the saleae software has a built-in decoder for such pulse/space length encoded protocols. It is not really commonly used.

Ok, good to know that it might not be the way to go for interpret datalink commands

[Madskp]

Although I have a hard time figuring out how you interpetered all that data from my screenshots 🙂

[B3OHACK3R]

Madskp wrote:

Although I have a hard time figuring out how you interpetered all that data from my screenshots ?

If you know how the data is transmitted you can easily find out just by looking at it.
Data is determined by the spaces (signal at 5V) between pulses (signal at 0V).

There are a few different space timings possible:
3.125 ms = logic 0
6.250 ms = “repeat” – same bit value as before
9.375 ms = logic 1
12.500 ms = start bit
15.625 ms = stop bit

The three pulses in the beginning can be ignored. Then there is the long start bit and afterwards you can start interpreting the data.

[Madskp]

Earlier on i this thread i did some testing where I connected the Beolab 3500 MK1 MCL connector on some of the products I had then.

I managed to do some control of a Beocord 3500 connected to a MCL2AV by controlling it from the Beolab 3500 IR eye.

However using the powerlink connection on my Beocenter 6 to connect to the Beolab 3500 I was not able to do anything

Since then I have got some more B&O stuff and have now tried to connect the Beolab 3500 MK1 to the powerlink connections on Both a Beomaster 4500 and a Beocenter 9300.

The cable for this connection is wired like this:

Powerlink pin 2 – > BL3500 pin 3, 5 and 7

Powerlink pin 3 – > BL3500 pin 1

Powerlink pin 5 – > BL3500 pin 4

Powerlink pin 6 – > BL3500 pin 6

In both cases I was able to control the sources on the Beomaster and the Beocenter, and the Beolab 3500 display was showing the source that was playing. When i blocked the IR eyes on the Beomaster and the Beocenter they would start up in a muted state like when the are controlled from a link room.

This setup more or less corosponds to the setup where a Beolink passive / ML-MCL converter is used to convert powerlink to MCL.

This explains why this works. In this case however the amplifier is not nessecary as the BL3500 is an active speaker.

 

[Author]

Posts