ARCHIVED FORUM -- March 2012 to February 2022READ ONLY FORUM
This is the second Archived Forum which was active between 1st March 2012 and 23rd February 2022
Hello BeoWorlders,
I am working on bringing modern automation and control to the existing BO products.
So far I managed to reverse-engineer NL, IR, Datalink and MCL protocols, I have fully integrated NL into HomeKit and will do the same with Datalnk, imagine controlling BeoGram from within the iPhone
my next target is MasterLink, I never had it installed nor used. But have acquired multiple devices over time. I was going through available information, it shouldn’t be that difficult, there is only one thing that bugs me. ML is based on RS485 (differential dual data wires D-/D+) and RS485 «architecture» is one master multiple(32) slaves. When master wants to talk to a particular node it is using some kind of addressing. I also saw in ML handbook list of errors and non-assigned address was one the the ML communication error. I never had ML installed and working hence is the question to you.
I. Can you address a particular slave as a link source?
2. Could you have multiple link slave sources?
3. When you plug ML products, could you turn them all at the same time or need go one by one with a reasonable delay.
That at would help me to understand what sort dynamic address allocation is used by master if any.
Thanks
At the lower level the Beolink protocol is quite complex and key to how it works is the "Lock Management Protocol". The lock management protocol is a vital part of BeoLink used to grant exclusive access to various services.
The protocol manages a key which must be possessed by any device wanting to execute a function that can alter the audio/video distribution, or is depending on it. Examples of functions requiring the key are selection of audio/video source, e.g., switch on a radio or change from radio to CD player; and control of source, e.g., change of track on a CD. The device possessing the key is called the lock manager by B&O.
Whilst the link uses differential data that is about as far as the comparison is when you compare it to RS485
To answer your exact questions:-
Regards Keith....
Thanks Keith. I have an overview of the BO lock management model. And without analysing telegrams my current assumption is that at the initialisation stage of a power master, every device is requesting a unique key which is then used to lock a command execution and prevent execution of a second command before completing the first one. Key question I have is how exactly the keys are generated and given that we can turn on all devices simaltuniously how devices manage to get a free time frame without data collision, I have few assumptions how I would have done it with random time frames and crc. But I better start collecting and analysing protocol telegram.
mczac:And without analysing telegrams my current assumption is that at the initialisation stage of a power master, every device is requesting a unique key which is then used to lock a command execution and prevent execution of a second command before completing the first one.
The lock management model is timed and when you power up all products on the Masterlink send a "Request Key" telegram, but as they don't get a response within a predefined time, so the slaves raise a "Key Lost" telegram. After a predefined time and receiving "Key Lost" telegram from each slave, the power master notifies the slaves that it has generated a key by broadcasting a "New Lock Manager" telegram.
The time between the first slave product sending a "Request Key" telegram and the power master broadcasting a "New Lock Manager" telegram is about 1.6 milli-seconds.
Keith, I have different timings from the whitepaper rather seconds than milliseconds, what is the source of information you are quoting?
mczac:what is the source of information you are quoting?
Measurements done years ago, I feel sure it was milli-seconds but I am happy to be corrected. It was a long time ago (20 years? ) when I studied the Masterlink in great detail.
If you have a protocol analiser you could check it..
ML data speed according to what I see is 19,200 bits/sec, so in 1.6 milliseconds only 30 full bits can be transferred, I think it is rather 1.6 seconds for initial handshake what you meant, but it is already a very good start!
mczac:ML data speed according to what I see is 19,200 bits/sec,
Yes, I think you are correct. 19,200 bits/sec does ring a bell as being correct for data speed. I wrote up some detailed notes many years ago, but currently I cannot find them and my memory is not what it was!!
From the MasterLink Handbook:
’The data speed is 19,200 bits/sec.
By way of comparison, the data speed for the Master Control Link is 160 bits/sec.’
MM
There is a tv - and there is a BV