Carawu, what's this about "old PC-programs for controlling cdmpro/cdm12 via i2s...."?
I don't know of them and they might lead to another avenue which could bring new ideas.
Guido, if you have a Linux kernel version 2.4 then I could send you a program to monitor the I2C bus. It uses two pins on the printer port to listen to bus activity.
The problem is the hardware to connect to the bus. The device used is a triple-channel digital isolator called an ADuM1300 manufactured by Analog Devices.
As far as I know it's an SMD only device so that's another headache, depending on your resources.
I used this "galvanic isolation" technique before turning to bit-bashing the PIC.
Why did I gave up with the Linux program? Because I was getting inconsistent results, possibly because of the coding of the controller, or more likely, because the program couldn't cope with the message format used in the application under investigation.
It would be good to see what that program makes of a "pure player", e.g. CD160/640/660.
I guess the easiest thing would be for me to invest in a cheap ebay machine and take a look myself. That would save you a lot of bother!
Anyway, I'm really pleased that my PCB has been identified in several CD players.
Regards,
ramchip
I don't know of them and they might lead to another avenue which could bring new ideas.
Guido, if you have a Linux kernel version 2.4 then I could send you a program to monitor the I2C bus. It uses two pins on the printer port to listen to bus activity.
The problem is the hardware to connect to the bus. The device used is a triple-channel digital isolator called an ADuM1300 manufactured by Analog Devices.
As far as I know it's an SMD only device so that's another headache, depending on your resources.
I used this "galvanic isolation" technique before turning to bit-bashing the PIC.
Why did I gave up with the Linux program? Because I was getting inconsistent results, possibly because of the coding of the controller, or more likely, because the program couldn't cope with the message format used in the application under investigation.
It would be good to see what that program makes of a "pure player", e.g. CD160/640/660.
I guess the easiest thing would be for me to invest in a cheap ebay machine and take a look myself. That would save you a lot of bother!
Anyway, I'm really pleased that my PCB has been identified in several CD players.
Regards,
ramchip
Thanks Carawu. I'll take a closer look at cdmpro when I get a chance.
Guido, it's good to know you are into PICs, I started with the 16F84 last year. Now I'm using the 16F74, a 40-pin beast! The coding is going well.
I talked myself into buying a cheap CD player on eBay. It arrived this morning. Unfortunately, it doesn't have the golden PCB that I wanted. It has one marked
CMKS-P3X. The player is a Philips, model number CD 163.
If you know where I can find a service manual, preferably a free online one, please let me know!
Regards,
ramchip
Guido, it's good to know you are into PICs, I started with the 16F84 last year. Now I'm using the 16F74, a 40-pin beast! The coding is going well.
I talked myself into buying a cheap CD player on eBay. It arrived this morning. Unfortunately, it doesn't have the golden PCB that I wanted. It has one marked
CMKS-P3X. The player is a Philips, model number CD 163.
If you know where I can find a service manual, preferably a free online one, please let me know!
Regards,
ramchip
ramchip said:
I'm using one in a preamp with support for i2c, rc5, kenwood xs, samsung vfd etc. rather use a small pic with i2c i/o exdenders than a beast
as for the player, seems to be a 1545 machine. inside:
cdm12.1 with
SAA7345GP-M5 decoder
TMP47C421AF display controller
MC68HC11ER20 controller
TDA1301TN1 servo
TDA1545 SMD dac
So here the 6811 talks to the tmp47.. for keyboard input and display time/track. I'm afraid that this is too new technology to be usefull with the other pcb. But what would you like to achieve with this player anyway, there are no keys to select a specific track (number)?
All new kit guido, very good!
What would be really useful for my purposes is a CMKS-81X circuit board (as shown in post #14 of this thread)
in a machine which has a track number selection feature. I wonder how many candidate machines fit the bill. The one that arrived today was advertised as a CD 162, but when it arrived it's clearly a CD 163.
It works well but it's not exactly what I need. Maybe I should try for one of these: Philips CD304MK2, Philips CD160, CD630, CD640 or CD660. Grundig 8100, Marantz CD273. Not sure which of these has track selection though!
Regards,
ramchip
What would be really useful for my purposes is a CMKS-81X circuit board (as shown in post #14 of this thread)
in a machine which has a track number selection feature. I wonder how many candidate machines fit the bill. The one that arrived today was advertised as a CD 162, but when it arrived it's clearly a CD 163.
It works well but it's not exactly what I need. Maybe I should try for one of these: Philips CD304MK2, Philips CD160, CD630, CD640 or CD660. Grundig 8100, Marantz CD273. Not sure which of these has track selection though!
Regards,
ramchip
guido said:CD650
Think non of the philips players you mention above. Hang on, i'll have a look in mine...
had a look, bit busy on those lines
No fun with a scope or logica analyser (old one for parallel busses).Maybe i could use your pic program to have a look. Would be interested in such a tool myself. Had a look on the net, but there's not much out there without extra hardware.
Except for one, but the zipfile is no longer available (i2csniff.zip)
Dont want to put too much time in it
I2C bus monitoring requires special attention to triggering the scope. If you're interested in doing this then check this website for a little circuit that detects the start condition for I2C messages. I used it to see what was happening on my bus.
http://www.i2cchip.com/monitoring_i2c_bus.html#Bus Monitoring
The guys at i2cchip can provide the Linux program i2cmon which is useful, but not optimised for old bus message formats, as I found out! Hence, the need to employ the PIC to decrypt what is happening bit by bit. My program is very much still under development and it's
being optimised for a relatively slow bus timing and a rather strange message format. I very much doubt that it would be of any use to you for looking at modern equipment. For that I would suggest using i2cmon as a starting point and take it from there.
Anyway, we are getting a little off-topic now so I better restate the point of this thread.
If anyone knows of a CD player which uses the CMKS-81X circuit board (as shown in post #14 of this thread) AND with the track selection feature, I would be very interested to know about it!
Regards,
ramchip
http://www.i2cchip.com/monitoring_i2c_bus.html#Bus Monitoring
The guys at i2cchip can provide the Linux program i2cmon which is useful, but not optimised for old bus message formats, as I found out! Hence, the need to employ the PIC to decrypt what is happening bit by bit. My program is very much still under development and it's
being optimised for a relatively slow bus timing and a rather strange message format. I very much doubt that it would be of any use to you for looking at modern equipment. For that I would suggest using i2cmon as a starting point and take it from there.
Anyway, we are getting a little off-topic now so I better restate the point of this thread.
If anyone knows of a CD player which uses the CMKS-81X circuit board (as shown in post #14 of this thread) AND with the track selection feature, I would be very interested to know about it!
Regards,
ramchip
As said before, the cd650 comes close. See the picture. Don't mind the pcb on the right, it's not standard
Neiter is the one in the back. The one on top of the main pcb is for the MAB controller. In newer versions it is replace with just a MAB, without pcb (this is the inside of mine before i replaced it. Current state of the machine is totally different anyway).
As for the i2c, the connectors to the front pcb are a bit different. There are two: i2c and powersuppy. RC5 is on one too. But you could still measure the i2c lines. Sorry for the bad quality picture.
The 650 has four
controllers on that bus. As said, it's busy on there. Think the display is updated continuously. Pressing keys would result in separate messages from display driver to display MAB controller, which then sends other messages to the one on the main PCB (i think).
I know i could decode the bus, but it would take me quite some time. Thought of triggering on the key being pressed, but since it's a two stage affair, this is also not that easy.
In the u.s. you should look for a magnavox cdb650. Actually, mine is a magnavox too.
If you ever find out the i2c command's, i'm interested.
Regards,
Guido
Neiter is the one in the back. The one on top of the main pcb is for the MAB controller. In newer versions it is replace with just a MAB, without pcb (this is the inside of mine before i replaced it. Current state of the machine is totally different anyway).As for the i2c, the connectors to the front pcb are a bit different. There are two: i2c and powersuppy. RC5 is on one too. But you could still measure the i2c lines. Sorry for the bad quality picture.
The 650 has four
controllers on that bus. As said, it's busy on there. Think the display is updated continuously. Pressing keys would result in separate messages from display driver to display MAB controller, which then sends other messages to the one on the main PCB (i think).I know i could decode the bus, but it would take me quite some time. Thought of triggering on the key being pressed, but since it's a two stage affair, this is also not that easy.
In the u.s. you should look for a magnavox cdb650. Actually, mine is a magnavox too.
If you ever find out the i2c command's, i'm interested.
Regards,
Guido
Attachments
Yes Andy, a complete machine would be needed to investigate further. Still need to know which model has a track selection feature as this is the message I'm trying to capture on the I2C bus.
Guido, thanks for the picture. I can see that the CD650 does come close. You've made extensive modifications by the look of it. Well done! Please tell me if track selection was a feature of the original machine. Btw, I live in England!
Regards,
ramchip
Guido, thanks for the picture. I can see that the CD650 does come close. You've made extensive modifications by the look of it. Well done! Please tell me if track selection was a feature of the original machine. Btw, I live in England!
Regards,
ramchip
ramchip said:
The inside looks different now. MAB pcb is gone, new pcb with pmd100 is added, including a clock. PCB on the right is gone, new pcb added with pooge output stage. The machine is used for experimenting.
As for track selection, i think you mean something like the picture, right? It's a hidden feature
And you actually don't need it, a player with a remote would do too: remote goes to display MAB, which decodes it and gives the commands to the decoder MAB on the main pcb.Regards,
Guido
Attachments
Okay, so it looks that I would benefit from seeing a Magnavox cdb650. The keypad connection to the decoder board is directly relevant to decrypting the bus messages.
However, somewhere there the documentation must exist that tell how the MAB chip communicates to other circuits. I don't mean the "datasheet" for the chip, that's not very informative. The designer's guide is really what I'm after, now that would save an awful lot of effort!
Regards,
ramchip
However, somewhere there the documentation must exist that tell how the MAB chip communicates to other circuits. I don't mean the "datasheet" for the chip, that's not very informative. The designer's guide is really what I'm after, now that would save an awful lot of effort!
Regards,
ramchip
ramchip said:
There is no connection from keypad to decoder board: there is a connection from keypad to keyboard/vfd driver (4-bit processor). This probably talks (informs about individual keypresses) to the MAB on the front pcb. This front MAB then informs the MAB on the main pcb what to do (also in case the front MAB received rc5 from a remote). At least, this is what i think on how it works. And all communication is over the same bus. Even worse, there is also an fts MAB on the same bus. And messages going the other way around to update the display.
So get a good i2c sniffer(read a modern logic analyser, usb based ones are out there for not too much cash).
The MAB datasheet is useless. It's an 8-bit controller which can be programmed with anything. The designers guide on how to setup the i2c/i2d communication is what you need, but i don't think you will find that outside philips (if it still exists, philips hasselt which made those machines is long gone).
One other method would be to get a machine with a motorola main pcb processor (few exist). There you can remove the prom and read the content (program is stored external and not inside the processor). Then you need to disassemble it to see what could come in. Possible if you have lots of time.
Guido
Don't know what an "fts MAB" is but the whole setup is too complicated because of the high number of controllers. The system I'm looking at has just two controllers and I've made good progress with the bit-bashing technique. The bus messages are quite clear now so I don't want to go down another route using costly equipment only to hit the same brick wall and wasting yet more time in the process.
If I had access to a straightforward CD player with that old Philips decoder board and some means of selecting a track I'm sure it would help me decrypt the messages completely.
I'm no stranger to disassembly of PROMs so your idea of looking at a motorola main pcb sounds interesting. However, unless the code is the same as in the MAB in the Philips decoder pcb then it might prove to be a bit of a dead-end.
To recap, I need either a simple player which uses the Philips decoder or possibly a motorola main pcb. More details on both aspects would be much appreciated.
Regards,
ramchip
If I had access to a straightforward CD player with that old Philips decoder board and some means of selecting a track I'm sure it would help me decrypt the messages completely.
I'm no stranger to disassembly of PROMs so your idea of looking at a motorola main pcb sounds interesting. However, unless the code is the same as in the MAB in the Philips decoder pcb then it might prove to be a bit of a dead-end.
To recap, I need either a simple player which uses the Philips decoder or possibly a motorola main pcb. More details on both aspects would be much appreciated.
Regards,
ramchip
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.