Hello,
I am new to this forum but I know this forum from lurking. I know that CD transports are a contentious issue sometimes and some of you are stripping them out of boomboxes. I also don't believe in using PC CD-ROM drives, I've been down that road myself before, and with SATA becoming more common, it gets harder to do.
I want to design a CD transport for the applications you folks would like to use it for, offering a completely open and easily reproducible platform using components that I will do my best to ensure are available for years.
The Toshiba TC94A29 looks like a good bet, the circuitry around it is nice and simple, and teaming it with an external micro, we can do stuff like CD-TEXT and all that on a nice VFD graphics display.
At the moment I am trying to secure information from Toshiba about this part, they are proving to be a little stubborn. The part has a micro on board, which I will simply program to provide a simple SPI interface so you folks can go wild and do whatever you like
But the devil is I will need to get the full details of the processor inside the chip as well as the assembler/compiler for it. The software is the easy part, getting the stuff out of Toshiba appears to be much harder.
If I can get the devtools and technical data from Toshiba, would you be interested in a totally opensource CD transport? One that can be built from parts and not be hard to get hold of in some countries? Of course many people are not up to working with fine pitch SMT components so I can provide pre-assembled PC boards.
Let me know what you think please
I am new to this forum but I know this forum from lurking. I know that CD transports are a contentious issue sometimes and some of you are stripping them out of boomboxes. I also don't believe in using PC CD-ROM drives, I've been down that road myself before, and with SATA becoming more common, it gets harder to do.
I want to design a CD transport for the applications you folks would like to use it for, offering a completely open and easily reproducible platform using components that I will do my best to ensure are available for years.
The Toshiba TC94A29 looks like a good bet, the circuitry around it is nice and simple, and teaming it with an external micro, we can do stuff like CD-TEXT and all that on a nice VFD graphics display.
At the moment I am trying to secure information from Toshiba about this part, they are proving to be a little stubborn. The part has a micro on board, which I will simply program to provide a simple SPI interface so you folks can go wild and do whatever you like
But the devil is I will need to get the full details of the processor inside the chip as well as the assembler/compiler for it. The software is the easy part, getting the stuff out of Toshiba appears to be much harder.If I can get the devtools and technical data from Toshiba, would you be interested in a totally opensource CD transport? One that can be built from parts and not be hard to get hold of in some countries? Of course many people are not up to working with fine pitch SMT components so I can provide pre-assembled PC boards.
Let me know what you think please
I'm always curious, built a variant of the Shigaclone a couple of years ago not using the standard board set most people used. This might make a good future project when I've finished my current crop - hopefully there is some interest here.
What mechanism are you considering as the basis for this design?
What mechanism are you considering as the basis for this design?
Transport? Well that all depends on what is available and cheap.
Currently I can get the boombox transports i.e. the cd clip design. I can get them in volume with what appears to be a clone Sony KSS 2xxx series pick-up. I can get other transports too, the thing is I will of course design it to fit a multitude of different transports. Tray mechanisms are quite simple, but will require an extra driver to control the tray motor, which is not a problem.
The processor I intend to use, I currently have the local distributor putting pressure on Toshiba to service my needs, so I will have an answer soon.
Unlike CD player kits, this one will be fully customizable, I am looking at even including XMP1.0 for the remote control, which means you can purchase a UEI remote for it from a dealer.
The design will consist of four ICs in total, the processor can either be fitted on the board, or those of you who wish to drive it with a PICMicro or whatever, can do so. I will leave that up to the individual's imagination.
It will consist basically of one IC to do the signal processing from the pick-up, the TC94A29 doing all the work, a BTL bridge driver IC, and then a general-purpose Freescale micro to control the show. The TC94A29 has a micro in it, but it is hopelessly too thin to do anything serious, as it was intended for cheap and nasty CD players with two-digit displays. To process the CD-TEXT info requires a smidgin more processing power.
Currently I can get the boombox transports i.e. the cd clip design. I can get them in volume with what appears to be a clone Sony KSS 2xxx series pick-up. I can get other transports too, the thing is I will of course design it to fit a multitude of different transports. Tray mechanisms are quite simple, but will require an extra driver to control the tray motor, which is not a problem.
The processor I intend to use, I currently have the local distributor putting pressure on Toshiba to service my needs, so I will have an answer soon.
Unlike CD player kits, this one will be fully customizable, I am looking at even including XMP1.0 for the remote control, which means you can purchase a UEI remote for it from a dealer.
The design will consist of four ICs in total, the processor can either be fitted on the board, or those of you who wish to drive it with a PICMicro or whatever, can do so. I will leave that up to the individual's imagination.
It will consist basically of one IC to do the signal processing from the pick-up, the TC94A29 doing all the work, a BTL bridge driver IC, and then a general-purpose Freescale micro to control the show. The TC94A29 has a micro in it, but it is hopelessly too thin to do anything serious, as it was intended for cheap and nasty CD players with two-digit displays. To process the CD-TEXT info requires a smidgin more processing power.
This project will probably hit a snag because a CD player uses licensed technology covered by patents. Consequently you're likely to be hit by non disclosure agreements and what have you... so you'll be liable to legal action.
Why not join my quest in reverse engineering the control mechanism - when you will be patent free! I think it's reasonably plausible though I've no experience with driving HE motors.
Why not join my quest in reverse engineering the control mechanism - when you will be patent free! I think it's reasonably plausible though I've no experience with driving HE motors.
Teac cd-5020a
Hi,
Please have a look at the TEAC drive: (CD-5020A). I think this can be really helpful.
Best regards,
Bins.
Hi,
Please have a look at the TEAC drive: (CD-5020A). I think this can be really helpful.
Best regards,
Bins.
Progress Update 9-12-2010
Due to having a child, and being enrolled in music school I have not done anything electronic related in over a year now. However now that said child is of manageable age, and I've completed my studies, I have revisited this topic.
After being rebuffed by Toshiba re the TC94A29 and with the knowledge that some of the CD patents have expired, I struck gold with a SONY chipset and some info from dark nets (read: p2p networks)
Having the all-important datasheet, and an application circuit, I am now pleased to announce I can design a player from scratch. The DSP does not include a microcontroller, which is a huge problem as far as the TC94A29 is concerned.
There are a couple of issues that arise:
1) The SONY DSP I intend to use, is expensive, it is roughly 24 euros. However you get what you pay for, it can do CD-TEXT and can be set into CD-ROM mode so I envisage that, teamed with a STA013 from ST Microelectronics, we can play back MP3 files in the same player. I will likely design the PC board for this option, but the chip can be omitted and a jumper setting informs the software the STA013 ain't there.
2) I intend to use a commercial microcontroller and a graphics LCD, the programming thereof could be an issue, as I use the semiconductor vendor's toolchains and debuggers which could cost you a fortune especially if used just to build a one-off. I use these things to earn a living so they're my standard toolbox.
3) The parts are not generally available, however here in South Africa, I can get them easily and plentifully from appliance spares outlets using my business account I have with them (I did repairs for many years to earn extra dosh in school) However, like I said previously, they're not cheap, and at over ZAR260 for just the CD-DSP, I'd be very miffed if I had to slip with a scope probe and short the thing out.
4) This thing will be extensively surface mount- a huge problem for many without access to hot air pencils/reflow ovens. I am sorry but through-hole components are getting harder and harder to get and I loathe using them when I can make ultra-compact assemblies using modest 0805 SMD components.
5) An additional component, a CPLD will be needed for glue logic with the STA013, and this will too, need to be programmed with a JEDEC file, meaning more development tools. For the sake of simplicity, this can be omitted, as it will substantially reduce costs.
I am going to try over Dec holidays to lay out a prototype PC board. I will be designing for a standard transport using a KSS-212 series optical pick-up, which are dime a dozen all over the place, and because I have a lot of them in my inventory.
The big trick to this I can see already, is going to be software development, as I will need to figure out how to sequence the DSP correctly to make it play a disc. On the face of it that seems to be the least of my worries, nonetheless, it seems reading the TOC is going to be the difficult one.
Until I am sure about security of supply I will not reveal which Sony DSP I intend using. All I can tell you is, that it is a relatively new device, and my first choice over hard-to-get and fading-into-obsolescence CXD11xx series DSPs
Due to having a child, and being enrolled in music school I have not done anything electronic related in over a year now. However now that said child is of manageable age, and I've completed my studies, I have revisited this topic.
After being rebuffed by Toshiba re the TC94A29 and with the knowledge that some of the CD patents have expired, I struck gold with a SONY chipset and some info from dark nets (read: p2p networks)
Having the all-important datasheet, and an application circuit, I am now pleased to announce I can design a player from scratch. The DSP does not include a microcontroller, which is a huge problem as far as the TC94A29 is concerned.
There are a couple of issues that arise:
1) The SONY DSP I intend to use, is expensive, it is roughly 24 euros. However you get what you pay for, it can do CD-TEXT and can be set into CD-ROM mode so I envisage that, teamed with a STA013 from ST Microelectronics, we can play back MP3 files in the same player. I will likely design the PC board for this option, but the chip can be omitted and a jumper setting informs the software the STA013 ain't there.
2) I intend to use a commercial microcontroller and a graphics LCD, the programming thereof could be an issue, as I use the semiconductor vendor's toolchains and debuggers which could cost you a fortune especially if used just to build a one-off. I use these things to earn a living so they're my standard toolbox.
3) The parts are not generally available, however here in South Africa, I can get them easily and plentifully from appliance spares outlets using my business account I have with them (I did repairs for many years to earn extra dosh in school) However, like I said previously, they're not cheap, and at over ZAR260 for just the CD-DSP, I'd be very miffed if I had to slip with a scope probe and short the thing out.
4) This thing will be extensively surface mount- a huge problem for many without access to hot air pencils/reflow ovens. I am sorry but through-hole components are getting harder and harder to get and I loathe using them when I can make ultra-compact assemblies using modest 0805 SMD components.
5) An additional component, a CPLD will be needed for glue logic with the STA013, and this will too, need to be programmed with a JEDEC file, meaning more development tools. For the sake of simplicity, this can be omitted, as it will substantially reduce costs.
I am going to try over Dec holidays to lay out a prototype PC board. I will be designing for a standard transport using a KSS-212 series optical pick-up, which are dime a dozen all over the place, and because I have a lot of them in my inventory.
The big trick to this I can see already, is going to be software development, as I will need to figure out how to sequence the DSP correctly to make it play a disc. On the face of it that seems to be the least of my worries, nonetheless, it seems reading the TOC is going to be the difficult one.
Until I am sure about security of supply I will not reveal which Sony DSP I intend using. All I can tell you is, that it is a relatively new device, and my first choice over hard-to-get and fading-into-obsolescence CXD11xx series DSPs
I've been relooking at this project. The Sony chips are expensive. Nonetheless I stumbled upon a chip used in a cheap boombox I found in a dime store. It plays MP3 files as well as regular audio. I didn't buy the thing until I was sure what was in there... (been burned twice already, the radios I've bought to hack, and then found I couldn't are piling up here - need to hock them at some stage)
The chip used is the ALi M5671. Extensive searching in Google got me the document I needed to see how the device looks inside at a glance. I found that it uses the common 8051 core (for which I have development tools and then some and have been using all my life since the age of 17) and, get this... the FLASH holding the executable code, sits outside.. so BOING!!! Reprogrammability and hacking is now possible without the pain in the @rse of programming pods. I am willing to bet the serial port has a bootloader listening on it a-la DALLAS DS89C430 in which case I shall modify my trusty serial bootloader I wrote in Delphi some years ago.
I am off to the store now to go and purchase a unit to take apart. I have written a nice letter to ALi asking them about the part and for technical info. I will sign whatever NDA they present, as I do not have to disclose the intenral info to create a saleable product. In any case, a lot of the stuff in that chip is hidden from the programmer, just a bunch of registers just like the Sony chips, which are public domain. With a working program, I can eliminate 100% of the guesswork (stuff like PID coefficients for the servos) and get this done... FAST!
Highlights of the device:
- Internal D/A converter, with external I2S interface (put on your own expensive DAC)
- Internal Servos - just need cheap 1/2 bridge driver IC (BA5496, etc...)
- Internal RF (laser pick-up literally connects directly to the chip, no interfacing)
- Internal DSP, which allows for volume to be adjusted, as well as WMA/MP3 playback
- I/O ports for LCD and push buttons
- Supports CD, CD-R, CD-RW. Can also process CD-TEXT and read CD-ROM
So, ladies and gentlemen, I believe I've found a solution, eventually, after some 18 months.
As for patents... First of all, the CD patents are busy expiring, Second of all I am just using the chip in a bespoke circuit, not violating anyone's patent. Third of all, royalties would likely be due if I used the COMPACT DISC logo anywhere. The solution to that here in South Africa (which is seeing wide adoption) is to drop the logo altogether and use the letters CD in a highly stylized way. I know how patent litigation works, considering I am a patent holder myself...
The chip used is the ALi M5671. Extensive searching in Google got me the document I needed to see how the device looks inside at a glance. I found that it uses the common 8051 core (for which I have development tools and then some and have been using all my life since the age of 17) and, get this... the FLASH holding the executable code, sits outside.. so BOING!!! Reprogrammability and hacking is now possible without the pain in the @rse of programming pods. I am willing to bet the serial port has a bootloader listening on it a-la DALLAS DS89C430 in which case I shall modify my trusty serial bootloader I wrote in Delphi some years ago.
I am off to the store now to go and purchase a unit to take apart. I have written a nice letter to ALi asking them about the part and for technical info. I will sign whatever NDA they present, as I do not have to disclose the intenral info to create a saleable product. In any case, a lot of the stuff in that chip is hidden from the programmer, just a bunch of registers just like the Sony chips, which are public domain. With a working program, I can eliminate 100% of the guesswork (stuff like PID coefficients for the servos) and get this done... FAST!
Highlights of the device:
- Internal D/A converter, with external I2S interface (put on your own expensive DAC)
- Internal Servos - just need cheap 1/2 bridge driver IC (BA5496, etc...)
- Internal RF (laser pick-up literally connects directly to the chip, no interfacing)
- Internal DSP, which allows for volume to be adjusted, as well as WMA/MP3 playback
- I/O ports for LCD and push buttons
- Supports CD, CD-R, CD-RW. Can also process CD-TEXT and read CD-ROM
So, ladies and gentlemen, I believe I've found a solution, eventually, after some 18 months.
As for patents... First of all, the CD patents are busy expiring, Second of all I am just using the chip in a bespoke circuit, not violating anyone's patent. Third of all, royalties would likely be due if I used the COMPACT DISC logo anywhere. The solution to that here in South Africa (which is seeing wide adoption) is to drop the logo altogether and use the letters CD in a highly stylized way. I know how patent litigation works, considering I am a patent holder myself...
Hi
I just received the SDK (software development kit) for this chip. It came complete with sample code to implement CD playback and a C compiler. The MP3 side of things wasn't included, which I am sure is not important here. MP3 isn't hi-fi anyway.
The device, upon closer inspection, has a DSP doing the MP3 and WMA decoding, in hardware by the looks of things, and is controlled via Programmed I/O and a register file.
I am waiting for the NDA, in order to sign it, and return it to ALi. When that is done I'll be receiving the balance of the documentation from them. I purchased the boombox containing one of these chips, on Saturday afternoon. The serial port has been exposed and located, according to docs I have, it listens for serial commands, just like the Dallas DS98C430. So we have a winner for sure.
The first step I will do this week, is to dump the original firmware so I have a reference point to work from.
For all development, I will use the internal DAC. I can output I2S to an external DAC or S/PDIF if I need to.
What I need to know is, what kind of jitter performance do you guys expect out of this? In digital circuitry, jitter is traceable to the clock sources. The boombox uses a cheap ceramic resonator which is +/-5% either side of fundamental clock frequency, which is pretty bad for hi-fi applications. Of course, using a good quality quartz crystal would be essential. I can have those made to my spec, and in sample qty cheaply and in 3 working days.
I still plan to employ a Noritake VFD graphics display in this project. It is expensive, but, other options do include graphics LCD displays. It all depends on what, you, the clients, want.
I just received the SDK (software development kit) for this chip. It came complete with sample code to implement CD playback and a C compiler. The MP3 side of things wasn't included, which I am sure is not important here. MP3 isn't hi-fi anyway.
The device, upon closer inspection, has a DSP doing the MP3 and WMA decoding, in hardware by the looks of things, and is controlled via Programmed I/O and a register file.
I am waiting for the NDA, in order to sign it, and return it to ALi. When that is done I'll be receiving the balance of the documentation from them. I purchased the boombox containing one of these chips, on Saturday afternoon. The serial port has been exposed and located, according to docs I have, it listens for serial commands, just like the Dallas DS98C430. So we have a winner for sure.
The first step I will do this week, is to dump the original firmware so I have a reference point to work from.
For all development, I will use the internal DAC. I can output I2S to an external DAC or S/PDIF if I need to.
What I need to know is, what kind of jitter performance do you guys expect out of this? In digital circuitry, jitter is traceable to the clock sources. The boombox uses a cheap ceramic resonator which is +/-5% either side of fundamental clock frequency, which is pretty bad for hi-fi applications. Of course, using a good quality quartz crystal would be essential. I can have those made to my spec, and in sample qty cheaply and in 3 working days.
I still plan to employ a Noritake VFD graphics display in this project. It is expensive, but, other options do include graphics LCD displays. It all depends on what, you, the clients, want.
Yes indeed, although I did plan to implement a UI directly from the M5671's CPU and I/O ports.
I've managed to find stock of the IC I plan to use, locally.. Cost is sub USD 7 each. I've just bought all the local stock of about 10 pcs. More is on the way...
Datasheet.... Check
Boombox that uses the device.... Check
Sample Software Project (found out yesterday its for Keil uVision).... Check
Keil uVision C Compiler... Check
In-circuit programmer software MTKFlasher... Check (was able to dump the FLASH... busy pushing that through a disassembler as we speak)
M5671 User's Manual... In Progress (having difficulty dealing with the Taiwanese... suspect language barrier is a problem. Will rewrite my request in Chinese and see if that helps). I need the user's manual to know what the registers inside the M5671 do.
This weekend, I hope to get the traditional "hello, world" test up and running, to confirm the C compiler works as the in-system programmer is fine.
I've managed to find stock of the IC I plan to use, locally.. Cost is sub USD 7 each. I've just bought all the local stock of about 10 pcs. More is on the way...
Datasheet.... Check
Boombox that uses the device.... Check
Sample Software Project (found out yesterday its for Keil uVision).... Check
Keil uVision C Compiler... Check
In-circuit programmer software MTKFlasher... Check (was able to dump the FLASH... busy pushing that through a disassembler as we speak)
M5671 User's Manual... In Progress (having difficulty dealing with the Taiwanese... suspect language barrier is a problem. Will rewrite my request in Chinese and see if that helps). I need the user's manual to know what the registers inside the M5671 do.
This weekend, I hope to get the traditional "hello, world" test up and running, to confirm the C compiler works as the in-system programmer is fine.
Further Updates (5/11/11)
Compiled "sample" software project intended to implement a car stereo, in Keil uVision...SUCCESS
Had a look at the software, it is rather complicated. While I can see all the function calls to do stuff like 'laser on', 'lens down', 'focus OK' and especially interesting stuff as how the CD-ROM is processed (using linear buffers and scanning for CDFS markers, pretty much the same way one would do it with ATAPI and a CD-ROM drive).
Also found the code to drive the LCD on the car radio, although in this case the LCD definitely does NOT match the one in the boombox.
Found various conditional compiles for various types of laser pick-up (Sony, Samsung, Sanyo)
Found code that supports reading USB memory stick via a CH3xxx USB HOST IC, but its disabled using conditional compiles.
Found code that does the CD-TEXT function... basically builds text into a buffer by extracting it from subcodes.
MP3 decoding is just like using the STA013, its a black box you feed with data coming from the CD DECODER block. It is used remarkably similar too... also has patch registers that need to be loaded with values before it will work. Since this is intended for HI-FI I am not going to use this feature. However, since this chip is much cheaper than the STA013, it can be pressed into service as a standalone MP3 decoder reading from Flash Drive.
Compiled "sample" software project intended to implement a car stereo, in Keil uVision...SUCCESS
Had a look at the software, it is rather complicated. While I can see all the function calls to do stuff like 'laser on', 'lens down', 'focus OK' and especially interesting stuff as how the CD-ROM is processed (using linear buffers and scanning for CDFS markers, pretty much the same way one would do it with ATAPI and a CD-ROM drive).
Also found the code to drive the LCD on the car radio, although in this case the LCD definitely does NOT match the one in the boombox.
Found various conditional compiles for various types of laser pick-up (Sony, Samsung, Sanyo)
Found code that supports reading USB memory stick via a CH3xxx USB HOST IC, but its disabled using conditional compiles.
Found code that does the CD-TEXT function... basically builds text into a buffer by extracting it from subcodes.
MP3 decoding is just like using the STA013, its a black box you feed with data coming from the CD DECODER block. It is used remarkably similar too... also has patch registers that need to be loaded with values before it will work. Since this is intended for HI-FI I am not going to use this feature. However, since this chip is much cheaper than the STA013, it can be pressed into service as a standalone MP3 decoder reading from Flash Drive.
Rick, please get in touch with me re the Shigaclone that was sent to you that you graciously offered to do some work on.
It was sent to you on 26th January and I've had NO COMMUNICATION from you since 22nd March, getting on for two months ago.
You were on this site last night so I KNOW you have read the many messages I've sent to you about it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.