Philips 14 bit/CDM1 Trouble Reading CDRs
Hi Guys,
First off, I am aware that there is an old thread relating to the Philips CD104, but I can't seem to find any discussion on why 14 bit Philips systems have problems reading CDRs.
I really like the 14 bit Philips TDA1540 DAC based CD Players - I really like the sound! I have owned several CD104s and a CDX all of them have struggled to read CDRs - 52x maxells, forget about it! After much research I discovered Japanese Taiyo Yuden CDRs written at 4X are a better option. But, although they play, the audio seems to completely break up in the worst cases or tick and cut out slightly in the better cases - You can hear that the laser is struggling.
Recently I bought a CD104 cheap, but it had issues! I did all the griplets but it still point blank refused to read any CDRs, it also skipped on some commercially made pressed CDs.
I decided to try altering the turntable height which vastly improved things (I suppose the disc was too close to the laser and not able to focus), I then got a TOC on a CDR and I was able to play the first track or two before the laser protested and the disk squeeled to a halt. I then realised the laser current pot might need gentle tweeking which I carefully did. This gentle adjustment made the CD player work like all the others I've ever owned, but as I said, like the others, CDRs display jitter like ticks and in bad cases break up.
Someone out there must know what I'm talking about! No 14 bit system I've owned seems to be happy with reading CDRs, somebody please tell me why! No TDA1541 based system with 7220 has given me problems reading CDRs.
I recently read somewhere that a properly adjusted CD104/CDX should read CDRs as good as newer machines but in my experience that hasn't been the case.
I must confess that I have done all the adjustments by feel and noises of distress put out by the struggling laser. I have been an electronics engineer for 30 years but for Roland musical instruments so I am no expert in CD players. I do assume though, that the aim of the exercise is to get the sharpest eye pattern possible by doing all the adjustments mentioned in the unit's service manual and CDM manual (which I have).
I just wondered if anyone out there can give me a pointer in the right direction as to what could sort out this reading issue.
So far I have done all the standard servicing tweeks - Lubed the motor bearing, adjusted TT height, tweeked laser, hardwired all the griplets, resoldered all boards and checked all the capacitors for value.
I'm now thinking maybe this is not a service issue, it's more of an electronics one, maybe I should look at improving things in the laser preamp section, like lower noise ICs?
Any help from you experts on the forum would be appreciated, please someone tell me if I should expect good reproduction from a CDR and if not why not: The CDM1 is a great drive with glass optics, is the problem with the preamp or the way the 1540 chip set processes the info? I assume the eye patterns on newer machines look sharper on a scope.
Hi Guys,
First off, I am aware that there is an old thread relating to the Philips CD104, but I can't seem to find any discussion on why 14 bit Philips systems have problems reading CDRs.
I really like the 14 bit Philips TDA1540 DAC based CD Players - I really like the sound! I have owned several CD104s and a CDX all of them have struggled to read CDRs - 52x maxells, forget about it! After much research I discovered Japanese Taiyo Yuden CDRs written at 4X are a better option. But, although they play, the audio seems to completely break up in the worst cases or tick and cut out slightly in the better cases - You can hear that the laser is struggling.
Recently I bought a CD104 cheap, but it had issues! I did all the griplets but it still point blank refused to read any CDRs, it also skipped on some commercially made pressed CDs.
I decided to try altering the turntable height which vastly improved things (I suppose the disc was too close to the laser and not able to focus), I then got a TOC on a CDR and I was able to play the first track or two before the laser protested and the disk squeeled to a halt. I then realised the laser current pot might need gentle tweeking which I carefully did. This gentle adjustment made the CD player work like all the others I've ever owned, but as I said, like the others, CDRs display jitter like ticks and in bad cases break up.
Someone out there must know what I'm talking about! No 14 bit system I've owned seems to be happy with reading CDRs, somebody please tell me why! No TDA1541 based system with 7220 has given me problems reading CDRs.
I recently read somewhere that a properly adjusted CD104/CDX should read CDRs as good as newer machines but in my experience that hasn't been the case.
I must confess that I have done all the adjustments by feel and noises of distress put out by the struggling laser. I have been an electronics engineer for 30 years but for Roland musical instruments so I am no expert in CD players. I do assume though, that the aim of the exercise is to get the sharpest eye pattern possible by doing all the adjustments mentioned in the unit's service manual and CDM manual (which I have).
I just wondered if anyone out there can give me a pointer in the right direction as to what could sort out this reading issue.
So far I have done all the standard servicing tweeks - Lubed the motor bearing, adjusted TT height, tweeked laser, hardwired all the griplets, resoldered all boards and checked all the capacitors for value.
I'm now thinking maybe this is not a service issue, it's more of an electronics one, maybe I should look at improving things in the laser preamp section, like lower noise ICs?
Any help from you experts on the forum would be appreciated, please someone tell me if I should expect good reproduction from a CDR and if not why not: The CDM1 is a great drive with glass optics, is the problem with the preamp or the way the 1540 chip set processes the info? I assume the eye patterns on newer machines look sharper on a scope.
Last edited:
Firstly you 100% need a scope to see what is really happening to the RF.
Can you expect good reproduction from a CDR... yes, but not on these early players. Looking at the RF and you will see that the amplitude is almost certainly lower from CDR than a pressed commercial disc, and also the jitter and general quality of the recovered signal will be far worse, looking blurry and ill defined on the scope.
This thread (half way down post #1) shows some real scope shots (and the Sony won't play CDR either). The diamond shape in the signal is very well defined. Also look at the relative amplitudes of the overlapping sine waves.
Sony CDP790 and KSS240 Restoration Project
The attached image (from the web) is a good example of how the CDR signal would look. Not only is it noisy and ill defined but also the relative amplitudes of the components that make up the signal are changed. The lowest amplitude sines are dropping down below the level at which the data slicer in the RF preamp section can accommodate them.
So unfortunately the bottom line is that the servo system is simply not designed to cope with this quality of signal. Yes, some folk may report limited success by altering laser power and so on but its not a fix.
Can you expect good reproduction from a CDR... yes, but not on these early players. Looking at the RF and you will see that the amplitude is almost certainly lower from CDR than a pressed commercial disc, and also the jitter and general quality of the recovered signal will be far worse, looking blurry and ill defined on the scope.
This thread (half way down post #1) shows some real scope shots (and the Sony won't play CDR either). The diamond shape in the signal is very well defined. Also look at the relative amplitudes of the overlapping sine waves.
Sony CDP790 and KSS240 Restoration Project
The attached image (from the web) is a good example of how the CDR signal would look. Not only is it noisy and ill defined but also the relative amplitudes of the components that make up the signal are changed. The lowest amplitude sines are dropping down below the level at which the data slicer in the RF preamp section can accommodate them.
So unfortunately the bottom line is that the servo system is simply not designed to cope with this quality of signal. Yes, some folk may report limited success by altering laser power and so on but its not a fix.
Attachments
Thanks Mooly,
Your answer is what I feared, but in testing the six 14 bit machines I've owned, some are a lot better than others, and a Japanese made CDR burned slowly makes a big difference. Do you think it might be worth modifying the preamp board before it gets to the Servo board by boosting the signal via a further op amp by enough to be read more clearly? With regards to jitter, where would that be introduced? The CDM1 or the SAA7010? If it is a failing of the laser assembly I think it's probably not modable but if it's to do with the electronics, then can these not be improved and tweeked a bit? I read a lot of the CD104 thread and you along with PlasticIsGood were helping someone with a machine only reading CDRs well and not pressed CDs, that leads me to believe the electronics rather than the CDM need a modification.
Your answer is what I feared, but in testing the six 14 bit machines I've owned, some are a lot better than others, and a Japanese made CDR burned slowly makes a big difference. Do you think it might be worth modifying the preamp board before it gets to the Servo board by boosting the signal via a further op amp by enough to be read more clearly? With regards to jitter, where would that be introduced? The CDM1 or the SAA7010? If it is a failing of the laser assembly I think it's probably not modable but if it's to do with the electronics, then can these not be improved and tweeked a bit? I read a lot of the CD104 thread and you along with PlasticIsGood were helping someone with a machine only reading CDRs well and not pressed CDs, that leads me to believe the electronics rather than the CDM need a modification.
Last edited:
The main cause of the jitter and noisy signal seen from CDR is caused by the fundamental difference in operation of these discs vs a pressed commercial disc.
The commercial disc is pressed (stamped) with the 'pits' and 'lands' both remaining reflective but at different 'heights' from the disc surface. Each change from a land to a pit and pit to land generates a logic 1. Logic 0 is generated while reading both pits and lands. These depressions are printed with razor sharp accuracy.
The CDR is all reflective but covered with a burnable dye with pits and lands created by burning away the dye at the appropriate points. This is imprecise compared to the pressed disc and gives slightly blurred edges to the transitions.
This means the data on the CDR disc is not as precisely defined and so needs electronics designed from the outset to be able to cope with this noisy jittery signal.
All this means there is a lot of variability when it comes to players not designed to read CDR but that 'almost work', and there are no hard and fast rules on whether anything can be done, and equally what may work on one player may fail on another.
The CDR/RW uses a dye that is burned away to create
The commercial disc is pressed (stamped) with the 'pits' and 'lands' both remaining reflective but at different 'heights' from the disc surface. Each change from a land to a pit and pit to land generates a logic 1. Logic 0 is generated while reading both pits and lands. These depressions are printed with razor sharp accuracy.
The CDR is all reflective but covered with a burnable dye with pits and lands created by burning away the dye at the appropriate points. This is imprecise compared to the pressed disc and gives slightly blurred edges to the transitions.
This means the data on the CDR disc is not as precisely defined and so needs electronics designed from the outset to be able to cope with this noisy jittery signal.
All this means there is a lot of variability when it comes to players not designed to read CDR but that 'almost work', and there are no hard and fast rules on whether anything can be done, and equally what may work on one player may fail on another.
The CDR/RW uses a dye that is burned away to create
Attachments
Hi Mooly,
Yes I do understand the technicalities of CDR and also that the more reflective the CDR surface the better, but what in electronics terms processes this noisy signal better on say a Rotel RCD855? Is it the preamp used after the CDM or the first chip to recieve the eye pattern like the Demodulator chip - Is it the way the early SAA7010 processes this noisy eye pattern recieved at pin 7 compared to the demodulator (SAA7210?) on an 855?
I presume that the laser output is no better or worse on a CDM1 compared to a CDM4.
Yes I do understand the technicalities of CDR and also that the more reflective the CDR surface the better, but what in electronics terms processes this noisy signal better on say a Rotel RCD855? Is it the preamp used after the CDM or the first chip to recieve the eye pattern like the Demodulator chip - Is it the way the early SAA7010 processes this noisy eye pattern recieved at pin 7 compared to the demodulator (SAA7210?) on an 855?
I presume that the laser output is no better or worse on a CDM1 compared to a CDM4.
Have you tried (although I suspect these are not now available) the 74 minute CDR which officially was the standard maximum size ?
I've no quick answers on what you could try. The reasons the discs don't play is simply that they do not meet Red Book standard (for whatever reason... there are several), and these early players were designed before the Orange Book recordable standards came into being.
You have possible amplitude issues with the recovered 3T to 11T signal components. To examine this, you need to look at the RF on a scope using a player that can handle CDR. This will highlight the problems (I have done this in the past). This will also show which burn speeds give the best results for a given blank CDR. I found higher speeds were significantly better than low (4x vs 16x) which is the opposite of what you often read.
The servo system may also be optimised better for CDR in some players... just guessing on that... but it seems likely.
If you are to consider tweaking a player then you have to at least see if you can see any problem areas with the RF. If it were purely a reflectivity issue then turning the laser power up (not recommended generally) would prove the point but there is more to it than that. My feeling is that when you examine the RF you will see that the relative levels of the highest amplitude sines vs the lowest are at a ratio that the window of the data slicer can not accept. Altering the level doesn't alter the ratio difference.
Its an interesting project but you will have to get really technical with the player and try and gather evidence as to why its not coping with CDR.
And remember at the end of the day, this is a known problem with the early players... you are trying to fiddle it to make it do something outside its original specification.
I've no quick answers on what you could try. The reasons the discs don't play is simply that they do not meet Red Book standard (for whatever reason... there are several), and these early players were designed before the Orange Book recordable standards came into being.
You have possible amplitude issues with the recovered 3T to 11T signal components. To examine this, you need to look at the RF on a scope using a player that can handle CDR. This will highlight the problems (I have done this in the past). This will also show which burn speeds give the best results for a given blank CDR. I found higher speeds were significantly better than low (4x vs 16x) which is the opposite of what you often read.
The servo system may also be optimised better for CDR in some players... just guessing on that... but it seems likely.
If you are to consider tweaking a player then you have to at least see if you can see any problem areas with the RF. If it were purely a reflectivity issue then turning the laser power up (not recommended generally) would prove the point but there is more to it than that. My feeling is that when you examine the RF you will see that the relative levels of the highest amplitude sines vs the lowest are at a ratio that the window of the data slicer can not accept. Altering the level doesn't alter the ratio difference.
Its an interesting project but you will have to get really technical with the player and try and gather evidence as to why its not coping with CDR.
And remember at the end of the day, this is a known problem with the early players... you are trying to fiddle it to make it do something outside its original specification.
Hi Mooly,
Many thanks for your time and input, I currently have three CD104Bs and a CDX, the CDX is much worse than the 104Bs but as I mentioned the last 104B I bought would hardly play a pressed CD never mind a CDR, some adjustment has made all the difference. I will look at the Eye pattern on my scope and try a TY disk at 16X. If the HF level is really low, I don't want to turn up the laser current too much as It would be better to increase the gain at the Pre Amp - What about altering the feedback loop on an op amp to increase gain slightly? Obviously if the slicer is having problems discerning the 1s and 0s, increasing the gain will not help - But swapping from Maxell to TY CDRs made a huge difference and I feel I am not a long way off from acceptable CDR audio. These CD104s really do sound great compared to anything else I've tried, I recently sold one modified to NOS to a guy who had a Musical Fidelity before, and he took one listen and said WOW! I thought the Rotel RCD855 converted to NOS were the bees knees until I compared one to the CD104B. I am determined to try and sort this problem out (if it can be sorted).
Many thanks for your time and input, I currently have three CD104Bs and a CDX, the CDX is much worse than the 104Bs but as I mentioned the last 104B I bought would hardly play a pressed CD never mind a CDR, some adjustment has made all the difference. I will look at the Eye pattern on my scope and try a TY disk at 16X. If the HF level is really low, I don't want to turn up the laser current too much as It would be better to increase the gain at the Pre Amp - What about altering the feedback loop on an op amp to increase gain slightly? Obviously if the slicer is having problems discerning the 1s and 0s, increasing the gain will not help - But swapping from Maxell to TY CDRs made a huge difference and I feel I am not a long way off from acceptable CDR audio. These CD104s really do sound great compared to anything else I've tried, I recently sold one modified to NOS to a guy who had a Musical Fidelity before, and he took one listen and said WOW! I thought the Rotel RCD855 converted to NOS were the bees knees until I compared one to the CD104B. I am determined to try and sort this problem out (if it can be sorted).
It is only the analog part (up to the output of the RF signal affected by the servo) that determines readability, nothing to do with the DAC part. The CDM-1 is not designed to play CDRs. Nevertheless, I have some (over 25 years old) CDRs that play perfectly, but most do not. Perhaps it is because of the reflectivity or the difference between dark and light "dots" which is much less than on aluminized CDs.
Thank you for your input! Mooly actually had a good solution! Record on a good media like Taiyo Yuden at 16X instead of 4X or 1X as I was doing following poor advice on other forums. I have now got a CD-104B with virtually no jitter or crackling from a CDR! No need to look for mods, as long as all the adjustments are done to the CDM1, all the Griplets have been hardwired and the capacitors are within spec all should be OK.
I have recently found that something other than the CDM1 setup is contributing to the poorer reading of CDRs. Out of three CDR-104s I have only one of them could read the CDR burnt at 4x without artifacts in the sound and interuptions (all three read the CDR burnt at 16x without issue). I got my scope out and perfectly adjusted the CDM1s on all three units but it made no difference to the two that had reading problems. I swapped the Servo board of the better 104 with the poorer one and the fault didn't move, nor did it with the power supply. The last board I swapped was the Decoder board and this surprisingly made the difference. I am not sure which component is causing CDRs to not read well but I'm going to try renewing the electrolytics and I'll post the results once that's done. If anyone knows what could be the culprit for these read errors please let me know.
I wonder if the error correction could be different... hard to say without doing an in depth investigation. Or servo time constants maybe.
As we keep saying though, these players are not designed for CDR from the outset and so anything is a bonus. I'm not sure setting up using a CDR is a good idea in the long run because you are optimising the player for an imperfect medium, possibly at the expense of playing Red Book standard discs.
As we keep saying though, these players are not designed for CDR from the outset and so anything is a bonus. I'm not sure setting up using a CDR is a good idea in the long run because you are optimising the player for an imperfect medium, possibly at the expense of playing Red Book standard discs.
I've plenty of players with CDM1 that have had no 'tuning' and they will all read CD-Rs, even poor ones. Most read errors come from poor media and poor recording. Choosing decent quality (e.g. Taiyo Yuden) discs and recording on a decent drive (even at high speed) will produce a disc that reads without CU (uncorrectable) errors and negligable C2 (correctable) errors. Try a CD-R/CD-RW drive rather than a combo and use decent recording software. (There's a complete art to this if you can be bothered!) There's a list of the better audio recording drives on the web somewhere - Plextors generally work well. Certain drives will allow access to low level error data so you can use Plextools or similar to scan CDs and CD-Rs and see exactly what's going on.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.