i recently scored an oddball philips cdd462 external cdrom drive for almost nothing..this is based on the older philips 16 bit cd players[cdm4/tda1543]..but with a extra parts thrown into it's white box[serial port on the back,and a few "extra" chips]it appears to work perfectly and really doesn't sound bad playing cd's,much better than i expected actually.can anyone point me at a webpage listing the many mods that were commonly done to these old 16 bit machines?i remember reading about many,but naturally i can't seem to find them now that the info might be of use to me i'm just looking for some ideas and to find out what might apply to this thing
Rodzilla said:i recently scored an oddball philips cdd462 external cdrom drive for almost nothing..this is based on the older philips 16 bit cd players[cdm4/tda1543]..but with a extra parts thrown into it's white box[serial port on the back,and a few "extra" chips]it appears to work perfectly and really doesn't sound bad playing cd's,much better than i expected actually.can anyone point me at a webpage listing the many mods that were commonly done to these old 16 bit machines?i remember reading about many,but naturally i can't seem to find them now that the info might be of use to me i'm just looking for some ideas and to find out what might apply to this thing
Hi
I recently modded a Philips CD614, same architecture as you describe
I changed rectifier diodes to Schottky Barier, installed low jitter clock and directly clocked pin 1 of the TDA1543. I also removed the muting transistors and improved the layout and decoupliong around the various chips.
the result i now a player with resolution and transparancy, a pleasure to listen to as also the stress has disapeared
cheers
Fin said:Hi Guido
Did you go Non-Oversampling?
How did you apply the clock to pin 1 of TDA1543?
Hi Fin
No, I couldn't as a Mitsubishi chip (datashet unknown) was in front, generating oversampled I2S.
Yes, I interrupted the line to TDA1543 pin1 and injected clock directly - result is stunning
cheers
Bricolo said:Directly the 11Mhe clock?
Does this only work because the 8x oversampled data is also at 11Mhz (IOW, we can't do this on a non os player)?
Hi
Yes, directly
In case it differs from the crystal clock: You can measure that frequency, buy or build a clock at that frequency and derive the crystal frequency.
Bricolo said:With an IC counter?
Hi
Well, with dividers, often you need divide/2 or so, that can be done with a single Dflipflop
Depends on the system
Originally posted by Guido Tent
Hi Fin
No, I couldn't as a Mitsubishi chip (datashet unknown) was in front, generating oversampled I2S.
Oh yes! Of Course! I remembered the Mitsibushi chip after I had posted.
Originally posted by Guido Tent
Yes, I interrupted the line to TDA1543 pin1 and injected clock directly - result is stunning
Is this better than reclocking BCK?
(I am thinking about my Arcam)
Did you reclock any other lines?
Fin said:
Oh yes! Of Course! I remembered the Mitsibushi chip after I had posted.
Is this better than reclocking BCK?
(I am thinking about my Arcam)
Did you reclock any other lines?
Hi
No, did not yet reclock other lines
The puret clock you can make at bck is the best - so better than reclocking
cheers
Guido Tent said:
Hi
No, did not yet reclock other lines
The puret clock you can make at bck is the best - so better than reclocking
cheers
can there be no delay/synchro problems? Since the Bclk is from the clock itself, and all the other lines are "processed" and delayed.
Hi Guido
Do you plan to?
In the Arcam, BCK is currently reclocked with a 74HC74 and then an inverter. Would it be worthwhile/correct to change this to a direct connection from an 11mhz XO2 devided by 2?
This is what I am concerned about and would like to understand before altering the clocking/reclocking. I think that the use of the inverter after reclocking in the Arcam has something to do with this.
Originally posted by Guido Tent
No, did not yet reclock other lines
Do you plan to?
Originally posted by Guido Tent
The puret clock you can make at bck is the best - so better than reclocking
In the Arcam, BCK is currently reclocked with a 74HC74 and then an inverter. Would it be worthwhile/correct to change this to a direct connection from an 11mhz XO2 devided by 2?
Originally posted by Bricolo
can there be no delay/synchro problems? Since the Bclk is from the clock itself, and all the other lines are "processed" and delayed.
This is what I am concerned about and would like to understand before altering the clocking/reclocking. I think that the use of the inverter after reclocking in the Arcam has something to do with this.
Fin said:Hi Guido
Do you plan to?
In the Arcam, BCK is currently reclocked with a 74HC74 and then an inverter. Would it be worthwhile/correct to change this to a direct connection from an 11mhz XO2 devided by 2?
This is what I am concerned about and would like to understand before altering the clocking/reclocking. I think that the use of the inverter after reclocking in the Arcam has something to do with this.
Hi
Try it, and see if it works - nothing to lose
If not, the HC74 clock input should be the cleanest possible
nice work from Arcam ....
cheers
Hi Guido
There are a few things in the existing distribution that I would like to understand before deciding on what to implement. If the original clock is disabled and the SPDIF is disconnected - this means that 4 of the 6 inverters in the hex inverter will be disabled.
This leaves 2 inverters working - the one for BCK and the one feeding the clock to the filter. I don't fully understand their purpose.
Are they both necessary?
Can the one feeding the filter be avoided if the filter and decoder are clocked directly from Guido's XO2?
There is no real need altering the already good clock distribution in the Arcam. Just replace the clock for something better.
There are a few things in the existing distribution that I would like to understand before deciding on what to implement. If the original clock is disabled and the SPDIF is disconnected - this means that 4 of the 6 inverters in the hex inverter will be disabled.
This leaves 2 inverters working - the one for BCK and the one feeding the clock to the filter. I don't fully understand their purpose.
Are they both necessary?
Can the one feeding the filter be avoided if the filter and decoder are clocked directly from Guido's XO2?
TDA1543 and tubes and no op-amp
Reviving this old thread, I just got my Magnavox player back from storage. It uses a TDA1543. I removed the I/V op-amp and substituted a resistive I/V load instead. But not to ground, but to a source of 2.2V I connected a 1.33k resistor. It seems happy to play into this load, producing about 1V of audio waveform. Note: instead of using an actual 2.2V supply, I built a resistor voltage divider, 3K from +5V and 2.4k to ground, connected to one of the output pins of the DAC. That's the thevanian equivalent of 1.3k to 2.2V. And it can source as well as sink current, something a voltage regulator or transistor circuit won't do. This audio then feeds a cathode follower tube circuit to then drive the output jacks on the back of the CD player.
Reviving this old thread, I just got my Magnavox player back from storage. It uses a TDA1543. I removed the I/V op-amp and substituted a resistive I/V load instead. But not to ground, but to a source of 2.2V I connected a 1.33k resistor. It seems happy to play into this load, producing about 1V of audio waveform. Note: instead of using an actual 2.2V supply, I built a resistor voltage divider, 3K from +5V and 2.4k to ground, connected to one of the output pins of the DAC. That's the thevanian equivalent of 1.3k to 2.2V. And it can source as well as sink current, something a voltage regulator or transistor circuit won't do. This audio then feeds a cathode follower tube circuit to then drive the output jacks on the back of the CD player.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Source & Line
- Digital Source
- old philips cdp mods