Having still not got round to actually doing any mods, I've got some more dumb questions. (Though I have actually now got a single sided layout for a regulator done and are waiting for a friend to make me a board)
1. When people talk about the DC blocking caps, I'm assuming that these are C655-658 (on a CD63)?
2. As I was looking at building a clock based on Ray's variation on the Kwak 7 (but adding my own reg as a power spupply), I was looking to source a spare crystal to be able to get it built up and tested on the bench before trying to take out the current crystal. Looking at Farnell, they have a number of 16.9344 MHz crystals and I notice that most list a frequency stability of +/- 30ppm, but they do have one (SMD one) that is +/-10ppm. I'm guessing that this will be a better spec than the one in the player and so theoretically I should get a better result?
3. They don't list an 8.4672 MHz crystal for the servo, and I seem to be struggling to find one online, so figure I have a couple of options. First is to use the crystal from the player, and second is to build another 16.9344 MHz clock, but add a divide by 2 stage on the output (using something like a type flipflop). I've discounted using a single clock for both stages with a divider as I've read on this thread that it doesn't work too well with a CD63. Any comments on either option or alternatives?
4. In the future I may look to go a bit further with the clocks, possibly looking a something like the flea or possibly Brent's micro clock (will 1 clock do both?). I've looked aroud and not actually found many clock designs to actually build yourself.
1. When people talk about the DC blocking caps, I'm assuming that these are C655-658 (on a CD63)?
2. As I was looking at building a clock based on Ray's variation on the Kwak 7 (but adding my own reg as a power spupply), I was looking to source a spare crystal to be able to get it built up and tested on the bench before trying to take out the current crystal. Looking at Farnell, they have a number of 16.9344 MHz crystals and I notice that most list a frequency stability of +/- 30ppm, but they do have one (SMD one) that is +/-10ppm. I'm guessing that this will be a better spec than the one in the player and so theoretically I should get a better result?
3. They don't list an 8.4672 MHz crystal for the servo, and I seem to be struggling to find one online, so figure I have a couple of options. First is to use the crystal from the player, and second is to build another 16.9344 MHz clock, but add a divide by 2 stage on the output (using something like a type flipflop). I've discounted using a single clock for both stages with a divider as I've read on this thread that it doesn't work too well with a CD63. Any comments on either option or alternatives?
4. In the future I may look to go a bit further with the clocks, possibly looking a something like the flea or possibly Brent's micro clock (will 1 clock do both?). I've looked aroud and not actually found many clock designs to actually build yourself.
I use one oscillator for both and it's sounding really quite stunning. Will try another at a later date, but not in a rush to do so.
Lee.
Yep I agree. I have run 2 seperate clocks previously but i'll not be doing it in this new player for a while. I will however be clocking the sys clock to the dac and decoder, the /2 to the servo and the /8 to the bit clock. I found direct clocking of the dac, filter and decoders a big step up over single master clocks in my CD960.
My player is working again!! Whoehoew!! 
Turned out to be the reg at the uP, at rf01. Didn't see that the ground pin was not connected to the ground, but to something else. And c902 had a bad joint and something else that looked suspicious. Now it sounds better than ever!! Sounds like a different player!
Tnks guys for helping me out! I really appreciate it!!
Did some extra mods to celebrate it ;-) Also tried Brents clock mod http://www.diyaudio.com/forums/digital-source/54009-marantz-cd63-cd67-mods-list-1203.html#post1792552, but didn't work on my player ("No disk"). How's that possible?
Turned out to be the reg at the uP, at rf01. Didn't see that the ground pin was not connected to the ground, but to something else. And c902 had a bad joint and something else that looked suspicious. Now it sounds better than ever!! Sounds like a different player!Tnks guys for helping me out! I really appreciate it!!
Did some extra mods to celebrate it ;-) Also tried Brents clock mod http://www.diyaudio.com/forums/digital-source/54009-marantz-cd63-cd67-mods-list-1203.html#post1792552, but didn't work on my player ("No disk"). How's that possible?
My player is working again!! Whoehoew!!
Tnks guys for helping me out! I really appreciate it!!
Did some extra mods to celebrate it ;-) Also tried Brents clock mod http://www.diyaudio.com/forums/digital-source/54009-marantz-cd63-cd67-mods-list-1203.html#post1792552, but didn't work on my player ("No disk"). How's that possible?
You need to establish if you have the MKII
Check the MKII service manual. You need to connect in front of the 10pF ceramic cap near the dac (dac side of it) to make it work as there a few components in circuit designed to make the decoder think it have an xtal directly connected. The change in circuit was probably to increase stability in the MKII model. Grab the SM supplement from rays pages. it highlights the different circuit on the MKII. It wont work it you clock the clock in pin direct on the decoder on the MKII
Well done for getting it running again too
edit: although the 10pF cap is shown close to the decoder in the SM, its actually close to the dac in the player
Attachments
Last edited:
To get this right in my head...
You need a 16.9344MHz clock with a divide 2 to get 8.4762MHz, then a further divide 4 to get 2.1168 MHz.
You connect:
1. The 16MHz signal to pin 28 of QD01 (XTI on the DAC) in place of the original crystal.
2. The 2MHz signal to pin 9 of QD01 (BCKI on the DAC) in place of the signal coming from pin 21 on Q503 (SCLK on the decoder)
3. The 8MHz signal to pin 19 on Q104 (XTLI on the servo)
Do you also need to connect the 16MHz signal to the pin 13 on Q503 (CRIN on the decoder)? And do you do anything with word clock signal (goes from pin 20 on Q503 to pin8 on QD01)?
Bartt as per my previous post, you need all the components in circuit in the MKII
In the MKII, RD14 is a 100r in the space of u193 and C521 is a 10pF in the space near the the DAC marked "RD14 or C521" You need both components between the clock and decoder.
However, its possible that if you lift the decoder pin 14 (CROUT) , you may be able to connect as per the MKI with the clock direct to pin 13. I've not tried this to prove it though.
In the MKII, RD14 is a 100r in the space of u193 and C521 is a 10pF in the space near the the DAC marked "RD14 or C521" You need both components between the clock and decoder.
However, its possible that if you lift the decoder pin 14 (CROUT) , you may be able to connect as per the MKI with the clock direct to pin 13. I've not tried this to prove it though.
Yes, exactly right.
EDIT- just to confuse things, you can use a completely seperate clock for the servo but the 16Mhz to the DAC, Decoder and the 2Mhz to the BCKI MUST be from the same source. Its also possible that any logic for dividing the clock signal will need to ensure that the clock remains in phase. You may also need to consider latency through divide circuits
If you want to clock the decoder direct, you must do as I said in the previous post and include the components in the MKII manual if you have a MKII.
I've not tried this mod on a MKI board but in theory it will work without the additional components.
My theory is this.......
It looks like the MKII was modified to present a load across the CRIN and OUT that looked like an xtal was directly connected. This most probably to make the circuit more stable.
Its possible that if all the components shown as changed in the MKII were removed you can clock direct to the CRIN pin but this is just a theory!!!
Last edited:
Wow! Ian you know your beans
Brent
MMM I detect sarcasm!!!!
I should point out that it was Brent that put me onto the MKII SM when I tried to clock the decoder direct and failed miserably!
Alan, you really should DL the SM from Rays site now
Lift the leg of C521 closest the DAC and attach the clock to the cap leg you now have floating above the top side!!!
yes but ive done the acoustica clock hack on pin 1 will that not affect c521MMM I detect sarcasm!!!!
I should point out that it was Brent that put me onto the MKII SM when I tried to clock the decoder direct and failed miserably!
Alan, you really should DL the SM from Rays site now
Lift the leg of C521 closest the DAC and attach the clock to the cap leg you now have floating above the top side!!!
alan