Marantz CD5400 clock replacement

[4real]

Sounds like real fun

Let me tell you want I might want to do to my CD-5400:

- replace opamps with AD8620
- remove muting, and maybee replace it with a relay
- remove or replace out caps.
- replace dac with a CS4398
- do some power supply tweaks
- replace crystal stuff

I guess, I'll first replace the opamps. But I'll take it slowly

[EC8010]

Looks like I may be able to postpone the screaming. Here's what I found on J013 (which definitely goes to pin 6 of IC14, the manual and real-life agree).

[EC8010]

I intend to start by improving the clock - that's a serious weakness. I might replace the muting with relays, and might well replace the coupling capacitors. I have a pair of AD8620, so they might go in. Frankly, the change that will make the biggest difference is the clock. Everything else is just polishing.

[4real]

Hmm, 16.89 is also not in the DAC datasheet. The closest is 16.9344 Mhz.... Could that be a match? That would also account for the 8.4672 Mhz crystal I guess (it's half of that.

But how to get on now? Rip out the that BU... thing, and replace it with a simple and good clock circuit?

[EC8010]

Yes, don't worry that the oscilloscope didn't measure frequency all that accurately; I'm sure it meant to say "16.9344MHz" but stumbled over the pronunciation. No need to rip out the BU... thing, just cut a track (once I've worked out where).

[4real]

Well, that's also fine But what about the connection from the BU to the TC94A14FA? And then I mean pins 34,35 and 36 specificly?

[EC8010]

That's trickier. At first, I was worried that the micro was really the master clock, but if you look, it uses the 16.9344MHz clock that arrives on pin 46. But the circuit diagrams and data sheets don't make sense to me. According to the Rohm BU2630 data sheet, pins 7, 8, and 9 are determinedly digital inputs, yet the Marantz diagram shows them connected to analogue signals. Something is wrong here.

What we expect to find is some sort of a master oscillator, followed by a frequency synthesizer to allow that frequency to be varied. We expect that variable frequency to go to the CD motor servo to control the speed of data read off disc and we expect it to go to the DAC to allow the DAC to cope with data having a variable sample frequency. We would expect that when variable pitch isn't engaged that the multiplication ratio of the frequency synthesizer would be an integer (2).

[4real]

Originally posted by EC8010
in 46. But the circuit diagrams and data sheets don't make sense to me. According to the Rohm BU2630 data sheet, pins 7, 8, and 9 are determinedly digital inputs, yet the Marantz diagram shows them connected to analogue signals. Something is wrong here.

Exactly what I found! So I'm not going nuts?

I dug a bit further and found that the pins are actually refering to TMP87PS71AF (IC71). There pins 34,35 and 36 are actually labled SDA, SCL and PLL, and this is the uC that controls the whole thing, and they are connected to a general purpuse IO pin. So I guess those are used to setup pitch stuff.

I guess, you can just leave it away

Might be some obfuscation in the manual to make it harder to understand?

So I guess, you'll just have to insert a fesh clock into the point labled "DSP" and cut the rest loose?

[EC8010]

Originally posted by 4real
I guess, you can just leave it away

Might be some obfuscation in the manual to make it harder to understand?

The manual is certainly clear as mud.

There's only one way of proving it. Put a fixed 16.9344MHz clock in to feed the DAC and the controller and see if it works. Watch this space...

[4real]

Do you have one?