Hi Lee,
After writing the above post I started looking in the service manuals for the 7001 and 8260 to find where the clock selector IC's were located. I then got out a spare board I have and tried to find the IC's, and did.
All I can say is that this mod is WAY above my level.
I can successfully replace the X101 oscillator with leads to a clock, but that's about it.
To do your mod, I'd have to closely look at what Brent did on his SA11.
Steve
Hi Steve.
It's not too hard to do I think. Just a case of removing the first clock buffer IC, and feeding the relevant clocks into the input of the second IC.
The SA11 is a different animal and uses a custom FPGA (I think) to control the clock lines. This IC is mounted on the mainboard and is not fitted to the Sony servo board under the mech.
The trickiest part I can see will be physically soldering coax cables to such a small contact area. Plenty of hotglue and sticky pad required methinks
Lee.
It's not too hard to do I think. Just a case of removing the first clock buffer IC, and feeding the relevant clocks into the input of the second IC.
The SA11 is a different animal and uses a custom FPGA (I think) to control the clock lines. This IC is mounted on the mainboard and is not fitted to the Sony servo board under the mech.
The trickiest part I can see will be physically soldering coax cables to such a small contact area. Plenty of hotglue and sticky pad required methinks
Lee.
Hi Lee,
I had a look at the schematic this week, and the frequencies going to the DAC are 16.9 and 8.47MHz, not 33.8 and 16.9. The master clock is divided by two twice, and then selected by IC735. So in theory, you need a third clock to cover all frequencies. Luckily the 8.47MHz isn't used if you tie the filter switch to GND permanently, and only 16.9MHz is fed to the DAC. That means you can get rid of IC735 as well.
However... I don't think adding a second clock will work. The decoder and DVD processor also use the 33.8MHz master clock, and data transfer to the DAC must be synchronous with this master clock. If you add a second 16.9MHz clock instead of using the divided master, this will probably lead to trouble. Both clocks will run out-of-sync. One thing you could do is use one C2 and use the divided output instead of IC734.
Ray
I had a look at the schematic this week, and the frequencies going to the DAC are 16.9 and 8.47MHz, not 33.8 and 16.9. The master clock is divided by two twice, and then selected by IC735. So in theory, you need a third clock to cover all frequencies
. Luckily the 8.47MHz isn't used if you tie the filter switch to GND permanently, and only 16.9MHz is fed to the DAC. That means you can get rid of IC735 as well.However... I don't think adding a second clock will work. The decoder and DVD processor also use the 33.8MHz master clock, and data transfer to the DAC must be synchronous with this master clock. If you add a second 16.9MHz clock instead of using the divided master, this will probably lead to trouble. Both clocks will run out-of-sync. One thing you could do is use one C2 and use the divided output instead of IC734.
Ray
Last edited:
aha that is a good point. And to be honest, I was going from memory about the dac's clock frequencies (never a great idea I know).
I'll settle for improving those IC's and power supply arrangements for now then.
In the sa11 Marantz have used that FPGA chip to get around this problem somehow, I need to look at that in more detail then i think.
Lee.
I'll settle for improving those IC's and power supply arrangements for now then.
In the sa11 Marantz have used that FPGA chip to get around this problem somehow, I need to look at that in more detail then i think.
Lee.
Yes, this stuff is getting more and more complex, with all these large chips that each need their own working frequency. The SA11 has an external clock input, so that could be a starting point. Anyway, there's always some master XO or crystal that can be replaced .
I'm going for removal of IC735
.I'm going for removal of IC735
Just thought I'd post an update on the latest mods I've just made.
I replaced the c2 with a c3. This is also supplied via an additional 5v super-reg. I got very big improvements in openness, detail and much better dynamics. Bass is now more powerful, treble is cleaner and there's a lot more air and space around instruments.
I'm surprised by just how big the improvements are to be honest.
Lee.
I replaced the c2 with a c3. This is also supplied via an additional 5v super-reg. I got very big improvements in openness, detail and much better dynamics. Bass is now more powerful, treble is cleaner and there's a lot more air and space around instruments.
I'm surprised by just how big the improvements are to be honest.
Lee.
Hi Lee,
Nice work! I'm hoping to get some work done soon, I already have part of the OS-CONs here. Waiting for the rest to arrive and then I can take the board out. I ordered some LT1963A regulators to replace the BA033 stuff that's on the board.
Did some tinkering with the clock schematic yesterday. It's not so easy to get rid of the extra divider and buffer IC. There are some inversions in the clock signal that have to be corrected. For example, Marantz used the inverted Q output (NOT-Q) of the 7474 instead of the 'normal' Q. This is because the signal is inverted again later on when it is switched by a NAND-gate. So it appears I will have to re-route some traces. Just have to see if that's possible in practise.
Ray
Nice work! I'm hoping to get some work done soon, I already have part of the OS-CONs here
. Waiting for the rest to arrive and then I can take the board out. I ordered some LT1963A regulators to replace the BA033 stuff that's on the board.Did some tinkering with the clock schematic yesterday. It's not so easy to get rid of the extra divider and buffer IC. There are some inversions in the clock signal that have to be corrected. For example, Marantz used the inverted Q output (NOT-Q) of the 7474 instead of the 'normal' Q. This is because the signal is inverted again later on when it is switched by a NAND-gate. So it appears I will have to re-route some traces. Just have to see if that's possible in practise.
Ray
Hi Lee,
The on-board logic of the 7001 is what I want to get rid of. The only frequency that's needed is 16.9MHz (if you glue the filter switch to GND), so the second flipflop-divider can go, plus some NAND logic behind it that switches between the two freq.
I'd rather go the other way and use an external clock with divider, or trim down the on-board logic and use that. Less logic = less jitter.
Regards,
Ray
The on-board logic of the 7001 is what I want to get rid of. The only frequency that's needed is 16.9MHz (if you glue the filter switch to GND), so the second flipflop-divider can go, plus some NAND logic behind it that switches between the two freq.
I'd rather go the other way and use an external clock with divider, or trim down the on-board logic and use that. Less logic = less jitter.
Regards,
Ray
Did a bit more this evening. Added a 12v and 8v reg. I also noticed with the std transformer (non KI) that the servo rail per reg was dropping nearly 1v under load. Clearly the winding is barely up to the job. I also noticed that my 5v regs were running a bit low. This was caused as I need about 5v drop out with these regs and the supply was about 8v which is a little low. I decided to isolate the original mains transformer from the board for pins 1 to 5 which are the ac8 and ac11 windings. I replaced the ac8 with a 2x9v 15va which is about 1.6A which happily is supplying all the 5v regs, 3.3v and vref. I was more concerned with the main servo 8v and 12v supply which seemed woefully inadequate. The ac11 rails are now powered from a 12v 30va transformer 2.5A.
The results at this point are staggering. I've not been this excited about a player in ages. Hopefully the new clock and a few more caps will arrive over the next few days so I can further release the potential.
The results at this point are staggering. I've not been this excited about a player in ages. Hopefully the new clock and a few more caps will arrive over the next few days so I can further release the potential.
Nice work there Ian.
I've investigated a bit further into what I've done re. transformers.
I've added an extra tx for the display and front panel, isolating the original psu for the servo 5v only.
I've also added a 15va 2x9v for the dac to isolate the dac supply from the servo psu too.
I've been planning to investigate the 8v (I think) servo winding for sagging as I think mine drops a little, although it does stabilise quickly.
Cheers, Lee.
I've investigated a bit further into what I've done re. transformers.
I've added an extra tx for the display and front panel, isolating the original psu for the servo 5v only.
I've also added a 15va 2x9v for the dac to isolate the dac supply from the servo psu too.
I've been planning to investigate the 8v (I think) servo winding for sagging as I think mine drops a little, although it does stabilise quickly.
Cheers, Lee.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.