Do not know what is going on. But there seems to be a lack of upper frequencies in right channel, as if they were there but not loud enought.
Thought that the recording was made this way...or/and room acoustics might be messing with my ears.
Did the rca swap and it seems "the lack" travels with it.
Checked all the components around the opamps (4xopa627ap), regulators(dexa), rca connector, dc offset is 25.3mv on left 24.4mv on right, all parts from regulators to rca connector resoldered to remove possible dry joints.......heck i don`t know what is going on.
Dac decoupling checked and all the parts resoldered again....same result......must be missing something.
One more important thing about this cd player....."the lack" was present before i started modding it, first reason i even started doing them.
Thought that the recording was made this way...or/and room acoustics might be messing with my ears.
Did the rca swap and it seems "the lack" travels with it.
Checked all the components around the opamps (4xopa627ap), regulators(dexa), rca connector, dc offset is 25.3mv on left 24.4mv on right, all parts from regulators to rca connector resoldered to remove possible dry joints.......heck i don`t know what is going on.
Dac decoupling checked and all the parts resoldered again....same result......must be missing something.
One more important thing about this cd player....."the lack" was present before i started modding it, first reason i even started doing them.
No.
In the CD63/67 architecture , adding a bit of RC separation to the servos helps - but only because the basic internal PSU and, especially, ground-plane architecure is a mess.
Otherwise - don't bother with analogue-style super-regulation for the purely-digital elements - it is a waste of time and resources. If it made any repeatable difference then something is seriously broken! (think about it...)
Digital circuitry requires really-good local decoupling (= SMD caps of appropriate type & value) directly across Vss / Vdd pins. And don't go adding big radial-leaded electolytics freehand to 'digital' elements in the hope that it helps; it makes for more noise elsewhere, in a way you have no control of.
tl;dr: keep your powder dry. Spend it where it matters.
In the CD63/67 architecture , adding a bit of RC separation to the servos helps - but only because the basic internal PSU and, especially, ground-plane architecure is a mess.
Otherwise - don't bother with analogue-style super-regulation for the purely-digital elements - it is a waste of time and resources. If it made any repeatable difference then something is seriously broken! (think about it...)
Digital circuitry requires really-good local decoupling (= SMD caps of appropriate type & value) directly across Vss / Vdd pins. And don't go adding big radial-leaded electolytics freehand to 'digital' elements in the hope that it helps; it makes for more noise elsewhere, in a way you have no control of.
tl;dr: keep your powder dry. Spend it where it matters.
Last edited:
Thanks Martin; a very 'no nonsense' explanation!
I'm using a separate TX and psu for the servo so that it has enough current and doesn't pollute any other circuits. I'll put the stock 7805 back in there and use the super reg for decoder analogue I think.
For decoupling I've been using Nichicon solid polymer types; do you think they're suitable?
I'm using a separate TX and psu for the servo so that it has enough current and doesn't pollute any other circuits. I'll put the stock 7805 back in there and use the super reg for decoder analogue I think.
For decoupling I've been using Nichicon solid polymer types; do you think they're suitable?
Since I'd always planned to use my audio transformers, I hadn't bothered any of the opamp circuit upgrades. Lately I decided to try out a few opamps though for a sound comparison with the txs. I currently have a pair of metal can lm4562ha's which sounded ok...until tonight when I upgraded alk the opamp components to 0.1%..wow what a change! The sound stage and instrument placement is so much more accurate! I have some lme49990 on the way...very keen to compare!
Had a question about using chokes and ferrites; just how much of the noise from say a standard LT317 can passive components like chokes and pps caps remove? When compared to a variety of super reg for example..
Had a question about using chokes and ferrites; just how much of the noise from say a standard LT317 can passive components like chokes and pps caps remove? When compared to a variety of super reg for example..
I think I found what had come wrong. I connected the 47R to the RCA output and then ground it by the NC contact of the relay which I think is wrong (in this case the RCA still has output signal and the 47R only ground a bit of the signal). What I should do is to put a 47R in series between the 4.7uF isolating cap and the RCA output and the NC contact should connect to the RCA output after the 47R and to ground. When the NC contact is closed then the output signal is completely short to ground. I will modify the circuit when I have the time.
Greetings, on resistance 47R how many Watts should be applied
I have now measured the DOS voltages another one of my SA-8260 players and have posted this below. In this DOS I used matched 2SK170's that Ray sent me. I was astounded to see the voltage matches at the gain stage resistors. The other channel in this DOS was not quite as good, but close.
Attachments
Nice
You wouldn't be kind enough to upload here the ExpressSCH schematics file?
Perhaps ExpressSCH/PCB is a bit easier to learn than Eagle.
Nice
You wouldn't be kind enough to upload here the ExpressSCH schematics file?
Perhaps ExpressSCH/PCB is a bit easier to learn than Eagle.
You can just download it yourself from the expresspcb.com website.
It a simple easy to use program for simple schematics. It's designed to link to their proprietary board software, which I used to use when I first started making boards.
I was just hoping for a free ride ...
I couldn't attach my schematic. But it's rather easy to make one up as it's a rather simple circuit.
Just removed the choke and X caps from the ac inlet to my cd67 and it made a significant improvement. Decided to do this after reading one of Martins articles and I have to agree with his findings.
A separate mod I did was to add a separate low noise regulator for the dac clock vref. Another improvement in detail.
I've ordered a bunch of 1761 regs to separate out the decoder supplies further. Still waiting for my dual lme49990 opamps to arrive from China too.
A separate mod I did was to add a separate low noise regulator for the dac clock vref. Another improvement in detail.
I've ordered a bunch of 1761 regs to separate out the decoder supplies further. Still waiting for my dual lme49990 opamps to arrive from China too.
as per Ray's suggestion, I changed the output caps from the blackgates to the Mundorf evo silvergoldoil. After about 100 hours of breaking in, the change is quite remarkable. Notably the soundstage has expanded and deepend and become more tangible. It was very hard to fit them given the position of the DOS board but more than worth the effort. I dont know about the rest of you, how far you have gone with this player, but mine really sings. Who could have imagined such a lowly Marantz could be taken to such hights? It has taken..four years I think, and mostly thanks to Ray: his knowledge and patience. I had the opportunity to compare mine to a stock Marantz: same model. The standard one is absolutely unlistenable compared to mine. I am going to turn my attention now to changing the caps in the speaker filter and that should just about end my quest for a magical audio experience. (famous last words!)
