Marantz CD63 & CD67 mods list

Jump to Latest
#6,381
Malefoda said:
May I ask you all a favor?
Make a kind of sheet about your system: picture and spec, level of price for it in a decent shop.
Maybe I can figure better what do want from your CDP.
Click to expand...

My spec is in my signature space, but to confirm:

CD63, heavy mods, 4562 at moment
Rod Elliott 88 pre-amp, alps Blue pot, op-amps to suit mood
LM3886 diy power amp, dual mono, some ok parts, ~30wpc
Tempest subwoofer, 270litres ported, -3db about 20hz modelled
Seas 2.5 way DIY speakers, mostly expensive parts
Wiring is Kimber (4PR, 4VS, Crystal Cu) and nice DIY interconnect
Mains: iso. tx on cd, some filtering, diy extension mods and nice leads & plugs
Supports: DIY flexi-rack with mdf shelves, plywood platforms for pre and CD
Room: 5 diffusers of varying complexity, and care taken elsewhere

Sounds better in some ways than £5k+ of shop system, but totally better than £2-3k normal shop system in almost every way. My opinions of course 😉

Simon
 

Attachments

  • wipsmall.jpg
    wipsmall.jpg
    85.3 KB · Views: 388
#6,383
SimontY said:


That's a really cool list Ray, but I can't see any obvious correlation between perceived sound quality and specs there! At least not at a glance. Sorry this is OT but... my preamp sounds nice with LM6172 followed by OPA2134. Two gain stages, although one might purely be a buffer at present. To my ears 2134 sounds a touch flat and boring and 6172 sounds eager at the top and bottom and these two chips balance themselves out well. It's very easy to hear some of the more difficult to decipher lryics, such as some Dire Straits ones that are often not so obvious. I think mixing op-amps can be a way to get the most out of them. At least I'll banish them from my cd63 soon 😉

edit: OP249 has, for the list, mediocre specs. And it sounds mediocre to me too. A winner for the objectivists there! And 2604 sounds poor and doesn't look good on paper either.
Click to expand...

Quagmire here used to get good results on the old Rotel pre with OPA2132 (later OPA627x2) followed by LM6172. The FET input and BJT output sounded better to my ears. Did you try that way around? (OPA2132 is similar enough to OPA2134.)

Also, I found any op-amp sucks and swallows as a buffer. 100% NFB. That's why the HDAM sounds crap, not because it's a bad circuit. (Sounds good if you use it as part of the filter by the way.) If you swapped it with an OPA627 as unity-gain buffer (or 'voltage follower', pick a name) that would probably sound just as bad.

The BUF634 has the same pin-out (where it matters) as a single op-amp. It's an OPEN-LOOP buffer as opposed to a closed-loop buffer like HDAM or G=1 op-amp. I put SMD ones on a brown-dog and made it into a dual buffer that slots into an op-amp socket. Sounded very good too.

If your gain is high enough you can use OPA637 too... (edit: as your gain stage)

BUF634 is better than G=1 op-amp, but if the op-amp is G=1 maybe you don't need it at all. If you feel you need a buffer, the BUF634 can go inside the feedback loop of the first op-amp and it's distortions cancel.
 
#6,384
Glenn2 said:
Quagmire here used to get good results on the old Rotel pre with OPA2132 (later OPA627x2) followed by LM6172. The FET input and BJT output sounded better to my ears. Did you try that way around? (OPA2132 is similar enough to OPA2134.)
Click to expand...

No but I will try it now! Thanks.

I will also think about making sure no op-amps have a gain of 1. Eventually I'll break free from this pcb and have an op-amp with gain, followed by another used as a filter for active crossover experiments, or rather 2 or 3.

Lovin the avatar 😉

Simon
 
#6,385
SimontY said:
No but I will try it now! Thanks.
Click to expand...

Hmm, sounded ok apart from swishing sound from the right channel, and when I turned it off I got horrid noises in the speakers! I think I ragged the pins a bit too much pulling it out though, or maybe it doesn't like that socket.

edit: nope, other one doesn't like it either. It's ok with the 10k pot on my power amp turned down a bit but otherwise it sucks in the right speaker cones and makes swishy noises. Probaby bad.

edit#2: 4562 also has the same problem following 2134! Swapped back it's happy. Strange, it seems what comes before affects how happy the chip is as WELL as what comes after!

edit#3: If I crank the gain up using the DIP switches I can use 2134 followed by 6172. The second one must have been working with gain = 1 and the 6172 probably isn't very happy at unity gain.
 
#6,387
SimontY said:


Hmm, sounded ok apart from swishing sound from the right channel, and when I turned it off I got horrid noises in the speakers! I think I ragged the pins a bit too much pulling it out though, or maybe it doesn't like that socket.

edit: nope, other one doesn't like it either. It's ok with the 10k pot on my power amp turned down a bit but otherwise it sucks in the right speaker cones and makes swishy noises. Probaby bad.

edit#2: 4562 also has the same problem following 2134! Swapped back it's happy. Strange, it seems what comes before affects how happy the chip is as WELL as what comes after!

edit#3: If I crank the gain up using the DIP switches I can use 2134 followed by 6172. The second one must have been working with gain = 1 and the 6172 probably isn't very happy at unity gain.
Click to expand...

Hmmm- the LM6172 is fussy. Sucking in a speaker cone suggests DC offset! If the pre is DC coupled on output and power amp DC-coupled on input tread very carefully.

The 6172 is BJT input with much larger bias currents - about a million times more. I can't remember the numbers but I think it's uA vs pA for a 2134. This can generate a large offset if resistance is not equal on inverting and non-inverting inputs.

If you have two op-amps, the gain should really be spread between the two of them.

Another thing in DC-coupled circuits is the position of the volume pot. If varying this varies the resistance seen by an op-amp input, the offset will swing as you change volume. Can be avoided by using a FET op-amp.

My old pre was DC-coupled throughout (power amp had cap on input though!) and I had FET input op-amp (627) after the volume pot, and BJT op-amp (6172) driving the output (with bias currents equalised) and had no problems. The only DC-blocking cap between the DAC in the CDP and the speakers was a film cap on the input to the power amp.

Sounds like your circuit is perhaps more prone to instability so maybe stick with what works. I don't want to be responsible for launching your woofers across the room!

Anyway a bit OT so I'll shut up now. Off for a bit of afternoon sun. 😎
 
#6,390
I dunno what's up guys, but thanks for sharing your knowledge. The LM6172 is happy when I crank the gain up. Only trouble is the gain's now too high so sound quality suffers. The op-amp combo sounds good despite the hardness etc. caused by excessive gain. Definately really enjoying 4562s in cd63 in combination with 2134 and 6172 in my preamp. I'll sort out a point to point wired job soon and consider the gain much more carefully. That's for another thread of course, so once again, sorry for being OT. Off to Bonnie Scotland for a week tomorrow so I'll not be able to post anymore off-topic drivel for a while! I'll be back with renewed enthusiasm for hi-fi though and a thirst for CD63 mods 😉

Simon
 
#6,399
Hello guys. Finally a post after reading about 640 pages (took me weeks). I've already modded my CD57 MK-II a little bit. All Black gate power supply (4 x 1000uF 25V bipolars), opamp supply (4 x 330uF 16V), DAC supply (4 x 33uF 16V bipolars), lots of extra capacitors (SMD & TH), LM7815/LM7915 +/-15V opamp supply, new 7805CT regulators placed, big copper outputs connected via coax, output caps removed, mute, digital out & headphone circuit components removed, new VAM 12.2 tray placed, extra weight on the CD holding mechanism, 230V input filter, grounded mains, and more.

I cant tell how it will eventually sound because I'm still waiting for my scirocco's to "burn in"and my high end gain clone has a new black gate PSU & HIQ NX in the feedback circuit. Until now it sounds wonderfull though. Especially with my new cables (170 euro's of them) and Alps blue volume control.

Can't wait for the LM4562's to arrive. I'll build a 5PPM clock when I finally have time.

About the PSU, everyone was talking about a few hundred pages ago. Is anyone still thinking about ordering the custom made torroids?
 
You must log in or register to reply here.
Top Bottom