Rotel RCD-951 RestoMod

[John3655]

Rotel Rcd-951 RestoMod, saw this player advertised not working, spares or repair for £50, bought it as something to do, a bit of fun, and a bit experimental trying film coupling caps instead of electrolytics.

The player is about 10-20 years old, it offers :-

• hdcd compatibility.
• an 18 bit dac.
• opa2604 opamps.
• reliable Sony KSS-213c mech.
• (shhh) it's full of black gate capacitors.
• Very spacious, high quality pcb. No tiny SMT components.
• Originally about £300-£500. Base unit in the range with 971 and 991 above.

It powered on but the tray was firmly stuck.
Bonus on first opening the case below, it came with a free cd!


Fairly dusty but no smoke smell, gave everything a good clean including the laser lens, rod, laser rack and tracks but despite re-lubing all, the tray would not run free.
Boiled the belt to shrink it slightly for more grip, after much head scratching eventually found that the little cog bar runner underneath the tray at the front was rotating but not freely, it had been shoved down out of its proper place, levering it up it clicked into place and the tray then was back to gliding in and out.

Capacitor porn! Black gates(BGs) everywhere, these caps are legendary, there's 2xFK (low iron) 100uf 25v line level coupling, and about 8 other F type powering the opamp 100uf and dac 47uf. The higher model RCD-971 contains even more.
The light blue ones nearest are Panasonic. What a quality thing though even the blue film caps around the opamps are mounted high on legs presumably for isolation.


On testing the laser read ok, and the player sounded quite good. I would imagine it would improve over time because who knows how long its been inactive and black gates take weeks to form nano tubes and come on song.

Decisions, decisions, leave the BG FK coupling caps in place, replace them with Nichicon ES or Audio Note Seiyru or polymers, anyway went with the big yellow Vishay 1813 MKT film caps 3.3uf nothing special, bypassed with Vishay 1837 MKP 0.1uf under the board. Using the Vcap coupling capacitor calculator, 3.3uf is ample for not much bass roll off at all at 20hz. The other aspect is film caps don't distort whereas electrolytic capacitors do. It's important to secure the film caps to the board with glue as any movement is likely to break the tracks off the board. If only they were made with longer more flexible legs. Polyester film caps can sound a bit grainy/fuzzy for the first few hours or ten, normally there's a point when they smooth out and bass deepens.

Black gates replaced with 35v Panasonic FC (blue and gold) and a few FM, they are tiny compared, is this sacrilege? Well the BGs go on to live another life where they may be appreciated more. One Muse KZ in the dac vref position. The Pacific chip, opamps etc are static sensitive.



The opamps, although they are served by 4 electrolytics they were not bypassed so added two ceramic 100nf's under the board. Maybe it's not necessary with BGs.

Changed out the 1n4003 regulator diodes for fast diodes SBYV27-100-E3, these are very overspec at 100v 2 amp. Might be wrong because there were other changes involved, but suspect this was very worthwhile.

Only now notice that that 2 watt resistor is bang up against the blue cap, so I need to go back in and move that away it must have been like that for 20 years so hopefully not too bad. Also a black spot on a ceramic cap.



Changed out the caps on the signal processor C164, C151 not pictured, always worthwhile.

Overall result, moving the coupling caps to film makes a massive change for purity (no distortion), soundstage, separation, scale, yadda yadda, anyway definitely very worthwhile for transforming any cd player.

Comments, extra suggestions, things I've missed, welcome.

 

[duncan]

What a fun project! It seems to be true, but is still surprising to me, that output couplers of 47uf or 100uf can sometimes be replaced with something of much lower value without an apparent loss of bass. You've done the math and I haven't. I suppose that's partly a function of the next component in the playback chain? i.e., high enough input impedance there means less sensitivity to the coupler value.

Wish I could hear your result, but will vicariously enjoy your description.

 

[John3655]

Yes the vcap calculator has the amps input impedance as a factor, my Roksan K3 has 23kohms and a Marantz pm6007 has 47kohms either are fine bass wise with a 3.3uf film coupling cap, could probably go even lower. Possibly only a concern with low input impedance valve amps.

I read that electrolytic coupling caps are over sized to minimise the distortion as well as give vanishingly low bass figures for the spec sheets. Philips late 90s players cd753 723 go with 47uf 25v, Marantz cd4000 5000 Cerafine, some older players like Philips cd48 used 100uf Silmic 1 bipolar, Rotel 100uf 16v Black gate, Technics go with 6.3v 47uf electrolytics truly tiny caps. That suggests that the higher voltage ratings might not be so necessary. The lower voltage also means that the plates might stay formed.

 

[John3655]

Was having trouble with the cd tray getting stuck, basically the issues that caused the demise of this player onto the second hand market in the first place.
It also wouldn't play some CDs.

Replacing the laser, fixed the play issues.
Had an an issue with the spindle motor being too low but found that was my bad fitting of the sprung suspension screws.

Changed the vref capacitor on the dac to a polymer 47uf bypassed with 100nf ceramic. My rudimentary understanding is low ESR at all frequencies.

Changed all the 47uf caps around the dac to 220uf.

Took the opportunity to investigate the 1000uf cap that was close to the 2.2watt resistor pictured above. It measured really badly for ESR (~10ohms) so I replaced it. Looking at the schematics it's the main power supply for the loader mechanism, maybe its the cause of the tray issues. Rotel rcd-951, rcd-971, rcd-991 cd player tray sticking issues suspect tired capacitor C251.

 

[John3655]

This player contains Oscons. I mean for a £300 player back in the 00's, quality.

The purple cap in the picture above, was wondering about replacing it but thought better of it as it's so close to the clock which I don't want to risk upsetting. Turns out it's an oscon, there's also one in the power line to the opamps.

The path of power is smoothing caps, regulator, smoothing/stability caps, line, inductor, oscon and small ceramic bypass, more reservoir caps, chip. Multiple stages of noise absorbers, amazing.

As oscons are solid polymer it shouldn't need replacing as they have a longer life than conventional wet electrolytics, having said that I haven't checked and should never assume, but don't want to take it apart again.

The tray has been behaving itself ejecting fast and smooth, and the amount of detail and flow is incredible. I even discovered the hdcd light came on with Sheryl Crows, Feels like home, album not a fault, it's an hdcd!

 

[John3655]

The KSL-2130 cd tray mechanism is still giving grief, one day all fine, tray nice and swift then two days later, stuck solid in, with just a whine sound from the motor.
The fact that the motor is whining, I think tells me it's spinning, which means the belt isn't gripping enough. Off with the cover again, taking the belt off, boiling it again, cleaning the pullies again, turning them by hand though, shows that it's hanging up just as the laser is fully lowered and the cd tray should be moving out, but no it won't move, no wonder the motor can't do it.
It seems to hang on the gear indicted in red below which seems a bit sharp, I filed it a little, probably not necessary as these plastics don't really wear.


Then just rotate it by hand between the laser sled fully up (red circle) and tray in position...

... and the fully down and tray out position. I also loosened the big black cog a tiny bit more than before so maybe it can move up and down fractionally if it needs to.
Time will tell whether it's cured, it will be a few days before I bother re attaching the top cover.

 

[John3655]

Yesterday, working fine, today turned it on, tray locked solid again. I'm now suspecting the lubrication, have been using a spray silicone grease but I think it dries too thick so I'm now trying a wet multipurpose grease.

However it had the problem before it came into my hands so idk maybe I should just get a new loader.

Been trying it over and over this afternoon, it's totally intermittent it'll eject 10 times then lock.

Overnight, had a thought, maybe the top position is being overrun ever so slightly and causing a gear clash (rather than mesh) on eject, which will absolutely stop the motor.
I've added a wedge of folded aluminium foil and whitetac to the top position, the sleg peg can still rest there but the cog can't rotate quite as far. Hopeful, as when loading its more cushioned and less of a thunk. If it works over a number of days, will get the glue out.

 

[John3655]

Well it kinda worked except the tray would eject on power on which means the tray closed position switch is not staying on when the player is off. Not ideal.

Added an insulation sleeve (red) to the closed position switch which should turn off the motor fractionally sooner.

This evening the player worked brilliantly with the tray gliding in and out smoothly and silently many times, and no lock ups, and it doesn't ram itself in anymore.

Also discovered that it's possible to put the tray in ever so slightly crooked like 1 degree, and that results in it dragging rather than gliding, exactly as I experienced earlier, so that's explained.

Saw a picture of the later 2006 rcd-06 circuit board, it's almost identical, no black gates as they had finished by then, the quad of 100uf black gates in the 951, are now a quad of 220uf FC's.

Well I hope this thread has shed some light on the old age issues of this loader, good luck with yours.

Sony Ksl-2130css kss-213 cd loader mechanism, lock, not ejecting, won't eject, tray dragging stiff rough.

 

[John3655]

Looking at the schematics for the higher up in the range rcd-991, noticed that one of the major differences was just about every electrolytic capacitor is bypassed with 100nf film or ceramic capacitors, the 951 just gets 47nf's if at all.

Changed the Vishay polyester caps I had put on the output coupling for 3.3uf Audyn Q4 polypropylene(~£4), why well mkp is better than mkt and the q4's have been incredible in some speakers I recapped recently. They are only slightly larger and as they're polypropylene I didn't bypass them. Put clear sleeving on the leads as they go close to other components, and since the photo they are hot glued to the PCB to secure them. The leads are long enough to not need extending and flexible enough not to put too much stain on the PCB tracks hopefully.

Dac power, C661 ceramic 0.047uf changed to 0.47uf that's quite big, 470nf a lot bigger than the 100nf that gets put on everything, and better fits with the ideal 1:100 ratio of bypassing when in combo with a 47uf electrolytic. Ceramic capacitors are available with huge capacity nowadays, well up to about 10uf although I did read in a datasheet that they don't discharge all that linearly. Makes me wonder about removing the electrolytics in favour of ceramics, that would be cutting edge.

C931 1uf on input to regulator, changed to 100uf just cause.

The big Rubycon 4700uf reservoir caps, added 100uf zlj high ripple electrolytics under the board.

Changed the opamp power supply group of 4 caps to silmic2's with a blob of special non resonant material between them, ok white tack, and bypassed with 0.47uf ceramics directly underneath the chips between pin 4 and 8.


Tonality is superb, crowd noises on live albums incredibly real.

What else can be done,
Change 100ohm output resistors for better although not in too much of a rush to do that as the existing resistors are all good quality.

Remove the quite big output caps, red next to the rca's at the bottom of the pic, these are 0.01uf. Not sure whether to as they look like quality caps.
Further suggestions welcome.

 

[John3655]

Underside of the main board showing the two electrolytics bypassing the main reservoir caps and the yellow 470nf ceramics bypassing the op amps. Is it ok to do this, well the manufacturer wouldn't do it that way because it's too manual, if they do it they'll make a place on the circuit board and you see that on a lot of higher end designs generally there are loads more capacitors on the higher end machines.


Swapped the two 220uf caps on power bus h3 for 470uf. There's one ceramic next to them C967 changed it from 47nf to 4700nf but I think that might have been too much as it didn't sound quite right (wary of oscillation), ended up settling on 470nf.
The ceramic on the clock chip c632 changed from 47nf to 100nf and the large hdcd chip c602 47nf to 470nf.

Initially when I changed caps I put Panasonic FR on all the dac supplies, but changed them again, putting a 100uf polymer on digital vdd c651 and hopefully a nice smooth Nichicon KZ 220uf on the analogue c652. Also changed the 100ohm output resistors to Vishay doubt that'll do much as the existing ones looked almost identical but I had them so might as well use them.

The circuit board is great not a single track has lifted they're thick and wide. Sound wise this player is now breaking new ground for me, best I've heard. Some audio layers get revealed by replacing electrolytics but there's another few layers when you start upgrading the ceramics particularly on a base level machine like this.

 

[kaguhara]

Hello, my Sansui CD-A77 / Alpha-77 was having some problems with the Disc tray as well with a similar KSS-2130CCM, initially can't read too but lubing the gears and rails solved it.

At first it was an "easy fix", CD tray sometimes closes by itself... did not have extra rubber and lubed the gears, and oh lord it got worse, now the sled the lens is sitting on is not rising anymore, wasted so much time thinking it was misaligned gears, turns out I put too much, removed the gears and cleaned it, put proper amount, and put actual new belt and it worked all the way to the sled rising.

Second problem to address was I wanted to replace the CD lens (could not because the new one was DoA) I whatever and use the old one anyway... aaand it's noisy as hell, i adjust the four screws that hold the rubber and spring, and it doesn't spin. I had to screw the pickup alllllll the way so it's flat when it tries to read.

Disassembled the motor part, lubed the switches for good measure, and oiled the motor. I noticed it seemed to spin slowly so it might have added to the sled problem.

Anyway, it was a lot of work and I learned a lot trying to bruteforce this mechanism, I hope this serves as extra info.