Marantz CD42 repair and mods

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Bernie,

I think the CD42 has a sonic signature somewhere between the CD63 and the much older ones, which definitely suits certain styles of music.

That CD63 is done, I don't think I'll be doing any more with it. It does sound good with output transformers in combination with good op-amps. The only thing is, on some loud CDs, it does seem to distort on loud transients. It might need an output resistor in series.

Simon
 
Hi Simon,

Yeah, when I played mainly rock & blues, the limitations of my original 52SE didn't show up so much. I think your upgraded 42 is much better, but as you say, the mods done were ( by your standards!) modest. Interesting about the distortion on transients - where do you think it's originating - are you overdriving your amp or is it somewhere before? I would have thought the transformer output would be silky smooth. I'd still like it, now my finances have recovered after the Volvo service! By the way, I've recently rescued a 63 fron the recycle centre - have you seen one refusing to read the CD and spinning backwards? The threads indicate the controller IC may be SNUFU'd. Have you done this repair before? Cheers Bernie
 
I find with rock music more resolution isn't always particularly welcome, unless it's accompanied by excellent bass weight and a smooth overall character. Perhaps stating the obvious there :)

I have no idea about the distortion, but it could be the op-amps on the previous stage, the transformers getting overloaded (I can't see why, they're used for doing something within their comfort zone I think)... or it could be the input stage of my pre-amp being a poor match. I will just have to try a couple of things (resistors here and there) and see if I can isolate the problem. I have already been advised to put some resistance on my pre's inputs so I might start there.

The CD63 is available if you'd like it, I'll e-mail about that but you can have it for little more than parts prices if that suits you. The resolution is massively higher than your CD42 and you'll have fun comparing it to that DAC of yours - who knows what will happen there. The clock, transformers and good op-amps can all be plucked out of course, should you prefer the standalone DAC.

As for that Volvo service you thought was bad... I just got my Mondeo through after rebuilding the top end of the engine, and then it needed 2 tyres, 2 wishboes and 2 handbrake cables to get through its MOT!

CD63s not reading and spinning backwards aren't exactly rare and can be due to various faults, not just processor-related. I'm not amazingly good at fault-finding, but I know that many fail due to bad connectors (to the front panel and moreso to the mech).

Simon
 
Snafu

Hi Simon

I'll look forward to your e-mail. Before the Lundahl upgrade, did you hear any transient distortion? Obvious question, but I'm good at those.

I bought the DAC on a whim; it only cost £60 or so and is very good for the price. Have you dipped into the Fikus Lampizator website? He seems to be a bit of a windbag :eek:, but does some interesting things. His baby is to add triode followers to the DAC output, skipping the op amp stages. Somewhat grandiosely, he claims to have approached "audio nirvana" but he has a bewildering collection of modded players, so he must be doing something right!

On the duff 63, it cost buttons and I have the service manual - I'll try not to fry this one. Cheers Bernie
 
I've had a listen to the CD52 I have here, and it's not bad! It sounds well on rock but dynamically it's very flat - very obviously lacking on Eric clapton Unplugged.

I've looked inside and changed the broken cog. I'm going to write a how-to guide with pictures for that job as people have asked about it a couple of times now.

I notice the CD52 has Elna Silmics around the opamps and it has, amazingly enough, a Black Gate 220/6.3V on the output of the main 5 volt regulator!

I've just put some LM4562 in and shorted out the output caps (less than 100mV per side in this one). I'm going to tackle those surface-mount filtering caps after the DAC next as those seem to shut the sound in.

Simon
 
It has to be said that there is a lot of misunderstanding about which type of cap to use and where.

For example = after regulators.
Read the specs for the 7805 type.
The cap after should not be low esr or large.
The one before can be.

In general, low esr caps can be used across devices (remote from the reg)
As for OSCONs, they are best for digital stuff.

Andy

.
 
It has to be said that there is a lot of misunderstanding about which type of cap to use and where.

For example = after regulators.
Read the specs for the 7805 type.
The cap after should not be low esr or large.
The one before can be.

In general, low esr caps can be used across devices (remote from the reg)
As for OSCONs, they are best for digital stuff.

Andy

.

Hi Andy, maybe you can clarify something that's always confused me. "After regulators" means directly after or anywhere after.
I tend to use 317 regs with tants directly on the output (as recommended) but just before the chip input I have some large, low esr caps for decoupling. Is that a problem?

Regards

Pete
 
As far as I read the info, that should be OK.
The low esr caps directly after the regulator causes resonant peaks in the response ( I think)
Similarly, a large cap immediately after the reg. does not allow it to "regulate".

Somewher, either on this forum or the net, there is a very good article about regulators.

Aha !!! Martin Clark Using 3-pin regulators off-piste: part 1

Andy

.
 
Last edited:
Just another Moderator
Joined 2003
Paid Member
Attached is a sim of two almost identical LM317 supplies. Both have low esr 100uF caps on the output. The green trace is with just the cap, the blue trace is with 0.25 ohms resistance added in series with the cap. The smaller the cap the worse the problem and the more series resistance has to be added to swamp the resonance.

The plots are showing the output impedance, mV can be read as milli ohms.

The LM317 can theoretically have large capacitance on the output, and I'm planning to try 1000uF.

Tony.
 

Attachments

  • low_esr_effect.png
    low_esr_effect.png
    15 KB · Views: 228
Yes, try the oscon; 10-68uF is quite large enough - it has to charge-up through the chips internal divider, so you don't want too long a start-up time. If the 47nF thingy is SMT, leave it in place too.

This dac was used in a lot of good equipment. DPA used it in various PDM1 / Little/Bigger Bit designs. They used an active, low-noise 2.5v voltage reference with buffer to drive these pins, so if you want to really experiment...

The full Philips 'CD7' chipset is to use the SAA7350 to generate the bitsream, and the TDA1547 solely to decode it. It's a pairing capable of stunning sound.

Hi martin, it;s quite an old thread but it sounds interesting. I'm wondering what should I do to bypass the internal I/V converters og SAA7350 and get the pure curent ourput of it, to convert then into voltage by driving other external circuits. I've done it in the past on a CD-48 carrying the TDA 1549 by using external high impedance buffer IC. The diference over here is that 7350 is a balanced output switching capacitor current output, while the TDA1549 was single ended. Another difference is that TDA 1549 had indepedent supply rails feeding the internal I/V converters (where you can easily disconect from DC), whiles the 7350 is fed from the same supply which feeds the switcing capacitor stage. I'm a bit confused what should I do. I got a feeling that it can be setup.
cheers
 
Andy

It's not something I ever tried for myself - you'd have to read the SAA7450 +TDA1547 datasheets side-by-side to work it out.

That said, a very quick look at the lattter shows there is a diagram or two - snapshot attached. Drop me an email if you can't find the datasheets and I'll send them over in reply.
 

Attachments

  • Screen Shot 2018-10-09 at 18.44.55.jpg
    Screen Shot 2018-10-09 at 18.44.55.jpg
    76.6 KB · Views: 127
Andy

It's not something I ever tried for myself - you'd have to read the SAA7450 +TDA1547 datasheets side-by-side to work it out.

That said, a very quick look at the lattter shows there is a diagram or two - snapshot attached. Drop me an email if you can't find the datasheets and I'll send them over in reply.

Hi Martin , thank for your kind reply. I already have the service manual of Philips cd951, as also the deck itself. I can't swear for the magic sound of this CD player, even after very long and extended mods. Somewhere I've read that all of the missfiring comes from the chippy intergrated I/V converters in Philips DAC's. Taking the chance I initiated to deactivate TDA1549 I/V's in another CD player (Marantz CD48) and pick the current output. The outcome was really better but only after using the MUSE 8920 J-fet as external IV. Just for comparison with the modded scheme, I could never name the intergrated TDA1549 I/V's as a crap. A bit tubbie, shy and flat, but plenty of musicality and character. Just wondering how the 7350 could sing without it's own I/V's..
 
Hi,

As a cheaper alternative to a cheap clock.....

I've found good improvements on some players just by disconnecting the clock out feed from the chip to the std crystal and feeding a clean isolated power supply to the crystal instead. I guess by the time the standard 5v feed (or whatever is used) is routed through and out of the chip it's pretty polluted. With a cheap lm317 reg you get quite an improvement for next to nothing.

Regards

Pete

Hi Pete,
can you indicate where to to connect the external DC line to feed the crystal on the attached chematic of CD42 ? I'm a bit confused with "Xin" and "Xout" pins of the chip . Should I remove any part(resisistor) around the crystal circuit? Thanks for your advice
 

Attachments

  • Clock pins SAA7350.png
    Clock pins SAA7350.png
    46.4 KB · Views: 85
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.