Philips CD104 tweaks

DK6400Brian said:
Replaced caps on the humming CD-100.

Turned out, that the replacement didn't cure the machine at all.
Resoldering the DAC board cured this case instantly. ;)


I think we have a problem with these machines.
Someone mentioned an issue with the treble not being exact in the middle as intended, but sounding somewhat off-phase or wishy washy.
I've got the same thing going here and noticed it for the first time a couple of days ago on "Clouds on my Tongue", song #10 on [Tori Amos, Under the Pink, 1994 Warner Music].

Wow. That's not what I would call "Precise treble performance", Gentlemen.
I had to go over it again and had it also played on a CD-207.
The problem remains on the CD-207 as well.

Is this a general issue with the TDA-1540/SAA70xx based DAC or are these DAC's so good that they discover bad CD-mastering ? :D

How well does the 14-bit NOS mod perform ?
Is the off-phase treble also an issue after the 14-bit NOS mod ?

Did you compare the CD100 and the cd 207 with an other player on the same amp and speakers ?

At my place I suffer from heavy RFI -EMC.
(2 fm stations 500 meter one GSM in 100 meter.)

At night soundstage is more open and rock centered.
On the day soundstage is shifted to the left and waggles

After modifiying amp (NAD 218) speakers (Magnepan IIIM)
Interlinks, passive volume( no preamp) with filters hexfets etc etc
etc etc !! finally I got rid of that damned RFI-EMC influence.
The modified cd104 nos with tda 1540 and org IV souds rock centered with wide open soundstage. So do all my modifiyed players (CD94 MK II f.i)

I compare with the DIY dac DI8M (ecdesigns with 8 tda541a one crown) and a modified CD 882 transport. (These are sounding for me the best !)

So I think the tda1540 is ok ! but RFI-EMC can ruine in my opinion the whole thing.

I hope this is helpfull info.

Best regards,

Onno
 
EMC-RFI

I forgot to tell this.

Take care of all the switching powersupply's like small ones for cordless telephones, PC's etc, dimmers, DECT telephones.
Replace them or switch them off by disconnecting.
Feed your amp, player,dac from a dedicated well filtered powerline.

BR.

Onno
 
Onnosr said:


Did you compare the CD100 and the cd 207 with an other player on the same amp and speakers ?


Yup. A portable CD-player from Philips from 2000/2001.
It had the same thing going with the same number from the CD.

I still need to try out the DVD player.

For your record, I'm using Sennheiser HD600 for intense listening.
I have a set of DIY ESL speakers, but still need to assemble the tube amplifiers before giving them some juice ;)

Regarding higher quality audio during nighttime.
My experience is, that the background noise levels differs throughout the day/night.

By the way. I made these terrific audiocables out of 8mm antenna-cable the other day. They're cheap and they're stunning great. :bigeyes:
Here...have a look:

An externally hosted image should be here but it was not working when we last tested it.


Kindest regards
Brian
 
Hi Guys,

Thank you for your responses regarding output stages and Viccpa for posting that circuit. Has anyone tried that simple fet output that Lucas calls Fetishizator ?

http://www.lampizator.eu/LAMPIZATOR/FETISHIZATOR/fetishizator.html

I haven’t got around to it yet but will give it a shot.

As for my valve output I managed to salvage some 1:4 step up transformers from an old polygraph machine that I wired in. I’m using a 20 ohm I/V in parallel with the 30 ohm primary to give about 11 ohms. I found the transformer I/V to make a big difference in the bass region, the bass seems clearly delineated, mids are fuller too. As for the highs, I haven’t compared it to anything yet.

Tubee and DK6400Brian,

You are right about the treble, my 4xOS TDA1541 machines betters the nos tda1540 in this department. However, I have been experimenting on ways if improving this. Shielding the dacs and bypassing the 3 psu pins on the 1540 with SMD NOP or COG capacitors have definitely helped. I used 50nF with 100nF X7R. Another tweak that improved the treble is to put ferrite beads on the little power supply chokes. I have done this on some of the IC’s and am in the process for applying it for all power lines on the decoder board. All this doesn’t really increase the detail, but rather, gets rid of the grain. Finally, a good clock with separate power supply and a reclocked LE input (from pin 12 of the SAA7000 to pin 2 of the dac) has definitely helped with the detail.

Cheers,

Richard
 
Richard.


I've listened intensively to the Tori Amos CD on our DVD-player and the specific problem with her S'es late in the number "Clouds on my Tongue" is still there. :yikes:

The conlusion is. Something was done wrong in the recording/mixing/mastering process of the CD rather than wrongdoing within the 14-bit DACs of Philips CD-100. :judge:
That suggests to me, that the combination of Philips CD-100, my homebrewed cable and the Luxman C-03B is good enough to pinpoint tiny faults in audiodetail. Well that's not bad at all.

A Philips CD-100 from 1983; An audiophile piece of machinery :worship: I rest my case !
 
Richard said:
Tubee and DK6400Brian,

You are right about the treble, my 4xOS TDA1541 machines betters the nos tda1540 in this department. However, I have been experimenting on ways if improving this. Shielding the dacs and bypassing the 3 psu pins on the 1540 with SMD NOP or COG capacitors have definitely helped. I used 50nF with 100nF X7R. Another tweak that improved the treble is to put ferrite beads on the little power supply chokes. I have done this on some of the IC’s and am in the process for applying it for all power lines on the decoder board. All this doesn’t really increase the detail, but rather, gets rid of the grain. Finally, a good clock with separate power supply and a reclocked LE input (from pin 12 of the SAA7000 to pin 2 of the dac) has definitely helped with the detail.

Cheers,

Richard

The CD300 i have needs to be treated that way, in time... I didn't know LE is that sensitive on reclocking, should give it a try. What frequency does it need? (could reclock it with dividers from the masterclock upgrade)
 
Hi Brian,

I’m inclined to agree with you. I’m using a Saba greencone tweeter at the moment and I haven’t been able to listen with better tweeters yet. I suppose a Raal would be a better magnifying glass. As for the TDA1540 I’m a firm believer in this dac. It simply sounds “correct” and very natural, with the mods I have done so far it outclasses my modified Philips CDM2 transport and Musiclabs (PCM1702) dac combination. However, as I mentioned earlier the TDA1541 (no A) in 4 x OS did appear to have a bit more detail in the top end. I plan on duplicating the output stage and chassis mods to try and do some more consistent evaluations between these dacs, but at the end of the day the TDA1540 still remains the main source in my system.

As for the interconnects you made, are the conductors stranded or single core copper? Have you tried the following recipe?
http://www.venhaus1.com/diysilverinterconnects.html

Tubee,

The reclocking of LE uses the 4.2336 MHz clock. See the attachment below:
Let us know what you think if you implement it

Cheers,

Richard
 

Attachments

  • no_oversampling.pdf
    27.8 KB · Views: 358
Richard,

So far, I can't think of any reason to trash the TDA1540.
Only thing I'm able to copy with regards to technical measurements is a bit of 3rd harmonic distortion which I think is about -40dB from signal.

Why this reclocking issue ?
The tremendous results you get. Isn't this just psychological ?
I mean....if you modify the electronics, you're not able to do an A/B-test anyway.
So when it works, isn't the great results not just a matter of the joy of being able to swap components ?

I'm really curious, because my CD-100 is completely unmodified, but if it's possible to squeeze some extra juice out of this pussy, I'd do it.
I love pussy, but I won't do anything just for the psychological amusement. It really has to be that extra good !

Modifying the output stage. Well I'm open for it.
Components are standard. Perhaps some handtrimming and finetuning could be good here. Components with tighter specifications/margin of error.
There's really some doubts out there if one should swap the opamps. One say this. Another say that. That's not good enough.

I also want to point my fingers at the "brag-factor".
I not into it. I'm interested in good sonic performance, not a brag war of brands of components.

What about internal shielding ?
Would shielding the output stage give something (in theory) ?

Regarding my homebrewed cable.
It's a 1.3mm solid signal core with double shielding (foil/mesh).
I have nothing to point my fingers at and I'm about to be manufactoring more of these during this December.
They're handmade and I'll be selling them upon order for $40.
They're connecting the CD-100 with the Lux C-03B and they're also able to do SDPIF or digital transfer if you like. They're that good.
 
Hi Brian,

Clocking the player and reclocking data lines does work, it’s not a psychological issue. I originally started on a Philips CD 202 and moved onto a Cd304, as the latter sounder better. A Kwak clock is the cheapest entry into this realm.

Regarding the mods, I approach all my audio mods just like my research, by having a control and trying to minimise confounders. In order to tests my mods to the CD 104/304 I managed to gather 3 decoder boards from 3 104’s. I keep 1 board standard as the control and modify the others, one tweak at a time, then compare them all to the control. It’s difficult to eliminate all confounders from these 20+ year old devices but getting my hands on 4 new tda1540s from the same batch helped.

When comparing opamps, sockets helped with changeovers and comparisons but some need extra bypassing, that’s when those extra boards come in handy. The other thing I noticed is that the audio chain needs to be transparent enough to pick up the differences in the tweaks.

As for shielding, there is a lot of evidence out there that suggest that is works in audio. As for shielding a valve output stage I’m not sure. However, I’ve taken apart lots of old valve based laboratory equipment and some critical circuits do have some form of shielding. Perhaps an electronic tech can comment on this.

I also want to point my fingers at the "brag-factor". I not into it. I'm interested in good sonic performance, not a brag war of brands of components.

There’s bragging and using a known component as a comparison. The goal of most DIYers is to better a standard component and to see if their diy creations rival high end devices. The only way to do this is to experiment and compare, it also helps to get together with some friends and have a shoot-out between standard and diy gear and get opinions in perceived vs evident differences.

Thanks also for the info on your cables, I’m using some similar diy interconnects. Silver/copper foil is also rather nice.

I have this innate need to tweak things hence suggest that you modify that pussy. One shouldn’t fear about curiosity killing the cat, open tube circuits and not waiting for capacitors to discharge will do that.

Cheers,

Rich
 
DK6400Brian said:
Richard,

Why this reclocking issue ?
The tremendous results you get. Isn't this just psychological ?
I mean....if you modify the electronics, you're not able to do an A/B-test anyway.
So when it works, isn't the great results not just a matter of the joy of being able to swap components ?

I'm really curious, because my CD-100 is completely unmodified, but if it's possible to squeeze some extra juice out of this pussy, I'd do it.
I love pussy, but I won't do anything just for the psychological amusement. It really has to be that extra good !

Modifying the output stage. Well I'm open for it.
Components are standard. Perhaps some handtrimming and finetuning could be good here. Components with tighter specifications/margin of error.
There's really some doubts out there if one should swap the opamps. One say this. Another say that. That's not good enough.

I also want to point my fingers at the "brag-factor".
I not into it. I'm interested in good sonic performance, not a brag war of brands of components.

What about internal shielding ?
Would shielding the output stage give something (in theory) ?

Regarding my homebrewed cable.
It's a 1.3mm solid signal core with double shielding (foil/mesh).
I have nothing to point my fingers at and I'm about to be manufactoring more of these during this December.
They're handmade and I'll be selling them upon order for $40.
They're connecting the CD-100 with the Lux C-03B and they're also able to do SDPIF or digital transfer if you like. They're that good.

I think the simple 5532 is not that bad either.... aside with the 1540. When you listen to a 1540 based player you will get extra artefacts on the sound which are not recorded. Extra harmonics, that's imo the reason a 1540 can sound more natural

For the reclocking issue: the cd300 has a lot of logic chips for Xtal, and the frequency is only a humble 4Mhz. Reclocking might not give a big difference in these players then. The low Xtal freq. considered could be a reason why this simple machine can sound that good, in special compared to Delta/Sigma dacs nowadays.

@ Richard: thanks for the schematic. Will repair a probable bad GND rivets first, and put some better caps in.

Reclocking does not give a huge difference in sound. Bass lines is often better contoured after reclocking, cleared up a little.

Indeed sockets help in opamp swaps. nice to do, some sound laid back, others detailed, other more upfront. All matter of taste, and has to fit with the rest of the gear. Often was the original opamp is not the badest choice at all:confused:

Modding can be deceiving: removing a source from a chain, mod it, replace in the chain and one difference is not to be forgotten: replacing the in time oxidised RCA sockets that only can give a sound improvement, nothing to do with the soldering at all.;)
 
Richard mate :D

Having an unmodified board as a comparison to all modifications made deserves respect and you're right. It's important that the chain is transparent to determine any changes so I'm about to give you standing ovations. I like your approach.

I need to do some research about this reclocking in order to be convinced instead of asking here right now.
I imagine some theoretical differences in timing resulting in jitter and distortion.

I have two CD-100s, two CD-207s.
None are modified apart from resoldering both CD-100 and changed caps in one of them.

Tweaking/adjusting/modifying the CD-100s is a 2009 project.
In the meantime I'll gather information so I know what to do, where to do it....and why !

Right now I'm waiting for a Sony WM-D6C walkman arriving today.
Just got hooked on cassettetape again and instead of having a fancy MP3 player, I want to record from the CD-100 and using the more fancy WM-D6C when I travel.

The search for dynamic CD's to record let me to the conclusion, that many newer CD's are terribly made.
They cannot be used seriously, much less being used in an analytic way on high end equipment to determine anything.

I agree. True High End is a matter of doing things right from the start. Tweaking or modifying mid range equipment corrects the flaws and leads eventually to a better way of handling the electronic signals. I'm aware of better manufactored components with longer expected lifetimes and tighter tolerances. Of course I would also use those.
In fact. The replaced caps in one of my CD-100s were Panasonic 105c caps with at least the double specified voltage. I did that to ensure a longer lifespan.

When I had the issue with the Tori Amos recording fresh in mind, I secretly wanted to "upgrade" to a TDA1541 player.
But it turned out eventually, that it came down to a bad recording instead and that the unmodified TDA1540 was in fact able to grap such tiny detail. Suddenly I was impressed again.
 
You can speak of distorsion yes, it is not as the original signal.

A 14 bit has more distorsion, 2^14 gives only a humble 16384 steps in 14 bit.
16 bit: 2^16 gives a lot better figure of 65536 steps. But the dac (substrate) layout, the way clock is distributed, output stage, mono dac vs stereo, all can give an influence.

The 1540 has the deepest bass in my setup, due to the subharmonics produced in this very dac. Subharmonics give a sort of "involving' thing to the music. Then you forget the treble rougness after a while. But all imo :)
 
An other thing: i looked in the schematic of my cd300, one dac has an symmetry adjusting pot, the other dac not. The symmetry circuit is connected to oscillator cap (for demodulator frequency) Has the CD104 such pot also? A cd304 14bit has not.

Anyone comments on that pot adjustment?
 
Hi Guys,

Thanks for the responses and thoughts. As Tubee indicated everything can give an influence. I have converted my CD 304 into a top loader (removed that whole resonant tray mechanism and cover) and have been experimenting with different disc clamps, I have tried machined brass pucks of different mass, a spare magnetic one from a Marantz CD94 Mk II and the original clamp with a small magnet. I reckon they make a difference, essentially the lighter the clamp the more open the sound. The guys over at the Shingaclone thread also did some experiments with clamps etc.

Brian, I’m also in the process of researching reclocking. There is a thread on DEM reclocking of the TDA1541A in which users reported large gains. I hope the following links are of use to you.
http://www.diyaudio.com/forums/showthread.php?threadid=11949

http://www.diyaudio.com/forums/showthread.php?threadid=75476

Pedja Rogic provides a lot of info on asynchronous reclocking, see about 1/2 way down on this page
http://www.pedjarogic.com/1541a/rev1.htm

The Sony WM-D6C walkman is old school, I still have my first cassette walkman (Toshiba) with my main system just to remember my humble beginnings in this obsessive hobby.

Tubee, no such pots on the CD304 or CD104 decoder boards.

As for modding and moving components in the chain, it’s good to be able to swap back and forth with a standard to compare.

For evaluating I usually use the track ‘Man in a long black coat’ by Bob Dylan. In this track there is a line “somebody out there is beating a dead horse” and it begs the question, is the tda1540 really worth all that effort? Or are we just beating an old high distortion dac? But with all the mods I’ve implemented it sounds very good, so good in fact (not to brag ;) ) that my standard Marantz Cd 94 Mk II sounded broken in comparison (this comparison was done prior to the 304s valve output mod). The transformer step-up and valve output puts it in a different league.

That’s why I’m curious to see if it’s just me being affected by the solder fumes or others are having similar findings.

Rich
 
It is so funny i have often swapped caps and so on. Discovered that some standard parts can sound very good, even better then some audiomongers declare THIS IS IT. I have become more cautious with mods after some bad results.

-B&G Nx cap near the opamp for instance.... simple blue philips lytic sounded A LOT better.
-Made a nice rectifier from hexfreds in my amp, after a try i refitted simple and cheap rec from BYW diodes.....
-Decoupling caps on cathode resistor: Simple philips blue axials instead of BG standard, Cerafine and so on. :scratch2:

What i will do on the CD300: only a few things, some caps replacing, maybe some beads added, resolder gnd connections, realign laser current, add gold RCA sockets and that's it.
 
Tray Motor board transistors?

Hi guys,

I've noticed that my CD104 has an earlier Tray Motor Control board with discrete components instead of the BA6109.....
One of the motor driver transistors has failed (the tray powers out, but not back in again) and I need to find a replacement. The number is either 2SC32840 or 2SC33840 but none of my references have this type number.
Can anyone tell me what logic lines SL1 and SL2 do? One is 'Tray Out' and one is 'Tray In' but which way round?
Can anyone help?
Thanks,
Richard
 
Re: Tray Motor board transistors?

mole42 said:
Hi guys,

I've noticed that my CD104 has an earlier Tray Motor Control board with discrete components instead of the BA6109.....
One of the motor driver transistors has failed (the tray powers out, but not back in again) and I need to find a replacement. The number is either 2SC32840 or 2SC33840 but none of my references have this type number.
Can anyone tell me what logic lines SL1 and SL2 do? One is 'Tray Out' and one is 'Tray In' but which way round?
Can anyone help?
Thanks,
Richard


Hi Mole42

2SC32840 (= BC328-40)

any TO92 (or SOT 54a) PNP silicium tor will do the job provided
the max current > 750 ma (or 1A) and V ceo > 45 volts

2SC33840 --> NPN etc. (= BC338-40)

(check EBC connections on tor)

SL 1 triggers the tray out
SL 2 triggers the tray in

BTW did you check all the (gray) cables to the connectors.
They "look" most of the times OK but in fact the are just broken.

Regards.

Onno