Interference in off-board CD clock PSU

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
This is a cross-post (also in top > other stuff > power supply design). I'm sorry if that's against the rules (new here), but I'm not sure where best to post this.

Let me explain a little of the problem:

I've reclocked my Marantz CD6000 with a Tent XO module. To increase performance, I am trying to use an off-board PSU. See here for pictures and a discussion of its construction.

The 12v output of the psu is fed via twisted-pair cabling (CAT5) into the CD player. The 12v powers one of Andy's (ALW) Jung super regulators which outputs 5v into the XO.

The problem is that when switching the kitchen/dining room lights off; the surge/spike/whatever causes the CD player to **** out and display 'Error'. This is how the CD6000 reports a clock error, as I found out when I tried to switch on the player without the clock psu powered on!

Basically, switching lights on/off causes some kind of abberation in the clock PSU which is passed through to the clock which causes (I guess) some kind of sync error and the player stops working until it's switched off and on again.

I'd really like some suggestions on how to fix this problem which never manifested when the old clock was sharing the player's internal toroidal power supply.

One idea of mine is to take a small wall-wart supply (240V AC -> 12V DC) apart and house it inside the cd player. I'd feed its 12v into the super reg and power the clock from there, but I'm concerned about noise from the wall wart in terms of both EMI and **** getting through the super reg into the clock.

Help!

Carl
 
The problem is that when switching the kitchen/dining room lights off; the surge/spike/whatever causes the CD player to **** out and display 'Error'. This is how the CD6000 reports a clock error, as I found out when I tried to switch on the player without the clock psu powered on!

Hi,


Maybe consult a qualified electrician ? (safety first)

Erm where are the fuses for the transformer secondaries ?

Tantalum caps aswell!

What other mods do you plan for the player?



Kind regards,


Ashley.
 
Hacker said:

The 12v output of the psu is fed via twisted-pair cabling (CAT5) into the CD player. The 12v powers one of Andy's (ALW) Jung super regulators which outputs 5v into the XO.

The problem is that when switching the kitchen/dining room lights off; the surge/spike/whatever causes the CD player to **** out and display 'Error'. This is how the CD6000 reports a clock error,

Help!

Carl

Hi

You have a common mode problem, wll, not you, but the CD player / power supply....

try connecting the + and - of the incomming power supply with 2 capacitors (say 10nF) to the cabinet (back), I recall the player has a metallic house.

It may be even better to use a coaxial cable to connect the power supply, and connect the (RCA or BNC) ground to the chassis

succes
 
Re: Re: Interference in off-board CD clock PSU

Guido Tent said:


Hi

You have a common mode problem, wll, not you, but the CD player / power supply....


Aha, I'll google that and study it.

Guido Tent said:
try connecting the + and - of the incomming power supply with 2 capacitors (say 10nF) to the cabinet (back), I recall the player has a metallic house.

You mean decouple both +/- to the chassis? I take it you intend me to use a film cap for this?

Guido Tent said:
It may be even better to use a coaxial cable to connect the power supply, and connect the (RCA or BNC) ground to the chassis

If I use coax, I need both inner and outer conductors in the cable to carry +12v and 0v return. I assume you mean that I should run 0v on the outer conductor and connect it to the chassis?

In the meantime, I've built a small internal PSU using a 16VA centre-tapped 9-0-9 transformer and a 4700uf smoothing cap. This is connected to the ALWSR. The primary is connected to the same terminals as the player's toroid.

So far, so good - no matter how much I've tried I can't make it skip, click, pop or generate an error. Admittedly, I've been playing it on silent while my girlfriend watches the TV and I don't know how it sounds, but at least it works.

Thanks for the advice, I may just try it out if the sound quality is poorer with this new PSU.

Carl
 
Re: Re: Re: Interference in off-board CD clock PSU

Hacker said:


You mean decouple both +/- to the chassis? I take it you intend me to use a film cap for this?

If I use coax, I need both inner and outer conductors in the cable to carry +12v and 0v return. I assume you mean that I should run 0v on the outer conductor and connect it to the chassis?

Thanks for the advice, I may just try it out if the sound quality is poorer with this new PSU.

Carl

Hello Carl

Film caps are fine, don't pay too much. In case of coax shield indeed serves as 0V.

best
 
Luke said:
I just bought one of those and have Keuis recomended mods for the CD63/67 (its the same damn machine with an extra dac).
The next thing Im considering is a clock like yours. How does it sound with a new clock?

To be honest, I only listened to it for very little time before blowing the damed player up. Well, ok, that's maybe a little over the top... let me explain.

I was adding better regulation to the analogue stage (using Andy's [ALW] super regulators, the best regs that money can buy imho) and managed to build one incorrectly, putting 18v down a 12v rail. Oops. That blew the HDAM and muting transistors and took some diagnosing, I can tell you.

I've pulled the transistors and removed the HDAM board and whilst it plays music, it sounds like a sack of **** (note that the reason is not due to lack of muting trannies; it's because without the HDAM the player can't power the preamp input stage properly).

I've got some AD8620 op amps on order from analog.com and I've already got a hold of some soic -> dil adaptors to mount them on. They will replace the blown HDAM circuitry and, hopefully, sound even better. Until then I can't really give an accurate description of how it sounds.

I'll come back here in a week and let you know. I'll take plenty photos, too - hopefully you and others can benefit from them :)

Cheers,
Carl
 
Luke said:
Hi Hacker,

I just bought one of those and have Keuis recomended mods for the CD63/67 (its the same damn machine with an extra dac).
The next thing Im considering is a clock like yours. How does it sound with a new clock?

cheers arthur

I've put an XO in the 63ki together with two ALW regs along with the 'more standard' mods. The XO goes right on the pcb, next to the dac (remove the caps and you can put the XO in with one extra hole drilled into the pcb for the XO vcc pen. Make the hole big enough so it does not touch the gnd layer of the pcb and connect vcc underneith the pcb).

One ALW reg then feeds the XO and the clock vcc pen of the DAC.
The other ALW reg feeds the vcc analog pen of the DAC.

Can't comment on which part of the mod contributes to what improvement (the player is too delicate to take apart/put together all the time) but i like the end result a lot. :D

Took some pictures too, i'll post some later

EDIT: oops, i think now you bought a cd6000. Then the clock distribution part is more difficult because of the TC chip.:smash:
 
guido said:
The XO goes right on the pcb, next to the dac

One ALW reg then feeds the XO and the clock vcc pen of the DAC.
The other ALW reg feeds the vcc analog pen of the DAC.

Can't comment on which part of the mod contributes to what improvement (the player is too delicate to take apart/put together all the time) but i like the end result a lot. :D


Hi

Good to see you place the clock close to the DAC. On the result: You need both, good clock, and good power supply

best
 
Lets say i at least have one of the two :D
 

Attachments

  • cd63 xo1.jpg
    cd63 xo1.jpg
    91.6 KB · Views: 182
EDIT: oops, i think now you bought a cd6000. Then the clock distribution part is more difficult because of the TC chip.

Hi Guido,

yes I have read this somewhere on Diyaudio, but the problem is I cant find a circuit to build a buffer and run 3 chips from one XO.
I think I will just build a normal XO to replace the existing clock and see if that makes a difference.

regards Arthur
 
Mmm,

I'll take it with me at the next diyaudio day.

Too many guido's around here....:dodgy:

End of threadjack :smash:

And an edit, busy post:

Think the other guido :clown: has an answer the question

If i had one, i would try to feed the DAC's indeed with the (buffered) clock and also the TC chip (that's 3). Maybe the clock to the TC needs an inverter to get the timing right (!).
Never found the datasheet on it.

Feeding the dac clock pen with the same PS as the XO should be better than leaving it to the main +5 (return currents :dodgy: :confused: )
 
Luke said:


Hi Guido,

yes I have read this somewhere on Diyaudio, but the problem is I cant find a circuit to build a buffer and run 3 chips from one XO.
I think I will just build a normal XO to replace the existing clock and see if that makes a difference.

regards Arthur

Hi

There's 2 Guido's |(well, there are more): GuidoB and GuidoT. We both live in the south of Holland.

As a buffer you may use HC04 inverters, but my (TentLabs) clock still is OK when driving 3 gates. Give each line a 47 ohm resistor close to the XO

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