Joe-HT said:Corrections! it's $37 clock only and $140 with reg shipped from Hong Kong. It's diyhifisupply.com and the site is up!
I'm reading like crazy and getting crosseyed!
ALW, is there a newer version of the 317at circuit you have on accoustica.com cd63 tweaks pages? Since I have the 317s, I'd like to use them in the most noise-free way possible.
hi. You may find this interesting:
pagehttp://www.diyaudio.com/forums/showthread.php?postid=1222372#post1222372
Regards, Lee.
I'm stil wondering what a low noise regulator would do. My AC on DC signal from the opamp power lines went from about 100mV to about 0,8160 - 0,9150 mV with standard 78CT15 and 79CT15, removed mute & headphone and all BG power supply with standard transformer when playing.
Sounds like verry low noise, I've even measured with 0,005% accuracy 2 days after calibration. Gotta love those Keithley 2000's. They really come in handy when your busy with small signals.
I't nice for pairing resistors too, but that's only about 0,002% acurate on 6,5 digits
Sounds like verry low noise, I've even measured with 0,005% accuracy 2 days after calibration. Gotta love those Keithley 2000's. They really come in handy when your busy with small signals.
I't nice for pairing resistors too, but that's only about 0,002% acurate on 6,5 digits
alcoholic said:
Hi,
You can use the new CD63KI list, it's not final yet because i'm still working on that player, but it contains a lot of updated part numbers.
jimh0612 said:
Thanks, your welcome! I'm hoping to update the site soon, with more text and in-depth explanation of the modifications, if I can find the time...
Regards,
Ray
Or in fact reading through a few pages of this there may be better alternatives?
Now as I am quite pushed for free time at the moment and want to enjoy my CD-63 without risking breaking it, I am thinking about paying for a level 3 upgrade from Sercal. Is it worth the money? They do not have any C1 clocks available so have suggested I supply my own. Which is the best clock for this upgrade, and should I be paying out for it, or is it a waste of money for the components he fits?
ThanksThis is his page
Now as I am quite pushed for free time at the moment and want to enjoy my CD-63 without risking breaking it, I am thinking about paying for a level 3 upgrade from Sercal. Is it worth the money? They do not have any C1 clocks available so have suggested I supply my own. Which is the best clock for this upgrade, and should I be paying out for it, or is it a waste of money for the components he fits?
ThanksThis is his page
Ah, I might wait a while then. One thing I saw where the original filter components. Are you going to add new filter components for removing the bump in higher frequencies? I think C605/606 would be 560pF.
One thing I was wondering about, why use ZL as wel as ZLH instead of all ZLH? The price difference is verry small.
My new component list will set me back 200 euro. Oh whell, it's not much when compared to my last stereo upgrades
One thing I was wondering about, why use ZL as wel as ZLH instead of all ZLH? The price difference is verry small.
My new component list will set me back 200 euro. Oh whell, it's not much when compared to my last stereo upgrades
alcoholic said:Ah, I might wait a while then. One thing I saw where the original filter components. Are you going to add new filter components for removing the bump in higher frequencies? I think C605/606 would be 560pF.
One thing I was wondering about, why use ZL as wel as ZLH instead of all ZLH? The price difference is verry small.
My new component list will set me back 200 euro. Oh whell, it's not much when compared to my last stereo upgrades
No, for the moment i'm going to stick to the original filter characteristic first, and see how it sounds. Feel free to use the ZLHs instead of ZL, I probably had those in stock and that's how they ended up on the list.
Regards,
Ray
Kwak clock driving 74xx or comparator
SimontY, Poynton, Brent
Hi guys
Don't know if my main question got lost in my waffle but while we're talking clocks, Poynton mentioned a Kwak clock oscillator driving a 74xx chip instead of a comparator.
Would this be better than the standard Kwak clock, and if so how would I do it?
Finer points are a bit over my head I'm afraid
Thanks in advance
Jim
SimontY, Poynton, Brent
Hi guys
Don't know if my main question got lost in my waffle but while we're talking clocks, Poynton mentioned a Kwak clock oscillator driving a 74xx chip instead of a comparator.
Would this be better than the standard Kwak clock, and if so how would I do it?
Finer points are a bit over my head I'm afraid
Thanks in advance
Jim
You'd just feed the 74xxx (74HCU04 would work well) a clean p+5v suply, and stick the Kwak oscillators output into one of the inputs. Tap output signal from the output of said gate.
NO idea how this compares with Elso's fast comparator, but it's one less PSU rail to worry about. Both comparator and logic gate will demand very tight HF decoupling/ layout for best results; the 7ns comparator specified is a brute in terms of ground-bounce and so on.
But - in the CD63 you can omit this stage anyway. Just build the kwak oscillator section, and couple the resulting sinewave output into the DAC X1 input via a series 10nF film cap +47ohms (at the Kwak output.) This works well (it did for me, anyway) becausI've e the DAC's clock input is a logic gate input anyway - it subsequently drives the other distributed outputs and the conversion latch inside the dac.
Since the signal you're transferring is basically a sine it brings the least EMI/EMC problems, too.
DO provide the dacs logic supply Pin 28 with the cleanest, low impedance 5v supply you can. More http://www.acoustica.org.uk/t/63/63hacks.html]here[/url]
NO idea how this compares with Elso's fast comparator, but it's one less PSU rail to worry about. Both comparator and logic gate will demand very tight HF decoupling/ layout for best results; the 7ns comparator specified is a brute in terms of ground-bounce and so on.
But - in the CD63 you can omit this stage anyway. Just build the kwak oscillator section, and couple the resulting sinewave output into the DAC X1 input via a series 10nF film cap +47ohms (at the Kwak output.) This works well (it did for me, anyway) becausI've e the DAC's clock input is a logic gate input anyway - it subsequently drives the other distributed outputs and the conversion latch inside the dac.
Since the signal you're transferring is basically a sine it brings the least EMI/EMC problems, too.
DO provide the dacs logic supply Pin 28 with the cleanest, low impedance 5v supply you can. More http://www.acoustica.org.uk/t/63/63hacks.html]here[/url]
by Thomo:
Thomo, Thanks and yes, that is #1 on my simple/effective regulator list. That's why I asked if Andy had any updates I should know about. The post linked above is over 5 years old! I assume an opamp reg would be quieter, but no idea how much!
I'm so impressed with the clarity and detail after my tentlabs clock install! But I have lost the smooth balanced sound I am used to! (Now very sharp and bright, like highs are bumped up.)
Sambessey, they do have more clocks in stock at diyhifisupply.
Alc, can I change those even moreto knock the highs back?
Simonty sir,
Would my CD63SE sound benefit from running a second clock (coax) out from my tent x02 over to pin 13 on the decoder (clock input)?
As always all help greatly appreciated!
Thomo, Thanks and yes, that is #1 on my simple/effective regulator list. That's why I asked if Andy had any updates I should know about. The post linked above is over 5 years old! I assume an opamp reg would be quieter, but no idea how much!
I'm so impressed with the clarity and detail after my tentlabs clock install! But I have lost the smooth balanced sound I am used to! (Now very sharp and bright, like highs are bumped up.)
Sambessey, they do have more clocks in stock at diyhifisupply.
Alc, can I change those even moreto knock the highs back?
Simonty sir,
Would my CD63SE sound benefit from running a second clock (coax) out from my tent x02 over to pin 13 on the decoder (clock input)?
As always all help greatly appreciated!
Yep, but no updates. I've posted some more thoughts on how to get the best out of 317s here though:
http://www.acoustica.org.uk/t/3pin_reg_notes1.html
They can be 20-30dB quieter; 40dB is possible if you are really, really careful. Extracting the improvementsthat discrete regs offer depends heavily on layout and the steps you take to control error sources though.
Also note it's not 'low noise' alone that decides how 'good' a regulator is. Maintaining a low value of output impedance over a wide(r) bandwidth is, I think, rather more important.
Joe-HT said:Simonty sir,
Would my CD63SE sound benefit from running a second clock (coax) out from my tent x02 over to pin 13 on the decoder (clock input)?
Hi,
You're asking the wrong person. I do the mods but I don't necessarily understand how they work
I don't think anyone's done what you're asking about, but I believe Brent has been considering trying it. There may be some sort of synchronisation issue to consider.
Certainly to feed the DAC a clock is great, then to feed the servo with a clock provides huge gains again. This part will help to control your unruly top end. Use of Black Gate caps will also help in this regard. No need to change the filtering IME.
Regards
Simon
I know in other systems, the digital filter can pollute signals and neighbouring circuits' power rails quite easily, and it can be sensible to seperate its clock from others.
However, it is important that the clock going to the DAC, digital filter and decoder are in sync. This is why the clock is typically sent to the DAC, then from the DAC it is routed to the digital filter, and then on to the decoder, despite the possiblity of interference and jitter being added.
What you can probably do is have several outputs from your clock, and feed each one seperately to the DAC, digital filter and decoder (or let the digital filter still feed the decoder as an intermediate step, or if you're feeling less brave/bothered).
That's probably more sensible, and cheaper than an extra clock.
Cheers,
Phil
However, it is important that the clock going to the DAC, digital filter and decoder are in sync. This is why the clock is typically sent to the DAC, then from the DAC it is routed to the digital filter, and then on to the decoder, despite the possiblity of interference and jitter being added.
What you can probably do is have several outputs from your clock, and feed each one seperately to the DAC, digital filter and decoder (or let the digital filter still feed the decoder as an intermediate step, or if you're feeling less brave/bothered).
That's probably more sensible, and cheaper than an extra clock.
Cheers,
Phil
bridging +/- power pins of opamp with small cap?
right now i am running the ad8610 opamps in a cd 67se. my question is, should i bridge the +/- power pins on the brown dog adapters with a small value of capacitance to help prevent oscillation problems since these are faster chips than the originals or, could this maybe degrade the sound quality? i have on hand, .01 mf orange drops, .1 mf panasonic stacked films, and a few .01 mf Dayton film/foil caps. i know the four Elna Cerafine decoupling caps are very close to the opamps. has anyone tried this approach? Thanks for any advice!
right now i am running the ad8610 opamps in a cd 67se. my question is, should i bridge the +/- power pins on the brown dog adapters with a small value of capacitance to help prevent oscillation problems since these are faster chips than the originals or, could this maybe degrade the sound quality? i have on hand, .01 mf orange drops, .1 mf panasonic stacked films, and a few .01 mf Dayton film/foil caps. i know the four Elna Cerafine decoupling caps are very close to the opamps. has anyone tried this approach? Thanks for any advice!
philpoole said:
Exactly my thoughts!
And since the digital filter is integrated into the DAC on the cd63se, I just need one wire from my clock/reg pcb to the decoder. I am a bit gun shy! I can lift a jumper and a resistor to isolate the trace and try it! Worst case; no laser focus, right?
Hmmm, care to speculate why u193 at the decoder has a resistor instead of a jumper? That means there are two resistors between DAC and decoder. Should I keep the same resistance between new clock and decoder? I plan to use coax, for no particular reason...
Thanks for letting me think out loud!
BTW- martin, it was your meddling over at acoustica.org that got me started modding 3 years ago! Thanks!
and Simon, don't be so humble! Didn't someone say you're the godfather of this monstrosity?
I think this will work, but may not necessarily be better. No harm in trying though!
If I were you, I'd use coax but only ground it at the clock end. If you use two grounds you create a ground loop. If you use one ground (i.e. from DAC) you avoid this, but that ground must carry the return current from the decoder clock feed as well as the DAC clock feed. I'm not entirely sure why this is bad, but I was advised once (I think from Martin) that it is probably better to just feed the DAC with your clock and let the DAC feed the decoder.
I may have misunderstood though.