Re: Re: Naim CD 555?
CD12 use PMD100+PCM1702x4 and cheaper than CD555. And the price is about $20000. But also sounds very good. The jitter is very low.
Here is CD12 test result:
http://stereophile.com/cdplayers/86/index6.html
and CD555 test result:
http://stereophile.com/cdplayers/207naim/index4.html
Telstar said:
CD12 use PMD100+PCM1702x4 and cheaper than CD555. And the price is about $20000. But also sounds very good. The jitter is very low.
Here is CD12 test result:
http://stereophile.com/cdplayers/86/index6.html
and CD555 test result:
http://stereophile.com/cdplayers/207naim/index4.html
I will give the PCM1704 one more try
.
In the meanwhile you can all put a direct out mod and dac swap to AK4395 (only $7.50) into your DEQ2496s (everyone with a dac has one of these by now right) and let me know what you think about my favorite chip.
Seeing now that there is such a devout group of supporters for the PCM1704 I will need to mod my Dac 2.7 further before letting it go so that I can resolve, at least in my own mind, that I am hearing a better example of what the 1704 can really do before commenting further. It is a simple matter to get rid of the opamp based I/V, dc servo, and filter with a Sowter transformer. I would also consider piggybacking some Ktypes but they are $30.00! Better wait on that to see if it actually gets a lot better.sendler said:
.
In the meanwhile you can all put a direct out mod and dac swap to AK4395 (only $7.50) into your DEQ2496s (everyone with a dac has one of these by now right) and let me know what you think about my favorite chip.
Hi Scott,
I'm not so sure that piggybacking these is a wise move. They are "colinear" and summed inside already to reduce R ladder errors. In the Denon DCD S-10, there are two PCM1702 for each channel, run opposite polarities to be summed again. From what I'm hearing, this seems to be extremely effective. I would think this is better than trying to parallel their operation. Besides, if parallel operation was better, don't you think they would have used them that way? I'm sure this was experimented with.
The PCM1702 and PCM1704 seem to be so close as to be the same part in reality. I'm sure what can be said for one would be the same for the other in practical terms.
-Chris
I'm not so sure that piggybacking these is a wise move. They are "colinear" and summed inside already to reduce R ladder errors. In the Denon DCD S-10, there are two PCM1702 for each channel, run opposite polarities to be summed again. From what I'm hearing, this seems to be extremely effective. I would think this is better than trying to parallel their operation. Besides, if parallel operation was better, don't you think they would have used them that way? I'm sure this was experimented with.
The PCM1702 and PCM1704 seem to be so close as to be the same part in reality. I'm sure what can be said for one would be the same for the other in practical terms.
Not bad when you consider what they are worth as spare parts from the service departments. Besides, industrial DACs can get way up there as well.
-Chris
I would like to thank all of you for the conributions, i am still - like many among us- not convinced that the newer generation chips are really better in anything beside cost. Therefore i ordered some parts and i' ll try to implement a simple DF1704-
pcm1704 dac.
I find the transformer I/V approach appealing too, but because of the very high frequencies in the dac output stream, i want first to be sure that the trans doesn't act like a short circuit somewhere in the mhz region and the core doesn't saturate because of them, causing extreme distortion.
Any links with measuments?
Greetings from Athens (Greece)
Konstantinos
pcm1704 dac.
I find the transformer I/V approach appealing too, but because of the very high frequencies in the dac output stream, i want first to be sure that the trans doesn't act like a short circuit somewhere in the mhz region and the core doesn't saturate because of them, causing extreme distortion.
Any links with measuments?
Greetings from Athens (Greece)
Konstantinos
anatech said:
Hi Chris,
Please can you suggest an alternative transformer than the Sowter 8347 (one 1702 x channel) preferibly european ?
The sowter seems nice to my inexpert eyes but it can not tolerate
a I/V resistor > 25ohms
( about 380mV on the secondary I believe) I would be "extract" more V RMS from my 1702, ideally 2000mv
is it possible?
Thanks
Paolo
Hi Paolo,
A solid state circuit.
If the transformer isn't absolutely top notch, it will be far worse in performance to a normal I/V converter.
Thinking about the transformer idea, I wonder. The output of the DAC is a current. A transformer will reflect impedances back, and this will probably vary. I have to say that I'm not convinced you will not have a very colored sound this way. I'm going to bet that proponents of doing this will call it a more natural sound.
Why is it that if something is simple, but distorted, "we" always fall back on the claim that it is more natural and true fidelity can not be measured? Well, we can measure distortion.
-Chris
Yes.
A solid state circuit.
If the transformer isn't absolutely top notch, it will be far worse in performance to a normal I/V converter.
Thinking about the transformer idea, I wonder. The output of the DAC is a current. A transformer will reflect impedances back, and this will probably vary. I have to say that I'm not convinced you will not have a very colored sound this way. I'm going to bet that proponents of doing this will call it a more natural sound.
Why is it that if something is simple, but distorted, "we" always fall back on the claim that it is more natural and true fidelity can not be measured? Well, we can measure distortion.
-Chris
The "24/96" versions of the Pass D1 ran piggybacked PCM1704-k's to boost the current output to a level similar to - but slightly higher than - the pcm63's. If you piggyback the DAC's, squeeze in a D1 I/V stage (which runs best at the higher current) and some quiet +/-30V supplies you'd have a relatively cost effective upgrade...
anatech said:
Hi Chris,
Ok I agree. Thus in your opinion 8347 is not top notch. Who is this top notch transformer? Lundhal ? wich model? Others?
I understand nada of transformeres ... :-(
Absolutely agree also here. I am not serching to "euphonize" the sound, I am just searching the best possible I/V converter
Curious to see others experts says that transformers are the right way......This is audio after all !
What's wrong with passive I/V conversion made with a little ( or very little resistor) ? I have not fear to amplify a lot the little signal, I am accustomed with phono stages !
Why is it that if something is simple, but distorted, "we" always fall back on the claim that it is more natural and true fidelity can not be measured? Well, we can measure distortion.
-Chris [/B]
Perfect agree also on this point.
About active I/V converters, is it good idea or not make it with a triode like E99CC ?
I am interested a lot into the techical reasons here, because I am searching to learn how these stages really works .
Thanks,
Paolo
Hi Paolo,
I haven't the experience that some other members have here. I have never designed and built an I/V converter yet. I have serviced CD players since they came out though.
I guess I could best give some guidelines that others may or may not agree with. You should generate the largest signal within the compliance range of that converter. Whether you use the output as a current or a voltage probably doesn't matter as much as the requirement that it remains a linear signal. We are not transferring power in a strict sense, just the information. Therefore, do this in the most linear fashion you are comfortable working with. Along with this, make sure you don't add more noise than you need to.
I've worked with signal transformers for years. What I know is that you don't use one without a specific reason to do so. So if you need to float one section from another, a transformer may be the right tool for you. Same with any component. It also follows that a tube may or may not be a good tool to use for conversion. This may go either way depending on the tube type you do use.
In a nutshell, generate the largest signal you can without running into clipping or distortion. This will reduce your amplification requirements (and noise). Don't focus on one aspect so much that other factors suffer. This is where many budding engineers in audio really mess up.
-Chris
I haven't the experience that some other members have here. I have never designed and built an I/V converter yet. I have serviced CD players since they came out though.
I guess I could best give some guidelines that others may or may not agree with. You should generate the largest signal within the compliance range of that converter. Whether you use the output as a current or a voltage probably doesn't matter as much as the requirement that it remains a linear signal. We are not transferring power in a strict sense, just the information. Therefore, do this in the most linear fashion you are comfortable working with. Along with this, make sure you don't add more noise than you need to.
I've worked with signal transformers for years. What I know is that you don't use one without a specific reason to do so. So if you need to float one section from another, a transformer may be the right tool for you. Same with any component. It also follows that a tube may or may not be a good tool to use for conversion. This may go either way depending on the tube type you do use.
In a nutshell, generate the largest signal you can without running into clipping or distortion. This will reduce your amplification requirements (and noise). Don't focus on one aspect so much that other factors suffer. This is where many budding engineers in audio really mess up.
-Chris
Thanks Chris, very precious raccomandations for me!
At least I'm luck in one thing: I have no difficult to evalutate ( in terms of who sound better)two different schematics, it is my best attitude
So I have "only"to build two different architectures and listen both.
Mission possible!
Cheers,
Paolo
At least I'm luck in one thing: I have no difficult to evalutate ( in terms of who sound better)two different schematics, it is my best attitude
So I have "only"to build two different architectures and listen both.
Mission possible!
Cheers,
Paolo
Sowter transformers
I plan to try the Sowter transformers unless someone who has actually heard it can recommend a better transformer. This mod could be built in 5 minutes.
.
http://www.sowter.co.uk/specs/8347.htm
.
I plan to try the Sowter transformers unless someone who has actually heard it can recommend a better transformer. This mod could be built in 5 minutes.
.
http://www.sowter.co.uk/specs/8347.htm
.
Ah, understood!
For the moment I will keep the stock Denon configurations ( one 1702 for channel only); I am not able to make those modify..... CS mother boards are not in mine possibility for now
Just curious : are the two DACS , brutally in parallel ?
If 2 is better than one, why not 4 or 8 and so on?
P.S.: with two DACS and double of I/V resistor, what is the output level Volts RMS? 2X or 4X ?
Cheers,
Paolo
For the moment I will keep the stock Denon configurations ( one 1702 for channel only); I am not able to make those modify..... CS mother boards are not in mine possibility for now
Just curious : are the two DACS , brutally in parallel ?
If 2 is better than one, why not 4 or 8 and so on?
P.S.: with two DACS and double of I/V resistor, what is the output level Volts RMS? 2X or 4X ?
Cheers,
Paolo
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.