| ashok |
Hi All,
In the near future DVD-A will probably be more popular than the 'older' audio CD. With DVD machine prices crashing it seems inevitable. DVD-A disc prices however may be the deciding factor.
For the DIY'ers , is there any chip that decodes 24 /96 as well as our old TDA1541A for 16/44.1 ? Basically that means is there any 24/96 chip equivalent to the TDA1541A in performance.
There aren't too many manufacturers and so the list of chips must be limited. Can anyone comment on a 'known good chip'.
Thanks.
Cheers. |
|
|
| arniel |
Hi,
Do you mean, in terms of specifications? Sure, almost any modern DAC will exceeed the specs (except some low-end ones), but in terms of sound quality, who knows!? That's a very subjective issue. |
|
|
| mcai7et2 |
I am also looking for a really musical, but realistically simple DAC to accept SPDIF and I2S inputs to fit into a CDP/DAC/Preamp project.
I have looked at the data sheets for the PCM1704 from TI, and I think that that could be manageable, providing someone with more talent with me can design and manufacture PCBs based on the data sheet.
This is the DAC used in the latest NAIM players and by all accounts is a great piece of kit. |
|
|
| Jocko Homo |
I made some boards that replaced both the SAA7220 and TDA1541 with B-B '1704 series parts.
And in case your are new.....I really hate both of thoses parts.
Jocko |
|
|
| tbla |
| quote: | | And in case your are new.....I really hate both of thoses parts. |
:att'n: how could we forget.....;) |
|
|
| ashok |
Hi Arniel,
I am sure they spec better than the old dac's. I want to know which one "sounds" best - subjective or not.
We will get many opinions . We can plonk for the one that most like ! It could be wrong but it's a good starting point.
Cheers. |
|
|
| arniel |
Ashok,
I understand your question... as you can probably guess, I am in no position to give you an answer!!!
This exercise is so challenging, partly because swapping DACs in the same circuit may not give optimised results. I've been looking at replacing the DAC in my Marantz CD42 for years now, and am still no closer to coming up with a definitive answer e.g. does one go for transformer coupled outputs, tube buffers, 5th order bessel filters, GIC filters??? And that's just the output stage...
Sorry this has been such a useless answer, and good luck!
Arnie |
|
|
| peufeu |
| quote: | Originally posted by Jocko Homo
And in case your are new.....I really hate both of thoses parts.
Jocko [/B] |
You mean, you hate PCM1704 ?
I was considering using it... can you be more precise ? I don't want to waste time and money on something flawed...
thanks |
|
|
| rfbrw |
| quote: | Originally posted by peufeu
You mean, you hate PCM1704 ?
I was considering using it... can you be more precise ? I don't want to waste time and money on something flawed...
thanks |
If you search the forum you will find JH hates the SAA7220 and the TDA1541. |
|
|
| Elso Kwak |
| quote: | Originally posted by ashok
Hi All,
In the near future DVD-A will probably be more popular than the 'older' audio CD. With DVD machine prices crashing it seems inevitable. DVD-A disc prices however may be the deciding factor.
For the DIY'ers , is there any chip that decodes 24 /96 as well as our old TDA1541A for 16/44.1 ? Basically that means is there any 24/96 chip equivalent to the TDA1541A in performance.
There aren't too many manufacturers and so the list of chips must be limited. Can anyone comment on a 'known good chip'.
Thanks.
Cheers. |
Hi ashok, Most modern DACs are backwards compatible meaning they can work on redbook CD's.
But to me it seems most manufacturers have taken the Delta-Sigma route. If that is sounding like TDA1541AS1?
:confused: |
|
|
| peufeu |
By the way, I'de be interested in that thread too. I have built a DAV with TDA1545 which sounds fantastic, and a friend has very good results with AD1875 (which shows even more potential, maybe...)
So is there a chip that sounds as good as the good old Multibits, while taking 24-96 ?
There is PCM1704, never heard it though... |
|
|
| Jocko Homo |
"As good as the (good) old multibit"?
Who ever said that they sound good?????? Not me.
Go buy the '1704. If you think that it costs too much, buy a AD1862.
Jocko......(who has both.....and the $$$$ tied up in surplus inventory to prove it.) |
|
|
| peufeu |
| quote: | Originally posted by Jocko Homo
Go buy the '1704. If you think that it costs too much, buy a AD1862.
|
So you mean you like 1704 a lot better than the other multibits, which quite answers my question. Now do you prefer AD1862 over it ? |
|
|
| Terry Demol |
| quote: | Originally posted by Jocko Homo
"As good as the (good) old multibit"?
Who ever said that they sound good?????? Not me.
Go buy the '1704. If you think that it costs too much, buy a AD1862.
Jocko......(who has both.....and the $$$$ tied up in surplus inventory to prove it.) |
Jocko,
Many people feel the PCM63 was the best of all those BB
DAC's. We used the '63 and disabled bipolar offset internal
compenasation current (2mA) and found significant improvement.
This is a worry WRT the 1704 as you can't disable it. However
the 1704 does measure slightly better. Have you done direct
comparison with 63 and 1704???
What DF did you use, we used PMD100 (HDCD) and now are
looking at running 63 in 0 x OS :-)
The PMD100 has great detail and resolution but I think it
may be a bit unmusical. I remember trying DF1704 and there
was mixed opinion compared to PMD100..... it's a can of worms!
Terry |
|
|
| peufeu |
| quote: | Originally posted by Terry Demol
Many people feel the PCM63 was the best of all those BB
DAC's. We used the '63 and disabled bipolar offset internal
compenasation current (2mA) and found significant improvement. |
What did you use for IV conversion ? A simple resistor ?
Maybe the internal current source did not like the variations in output voltage caused by the resistor ?
I think the Folded Cascode circuit is much superior to a simpel resistor... |
|
|
| ashok |
Thanks to all for your replies.
Hi Peufeu,
Can you post your folded cascode design ? I can build it and compare it with my resistor loaded TDA1541 output. I have been looking for alternative solutions but never could decide on one. I also agree that the ideal situation would be to have a virtual zero ohm load at the current output .My load is 50 ohms but might make an audible difference if that was a virtual zero. Reducing the load to 10 ohms would be throwing away precious signal.
Will be waiting to see your folded cascode solution.
Cheers. |
|
|
| peufeu |
Well, it's not exactly "mine" (Rudolf Broertjes and I thought about it as the same time) but here it is :
link
Try the third version. I have not tried it yet, but the PCBs are ordered already... want an Eagle file for your PCB ?
With it you can use 1k load and get more signal ! |
|
|
| jean-paul |
Did I hear Eagle files !?!? ;) Please post them here, saves us a lot of time designing one.
I thank you in advance,
Jean-Paul |
|
|
| Terry Demol |
| quote: | Originally posted by peufeu
What did you use for IV conversion ? A simple resistor ?
Maybe the internal current source did not like the variations in output voltage caused by the resistor ?
I think the Folded Cascode circuit is much superior to a simpel resistor... |
Jocko,
We went away from R I-V years ago because the '63 has more
distortion with increased OP compliance. You can directly
compare this by using a good OPA I-V with something like a
627 and add varying resistance in front of it to simulate an R
I-V. The '63 starts to change it's sound with as little as 15R.
WRT Folded Cascode topology, I agree. We have built many of
those cascode I-V's in various topologies with Jfets, BJT's MOS,
complementary, SE, folded, CFP, non CFP, etc etc.
However to answer your question, when we first checked the BPO
phenomenom it was with discrete CFB opamp I-V (closed loop).
There was no voltage swing at dac OP, and it still degraded the
sound, quite strange.
Terry |
|
|
| Jocko Homo |
Who's folded cascode???????
Never compared the '63 to the '1704.
'1704 vs. '1862.
The '1862 was the last multibit that AD made, so that only leaves B-B as a supplier of anything higher than 20 bits.
Did compare the '1702 vs the '1862. Yeah, they sounded a bit diferent. Not enough for me to lose sleep over. I had lots of '1862s around (still do), so we made few '1702 versions. Just a few for the hell of it.
Filters:
No way to shove a PMI one in my designs (maybe I shoud have taken that job offer........). Anyway....we were able to compare a CDP with that HDCD **** to some of ours. Once I replaced the stock clock with a better one, they were close. As I refined the modified clock, the one with the '1704 was better. The first clock had a bit of sibilance.That went away with the better one.
The HDCD was more veiled, and on some really nasty sounding CDs, it mucked things up enough as to make some people feel that is was more pleasing to listen to/suffer through.
Long story short:
Get rid of op-amps.
Get a good clock.
The rest contributes less.
Jocko |
|
|
| Terry Demol |
| quote: | Originally posted by Jocko Homo
Who's folded cascode???????
|
Geeze, I think it must have been ours ;)
When I first saw your GB stage here we had already
implemented and tested about 10 different topology iterations
of this I-V approach :)
| quote: |
Never compared the '63 to the '1704.
|
Me neither, however I have a feeling the newer
dacs such as AD1955 may be better anyway.
I have a proto nearing completion with 3 x AD1853
in parallel just to see how these newer generation go
against the old "Chevs" such as 1541A/S1 and 63K
| quote: |
Filters:
No way to shove a PMI one in my designs (maybe I shoud have taken that job offer........). Anyway....we were able to compare a CDP with that HDCD **** to some of ours. Once I replaced the stock clock with a better one, they were close. As I refined the modified clock, the one with the '1704 was better. The first clock had a bit of sibilance.That went away with the better one.
|
That was our feeling also. Did you use slow rolloff on
DF1704 ?
| quote: |
The HDCD was more veiled, and on some really nasty sounding CDs, it mucked things up enough as to make some people feel that is was more pleasing to listen to/suffer through.
|
Yeah, I think that PMD100 just has something about it that
is unmusical. This is playing non HDCD discs.
And then we've got the latest batch of super ASRC's such as
AD1896.... life gets too bloody complicated.
Next life it's definately going to be collecting stamps :)
Terry |
|
|
| peufeu |
| quote: | Originally posted by Terry Demol
Geeze, I think it must have been ours ;)
When I first saw your GB stage here we had already
implemented and tested about 10 different topology iterations
of this I-V approach :)
|
I think this schematic is older than the Net itself... wouldn't be surprised to find some in Tektronix schematics from the sixties...
| quote: |
I have a proto nearing completion with 3 x AD1853
in parallel just to see how these newer generation go
against the old "Chevs" such as 1541A/S1 and 63K
|
I'm very interested un that ! Can you post your progress ?
Thanks ! |
|
|
| Jocko Homo |
Never tried the slow roll-off. Did you???
FWIIW:
Most of the I/V desings that you guys are talking about here are variations of mine......which was done back in '89. The guys that you are refering to have corresponded with me long enough to get most of the hints.
So.....they get the credit for having the initiative and imagination to build on what info that I have supplied them.
I will claim the credit for having one of the first CDPs on the market that used one. At the time, there were about 3-4 other companies that used something similar. No idea who was actually first.
A dealer in Chicago thinks that he gave Mark Brasfield (MSB Designs) the idea for his by telling him about mine. Who knows, or cares.......
Jocko |
|
|
| tbla |
| quote: | | I will claim the credit for having one of the first CDPs on the market that used one |
...what brand...? :deerman: |
|
|
| Elso Kwak |
| quote: | Originally posted by tbla
...what brand...? :deerman: |
What did you eat or drink Troels???
:confused: |
|
|
| mcai7et2 |
How about a really simple PCB design for this DAC, one which people could add their own I/V stage and analog filtering to?
I think there would be a lot of interest in a board with a very simple, modern DAC design which they could build into either an existing CDP or as a stand alone unit. |
|
|
| tbla |
| quote: | | What did you eat or drink Troels??? |
nothing.....YET. :apathic: |
|
|
| Blitz |
| quote: |
I have a proto nearing completion with 3 x AD1853
in parallel just to see how these newer generation go
against the old "Chevs" such as 1541A/S1 and 63K
|
Any news on this ? |
|
|
| Terry Demol |
| quote: | Originally posted by Blitz
Any news on this ? |
Just got back from 2 weeks holidays camping
down in Victoria. Gotta get my brain back into gear:)
There is still a lot of work to do.
Cheers
Terry |
|
|
| Blitz |
| quote: | Originally posted by Terry Demol
Jocko,
Many people feel the PCM63 was the best of all those BB
DAC's. We used the '63 and disabled bipolar offset internal
compenasation current (2mA) and found significant improvement.
Terry |
How did you do that ? I have a sheldon DAC with passive I/V and wonder if he did it as well... |
|
|
| Terry Demol |
| quote: | Originally posted by Blitz
How did you do that ? I have a sheldon DAC with passive I/V and wonder if he did it as well... |
Ref to data sheet, disconnect pin 5 (BPO) connection to pin 6
(I out). There will now be 2mA offset to compensate for.
Cheers,
Terry |
|
|
| Blitz |
| Ok, the sheldon DAC has them separated with pin 5 to ground, which is I guess is the setup you use as well, right ? |
|
|
| Terry Demol |
| quote: | Originally posted by Blitz
Ok, the sheldon DAC has them separated with pin 5 to ground, which is I guess is the setup you use as well, right ? |
That's right
T |
|
|
|
