Hi jean-paul
reading the SAA7220 bit you gave (thanks), I see that it specifies that the filter compensates for a 3rd order Bessel filter.
This means that the output cannot be used directly in a standard (say 1500) I/V resistor, such as the 1543 I have tweaked. In fact I do find the highs a bit far away. So this might be why.
Can that be the reason?
(The 1543 NOS version I have that bypasses the 7220 is too bright, the opposite in sound, and it has a bright spot in the center).
reading the SAA7220 bit you gave (thanks), I see that it specifies that the filter compensates for a 3rd order Bessel filter.
This means that the output cannot be used directly in a standard (say 1500) I/V resistor, such as the 1543 I have tweaked. In fact I do find the highs a bit far away. So this might be why.
Can that be the reason?
(The 1543 NOS version I have that bypasses the 7220 is too bright, the opposite in sound, and it has a bright spot in the center).
Noise on only one channel........
Hmmm.........
Having built hundreds of Magnavox CDPs, the '1541A is the weak link. They get noisy and/or distorted when they are fixin' to fail. Take a heat gun and freeze spray to one............you'll find out. At one time...........on a slow day.......I graded all of the '1541A chips in the shop based on what temperature they crapped out at.
Only replaced one '7220 in all those years.
Jocko
Hmmm.........
Having built hundreds of Magnavox CDPs, the '1541A is the weak link. They get noisy and/or distorted when they are fixin' to fail. Take a heat gun and freeze spray to one............you'll find out. At one time...........on a slow day.......I graded all of the '1541A chips in the shop based on what temperature they crapped out at.
Only replaced one '7220 in all those years.
Jocko
triode_al said:
No, The 1541 does not like DC at the outputs, therefore a resistor of 1500 will give distortion, since i-out is -2mA at digital silence. A huge resistor gives too much dc.
For a 1541 you should stay under 50 ohms, maybe even lower.
This means that the voltage over the resistor is little, so you need to amplify it again. But it also means that the dc is low enough not to give distortion.
Correct Guido,
Apologies 😱 that my Dinglish sentence gave the wrong impression: the 1500 ohms is 1543 only; 1541 should have some quite lower, indeed maybe 10 ohms if one likes to amplify the resulting milliVolt to microVolt signal. But of course anyone who made a MM-input to a RIAA stage can do that. Maybe I'll try it myself!
I have had a 1541 pop out on one channel just like Jocko describes. My fault: soldering errors of course that could not be remedied.
Apologies 😱 that my Dinglish sentence gave the wrong impression: the 1500 ohms is 1543 only; 1541 should have some quite lower, indeed maybe 10 ohms if one likes to amplify the resulting milliVolt to microVolt signal. But of course anyone who made a MM-input to a RIAA stage can do that. Maybe I'll try it myself!
I have had a 1541 pop out on one channel just like Jocko describes. My fault: soldering errors of course that could not be remedied.
Re: Noise on only one channel........
You melted them all and then selected the ones that went glass transparent at a higher temperature?
😀 😀 😀
Jocko Homo said:
You melted them all and then selected the ones that went glass transparent at a higher temperature?
😀 😀 😀
Not quite, bub......
If you got them hot enough, they all became noisy. I picked out the ones that became noisy at temperatures close to operating point. (My stuff tends to run hot..........all those current sources and such in that fancy-schmancy I/V stage that I came up with.)
Jocko
If you got them hot enough, they all became noisy. I picked out the ones that became noisy at temperatures close to operating point. (My stuff tends to run hot..........all those current sources and such in that fancy-schmancy I/V stage that I came up with.)
Jocko
the zanden cd player does! WOW !!!
Check out www.zandenaudio.com and read through the specs of their cd player.
Check out www.zandenaudio.com and read through the specs of their cd player.
In a previous post I saw it mentioned that the CD850 Philips uses a TDA1541... Not true I believe. It was I think the first bitstream player by Philips. I owned one. No TDA1541...
Mission PCMII uses TDA1541A if not mentioned yet... Also Arcam Alpha (original) and the Alpha Plus as well as their Black Box 1 & 2 DACs.
There was also a Sherwood, CD3020 I think using a Sony mech like Adcom did.
There were also some Dual CDPs that were actually Rotel RCD855 players thus TDA1541 as well.
The Quad CD66 btw used a standard looking Philips board with LM833 or TLO72 but always sounded better then the Rotel 855/955 to me.. Found out recently why this may have been when I peeled the "Quad Selected" sticker of a dac chip. It is an S1. Quad is legendary on spares (well they were when they were Brittish) so they may be a source for TDA1541A S1 chips...
Can not think of any other machines not already listed for now.
Guillaume
Mission PCMII uses TDA1541A if not mentioned yet... Also Arcam Alpha (original) and the Alpha Plus as well as their Black Box 1 & 2 DACs.
There was also a Sherwood, CD3020 I think using a Sony mech like Adcom did.
There were also some Dual CDPs that were actually Rotel RCD855 players thus TDA1541 as well.
The Quad CD66 btw used a standard looking Philips board with LM833 or TLO72 but always sounded better then the Rotel 855/955 to me.. Found out recently why this may have been when I peeled the "Quad Selected" sticker of a dac chip. It is an S1. Quad is legendary on spares (well they were when they were Brittish) so they may be a source for TDA1541A S1 chips...
Can not think of any other machines not already listed for now.
Guillaume
There are also interesting standalone DAC units such as the Philips/Marantz DFA888 which houses a TDA1541A. This is an 'integrated ' amplifier.
The reciever by the way is a Yamaha YM3623 chip.
The reciever by the way is a Yamaha YM3623 chip.
Marantz PM75 amp also has TDA1541(A) built in....
Van Den Hul made a few custom tube pre amp/ dacs using TDA1541A that he claims to have handselected..... (Not sure if he started with crowns or not, or hand selected from standard stock.)
Van Den Hul made a few custom tube pre amp/ dacs using TDA1541A that he claims to have handselected..... (Not sure if he started with crowns or not, or hand selected from standard stock.)
A current source is possible, and the resistor you mention can perform such a function of course. Though a current source is probably more un-heard (?).
This is what Pedja for example does: a FET is used, see J1 (2SK170). I add Pedja's scheme (I hope he doesn't mind, it is an old schema of his, and just for illustration on the current source.
This is what Pedja for example does: a FET is used, see J1 (2SK170). I add Pedja's scheme (I hope he doesn't mind, it is an old schema of his, and just for illustration on the current source.
Bernhard said:
Eh, 5/2mA=2k5?
At idle (2mA) that's ok, but with music (0 - 4mA) that's not going to work, the dc component is back (+5V - -5V).
Think i need to correct myself: it does not like a voltage at the outputs. Opamp i/v keeps it at virtual ground. Resistor i/v does not, so the resistor needs to be small to keep the voltage low.
AC (music) or DC. A to high voltage difference between gnd and output gives distortion. So not only at dig zero, also with music.
As far as i know Pedja's design injects 2mA with the fet to compensate for the -2mA at dig0. So change the 'dac' from 0to-4 to +2to-2mA outputs. The output stage does not have to handle the 2mA current at dig 0.
AC (music) or DC. A to high voltage difference between gnd and output gives distortion. So not only at dig zero, also with music.
As far as i know Pedja's design injects 2mA with the fet to compensate for the -2mA at dig0. So change the 'dac' from 0to-4 to +2to-2mA outputs. The output stage does not have to handle the 2mA current at dig 0.
On this website is written the same, TDA1541 does not work with large I/V resistor value, but others do, like PCM61 and perhaps PCM56.
http://pw2.netcom.com/~wa2ise/radios/tubedac.htm
http://pw2.netcom.com/~wa2ise/radios/tubedac.htm
guido said:
I use 33R.
And "some" gain on the output stage.😀
OPA637, biased to class-A.
And it sounds good.😱
- Status
- Not open for further replies.