Denon DCD S10 (2 AD8620, 4 opa627BP, 8 opa637BP, Elso kwak clock 6)
Rotel 855 (non/os, 2 AD8620 ,Elso kwak 7)
Magnavox CD650(non/os, 2 AD8620, TDA1541A S1)
When I listen some violin CD, I fell the Rotel 855 and Magnavox CD650 are sweeter than the Denon DCD S10, but listen femal voice the Denon DCD S10 is less harsh than the Rotel 855 and Magnavox CD650. Will the TDA1543 nos/os be less harsh than TDA1541A nos/os?
Rotel 855 (non/os, 2 AD8620 ,Elso kwak 7)
Magnavox CD650(non/os, 2 AD8620, TDA1541A S1)
When I listen some violin CD, I fell the Rotel 855 and Magnavox CD650 are sweeter than the Denon DCD S10, but listen femal voice the Denon DCD S10 is less harsh than the Rotel 855 and Magnavox CD650. Will the TDA1543 nos/os be less harsh than TDA1541A nos/os?
I just changed the two NE5532s on non/os Rotel 855 and Magnavox CD650. Should I build a new I/V stage?
When you make a TDA15xx cdplayer non os the analog filter has to be recalculated. Use the Search function for more info, there are threads about the subject. The harshness comes very likely from the unchanged filters, it has nothing to do with the used chip.
When you're brave please listen to the cdplayers once without analog filtering. So use the opamps but remove filter components. When the harshness is gone you have an indication you're on the right path.
Warning: I wouldn't recommend to listen without analog filtering permanently. There could be ( there are in fact ) some RF components in the output signal that might disturb your amp. That being said I can tell I use only 6 dB/oct filtering with TDA1543 non os DAC's and I get away with it. Although I haven't built it, the discrete outputstage Rbroer designed looks promising and it is simple to build. Should be better than a opamp stage anyway.
When you're brave please listen to the cdplayers once without analog filtering. So use the opamps but remove filter components. When the harshness is gone you have an indication you're on the right path.
Warning: I wouldn't recommend to listen without analog filtering permanently. There could be ( there are in fact ) some RF components in the output signal that might disturb your amp. That being said I can tell I use only 6 dB/oct filtering with TDA1543 non os DAC's and I get away with it. Although I haven't built it, the discrete outputstage Rbroer designed looks promising and it is simple to build. Should be better than a opamp stage anyway.
Crossover point?
Hi Jean-Paul,
What is the crossoverpoint (-3dB) of your 6dB/octave filter?
😕
jean-paul said:
Hi Jean-Paul,
What is the crossoverpoint (-3dB) of your 6dB/octave filter?
😕
jean-paul said:
That's being said, I'm perfectly happy without any filtering at all in my NOS TDA1543 DAC. There is no sign of harshness or fatiguing or irritation, totally smooth highs.😉
mikeliu said:
Hi Mikeliu,
I was never able to elicit any harsh sounds from the TDA1541! In fact I found it too mellow, "polite", uninvolving....
The TDA1541 does not sound harsh, NON-OS, in my Phlips CD-650 with orginal analog filtering still implemented, that is for oversampling.😎
If you find the TDA1541 harsh you will find the TDA1543 unbearable.......
I am not familiar with the AD8620 yet. May I suggest using the OPA2604 instead? Especially as a IV-converter it can't be beaten I think. I tried a lot of IC's !😉
Just a thought:
You did lower the supply voltage to +/- 12V for the AD8620 in these players?
Then you should try AD8620 I think. Please don't suggest other chips when you don't know the used type. Leave it where it is Mikeliu, AD8620 is very good.
Very good point as most cdplayers use +/- 15 V which is too much for AD8620. I assume you did correct the supply voltages otherwise you probably have the main reason for the nasty sound. Nevertheless I think it is better to recalculate the analog filter ( or leave it out when you like to live dangerous ).
Elso, one DAC has a filter with -3 dB at 48 kHz just to prevent very high frequencies from going in my amp. Never bothered to implement sinx/x filtering. I don't have any problem with the high roll off.
To be honest, most of my TDA1543 dac's didn't have filtering at all. Maybe I put 2 caps at the output of that series just for peace of mind 😉
How can you suggest a in your opinion good type when a even better type is used ? ( that you don't know ). I know OPA2604 too and although it is good ( but not superb, only OPA627/637 get that qualification if you ask me ) AD8620 is better IMO. Too bad it is only available in SMD. Since I know your love for SMD I think you'll never try it 😀
Building discrete is the best solution however if I have to believe people that design discrete I/V stages for TDA1543. When the hot weather has gone I will start with non os DAC's again. I sold all of them including the prototypes for I had enough of building them. That doesn't mean I don't hear them anymore. I would like to build one with TDA1541A and discrete I/V when time permits.
Building discrete is the best solution however if I have to believe people that design discrete I/V stages for TDA1543. When the hot weather has gone I will start with non os DAC's again. I sold all of them including the prototypes for I had enough of building them. That doesn't mean I don't hear them anymore. I would like to build one with TDA1541A and discrete I/V when time permits.
Better or Better
Hi Jean-Paul,
Oooh, it is all a matter of taste. I did not say the OPA604 is better, just that I have good results with it, meaning I like the sound.
I did not like the OPA627 at all for IV-conversion.
If you find AD8620 better OK with me.
I have two samples of the AD8610 (single version of the AD8620) waiting to try.
Here and on the Audio Asylum there are endless threads about what is the "best" IC. People will never achieve consensus about that.
I am not too fond of SMD but my Asynchronous Reclocker uses three MC74VHC74 in SMD. And my 45 MHz KWAK-CLOCK uses the AD8611 also only avalable in SMD. I can handle SMT, but you must realize it was intended for machine insertion.....
🙄
jean-paul said:
Hi Jean-Paul,
Oooh, it is all a matter of taste. I did not say the OPA604 is better, just that I have good results with it, meaning I like the sound.
I did not like the OPA627 at all for IV-conversion.
If you find AD8620 better OK with me.
I have two samples of the AD8610 (single version of the AD8620) waiting to try.
Here and on the Audio Asylum there are endless threads about what is the "best" IC. People will never achieve consensus about that.
I am not too fond of SMD but my Asynchronous Reclocker uses three MC74VHC74 in SMD. And my 45 MHz KWAK-CLOCK uses the AD8611 also only avalable in SMD. I can handle SMT, but you must realize it was intended for machine insertion.....
🙄
It is, I am sure a lot of people don't like OPA604/2604 at all. I only objected to the comparison/suggestion that's out of balance.
One has to know both to do a suggestion was my point.
First time I hear that. Never found a better one. In CD80 I did extensive testing with all DIL types I know and it was absolutely the best. It needs some time to settle though and it needs adequate decoupling and a proper supply ( don't they all ? ).
We agree, let the machines solder them. Don't like them either but slowly but steady we are forced to use them since the market demands them ( read: they are cheaper to produce ).
Thank everyone. Today I replaced the two AD8620 with Opa2132BP and Opa627, and compared with Denon S10 the harsh still is there. Any analog filter recommend will be appreciated!
- Status
- Not open for further replies.