Hi
Last month, I bought a MHDT labs parasidea non oversampling dac which I find really great for the price. Originally, there's a OPA2604 for the TDA1545 I/V conversion. This IC for me is only mediocre so I tried OPA627 to replace the OPA2604. After many days of listening, I really didn''t like the OPA627 sound in this dac. Surprisingly, the sound was harsh in the mid -high frequencys with this op amp.. There's a tube follower after the I/V conversion. I also tried an AD826 with good results. Some peoples talked to me about LM6181 for doing I/V conversion. I never tried this op amp. Any recommendations about what should be the best op amp to try for I/V?
Thanks
Legarem
Last month, I bought a MHDT labs parasidea non oversampling dac which I find really great for the price. Originally, there's a OPA2604 for the TDA1545 I/V conversion. This IC for me is only mediocre so I tried OPA627 to replace the OPA2604. After many days of listening, I really didn''t like the OPA627 sound in this dac. Surprisingly, the sound was harsh in the mid -high frequencys with this op amp.. There's a tube follower after the I/V conversion. I also tried an AD826 with good results. Some peoples talked to me about LM6181 for doing I/V conversion. I never tried this op amp. Any recommendations about what should be the best op amp to try for I/V?
Thanks
Legarem
Lm6181
Hello Legarem
I'm not sure that the current feedback op amp LM6181 is a good solution:this is not an audio op amp.
Moreover National Semiconductor advises an other product:LMH6682.
Normally the high speed op amp(designed for video applications,ADSL modems,line driver ) have a disagreeable sound.
Why not tests whith a discrete circuit?
Hello Legarem
I'm not sure that the current feedback op amp LM6181 is a good solution:this is not an audio op amp.
Moreover National Semiconductor advises an other product:LMH6682.
Normally the high speed op amp(designed for video applications,ADSL modems,line driver ) have a disagreeable sound.
Why not tests whith a discrete circuit?
Re: Lm6181
I can agree to a certain level regarding your remark about discrete circuits.
But think again when saying that high speed opamps have a disagreeable sound. Did you hear AD844 and AD815 in audio applications ?
Legarem, did you decouple the OPA627 as recommended ( including a film cap from the + to - pins ) ?
parsecaudio said:
I can agree to a certain level regarding your remark about discrete circuits.
But think again when saying that high speed opamps have a disagreeable sound. Did you hear AD844 and AD815 in audio applications ?
Legarem, did you decouple the OPA627 as recommended ( including a film cap from the + to - pins ) ?
High speed op amp sound
Hi jean paul
I don't know these Analog Devices op amp,but only the National semiconductor's H.S. amp like LM6171, LM7171,LM6165,LM6103:it does not appeal to their sound to me.
I prefer discrete circuit because they are more easy adaptable to the requirements(sounds or others) that we have need.
Hi jean paul
I don't know these Analog Devices op amp,but only the National semiconductor's H.S. amp like LM6171, LM7171,LM6165,LM6103:it does not appeal to their sound to me.
I prefer discrete circuit because they are more easy adaptable to the requirements(sounds or others) that we have need.
I agree that a discrete circuit should normally be better than an op amp. First, I have to find a good discrete circuit. Second, in the Paradisea, there's not to much place so it will not be easy to implement one.
About decoupling OPA627, there's a 0.1 mfd from V+ to ground and V- to ground near the op amp.
I am also tempted to try an I/V with a resistor but in the past when I tried this topology in another dac (AD1862) I found that it lacked dynamic.
Legarem
About decoupling OPA627, there's a 0.1 mfd from V+ to ground and V- to ground near the op amp.
I am also tempted to try an I/V with a resistor but in the past when I tried this topology in another dac (AD1862) I found that it lacked dynamic.
Legarem
Change them for 10 uF 50V Panasonic FC electrolytics very close to the supply pins.
Add one 0.1 uF MKT filmcap ( 5 mm pitch like Wima MKS 2 ) from + to - directly ( not to ground ) on the chip or under the pcb. You'll like the result.
I believe it was CarlosFM who said values of up to 220 nF can be used. Never tried that, I mostly use 10 nF or 100 nF.
Add one 0.1 uF MKT filmcap ( 5 mm pitch like Wima MKS 2 ) from + to - directly ( not to ground ) on the chip or under the pcb. You'll like the result.
I believe it was CarlosFM who said values of up to 220 nF can be used. Never tried that, I mostly use 10 nF or 100 nF.
It is the questionable quality of tantalum caps in the past ( they often blew when used directly on supply lines ) and the mining/war scandal that keeps me from using them.
Besides that they sound worse than conventional electrolytic caps. At least they did the last time I tried them.
When you send me a PM with your adress etc. I can send you some 1 uF/25V caps with sealed ends I recently obtained. You are the guinee pig then
I will appreciate your report on them though.
Besides that they sound worse than conventional electrolytic caps. At least they did the last time I tried them.
When you send me a PM with your adress etc. I can send you some 1 uF/25V caps with sealed ends I recently obtained. You are the guinee pig then
I will appreciate your report on them though.Tantalum cap reliability
Jean paul,
what you say is true.In the past these were not much reliable,a short circuit was one what much frequent one .
From approx.1990 this has not happened to me.
A five band dynamic expander/compressor(built for a musician friend) in 1993 is still very working after many years:each op amp(to the inside they are 24 NE5532)are bypassed with two 4.7 uF 25V tantalum cap.
Total 48 Sprague cap.
No capacitor has never given a problem
Jean paul,
what you say is true.In the past these were not much reliable,a short circuit was one what much frequent one .
From approx.1990 this has not happened to me.
A five band dynamic expander/compressor(built for a musician friend) in 1993 is still very working after many years:each op amp(to the inside they are 24 NE5532)are bypassed with two 4.7 uF 25V tantalum cap.
Total 48 Sprague cap.
No capacitor has never given a problem
Modern low esr electrolytic
For my experience,a well filtered and coupled power supply have the same importance of the audio circuit.
In more situation,the low esr electrolytic like Panasonic it gives the impression to be better than tantalum.
In my DAC,for the digital supply,the tantalum cap bypassed with a ceramic is the better solution.The market offers many solutions and we must use that best one for what we want to obtain.Some op amp are very sensitive to the supply quality .
For my experience,a well filtered and coupled power supply have the same importance of the audio circuit.
In more situation,the low esr electrolytic like Panasonic it gives the impression to be better than tantalum.
In my DAC,for the digital supply,the tantalum cap bypassed with a ceramic is the better solution.The market offers many solutions and we must use that best one for what we want to obtain.Some op amp are very sensitive to the supply quality .
When I looked at the circuit, I made a mistake. The V+ and V- were not decoupled at all. There's a +19V and -19V with 100 ohms resistors in series with the V+ and V- to lower them near +15 and -15 V. The +19 and -19V is coming from shunt regulators made with TL431. I tried again the OPA627 with 10 mfd tantalum on each rail and I placed a 0.47 mfd film cap between the two rails. All the caps are soldered to the brown dog socket. Wow !!! the sound is now seriously better than most op amps tried.
I wish to try the AD8065 and perhaps try Black gates to replace tantalum caps.
Legarem
I wish to try the AD8065 and perhaps try Black gates to replace tantalum caps.
Legarem
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.