Looks good. If one needs stacked AD844's, I would make them parallel to each other and having independent local bypass caps on each power pin.
I agree with Abraxaito (of course)
The information of the pre-echo originates in your dark matter (that DSP) and not from the sound field. So at least in audio, dark matter occurs.
I must say I have seen worse residues of filters (SAA7220).
The interesting thing is that the digital filter can be filtered analogly, at the frequency that this artefact has. Probably, for instance with a simple bridged nulling filter.
What is the frequency it has? It looks like 20 KHz *36 = 720 KHz.
Update
Hi folks... Haven't been on here in some time. The PCM1704 DAC with the DDNF and OPA627 buffer still reigns supreme here. M3 DAC (parallel TDA1541A S1 Crowns) has made some progress. It now just about matches the sound quality of the 1704 based DAC. It uses a pair of the Red Baron boards and has a DDNF I/V board to each DAC. I sum them after the coupling caps going into the OPA627 buffer. I set the OPA627 buffer to a gain of 1.5. Experiments with a BUF03 buffer did reveal some issue in clarity in the critical lower midrange. So I have moved on from using the BUF03's in critical applications. If you don't have ribbon tweeters then likely you will not notice that. I believe the current configuration will be the final for the M3 DAC experiment. In other news... I took a rather nasty lightning strike the Monday following the US Thanksgiving holiday. It came in the DSL cable. I lost my modem, router, printer, a Bluray player and the "A" board on my plasma television. Sadly last week my Class D audio amp went too. Not clear if lightning had anything to do with that. I will have to hold off any critical DAC comparisons until I have my amp back or get a replacement.
Hi folks... Haven't been on here in some time. The PCM1704 DAC with the DDNF and OPA627 buffer still reigns supreme here. M3 DAC (parallel TDA1541A S1 Crowns) has made some progress. It now just about matches the sound quality of the 1704 based DAC. It uses a pair of the Red Baron boards and has a DDNF I/V board to each DAC. I sum them after the coupling caps going into the OPA627 buffer. I set the OPA627 buffer to a gain of 1.5. Experiments with a BUF03 buffer did reveal some issue in clarity in the critical lower midrange. So I have moved on from using the BUF03's in critical applications. If you don't have ribbon tweeters then likely you will not notice that. I believe the current configuration will be the final for the M3 DAC experiment. In other news... I took a rather nasty lightning strike the Monday following the US Thanksgiving holiday. It came in the DSL cable. I lost my modem, router, printer, a Bluray player and the "A" board on my plasma television. Sadly last week my Class D audio amp went too. Not clear if lightning had anything to do with that. I will have to hold off any critical DAC comparisons until I have my amp back or get a replacement.
"Class D audio amp went too. Not clear if lightning had anything to do with that."
Probably blew up in protest to it's own switching noise.
Technics showing the class-d way with the SE-R1 ($30k), twice the switching frequency (1.5MhZ), to easier be able to filter it all out without going near the audio band.
Cheers George
Probably blew up in protest to it's own switching noise.
Technics showing the class-d way with the SE-R1 ($30k), twice the switching frequency (1.5MhZ), to easier be able to filter it all out without going near the audio band.
Cheers George
Class D
Hi George, When you have Magnepan speakers you have to have power. The "Class D Audio" Amp developed a fault in the left channel (Ch2) on the boards module. A nasty popping sound. I sent it back to the vendor for repair. If possible.... I am afraid I am a poor audiophile so if I had $30K sitting around I'd likely get a Ford Mustang with a V8. I can't actually disagree with you, in many ways Class D is a flawed scenario. Technics is on the right track. I have measured a lot of RF noise here with the amp on. Easy to do with a Ham radio. The maggies make for a large pair of antennas. Considering giving a Crown XLS 1502 pro amp a roll in the system. Being pro it is hard to say if it is good enough with the ribbons. The power to dollar ratio is very high. My only other option is an old Moscode 600 (heavily modded = not euphonic) that has a bad output Mosfet. At least that is what I think at this point. That is one power hungry amp.
"Class D audio amp went too. Not clear if lightning had anything to do with that."
Probably blew up in protest to it's own switching noise.
Technics showing the class-d way with the SE-R1 ($30k), twice the switching frequency (1.5MhZ), to easier be able to filter it all out without going near the audio band.
Cheers George
Hi George, When you have Magnepan speakers you have to have power.
The "Class D Audio" Amp developed a fault in the left channel (Ch2) on the boards module. A nasty popping sound. I sent it back to the vendor for repair. If possible.... I am afraid I am a poor audiophile so if I had $30K sitting around I'd likely get a Ford Mustang with a V8. I can't actually disagree with you, in many ways Class D is a flawed scenario. Technics is on the right track. I have measured a lot of RF noise here with the amp on. Easy to do with a Ham radio. The maggies make for a large pair of antennas. Considering giving a Crown XLS 1502 pro amp a roll in the system. Being pro it is hard to say if it is good enough with the ribbons. The power to dollar ratio is very high. My only other option is an old Moscode 600 (heavily modded = not euphonic) that has a bad output Mosfet. At least that is what I think at this point. That is one power hungry amp. 
my friend I do not sell anything
. After tested it well, I can also share gerber file and so you can carry in your own.In other news... I took a rather nasty lightning strike the Monday following the US Thanksgiving holiday.
Sorry to hear your lost. Ever thought about using UPS for your devices? I have one used APC 1000VA UPS. It's inexpensive ($20 for the UPS and another $20 for batteries) and provides enough protection for my audio devices. Some say UPS will degrade the sound, but I found this APC has minimal impact on sound quality.
thanks, waiting for the gerber file thenmy friend I do not sell anything
abraxalito,
Today I tried 10uf/1210 X7R on the buffer op amp and the result is great. Many years ago I used 10uf electrolytic but it did not satisfy me. 100nf film or C0G is still better than electrolytic here. However, 10uf X7R is different story. The bass is clean and powerful. After few hours of run-in, mid and high improved, they are now equal to the effect of 100nf C0G. Thanks for the tip.
@canvas ... Hi canvas. Interesting read this:
Have you tried to parallel a 100 nF C0G with the 10 uF X7R? I personally have good experiences with C0Gs (in another context though) and I reckon that the combination could be very fine ... As it is I'm also looking for some superb (the very "best" what that may be) 10 uF capacitors, and the electrolytics I've listened to have - to my ears - not had the fullness or other qualities of their larger value siblings (I have tried some including NP types). Might I ask which X7Rs you have tried (& listened to)?
Cheers,
Jesper
Have you tried to parallel a 100 nF C0G with the 10 uF X7R? I personally have good experiences with C0Gs (in another context though) and I reckon that the combination could be very fine ... As it is I'm also looking for some superb (the very "best" what that may be) 10 uF capacitors, and the electrolytics I've listened to have - to my ears - not had the fullness or other qualities of their larger value siblings (I have tried some including NP types). Might I ask which X7Rs you have tried (& listened to)?
Cheers,
Jesper
Thanks for sharing your feedback. But don't stop at just a single cap, try adding two three or four
I have some circuits where I've piled up caps atop a chip to about 10mm height.The weakness of a 10uF electrolytic is going to be its high ESR. Typically will be an ohm or so. 10uF ceramics have ESRs considerably lower.
But don't stop at just a single cap, try adding two three or four
I am thinking about stacking five 2.2uf X7R instead of one 10uf. I am not sure if this will further lower the ESR, since MLCCs are multi-layered. One thing about 100nf C0G. I found it sounds more balanced when soldered on both rails (one end at + and the other at -). According to some datasheets, this will improve distortion and PSRR.
Poting
Have been doing this for many years, but please also use these on DAC's power supply pins - especially delta-sigma DACs benefits as they switch to rail and noise there becomes instant jitter quicker than instant coffee. In fact, use them everywhere, not just on opamps - that is why I bought 500 last time.
.
What load did you have on the output of the AK4396? The caps on the rail are there to reduce the induced rail noise (caused by the internal opamps operating in classAB). But you can also reduce the rail noise by reducing the output loading. So if adding caps blew you away you may find reducing the load (increasing its impedance) gives you a similar enhancement but primarily to the bass (where the caps are having little effect).