Hi all,
I do own a logitech transporter which uses a AK4396 DAC followed by the same output stage as is found in the schematic of the AK4396 evaluation board, which can be found here: http://www.asahi-kasei.co.jp/akm/en/product/ak4396/akd4396-sbw2-02e.pdf (page 30)
I am looking for a better sounding output stage. With the emphasis on sounding not measuring. Since my amplifier has RCA input only I am not interested in balanced outputs.
There are a few things crossing my mind here:
- omitting the low pass filtering since the out of audio band noise is already very very low, just use one discrete (burson audio) opamp as a buffer
- using a non opamp discrete output
- omitting the filtering and the dc blocking cap and use a discrete opamp as a differential amplifier
What do you guys think?
Any good suggestions?
Ronald
I do own a logitech transporter which uses a AK4396 DAC followed by the same output stage as is found in the schematic of the AK4396 evaluation board, which can be found here: http://www.asahi-kasei.co.jp/akm/en/product/ak4396/akd4396-sbw2-02e.pdf (page 30)
I am looking for a better sounding output stage. With the emphasis on sounding not measuring. Since my amplifier has RCA input only I am not interested in balanced outputs.
There are a few things crossing my mind here:
- omitting the low pass filtering since the out of audio band noise is already very very low, just use one discrete (burson audio) opamp as a buffer
- using a non opamp discrete output
- omitting the filtering and the dc blocking cap and use a discrete opamp as a differential amplifier
What do you guys think?
Any good suggestions?
Ronald
Bill Fuss said:
Hi Ronald,
This post is 2 months old now, since then I have experimented with 4 other trafos, and found the transformer solution simple, elegant, transparent, extremely dynamic, with the largest, deepest, widest, and more layered soundstage I have ever encountered.
The chip has differential outputs so the differential nature of transformers is a match made in heaven. They happily operate in single ended configuration as well as balanced, transformers just dont care.
To any naysayers, don't pass judgement until you have tried a couple high quality trafos in a proper implementation, you will not be disappointed.
Best regards, Bill
Hello Bill,
Thanks for your response
Do you just use a transformer after de AK4396 and that's it? Or is it followed by a buffer stage of some kind?
Which transformers exactly do you use?
And how about DC, is there a cap between the AK4396 output and the transformer in your solution?
Regards,
Ronald
Thanks for your response
Do you just use a transformer after de AK4396 and that's it? Or is it followed by a buffer stage of some kind?
Which transformers exactly do you use?
And how about DC, is there a cap between the AK4396 output and the transformer in your solution?
Regards,
Ronald
tingtong5 said:Hello Bill,
Thanks for your response
Do you just use a transformer after de AK4396 and that's it? Or is it followed by a buffer stage of some kind?
Which transformers exactly do you use?
And how about DC, is there a cap between the AK4396 output and the transformer in your solution?
Regards,
Ronald
Hi Ronald,
No caps in the signal path and no buffers, just current limiting resistors on the primaries to protect the chip output, and direct out to the RCAs from the secondaries using 600/600 ohm transformers. About as simple and elegant as it can get.
The 4396 has true differential output which is perfect for feeding a transformer unless there is some residual DC offset, and that can be dealt with in the primary configuration. I have not found any DC to be concerned with.
I personally use first order filters on primaries and secondaries with an F3 around 150Khz, very small cap across the primaries, and RC filter across to secondaries to kill any ultrasonic resonance (ringing).
No so called "loading" is required or wanted on the output of a 600 ohm transformer.
I have tried a pair of inexpensive Jensen JT-123s (not very good), UTC A-20 (fantastic) A-22 ( limited bass reach but still outstanding), Sescom MI-14 (fantastic) and ms-11 taken from mic splitter boxes (so far outstanding but harder to implement).
There are so many good choices from different manufacturers, new and vintage, they are easily found on Ebay.
Look for at least +15dbu capable @ 20 hz and as wide a bandwidth as possible, and 500 or 600 ohm windings. I have a few catalogs for vintage Sescom, UTC, and Triad I can email, but they are too large to post. Email me if you would like a copy.
Best, Bill
I'm with Bill.
I use the Lundahl LL-1690 out of the AK4393 used in the DCX2496. Granted, I only use this output to feed my bass, but very good bass at that. No passive filters of course; not needed in that application.
Get a good transformer, and you won't be disappointed. The LL-1690 runs $125 each.
I use the Lundahl LL-1690 out of the AK4393 used in the DCX2496. Granted, I only use this output to feed my bass, but very good bass at that. No passive filters of course; not needed in that application.
Get a good transformer, and you won't be disappointed. The LL-1690 runs $125 each.
Ever considered That1200?
http://www.thatcorp.com/1200-series_High_CMRR_Balanced_Line_Receiver_ICs.html
http://www.thatcorp.com/1200-series_High_CMRR_Balanced_Line_Receiver_ICs.html
connecting transformer to AK4396 - how?
"To any naysayers, don't pass judgement until you have tried a couple high quality trafos in a proper implementation, you will not be disappointed." (Bill)
Now I want to find out what´s behind Bill´s advice and try some Lundahls I shot on ebay:
I did a lot (or too much) of reading about transformers but nothing gave me a clear picture of how to connect them to an AK4396. And what´s about the DC portion on the DAC´s output?
Maybe you´d like to help me ...
"To any naysayers, don't pass judgement until you have tried a couple high quality trafos in a proper implementation, you will not be disappointed." (Bill)
Now I want to find out what´s behind Bill´s advice and try some Lundahls I shot on ebay:
- LL 1517 http://lundahl.se/pdfs/datash/1517.pdf
- LL1539 http://lundahl.se/pdfs/datash/1539.pdf
- LL1540 http://lundahl.se/pdfs/datash/1540.pdf
I did a lot (or too much) of reading about transformers but nothing gave me a clear picture of how to connect them to an AK4396. And what´s about the DC portion on the DAC´s output?
Maybe you´d like to help me ...
Hi Guys
We've designed our own QuadFilar transformer to be specifically used for differential outputs of "voltage" DACs such as CS4398. It needs no additional components, passive or otherwise. We have now run it in a number of different players including the CS4398 used in the Oppo Blu-Ray player (BDP-83, not SE).
BTW, here is one report/feedback just received:
"It sounds fantastic! Heaps of detail, heaps of bass - just scary good."
When it comes to "current" DACs we prefer a proper I/V Converter, no feedback, extreme bandwidth and totally immune to overload at HF. Indeed some, but not all, so-called digital sound can be ascribed to slew-rate-induced distortion. People like Charlie Hansen has mentioned this as indeed so have I for years. But for "voltage" DACs the transformer solution is a natural IMO. But it has to be an exceptional transformer, very exceptional. The other source of digital sound is ultra-low frequency instability in the clocking signal. But that's another story.
BTW, may I suggest to Hurtig that maybe he should not be so... hurtig.
Not everybody is going to understand that pun.
Cheers.
Joe R.
We've designed our own QuadFilar transformer to be specifically used for differential outputs of "voltage" DACs such as CS4398. It needs no additional components, passive or otherwise. We have now run it in a number of different players including the CS4398 used in the Oppo Blu-Ray player (BDP-83, not SE).
BTW, here is one report/feedback just received:
"It sounds fantastic! Heaps of detail, heaps of bass - just scary good."
When it comes to "current" DACs we prefer a proper I/V Converter, no feedback, extreme bandwidth and totally immune to overload at HF. Indeed some, but not all, so-called digital sound can be ascribed to slew-rate-induced distortion. People like Charlie Hansen has mentioned this as indeed so have I for years. But for "voltage" DACs the transformer solution is a natural IMO. But it has to be an exceptional transformer, very exceptional. The other source of digital sound is ultra-low frequency instability in the clocking signal. But that's another story.
BTW, may I suggest to Hurtig that maybe he should not be so... hurtig.
Not everybody is going to understand that pun.
Cheers.
Joe R.
I agree with this and what Bill said.
I have quality transformers direct out with the CS4398 and the sound is very transparent, very dynamic and very open
Curious about which Sowters and are you using I or V outs on AD1865?
Cheers.
Joe R.
I out, V out is horrible.
Which Sowter model number? Do you know if you are close to 2V output?
Cheers.
Joe
To be precise I am am using the top of the line model 9545 and a 4k7 resistor on the secondary. The primary windings in // and the secondaries in series.
I am not close to 2V output, for that you need higher valued resistor but who actually needs such a "hot" CDP?
The arrangement is very similar to Audio Note higher priced DACs.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.