help with this AKM dac, please - diyAudio
Go Back   Home > Forums > Source & Line > Digital Source

Digital Source Digital Players and Recorders: CD , SACD , Tape, Memory Card, etc.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 17th July 2006, 10:26 PM   #1
diyAudio Member
 
Join Date: Dec 2003
Location: Buenos Aires, Argentina
Default help with this AKM dac, please

Hi everybody.
I wish to know if it`s feasible to use the dual differential outputs of this dac without resorting to additional analog filtering:

http://www.asahi-kasei.co.jp/akm/en/...k4385_f01e.pdf


I´m trying to use it to feed Burr Brown`s TPA6120 in dual diff manner, and using this powerful chip as the output stage of my cdp-dvd player.
For what I know, the dac is a voltage output one, so no IV conversion is needed; out of band noise appears to be benign, but maybe needs more filtering than the provided by the internal ones.
Sorry, I`m not an expert in this issue.
And I don`t know how to implement a LPF around tpa6120.

Actually, the dac is feeding a single opamp lpf output stage, very similar to the one depicted in the PDF.
But it would be great to implement TPA6120 there as an experiment, to drive longer cables, etc.

Thanks in advance for the input.
  Reply With Quote
Old 22nd July 2006, 04:36 PM   #2
diyAudio Member
 
Join Date: Oct 2001
Location: .
You cannot use the TPA6120 as a LPF in the manner shown in the AKM datasheet. Capacitors are not allowed in the feedback path. What you can do is place a passive low pass filter before the TPA6120.
  Reply With Quote
Old 24th July 2006, 05:12 PM   #3
diyAudio Member
 
Join Date: Dec 2003
Location: Buenos Aires, Argentina
Yes, Rfbrw, I´ve notice that in the datasheet. TPA6120 is a cranky animal.
Thanks for the suggestion about lpf before the chip; by coincidence I was doin´some experiments about the issue this weekend.
I´d never before had soldered SMD, so with lots of patience I´d mounted the TPA6120 ¨dead bug style¨ -the thermal pad is nicely exposed in this sort of arrangement-, using some tiny copper sheet grounded, and soldered to the pad, as my ground plane.
Then, the feedback loop was mounted almost ¨in the air¨-in fact, not touching the PCB in any sort of way-, with little resistors minimally leaded. I guess I´ve made the tightest possible loop around the chip with the disponible resources -what a pain in my eyes...-.
Decoupling was straightforward this way, because you have the ground only some 1/2 mm away.

Then, I made some simple first order balanced LPF at the input of the TPA6120...
I plugged the engender expecting the worst -you know, severe oscillation, DC at the outputs etc- (I ´ve searched a lot about that chip, and the readings were not very reassuring: a lot of people talked about high dc output, etc-.
But, much to my surprise, the chip performed stellarly: no heat, no DC -only some 0.4/0.6 mV in each channel-, no on or off ¨thumps¨...
And the sound: I was expecting something harsh. Again, TPA6120 surprised me with a round nice sound, very very detailed but warm and easy-listening...
I was doing some listening just yesterday night, and the experience was relaxing. Much, much better than the prior implementation -some OP275 lpf filtering/buffering, in a config identical to the one in the datasheet of the AKM dac-.
I wish I had some photos to show the strange look of the TPA solded upside down with the tiny resistors around: it looks like a dead spider -no kidding-...
Now I`m concerned about the capacity of the AKM dac to drive the LPF... I have used 340r series/1680 pF shunt. This low value of R, not to disturb the gain setting of TPA6120 -the datasheet suggests that it`s better to adjust the gain with the input resistor, which adds, I believe, in series to the LPF ones.
What do you think about it?
Well, I´ll post about new findings when I get time to do some more experiments.

Thank you very much, Rfbrw
  Reply With Quote
Old 24th July 2006, 07:36 PM   #4
diyAudio Member
 
Join Date: Dec 2003
Location: Buenos Aires, Argentina
Default a doubt...

In the datasheet of the AKM dac I noticed that load resistance is stated at 4kohms, so maybe I`ll do some redesign of the feedback around TPA6120 with 4,7 k resistors, -splitting the input resistor in two sections, with a shunt cap in the middle as LPF- to avoid an excessive load on the dac chip.

Now, series resistance at dac output is some 1,6 k...
Did any of you know if is it worth to redesign the whole thing to obtain a better impedance matching, to easy the life of the AKM dac?
I don`t know if a 4k load resistance part has any margin to deal with heavier loads, or if it is in peril.
  Reply With Quote
Old 27th July 2006, 04:30 PM   #5
diyAudio Member
 
Join Date: Dec 2003
Location: Buenos Aires, Argentina
Thumbs up well, I´ve been reading a lot...

...about TPA6120, fully differential configurations regarding the AKM4385 dac, etc.
There are lots of info in prior forums regarding differential dacs, analog filtering, etc, with designs from Thorsten, Dimitri etc.

For what I`ve understanded, an optimum match between AKM4385 dac and TPA6120 preamp is not possible, at least from a technical point of view.

But if you use some four single opamps -or two duals- performing lpf function between the DAC and the TPA6120, yes, it`s definitively possible. Impedances are much better matched, so a better resistor configuration is feasible around TPA6120 to obtain practically zero dc output.

But lazy as I`m -and short of time, and short of space inside the little Sony cdp-, and in a maniac desire to maintain simplicity and to avoid more opas in the signal chain-, I decided to couple the dac with the TPA6120 with only a 1st order lpf between.

The dac has 4kohms output load capability, so I was constrained to use some large resistors -4k- in the R position of the simple RC filter.
More so: these resistors define part of the gain scheme of TPA6120.
To avoid the effects of the capacitance of the C of the lpf filter over the TPA6120 inputs, I used some 1k input resistors -sort of ¨stoppers¨- directly at the input of TPA6120.
So, TPA6120 ¨sees¨ some 5k Rin...
To maintain a gain of two with tpa6120, I changed the feedback resistors to 4,7k -the TI datasheet recommends not to exceed 10k in this position-.
In the configuration suggested by TI, is possible to obtain cuasi-zero DC at tpa6120 output; but once you start to mess with those values, DC appears, -25mV in this case- so I decided to put some 47uF caps at the output.
Well, the thing works, and sounds very good...
  Reply With Quote
Old 27th July 2006, 04:55 PM   #6
diyAudio Member
 
Join Date: Dec 2004
Location: Roosendaal
Hi Frederico,

I have little experience with the TPA chip, but more with the AKM4385.

As long as you keep the load above 600 Ohm you will not run into trouble.

You can find a lot of examples in the groupsection on Yahoo (M-Audio_SuperDAC_2496 )

Greetings, Harry
  Reply With Quote
Old 27th July 2006, 05:04 PM   #7
diyAudio Member
 
Join Date: Dec 2003
Location: Buenos Aires, Argentina
Default tpa6120 as output stage

Frankly, I was expecting the worst regarding TPA6120. It´s extremely sensible to layout issues, and I`m no master at the pcb art -I mounted the chip ¨dead coackroach style¨, heatsinking it wit some copper sheet-.

But maybe that kind of layout has the virtue of isolate the input pins of the chip from PCB parasitics -Kwei reccomends in some instances that method of mounting as a good way to achieve nice sound- . I don´t know, but the chip is apparently stable, with minimal power consumption, and no heat.

Sound: the thing that most caught my attention was the rendering of all kind of percussions. Bass drum is very well portrayed, snare drums have very nice attack, and hi hat and splashes sound very natural and fast.
In this aspect, I always rate my Thorens TD125/Shure v15 IV as clear winner, but this time cd reproduction has come very close...

Another thing that this TPA6120 does well is to deal with intrincate mixes, but it does so without been ¨analytical¨ in a harsh way. In my opinion, is very easy-listening, not ear fatiguing. But has lot of detail, and sounds very...analog.
There is not a trace of artificial sibilance in the voices -something that with a pair of schemes based in opamps that I did before, was very noticeable-.

Well, I was very happy with the modification



Cheers!
  Reply With Quote
Old 27th July 2006, 05:09 PM   #8
diyAudio Member
 
Join Date: Dec 2003
Location: Buenos Aires, Argentina
Default originally posted by harryeng

Quote:
I have little experience with the TPA chip, but more with the AKM4385.
Quote:
As long as you keep the load above 600 Ohm you will not run into trouble.
Thank you very much for the info, Harry!
I´ll check out that Yahoo info.

Now you give me reasons for some more experimentation...



Saludos!
  Reply With Quote
Old 27th July 2006, 08:59 PM   #9
diyAudio Member
 
Join Date: Dec 2003
Location: Buenos Aires, Argentina
Default harry...

...for what I´ve seen in the Yahoo Groups, they were using AK4393, a higher end part -with somewhat better specs- than AK4385.
AK4393 datasheet states a load resistance of some 600 Ohms; but for AK4385 the value is 4k.

Thanks indeed for the help!
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



New To Site? Need Help?

All times are GMT. The time now is 04:05 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2