Re: Re: Looking for an active I/V stage for TDA1543-based DAC
Ok, I'll be more specific....
The I/V section from the datasheet is very simple and uses a generic opamp. I'm sure that after spending a couple of weeks I'll develop something better. What I am looking for is a quality solution to start with and tweak too, something with a Class-A biasing, virtual ground, quality caps for bypassing and so on.
I searched on the forum and came up with this schematics, going to try it out soon and post the results:
Elso Kwak said:
Ok, I'll be more specific....
The I/V section from the datasheet is very simple and uses a generic opamp. I'm sure that after spending a couple of weeks I'll develop something better. What I am looking for is a quality solution to start with and tweak too, something with a Class-A biasing, virtual ground, quality caps for bypassing and so on.
I searched on the forum and came up with this schematics, going to try it out soon and post the results:
Attachments
Hi IpsilonSound
How are finding your dac thus far?
Did you connect it up to your amp successfully? [single supply]
I shall be interested in your ongoing active IV quest, been considering investigating something active myself.
Though space and power supply are also at a premium in my
application.
Cheers
Setmenu
How are finding your dac thus far?
Did you connect it up to your amp successfully? [single supply]
I shall be interested in your ongoing active IV quest, been considering investigating something active myself.
Though space and power supply are also at a premium in my
application.
Cheers
Setmenu
Peter Daniel said:
Thanks, Peter... I've seen the thread already and considered the chip for I/V conversion. The THD is just .0003%, it's unity gain stable and needs only one power supply which will facilitate biasing to ClassA. The OPA2604 would be a nice dual chip for this task, I'll order a couple of samples of them to try them out with my DAC.
setmenu said:
Hi Setmenu,
I haven't managed to solve this issue so far 100%. Omitting the ground or putting the bypass cap between the ground and the pot solves the single power supply issue for DAC and PPA. But the sound is not up to what I could live with. It's affected, not as smooth and dynamic. Using transformers will fix the problem and , but transformers are too large to be put inside the PPA case.
I guess, the only option would be active stage with opamps.
I would hate to reject the idea of PPA+DAC under a single case, the PPA is already large to be called 'portable'.
By the way, you got real talent to put so much into such a tiny case.
Cheers!
Hi IpsilonSound
I too found the shared battery less than satisfactory.
That said the added weight /bulk of duel packs is not too bad,
about 60 grams extra depending on capacity,I tend to allow for
about 10hours run time as a minimum.
The whole setup weighs in at about 570 grams + bag.
My previous outboard dac setup was more like 1000gram +bag
Mind all the above is insane compared to an Ipod + buds ,hehe.
So your is setup is intended as the above, for portable use?
If so, I am presuming you mean more for transportable rather
than something to be carried about when going for a walk!
So how is it all sounding thus far?
Do you like the sound of the passive dac?
I am pretty happy with mine , it is certainly is not perfect! But it has been pressed into service .
It also chucks out a fair bit of heat especially boxed up and inside a soft bag during warm weather..[1543 running at 8v]
It would be interesting to built something else , say delta sigma
based , [CS 43122?] but I have not done much studying of these
dacs as yet.
It would still have to fit into a tiny space , and I have a feeling
that things could be more 'tetchy' with this technology?
But at the moment being steeped in ignorance [compared to folk here!!] I am just pleased to have hit my target and completed a
project that is now actually being used
Cheers
Setmenu
I too found the shared battery less than satisfactory.
That said the added weight /bulk of duel packs is not too bad,
about 60 grams extra depending on capacity,I tend to allow for
about 10hours run time as a minimum.
The whole setup weighs in at about 570 grams + bag.
My previous outboard dac setup was more like 1000gram +bag
Mind all the above is insane compared to an Ipod + buds ,hehe.
So your is setup is intended as the above, for portable use?
If so, I am presuming you mean more for transportable rather
than something to be carried about when going for a walk!
So how is it all sounding thus far?
Do you like the sound of the passive dac?
I am pretty happy with mine , it is certainly is not perfect! But it has been pressed into service .
It also chucks out a fair bit of heat especially boxed up and inside a soft bag during warm weather..[1543 running at 8v]
It would be interesting to built something else , say delta sigma
based , [CS 43122?] but I have not done much studying of these
dacs as yet.
It would still have to fit into a tiny space , and I have a feeling
that things could be more 'tetchy' with this technology?
But at the moment being steeped in ignorance [compared to folk here!!] I am just pleased to have hit my target and completed a
project that is now actually being used
Cheers
Setmenu
setmenu said:
Hi Setmenu,
If both DAC & PPA are in the same case, it can be used as a portable, I think, too big for a pocket but can be easily carried in a small bag. I tried it already, listened to it while driving a car, took it to the park with me, listened to it at work, and it's ergonomically great. The CD player with optical can be attached to the PPA case with a velcro strip. I only wish the optical cable was shorter, mine is 3ft and that's the shortest one I could find.
setmenu said:
Yes, I'm very happy with the sound overall, but still believe there's room for improvement. The bass could be a bit tighter and that can only be fixed with an active gain stage, I guess. Also, cannot decide between the super-e or single non-polar BG cap combination per channel. The caps are still under burn-in, I'll give them another week before final decision. Even now it sounds better than my portable and DVD player. I could pick up some tiny sound cracking from original LP recorded on a compact disk which I never heard before. It's the "Bahama-Mama" song by Boney-M, track#4, album "Oceans of Fantasy" 1975, the CD was made in Hamburg, Germany by BMI, if i remember correctly. Just for curiousity, try to listen to it and notice a bit of tiny record cracks at the first 3-4 seconds of intro to the song
I never noticed those original analog 'imperfections' with my DVD as a source or NAD, though listened to this cd a bazillion of times Maybe the TDA1541-based DAC sounds better, but it cannot be used as a portable and TDA1543 is a great compromise.
Also, i would recommend using CS8414, since it's a slightly better chip with improved jitter and error handling. It's packaged SOIC only, though, and needs a DIY or BrownDog adapter unless one wants to solder it forever to the board. Would be real nice to compare the sound CS8414 and with overclocking. Everyone tried overclocking with CS8412 and the result was positive.
Some recent experiments...
I tried optical so far, plan to connect digital thru coax soon and compare the sound. The TORX141 optical receiver was powered by 4.9V before and it's too much for it, the recommended voltage is 3.6, but I don't have a proper regulator yet. As a workaround, I put a rail splitter to make 2.5V to feed to the receiver yesterday. I also removed 2 0.01uF caps and 75ohm parallel resistor before input for CS8414. This setup is used everywhere with digital coax. The sound, I think, became more hissy, still good, but with more high-frequency added. I think I'll roll back to the previous setup with 2 caps and a resistor. They are filtering the noise coming with digital, as said in the CS8414 datasheet. Too bad, it doesn't say anything about optical connection to the chip, is it universal? Doede in his DAC uses 2 caps for both optical and coax and 75ohm resistor for coax only. I'll go the same route then...
My THS4081 and THS4082 opamp chips should arrive anytime soon, as well as a set of LM6172 and LM6171. I look forward to building a I/V stage with them, though, it may not be so easy to do at first, since I plan to add Class-A biasing, virtual ground, bypassing to the stage, bass boost, and so on... Maybe a good idea to swith my extra working PPA amp board with these chips and see how they behave in this setup, they may turn out to be a kind of cranky ones to work with
will ask Tangent then, he's great with opamps!
Peter, I know that you've mentioned it before somewhere, sorry for asking again. Which overclocking setup did you try, Tent or KWAK? Just curious, since I don't mind skipping this part of experimenting for DAC for several reasons(part availability, affect on batteries, more complexity/parts-shared space issue).
How would you characterize the sound before and after overclocking?
Also, have you tried using ADP3301 regulators with your DAC? the highest voltage they have is 5V(ADP3301AR-5). their specs are not bad and I'm thinking of putting 3 of them on the board, one for optical receiver(3.3v), 2 for CS8414(separate digital & analog 5V Vdd's) replacing the LM78xx one.
many thanks!
How would you characterize the sound before and after overclocking?
Also, have you tried using ADP3301 regulators with your DAC? the highest voltage they have is 5V(ADP3301AR-5). their specs are not bad and I'm thinking of putting 3 of them on the board, one for optical receiver(3.3v), 2 for CS8414(separate digital & analog 5V Vdd's) replacing the LM78xx one.
many thanks!
I tried both.
The sound with reclocking is more "digital' , lacking the natural smoothness and effortlessness of not reclocked signal.
Not to mention that initially I was quite impressed with reclocking, it only a year later that I found it's not the "right" sound
But try it yourself, I would be interested in hearing others opinion on that.
I think PS has very big influence on the sound of those DACs. My AC, extensively filtered supply, seems to be sounding better than battery now.
The sound with reclocking is more "digital' , lacking the natural smoothness and effortlessness of not reclocked signal.
Not to mention that initially I was quite impressed with reclocking, it only a year later that I found it's not the "right" sound
But try it yourself, I would be interested in hearing others opinion on that.
I think PS has very big influence on the sound of those DACs. My AC, extensively filtered supply, seems to be sounding better than battery now.
I am using Linear Technology's LT1963 [LDO] regs for my units.
They quote low noise and comparative lack of cap sensitivity for
these.
Plus they are available BIG rather than silly little sot23 packages!
I have not tried the soic AD devices yet , though I do have some.
I have not done such exhaustive comparisons as Peter , but
have certainly benefited form the fruits of his labour , cheers
again Peter .
Regards dac sound, I still find the HF roll off bothers me a little.
But I do seem to get the impression that some consider the 'cure' is worse than the 'illness' though
I am certainly interested in experimenting with some sort of 'active' IV conversion approach.
I wonder what the effects of such an approach will have on the
HF performance, as most mention bass and dynamics as affected
areas.
From what I understand HF roll off is the result of the NOS approach.
Is this why some people prefer the NOS sound ?
I do understand that my implementation of the TDA is probably not a patch on some here, but never the less, unless corrected,
they all must have this HF roll off ? 
To single out a fave sound....Cymbals
Cymbals need HF or
Starting to ramble now
Setmenu
They quote low noise and comparative lack of cap sensitivity for
these.
Plus they are available BIG rather than silly little sot23 packages!
I have not tried the soic AD devices yet , though I do have some.
I have not done such exhaustive comparisons as Peter , but
have certainly benefited form the fruits of his labour , cheers
again Peter
.Regards dac sound, I still find the HF roll off bothers me a little.
But I do seem to get the impression that some consider the 'cure' is worse than the 'illness' though
I am certainly interested in experimenting with some sort of 'active' IV conversion approach.
I wonder what the effects of such an approach will have on the
HF performance, as most mention bass and dynamics as affected
areas.
From what I understand HF roll off is the result of the NOS approach.
Is this why some people prefer the NOS sound ?
I do understand that my implementation of the TDA is probably not a patch on some here, but never the less, unless corrected,
they all must have this HF roll off ?
To single out a fave sound....Cymbals
Cymbals need HF or
Starting to ramble now
Setmenu
I didn't observe any HF roll off. The highs are actually very good.
I've built once an oversampling DAC and it seemed to me it had more HF roll off, that my current one (TDA1543)
http://www.diyaudio.com/forums/showthread.php?s=&threadid=17353
I've built once an oversampling DAC and it seemed to me it had more HF roll off, that my current one (TDA1543)
http://www.diyaudio.com/forums/showthread.php?s=&threadid=17353
Setmenu, I agree with Peter. The rolloff does exist but it's around 20kHz, which is hardly be detected by a human being, only by using equipment, I think.
I built an alternative passive I/V stage earlier with a +3dB correction, but didn't notice anything better in the high frequency area, except the sound wasn't as rich as using the Nonoz 3 BG-based stage.
Overall, the sound of the DAC doesn't lack the high frequency and cymbals sound pretty good, I just listened to Brahms Hungarian Dance #5 this morning and there's plenty of plates and cymbals there.
Ho do you connect your digital input to CS8412?
Could you point me to the schematics that you used? it may be something else that causes roll off in the treble, that can be changed/improved.
The power supply plays the big part for sure. I haven't tried your regulators yet, still using LM7xx. I tried AN8008 for TDA1543 but it gets real hot if connected to the 24V adapter which is used for PPA. So i replaced it again with LM one.
Thanks
I built an alternative passive I/V stage earlier with a +3dB correction, but didn't notice anything better in the high frequency area, except the sound wasn't as rich as using the Nonoz 3 BG-based stage.
Overall, the sound of the DAC doesn't lack the high frequency and cymbals sound pretty good, I just listened to Brahms Hungarian Dance #5 this morning and there's plenty of plates and cymbals there.
Ho do you connect your digital input to CS8412?
Could you point me to the schematics that you used? it may be something else that causes roll off in the treble, that can be changed/improved.
The power supply plays the big part for sure. I haven't tried your regulators yet, still using LM7xx. I tried AN8008 for TDA1543 but it gets real hot if connected to the 24V adapter which is used for PPA. So i replaced it again with LM one.
Thanks
Peter Daniel said:I didn't observe any HF roll off. The highs are actually very good.
I've built once an oversampling DAC and it seemed to me it had more HF roll off, that my current one (TDA1543)
http://www.diyaudio.com/forums/showthread.php?s=&threadid=17353
Obviously it depends on the whole system and ones own preferences as to how one perceives the sound of any piece of
equipment.
Compared to many here my own setup is indeed humble.
The fact that my 'good' system is a headphone based one is also
probably rare on this site.
But the Stax electrostatics to my ear do trebles very nicely and it is
possible to detect variations in HF presentation very clearly with
them.
The nos dac does sound very nice with the Stax phones, it is just
that there does seem to be a little something less than I would
prefer.
But the above is subjective.
On the objective front, I do believe I am correct in thinking that the nos approach does not result in a 'flat' response .
I believe it usual to expect a little roll off at the lowest extremes
coupled with a 3db point at around 20KHZ.[for the 1543 at least?]
IpsilonSound
The roll of is around 3db at 20K but begins at around 10-12 I believe.
So this is a small but real reduction in HF over a pretty wide spectrum.
To my ear it changes the signature of vocals, instruments,ambience..
Not a lot, but enough to perk my interest.
How much this bothers me depends on the recording.
Sometimes it can be beneficial.
I do not doubt my circuit could honed to produce finer results,
compromises have been made for portable use, only using a optical input for starters.
But overall it performs an amazing job considering!
Compared to my Arcam CD23 it is quite a close run thing.
Once acclimatized to the fact it will be/is different I can happily listen and just enjoy music.
The only thing I avoid is playing sine wave sweeps
Setmenu
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.