Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Is it worth the trouble to build a DAC with TDA1541?
New Posts  All Forums:Forum Nav:

Is it worth the trouble to build a DAC with TDA1541? - Page 2

post #16 of 27
Thread Starter 
Thanks for the inputs guys.

Quote:
Originally Posted by erin View Post
If you want to design one from scratch on vero boards you would most likely at best get something that sounds as good as the above kits but it would take you much, much longer to build than the kits, and at worst would not sound good due to the track layout etc.

Save yourself the time and get a kit.

.
not really. If I do build it I want to use all teflon for the by-pass caps and likely all battery power supply. Then it is much easier to just use perf board.

The "king of dac chip" gets the special treatment I guess. Then I won't need another dac for a long long time. The DAC technology keeps on changing, I want to make something that I can just stick with.

one more question: is it better to go "4x 1541 in parallel, no buffer" , or "1x 1541 followed by a 5687 srpp stage"?
post #17 of 27
Teflon? You are aware that you need 14 bypass caps?

Quote:
Originally Posted by AudioCats View Post
one more question: is it better to go "4x 1541 in parallel, no buffer" , or "1x 1541 followed by a 5687 srpp stage"?
Neither. 4x chips still won't give nearly enough voltage or current. And, SRPP is totally inappropriate. It was a fashionable tube stage in the 90's, but can be bettered by a simple grounded cathode stage in this position.
post #18 of 27
Quote:
Originally Posted by AudioCats View Post
Supposedly the 1541 is one of the oldest dac chips, is it worth the trouble to build a DAC with it? ( I have read that some folks think it is still the best out there.)
I guess it depends how you feel about solid state progress over the last 30 years. As far as I'm aware basically every piece of IC has improved by orders of magnitude in that time. Perhaps DACs are the one exception that actually get worse with technological progress and increased sophistication. But it's worth a little research imo before embarking on what seems to be a very antiquated project.

Having said that I do like tube amps, and not because I believe that they are doing a better job amplifying the signal. I actually like tubes because they are so antiquated, kind of like a classic car.
post #19 of 27
I dunno, but I would be a little surprised if just throwing a few teflons at something that seems to need pampering to sound its best working as intended. you need to supply a high grade power supply, regulation and good quality clock to something like this, especially if you plan on using NOS. teflons will just make the track lengths longer with more induction IMO I cant see that solving more problems than it causes personally. with these it seems the surrounding circuitry is at least as important as the chip itself. high grade SMD films by evox/rifa or wima to make up the entire circuit is much more likely to do as intended IMO and would be cheaper, even allowing you budget to get the boards made if you are up for a project in eagle or whatever. there are actually a couple of high end projects on DIYAUDIO at the moment using these chips, I know you are keen to do this yourself, so I wont push, but I really think that for a build like you are wanting you would be better to use another chip not so fussy, perhaps a sabre 9022 or PCM1704. also i'm under the impression that bypassing high speed circuits needs to be in much closer proximity than largish axial lead caps will allow for it to do the job, someone correct me if i'm wrong.
post #20 of 27
Quote:
Originally Posted by Spacehead View Post
You seem to have a lot of knowledge about these DACs. I have one Technics CD player, that has PCM54 DAC. Would it be possible to extract that circuit from that cd player and some how utilize the PCM54 dac chip?
The PCM53/54 DACs sound awesome so I understand why you would want to use it but they are parallel input not i2s so are quite difficult to work with these days unless you are able to design some "glue logic" or able to program an FPGA to convert SPDIF or I2s to parallel input.
Good luck with that project.
post #21 of 27
Quote:
Originally Posted by AudioCats View Post
Thanks for the inputs guys.



not really. If I do build it I want to use all teflon for the by-pass caps and likely all battery power supply. Then it is much easier to just use perf board.

The "king of dac chip" gets the special treatment I guess. Then I won't need another dac for a long long time. The DAC technology keeps on changing, I want to make something that I can just stick with.

one more question: is it better to go "4x 1541 in parallel, no buffer" , or "1x 1541 followed by a 5687 srpp stage"?
Mate, if you looked at the link i gave you, The analogmetric kit has enough room for you to use whatever capacitor you want. You could also run it off batteries if you wanted. You can also just buy the PCB from them for about $20.

There are enough fans of the TDA1541 for you to easily google and see the popularity of this DAC. Yes it is a good one. Use it. No matter which DAC you choose there will always be someone to tell you that your choice is wrong and others to tell you you are right.

If you want to build your own from scratch go ahead and do it. From my own experience it is much easier to solder onto a PCB than to perf board.

These kits are are good although the PCB layout could have been a little better, its good enough.
They are much better than most Philips PCB layouts using the TDA1541.

Should you use SRPP? Well that depends upon whether you like tube sound or not. Or what tube you are prepared to use?

I think you would be better off trying an op-amp first.
4X TDA1541 is excessive for a first go.
2X TDA1541 will give you the benefits of paralleling without the $ cost of four ICs. I personally think the buffered output sounds better than passive.
post #22 of 27
I also agree with qusp above.

If you want to see a good TDA1541 design using SMD caps for the decoupling capacitors look at this CD player Whoor!!!!!

I'd like to hear this one:

6moons audio reviews: Abbingdon Music Research CD-77 and AM-77
post #23 of 27
I have 2 dacs here that use that chip in different ways. You should really take a look at the Audiosector dac, available as a kit - and I can tell you it sounds very good.

Theres also the DDDac which tackles the current output by stacking a bunch of the chips together.

But, really while you will get a working version by building on perfboard, I think it would be hard to get it sounding its best in that set up. I think for any kind of high speed digital stuff, very careful layout is required, and most likely at least 2 layer boards.

the chip, when properly implemented sounds great.... really on a par with anything else out there. If you are going to have a go I would suggest using a Pass D1 output stage for the I/V conversion.

Heres some links for you:

audiosector thread

DDDAC 2000

Pass D1 output stage

Pass D1 I/V for TDA1545 - diyAudio

The Pass D1 For Tda1543 - diyAudio


and theres lots there but you will need to search....


Watch your power supplies - they make a large difference!! Shunts all the way....


Fran
post #24 of 27
Quote:
Originally Posted by rds View Post
I guess it depends how you feel about solid state progress over the last 30 years. As far as I'm aware basically every piece of IC has improved by orders of magnitude in that time. Perhaps DACs are the one exception that actually get worse with technological progress and increased sophistication. But it's worth a little research imo before embarking on what seems to be a very antiquated project.

Having said that I do like tube amps, and not because I believe that they are doing a better job amplifying the signal. I actually like tubes because they are so antiquated, kind of like a classic car.
Despite that i know absolutely nothing, this what i suspect.

I have a circa 1989 Marantz CD-80 CDP and a PM-75 amp that also has a DAC. They both use the TDA1541 and both DACs sound identical. They sound way too dark for my tastes.
post #25 of 27
Part of what you are hearing may be the output (I/V) stage. Probably NE5532/34 opamps implemented straight out of the datasheet.

Speaking of output stages, I use this one currently in my rotel rcd855 w/tda1541a:

Less simple I/V for TDA1541 - diyAudio

My primary concerns were low input impedance and DC coupled. Running more current thru the stage could get the impedance even lower. I would guess in my implementation it is probably in the range of 5 ohms or less (hopefully), but I haven't tried to model it or calculate it. Not even sure how to go about that
post #26 of 27
Thread Starter 
Thanks for the inputs guys.


Quote:
Originally Posted by dsavitsk View Post
Teflon? You are aware that you need 14 bypass caps?



.
actually a bit more. It seems that the msb by-pass should be larger, so I might do 0.6uf, 0,4uf, 0.2uf, 0.2uf, 0.1uf, 0.1uf, 0.1uf (msb to lsb). I will use 0.1uf/200v FT2's, that will be 17 caps per side, 34 per 1541. I know, they are huge. That is why I don't want to use commercial boards.

It turns out that I can't do 4x1541 anyway, not when I want to stay with teflon-only. I only have about 100 FT2 laying around. I do also have 30 some FT3 0.22uf for the tube buffer output.

going with vitQ's will be much easier, but they require teflon by-pass anyway so the size won't be that much smaller.

I have 5687 right now, what tube do you guys recommand? Do I need interstage transformer to do the I/V, or is a fancy resistor good enough?

does anybody know the (ball park) current draw of each supply voltage? It will help me to figure out how many batteries I might need.
post #27 of 27
Quote:
Originally Posted by AudioCats View Post
Thanks for the inputs guys.




actually a bit more. It seems that the msb by-pass should be larger, so I might do 0.6uf, 0,4uf, 0.2uf, 0.2uf, 0.1uf, 0.1uf, 0.1uf (msb to lsb). I will use 0.1uf/200v FT2's, that will be 17 caps per side, 34 per 1541. I know, they are huge. That is why I don't want to use commercial boards.

It turns out that I can't do 4x1541 anyway, not when I want to stay with teflon-only. I only have about 100 FT2 laying around. I do also have 30 some FT3 0.22uf for the tube buffer output.

going with vitQ's will be much easier, but they require teflon by-pass anyway so the size won't be that much smaller.

I have 5687 right now, what tube do you guys recommand? Do I need interstage transformer to do the I/V, or is a fancy resistor good enough?

does anybody know the (ball park) current draw of each supply voltage? It will help me to figure out how many batteries I might need.
The TDA1541 is the ultimate tweakers DAC where you can tune the sound to your own taste. No one variety of capacitor will be universally declared "the winner" you will have to find out for yourself which caps sound best to you for the decoupling pins.

The 5687 is a very nice sounding tube for a pre amp. I dont know how it will sound in a DAC output stage.

For current draw check the datasheets.

Allow me to refer you to this site which will give you some ideas:
Updates lampizator references postings
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Is it worth the trouble to build a DAC with TDA1541?