Next week I'll start a new PCB design of my version of the well known non OS DAC with 8412 and TDA1543. It'll have a input transformer ( again ! ) and a discrete power supply for the TDA.
Since I am planning to have the PCB's made professionally I have to order a large amount of PCB's. The old version of my DAC can be seen on www.dhtrob.com under Projects ( Nonodac ). The new board will be better layed out for BG/OSCON etc. and it'll have a ground plane but it will have strong similarities with the old one of course. Output will still be unbuffered and passive and the BG's will be used in Super E. Sorry I had already made a few but I don't have pictures of that version. The old one sounded quite nice as you can read at the mentioned site.
My question is if there are DIYers out there that would like to have such a PCB. They won't be cheap that's for sure but that comes with the quality I have in mind. It is not my meaning to get rich of selling these PCB's but merely wanting to order a large quantity so that the price will go down. Besides that noone else is offering PCB's with input transformers so I guess I am not in the way of anyone.
Call it a kind of group buy or something. Please realise that production depends on how many want to have such a board. So there is a risk that this project won't be realised
Detailed information and prices will follow. This PCB is for experienced builders that don't need much except for a PCB and a schematic.
Since I am planning to have the PCB's made professionally I have to order a large amount of PCB's. The old version of my DAC can be seen on www.dhtrob.com under Projects ( Nonodac ). The new board will be better layed out for BG/OSCON etc. and it'll have a ground plane but it will have strong similarities with the old one of course. Output will still be unbuffered and passive and the BG's will be used in Super E. Sorry I had already made a few but I don't have pictures of that version. The old one sounded quite nice as you can read at the mentioned site.
My question is if there are DIYers out there that would like to have such a PCB. They won't be cheap that's for sure but that comes with the quality I have in mind. It is not my meaning to get rich of selling these PCB's but merely wanting to order a large quantity so that the price will go down. Besides that noone else is offering PCB's with input transformers so I guess I am not in the way of anyone.
Call it a kind of group buy or something. Please realise that production depends on how many want to have such a board. So there is a risk that this project won't be realised
Detailed information and prices will follow. This PCB is for experienced builders that don't need much except for a PCB and a schematic.
Did you consider Elib to make PCBs? He mentioned he can supply boards at $2: http://www.diyaudio.com/forums/showthread.php?s=&postid=236099#post236099
Will be reclocking implemented too?
Will be reclocking implemented too?
Peter Daniel said:
Because of the stray RF it causes and the lack of really understanding the benefits of asynchronous reclocking. I do believe ( oops wrong word ) in the benefits of synchronous reclocking but if I implemented this I would be in the way of other people that offer boards just like implementing a tube buffer etc.
Just a simple and tidy DAC as once designed by members of the amp board forum ( which was an open source project after all ) with some more whistles and bells. When any kind of buffering ( SS/tube ) is wanted by the builder he\she is free to add that stage on a separate PCB or even hardwired.
I'll send a mail to Elib for more information about the quality of the chinese made boards.
Then this won't be a suitable PCB for you 
Please connect your scope to the DAC that has asynchronous reclocking and see. Remember my version won't have an active output stage. In fact the last ones I made had no filtering except for one cap. To be honest I liked the version without filtering most despite the risk of RF. My amp did not protest.
You see I am not a friend of RF. Not in my amp anyway.
Please connect your scope to the DAC that has asynchronous reclocking and see. Remember my version won't have an active output stage. In fact the last ones I made had no filtering except for one cap. To be honest I liked the version without filtering most despite the risk of RF. My amp did not protest.You see I am not a friend of RF. Not in my amp anyway.
Both my DACs don't have active stage, just resistors and no filters. I never noticed lack of filters. I still have both DACs so I will have another listen
But what I remeber adding asynchronous reclocking (suggested by Elso) brought the TDA1543 DAC to a next level. Actually I still prefer it to my TDA1541 S2 in my Marantz (without reclocking). Highs are a bit more rough, but this doesn't bother me either
What I really like is a great resolution of a passive I/V converting and I suspect that OPA 627 in Marantz isn't the best choice.
But what I remeber adding asynchronous reclocking (suggested by Elso) brought the TDA1543 DAC to a next level. Actually I still prefer it to my TDA1541 S2 in my Marantz (without reclocking). Highs are a bit more rough, but this doesn't bother me either
What I really like is a great resolution of a passive I/V converting and I suspect that OPA 627 in Marantz isn't the best choice.
Konnichiwa,
And what has that to do with price of tea in china, or the subjective sonics for that case?
BTW, this is going to bake you all guy's noodle....
The Shanling CD T-100 I currently use in front of an Ack!Dac upsamples (via CS8420) the Digital Output. The Ack!Dac lacks the "anti sinc" Filter I suggest.
In my current system (MF Audio TVC, Shanling SP80 Amplifier with the solid state/digital pot "preamp" bypassed, Supravox open baffles, Bastanis Prometheus open baffles or Beauhorn B2.2 Revelation) I prefer more often than not the Upsampled output to the Non Oversampling DAC. Upsampled to 96KHz vs unupsampled 44.1 to the NOS Ack!Dac has more detail and sounds more natural on cymbals and brushes and the soundscape gains depth. This is the only time I have not noticed any degradation going hand in hand with this. I will have to try a passive anti sinc filter on the Ack!Dac one day....
At the same time switching the Upsampling in within the Shanling CD Player (Pacific/Microsoft PDM200 Filter, parallel PCM1704 DAC's, appnote analogue stage plus alve buffer), reliably FUBAR's the sound and not just a little either.
Go figger....
Sayonara
jean-paul said:
And what has that to do with price of tea in china, or the subjective sonics for that case?
BTW, this is going to bake you all guy's noodle....
The Shanling CD T-100 I currently use in front of an Ack!Dac upsamples (via CS8420) the Digital Output. The Ack!Dac lacks the "anti sinc" Filter I suggest.
In my current system (MF Audio TVC, Shanling SP80 Amplifier with the solid state/digital pot "preamp" bypassed, Supravox open baffles, Bastanis Prometheus open baffles or Beauhorn B2.2 Revelation) I prefer more often than not the Upsampled output to the Non Oversampling DAC. Upsampled to 96KHz vs unupsampled 44.1 to the NOS Ack!Dac has more detail and sounds more natural on cymbals and brushes and the soundscape gains depth. This is the only time I have not noticed any degradation going hand in hand with this. I will have to try a passive anti sinc filter on the Ack!Dac one day....
At the same time switching the Upsampling in within the Shanling CD Player (Pacific/Microsoft PDM200 Filter, parallel PCM1704 DAC's, appnote analogue stage plus alve buffer), reliably FUBAR's the sound and not just a little either.
Go figger....
Sayonara
As I understood the higher frequency the better. That means a low jitter 80 or 100 MHz XO has to be used. That frequency and it's harmonics WILL show up at the output in my implementation.
And where do I find a low jitter 80 or 100 MHz module that is suitable ? I guess the answer is building a discrete clock which takes space and will stress the design/test stage severely. The version I will redesign is in fact a redesign with another CAD program of a built and tested DAC that was plain good.
It also adds to the total cost of the project and it delivers a risk of people having problems with the RF etc.
And where do I find a low jitter 80 or 100 MHz module that is suitable ? I guess the answer is building a discrete clock which takes space and will stress the design/test stage severely. The version I will redesign is in fact a redesign with another CAD program of a built and tested DAC that was plain good.
It also adds to the total cost of the project and it delivers a risk of people having problems with the RF etc.
I'm using a crystal close to 100Mhz. The reclocker might add to the price, but it also can be optional on a board. Somebody who wants to use it, can install all the parts, and somebody who doesn't want it, leaves the spaces empty. With careful bord layout, it can be somwhere on a side and wouldn't interfere with a main circuit.
I'm also using separate transformer for reclocking circuit. From what I remember it was major improvement over the bare bone TDA 1543 circuit.
I'm also using separate transformer for reclocking circuit. From what I remember it was major improvement over the bare bone TDA 1543 circuit.
????? What does NAD output mean ???????
That somebody has obviously other experiences than me. OPA627 is way better than OPA2134. I did test this. That doesn't mean OPA2134 is bad. When you need a double opamp it is OK, maybe OPA2604 is a better choice but that depends on taste/implementation I guess. Nowadays I would use LM6172 but only after careful examination of the cdplayer/schematic in question. Of the oldies I did like OP249 too.
IMO it is a pity that single opamps are so seldom used in good cdplayers as the creme-de-la-creme is found amongst single versions. That makes way for building discrete, I know
That somebody has obviously other experiences than me. OPA627 is way better than OPA2134. I did test this. That doesn't mean OPA2134 is bad. When you need a double opamp it is OK, maybe OPA2604 is a better choice but that depends on taste/implementation I guess. Nowadays I would use LM6172 but only after careful examination of the cdplayer/schematic in question. Of the oldies I did like OP249 too.
IMO it is a pity that single opamps are so seldom used in good cdplayers as the creme-de-la-creme is found amongst single versions. That makes way for building discrete, I know
www.dhtrob.com seems down but I found this listening session:
http://home.zonnet.nl/horneman/dhtrob.htm
http://home.zonnet.nl/horneman/dhtrob.htm
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.