NOS Nakamichi DAC-101 2X TDA1541A-S1

[fragglan]

Hello! I have an old Nakamichi DAC-101 that i would like to run in NOS. The digital filter YM3414 (4 or 8times OS) split the I2s signal to one left and one right . The left signal comes to the dac TDA1541A-S1 like one "left normall" and one "left inverted". The right signal goes to it's dac in the same way as the left do. I need to find a device that can replace the digital filter. I only want the I2s signal to be splited in left and right without filtering or OS. I send the schematic for you people to see.

[amc184]

The digital filter YM3414 (4 or 8times OS) split the I2s signal to one left and one right.

Unfortunately not. The signal bus between the receiver and the digital filter is not I2S, but a time multiplexed right justified format.

I only want the I2s signal to be split into left and right without filtering or OS.

There is no 'only' about it.

The YM3414 LSI currently in your DAC does another couple of jobs apart from oversampling and filtering. It de-multiplexes the data and changes the digital format. The glue logic after the filter LSI further alters the digital bus and improves its timing accuracy.

Replacing this functionality is a big job. I expect you've seen the threads on removing the SAA7220 filter in a lot of players, which is simple to do (this is because it only filters and oversamples, it doesn't change the digital bus format).

I am genuinely sorry to rain on your parade, but unless you have a lot of experience designing with glue logic, or preferably CPLDs, this is not a project I would even contemplate starting.

[fragglan]

Originally Posted by amc184

Unfortunately not. The signal bus between the receiver and the digital filter is not I2S, but a time multiplexed right justified format.



There is no 'only' about it.

The YM3414 LSI currently in your DAC does another couple of jobs apart from oversampling and filtering. It de-multiplexes the data and changes the digital format. The glue logic after the filter LSI further alters the digital bus and improves its timing accuracy.

Replacing this functionality is a big job. I expect you've seen the threads on removing the SAA7220 filter in a lot of players, which is simple to do (this is because it only filters and oversamples, it doesn't change the digital bus format).

I am genuinely sorry to rain on your parade, but unless you have a lot of experience designing with glue logic, or preferably CPLDs, this is not a project I would even contemplate starting.

Ok, i'm happy to get the answer so fast, and you are absolutly right... i'm not that experienced to "fix" with this kind of things, and yes i thaught it would be that easy just to find a spliter IC. My plans are now ruined, and i thinking about desolder the dac's. I really would like to hear these beauty's nos.
I allready have the S1 single crown in my digital amplifier (marantz PM-95) and i just love how it sounds. I have to try building the DIY-dac with these two chip's, or is it a waste of high performance dac's? Is the TDA1541A good enaught for this? Advices is thankfully recived.

[guido]

I can't read that rtf file and the online conversion tools either. So i don't know what is before the digital filter. In case it's I2S, i worked on a dac some time ago which did accept I2S and used two TDA1541's the way you want them to be used (so one DAC per channel). Allthough it's not perfect (see thread). No filter/oversampling.

It does require some glue logic (most important part is a shift register), so i think it's too complex. Thread is here on this forum.

But, i did see something similar once in a chinese DAC design being sold on the net. And they used a CPLD, so a one chip solution. Maybe that would suit you.

Personally i would replace the YM with a PMD100. Should be a much better filter and you can also use the TDA's differential. But this also requires some additional glue logic and going beyond the TDA's specs

[amc184]

So i don't know what is before the digital filter.

Neither do I (I'd put my money on a YM3623 or similar), I based my assessment on what the YM3414 accepts, which is very dissimilar to I2S.

I have to try building the DIY-dac with these two chip's, or is it a waste of high performance dac's?

Using the S1s for a DIY DAC is a good idea, and there are kits avalible for using two TDA1541As, but usually just in parallel, rather than the differential configuration your Nakamichi uses.

Another idea would be to build a new front end for the Nakamichi, but again this would be a lot of work.

[fragglan]

Originally Posted by guido

I can't read that rtf file and the online conversion tools either. So i don't know what is before the digital filter. In case it's I2S, i worked on a dac some time ago which did accept I2S and used two TDA1541's the way you want them to be used (so one DAC per channel). Allthough it's not perfect (see thread). No filter/oversampling.

It does require some glue logic (most important part is a shift register), so i think it's too complex. Thread is here on this forum.

But, i did see something similar once in a chinese DAC design being sold on the net. And they used a CPLD, so a one chip solution. Maybe that would suit you.

Personally i would replace the YM with a PMD100. Should be a much better filter and you can also use the TDA's differential. But this also requires some additional glue logic and going beyond the TDA's specs

The file i paste in the thread is a zipped word document. I send the file with the digital receiver. It's the pioneer PD0037. hope you can open this. Do you think it would improve the sound to run the digital filter in 4xOS ? What i heard, heavy OS make the high frequency's to fade a way,with less details. I maby forgot to say this is a dac for cars, but i thinking in build a new power source to run it with my stereo system. Or.. as i said before, take out the dac's and build something new. That would really be a shame when this is a good sounding dac. I bought it just for the two s1 single crown chip's, and to make it to NOS. I have to think some days...

[guido]

I can't open rtf. Don't have office installed at home and never found the need to do so. If i really need office, i'll use the pc at work (where i can also print stuff).

As for the DF, i don't know this one. Played with different settings on the pmd100, but i could not hear a difference (yet). But it was only a test.

[amc184]

Why are you posting pictures within rich text format documents that are zipped? It's much easier for everyone trying to help you if you just post JPEGs.

The fact that the receiver is a PD0037 doesn't really change any of what I said before though.

[amc184]

Anyway, here are the schematics as JPEGs for this discussion:

[guido]

Should have seen this when i was working on the PMD100 to TDA1541a connection. Same mode on the TDA1541 and it shows glue logic to create LE pulse.

Nothing to find on the pioneer chip. But i don't see some pins to select a mode or so. S1,S2 seem to be input selection. So i agree, seems difficult to adapt. Or a CPLD and some time...