Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Digital Line Level DACs, Digital Crossovers, Equalizers, etc.

Building the ultimate NOS DAC using TDA1541A
Building the ultimate NOS DAC using TDA1541A
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 11th October 2018, 08:55 PM   #6661
ryanj is offline ryanj  Australia
EE Hack
diyAudio Member
 
ryanj's Avatar
 
Join Date: Jul 2010
Location: Seaford
Building the ultimate NOS DAC using TDA1541A
I mentioned this issue of the LSB with Ians pcb for the 1541 a few years back. In the end I thought it wasn't a problem because it sounded so much better than i2s. All this time I was actually listening to a 15bit dac. ...until EC came along and shared his circuit.



Quote:
Originally Posted by ryanj View Post
Hi Fellas,

Dont mean to litter the forum with scope shots, but just wondering what exactly is going on here (aside from the noise from BCK on the Analog outputs) It looks like the sample starts before the least significant bit is clocked in? According to the data sheet (TDA1541A) this shouldn't be so - Unless i'm not clear about something.

Also notice how much BCK is visible on the Analog outputs without I2S attenuation - Hope to change that soon.

Bandwidth limited,
Top trace is BCK using only 50mhz probe - hence the slow rise time.
Bottom trace 11025hz square wave at full scale current output.
Simultaneous mode.

Ryan
  Reply With Quote
Old 11th October 2018, 09:34 PM   #6662
koldby is offline koldby  Denmark
diyAudio Member
 
koldby's Avatar
 
Join Date: Jun 2003
Location: Ruds Vedby
Quote:
Originally Posted by batteryman View Post
A crude way would be a monostable and gate to simply delay the leading edge of LE by about 2clock periods - ie 1.4us, (and also to invert CLK)
I just got an idea to make it easy to get Ians coverter to work as it should, and went into the lab. to test it.
What about using the LE (or word clk ) input to Ians board as LE to the TDA1541A
See the picture.
It looks as if this would work just perfectly. You still have to invert the CLK of course.
I am sure some of the clever heads in this thread will correct me , if I am wrong
Attached Images
File Type: jpg IMG_20181011_232506.jpg (93.0 KB, 154 views)
__________________
What is Home Theatre without good sound?
  Reply With Quote
Old 12th October 2018, 06:14 AM   #6663
ecdesigns is offline ecdesigns  Netherlands
diyAudio Member
 
Join Date: May 2006
Location: Holland
Hi batteryman.

Quote:
Yes, looks like Ian has got it wrong but will inverting BCK do anything to cure the peak break up?
Ian may not have designed this converter to be compatible with the TDA1541A like rfbrw mentioned, but for other D/A converter chips like the AD1865 for example.

AD1865 clocks in data on the -rising- edge of BCK and this is exactly what Ians converter does, so nothing wrong here.

AD1865 latches its outputs on the -falling- edge of LE, Ians decoder also generates this signal correctly.

Just have a look at AD1865 datasheet page 7, this should clear things up.


TDA1541A just happens to clock in data on the falling edge of BCK and latches its outputs on the rising edge of LE, just the other way around. TDA1541A doesn't like it when LE changes state while data is still being clocked in either.

So if we want to drive a TDA1541A (SIM) with Ian's converter, we have to use some glue logic to make it work.

> BCK has to be inverted.

> LE signal has to be modified like i described in post #6630.
  Reply With Quote
Old 12th October 2018, 06:41 AM   #6664
batteryman is offline batteryman  United Kingdom
diyAudio Member
 
Join Date: Jun 2011
Location: SCOTLAND
Quote:
Originally Posted by ecdesigns View Post
Hi batteryman.

Ian may not have designed this converter to be compatible with the TDA1541A like rfbrw mentioned, but for other D/A converter chips like the AD1865 for example.

AD1865 clocks in data on the -rising- edge of BCK and this is exactly what Ians converter does, so nothing wrong here.

AD1865 latches its outputs on the -falling- edge of LE, Ians decoder also generates this signal correctly.

Just have a look at AD1865 datasheet page 7, this should clear things up.

TDA1541A just happens to clock in data on the falling edge of BCK and latches its outputs on the rising edge of LE, just the other way around. TDA1541A doesn't like it when LE changes state while data is still being clocked in either.

So if we want to drive a TDA1541A (SIM) with Ian's converter, we have to use some glue logic to make it work.

> BCK has to be inverted.

> LE signal has to be modified like i described in post #6630.
There is a specific OB jumper on the board specifically for use with the 1541 and connection diagrams for normal and a pair in balanced, but he seems to have overlooked the difference between it and the other dac chips.

What do you think of inverting BCK and then using Koldby's suggestion of feeding LE from the board's input, bypassing the board?

Now, correcting these shortcomings may improve the sound, but may not explain the distortion on peaks but would the wrong format? E.g.right justified, or normal pcm, not I2S
  Reply With Quote
Old 12th October 2018, 07:43 AM   #6665
Hanze Khronye is online now Hanze Khronye  Australia
diyAudio Member
 
Join Date: Jan 2018
I have read that that Ians PCM board doesn't work quite right with TDA1541A, that it does function but results in 15bit performance. No doubt many of you have heard/read the same.

Perhaps this can be fixed with a firmware upgrade to Ians board (I2S to PCM for use with TDA1541A), perhaps Ian can, and will assist.

To batteryman, I have experienced what you speak of, and it presented in the same way. In my instance it was the peak to peak signal waveform was crashing into the DC rail voltage.

With hope that it all works out.

Hanze.

Last edited by Hanze Khronye; 12th October 2018 at 08:13 AM.
  Reply With Quote
Old 12th October 2018, 07:51 AM   #6666
3lite is offline 3lite  Poland
diyAudio Member
 
Join Date: May 2016
Definitely. It should be an easy fix for Ian if he wants to fix it.

Quote:
Originally Posted by batteryman View Post
Now, correcting these shortcomings may improve the sound, but may not explain the distortion on peaks but would the wrong format? E.g.right justified, or normal pcm, not I2S
Yes, it does explain it. Again, MSB, hint, hint...
  Reply With Quote
Old 12th October 2018, 08:24 AM   #6667
Hanze Khronye is online now Hanze Khronye  Australia
diyAudio Member
 
Join Date: Jan 2018
Quote:
Originally Posted by 3lite View Post
Definitely. It should be an easy fix for Ian if he wants to fix it.



Yes, it does explain it. Again, MSB, hint, hint...
I believe that Ian is a good guy, he has offered much and tailored to the individual, and has delivered.

It should be a 'walk in the park' for him to address this issue.

Hanze.

Last edited by Hanze Khronye; 12th October 2018 at 08:30 AM.
  Reply With Quote
Old 12th October 2018, 08:39 AM   #6668
koldby is offline koldby  Denmark
diyAudio Member
 
koldby's Avatar
 
Join Date: Jun 2003
Location: Ruds Vedby
Quote:
Originally Posted by Hanze Khronye View Post
I believe that Ian is a good guy, he has offered much and tailored to the individual, and has delivered.

It should be a 'walk in the park' for him to address this issue.

Hanze.
Sure it would be no problem to fix that on NEW i2stopcm boards, but all the boards that has been sold to people , that wants to use them with TDA1541A , must use a hardware fix.
It is not as difficult as it sounds, though. I believe the solution I suggested in post #6662 will be an easy fix and even be better, as the LE transition is many clk cycles after the clk to TDA1541A has stopped.
__________________
What is Home Theatre without good sound?
  Reply With Quote
Old 12th October 2018, 09:05 AM   #6669
batteryman is offline batteryman  United Kingdom
diyAudio Member
 
Join Date: Jun 2011
Location: SCOTLAND
Quote:
Originally Posted by 3lite View Post
Definitely. It should be an easy fix for Ian if he wants to fix it.

Yes, it does explain it. Again, MSB, hint, hint...
That's what I thought, but there was a suggestion it was not Ian's board but the wrong data format........

I didn't accept that as it is only the peaks that cause the distortion - ie when the MSB switches.

Also, regarding comments that the LSB isn't significant and you cannot hear the difference between 14 and 15 bits, surely, the LSB is set for every odd numbered value from 1 to max so you cannot just dismiss it. MSB -1 you could if you accepted a resolution reduction.

I'm sure we'll hear from Ian about a firmware update soon.
  Reply With Quote
Old 12th October 2018, 09:07 AM   #6670
batteryman is offline batteryman  United Kingdom
diyAudio Member
 
Join Date: Jun 2011
Location: SCOTLAND
Quote:
Originally Posted by koldby View Post
Sure it would be no problem to fix that on NEW i2stopcm boards, but all the boards that has been sold to people , that wants to use them with TDA1541A , must use a hardware fix.
It is not as difficult as it sounds, though. I believe the solution I suggested in post #6662 will be an easy fix and even be better, as the LE transition is many clk cycles after the clk to TDA1541A has stopped.
This certainly avoids having to ship the board back to Ian with postage costs for both parties (Ian ought to pay his customers shipping though)

I'll try the hardware fix this weekend and report back.
  Reply With Quote

Reply


Building the ultimate NOS DAC using TDA1541AHide 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

Forum Jump


New To Site? Need Help?

All times are GMT. The time now is 07:52 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki