dac I/V convertion with very low distortion - Page 15 - diyAudio
Go Back   Home > Forums > Source & Line > Digital Source

Digital Source Digital Players and Recorders: CD , SACD , Tape, Memory Card, etc.

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 12th August 2012, 06:57 PM   #141
diyAudio Member
 
Join Date: Apr 2003
Location: Eastern Pennsylvania
Quote:
Originally Posted by smms73 View Post
As the doubt about the need of matched transistors appears very often, i post here the same circuit with bc327 for pnp and bc547c for all npn, they are very different and the thd is 0.000008% (7,8ma/100k), its no need for matched transistors. hope this end the subject, and we can move on.
Okay, today I took the time to enter the circuit in my simulator. I must say, I'm impressed. I randomly substituted all manner of various transistors throughout the circuit, and found the THD indeed remains in the vicinty of -140dB, although the specific profile of hamonics vary. This is highly unusal behavior for an open-loop bipolar circuit in my experience. Well done.
__________________
Ken
  Reply With Quote
Old 12th August 2012, 07:13 PM   #142
diyAudio Member
 
Joachim Gerhard's Avatar
 
Join Date: Mar 2009
A circuit idea like this appears every 10 years maximum and that in a surrounding where many search the holy grail in I/V converters. Now that i see it it is so obvious as it would have being there all the time. Certainly a stroke of genius in my book. Lets see what it brings in a practical implementation.
  Reply With Quote
Old 12th August 2012, 08:14 PM   #143
MiiB is offline MiiB  Denmark
diyAudio Member
 
Join Date: May 2005
Location: Denmark
I tend to agree....Very very nice Concept...
  Reply With Quote
Old 12th August 2012, 10:10 PM   #144
diyAudio Member
 
dirkwright's Avatar
 
Join Date: May 2011
Location: Virginia
Quote:
Originally Posted by Ken Newton View Post
Ideally, voltage gain would be much higher, increasing as the virtual ground impedance of the grounded-base stage input is reduced. But what matters here is current flow. The circuit conveys DAC output current to the 400R I/V resistor, while isolating the DAC output from the voltage swing developed across the I/V resistor.
OK, then I don't understand this circuit.
  Reply With Quote
Old 12th August 2012, 10:33 PM   #145
diyAudio Member
 
Joachim Gerhard's Avatar
 
Join Date: Mar 2009
Infinte gain ? It just has not happened. Infinity is in the future.
  Reply With Quote
Old 12th August 2012, 11:23 PM   #146
diyAudio Member
 
Join Date: Apr 2003
Location: Eastern Pennsylvania
I'm refering to voltage gain as seen from between the grounded-base emitters and the I/V resistor. Let's take a couple examples for illustration. If the output voltage across the 400R I/V resistor were 1VRMS, it would mean the peak current from the DAC were 3.5mA. Now, if the input impedance of the grounded-base emitters were 2R, then the peak input voltage devloped at those emitters (and at the DAC's current output pin) will be: 2R x 3.5mA = 7mV peak = 4.9mVRMS. The overall voltage gain would then be: 1V / 4.9mV = x204 = 46dB.

If the grounded-base input impedance were 1R, then the voltage gain would be x408, or about 52dB. The lower the input impedance, the greater the apparent voltage gain. Ideally, there would be zero apparent input voltage, as the input is supposed to be an A.C. ground with an impedance of zero Ohms. This could give the appearance of infinite voltage gain (divide by zero error aside), but this would be an illusion, as the circuit is conveying current not amplifying voltage.

The apparent input to output voltage gain is not really of concern here. The conveyance of current by the circuit is what matters. DAC signal current is input to an (ideally) zero impedance virtual ground and conveyed through an I/V resistor, which in turn produces the signal voltage according to Ohm's law.
__________________
Ken

Last edited by Ken Newton; 12th August 2012 at 11:49 PM.
  Reply With Quote
Old 12th August 2012, 11:44 PM   #147
diyAudio Member
 
Joachim Gerhard's Avatar
 
Join Date: Mar 2009
The most simple solution for that is an Opamp setup for series feedback. Going into Pin 2 and grounding Pin 3. The "real" part of even a low input impedance current feedback Opamp like the LME49713 is around 50 Ohm though. The "Zero" Ohm input impedance is a result of the feedback. The topology presented here could also being setup for series feedback but then we loose the open loop character. It is the old discussion about what sounds best and what measures best.
  Reply With Quote
Old 13th August 2012, 02:29 AM   #148
regal is offline regal  United States
diyAudio Member
 
Join Date: Jan 2004
Location: MD
Quote:
Originally Posted by Ken Newton View Post
Regal,

What is the source of that 670R figure? I'm not disputing it, it's just that the output impedance of the PCM179X series has been shrouded in mystery because T.I. hasn't seen fit to document it in their datasheets. By the way, the number I've seen reported (unofficially and indirectly) is 100k, supposedly from T.I. engineers in Germany.
The same way one measures the input impedance of an I/V stage just in reverse, we know the ipk-pk output of the PCM179x, feed it an I2S signal encoded with a sine wave from LSB to MSB, DAC output to and I/V of known input resistance (10 ohm resistor to ground), scope on top of resistor. Ohm's law from there.

The "unofficial" 100k number comes from just measuring the 4 bits that are supposedly R2R (Ti calls them current segment on their functional block diagram) but the base of the architecture is the "advanced segment modulator."
  Reply With Quote
Old 13th August 2012, 09:12 AM   #149
Calvin is offline Calvin  Germany
diyAudio Member
 
Calvin's Avatar
 
Join Date: Nov 2004
Location: close to Basel
Hi,

I don´t know why, but TI is very secretive about the output impedance of their audio DACs. I got told figures of 2M||22pF for the PCM1792/1794a, but I regard those numbers as simply wrong. I use 2k||22pF instead.
Unfortunately the output impedance varies widely with different DACs.
The older Colinear 20Bit PCM63 is specced at 670Ohms. The 24Bit PCM1704 is specced at 1kOhm. Non-Audio current output DACs like the DAC5675, DAC902, etc are specced at 200k-300k, 5pF-12pF. So I assume that the PCM179x-series has similar to the other audio-DACs rather lowish impedance values.
In simulation, I use current sources with parallel R and C. The offset current beeing as stated in the datasheet for current sourcing DACs-6.2mA or -3.5mA and the amplitude 3.9mA, resp. 2mA.
Of course this IV-stage needs a dedicated current sink towards the negative supply to null the DACs centre current, for example a simple source resistor degenerated JFET.
Did a sim of the IV-stage also, using the BC807/BC817 Philips/NXP models (SMD equivalent to BC327/337). THD seems really low, but input impedance rises above 1MHz with some weird up-and downs, especially around 66MHz, where seemingly a resonance builds up. Everything else sims well. There´s a input offset of a couple of tens of mVs which may be cancelled with a simple Poti betwee the supply lines, the wiper feeding into the bases of Q2/Q3. There´s also an output offset of a couple of hundreds of mVs, which may not be serious, if a following buffer stage is AC-coupled. Else it must be taken care of.

jauu
Calvin

Last edited by Calvin; 13th August 2012 at 09:28 AM.
  Reply With Quote
Old 13th August 2012, 01:41 PM   #150
smms73 is offline smms73  Portugal
diyAudio Member
 
smms73's Avatar
 
Join Date: Jul 2005
Location: Lisboa, Portugal
Quote:
Originally Posted by Calvin View Post

In simulation, I use current sources with parallel R and C. The offset current beeing as stated in the datasheet for current sourcing DACs-6.2mA or -3.5mA and the amplitude 3.9mA, resp. 2mA.
Of course this IV-stage needs a dedicated current sink towards the negative supply to null the DACs centre current, for example a simple source resistor degenerated JFET.
Did a sim of the IV-stage also, using the BC807/BC817 Philips/NXP models (SMD equivalent to BC327/337). THD seems really low, but input impedance rises above 1MHz with some weird up-and downs, especially around 66MHz, where seemingly a resonance builds up. Everything else sims well. There´s a input offset of a couple of tens of mVs which may be cancelled with a simple Poti betwee the supply lines, the wiper feeding into the bases of Q2/Q3. There´s also an output offset of a couple of hundreds of mVs, which may not be serious, if a following buffer stage is AC-coupled. Else it must be taken care of.

jauu
Calvin
Hi calvin,

For compensate the output offset current of the dac, it only needs to increase the upper current source, this also serves to trim the output voltage offset.

thanks to bring to my attention the problem of input impedance, add c4 and c5 should resolve the problem.
The output capacitor C1 is now 4n4 for filtering hi freq noises from dac, below is the circuit with some alterations. And the frequency response, as you see there is no instability now.

Only today i realize that pcm1794 has 7.8ma peak to peak, i was thinking it was 7.8ma peak, so all prior result of harmonic distortion have been made at 15,6ma peak to peak, at 7,8ma p-p the distortion is smaler, this version i apresent today has:

0.000006% at 10kohm input
0.000024% at 1Kohm input
0.000043% at 500ohm input
0.000096% at 200ohm input
Attached Images
File Type: jpg comp.jpg (369.7 KB, 289 views)
  Reply With Quote

Reply


Hide 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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
ultra low noise low distortion preamplifier needed neazoi Solid State 13 11th October 2011 01:03 PM
IRF610 I/V convertion from Pass Labs D1 dac - single ended schiller Digital Line Level 5 15th October 2009 04:33 PM
Distortion+low volume.. GeirW Tubes / Valves 5 25th November 2004 12:04 PM
Best low-noise low-distortion buffer? borges Solid State 7 9th December 2003 12:40 PM


New To Site? Need Help?

All times are GMT. The time now is 06:36 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2