Using the AD844 as an I/V - Page 18 - 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 26th January 2013, 02:28 AM   #171
diyAudio Member
 
georgehifi's Avatar
 
Join Date: Sep 2003
Location: Manly Australia (Jewel of the Pacific)
Send a message via MSN to georgehifi
I tried once got a headache and never looked at it again.

Cheers George
__________________
Avatar : Production Lightspeed Attenuator
www.lightspeedattenuator.com
  Reply With Quote
Old 26th January 2013, 03:04 AM   #172
diyAudio Member
 
Join Date: Jun 2003
Location: Hamilton, was Ottawa (Canada)
Quote:
Originally Posted by jcx View Post
LTspice is good too, some say better if you are looking for "an engineer's Spice"
Mmmm, would that be good for a dry rub on ribs?

  Reply With Quote
Old 26th January 2013, 03:12 AM   #173
diyAudio Member
 
Join Date: Jun 2003
Location: Hamilton, was Ottawa (Canada)
Google-ogled the Interwebs for "ad844 iv":

Pedja on 844:
DIYHiFi.org • View topic - Pedja's AD844 I/V stage

…you can use either buffered (pin 6) or unbuffered output (pin 5). But you shouldn't connect these pins one to another, as it will short the buffer input and output. (You can put some small cap across the diamond to improve on its stability but that would be another issue.)

Purely sonically, the buffer you have within AD844 is the good one but, things being said, all the other IC diamond buffers I tried measured better.


I did used buffered output of AD844 (pin 6). Unbuffered output sounds more natural but in most cases it lacks the dynamics of the buffered output.


More from Pedja:
DIYHiFi.org • View topic - Let's talk audio design

As for the IC transistor matching and linearity, AD844, which is also similar (diode connected first transistors instead of the real diamond structure), also varies a bit from sample to sample, its current mirrors can show from 0.03% to 0.07% THD under the same conditions as described above. Stereophile measurements of Ayre CX-7 point out the same issue. Anyhow, taking into account internal bias of AD844 (6.5mA for all the arrays), I am convinced that this part can easily do much better, only if someone at AD would like to make possible external biasing. I am not sure they think in this direction though. Other than that, from conversation with Maxim, I indeed realized that linearity wasn’t any priority in case of their 435/436 and for the audio they simply recommend use of feedback.




CG on why many audio opamps are not ideal as IV:

DIYHiFi.org • View topic - I/V on the PCM1704

As has been said a lot of times, current output DACs, audio or not (back to my day job) produce minimum distortion when used into very low impedance loads. Ideally a virtual ground. That's because the current sources used within aren't perfectly linear - they're designed to be monotonically accurate and fast.

Probably 99.985% of every current output DAC implementation used in commercial products for audio has used an opamp as an I/V converter. Why? Because it appears to be the right thing - it says so right there in the data sheet - and they are cheap and repeatable. (So is bad ethnic food...) The problem is that the virtual grounds supposedly found in inverting opamp stages really are only inverting grounds at audio frequencies. Unfortunately, the switching currents at the DAC output are much higher in frequency than that. So much for the virtual ground. If you want to read more about that, Google the name Barrie Gilbert and read what he has to say on the subject. If he isn't qualified enough on the subject for you, you're certainly wasting your time here.

Common base circuits can be made very, very linear and easily enough. Yeah, you can make an opamp work in the circuit, but you have to go to great lengths and use an expensive part. Even then, it may well not sound as good.

Better you should build a simple I/V converter and spend your time and money on improving the power supply and regulation - a whole other topic."


Hmm, didn't Gilbert have something to do with this ... ;-)



Zinsula made an open loop pre with ad844:

DIYHiFi.org • View topic - Op Amp based open loop line amp

He is using it opamp style but Charles Hansen does talk about loading pin 5:


We have been using a circuit like this for several years in our less expensive products. The sound is *far* beyond normal op-amps (but still not as good as zero-feedback discrete circuits).

One problem with your circuit is that the input offset current will cause problems. Another problem is that when you load down pin 5 with low resistor values (such as the 5k you have shown) then some problems crop up. I will leave it to you to find solutions to these problems
.”

Zinsula replies:

Thank you for the hints, Charles. I'm afraid that I will not see a solution to the problems on a theoretical way, so I have to buy these parts.

Regarding the offset, I think at injecting/sinking current with a ccs.

Regarding the other problem, I'm a little disappointed as I thought to place a pot at pin 5 loading the internal mirrors and having a buffered volume control.
If it is a current limit problem of the mirrors......I do not know their maximum current though. In the AD846 data sheet, which is a similar part, they claim maximum 1mA. So I could increase the resistor at the inverting input(s) to avoid that a signal of say max. 5V will saturate the mirrors even when pin 5 is connected with a very low resistor to ground. Although I believe that AD844 will allow higher currents in the mirrors.
Unfortunately, I can't really find out this from the datasheets. Maybe I'm just not bright enough...



Here's Zinsula's AD844 circuit:

zinsula AD844 conceptual schematic 01.gif


He goes on:

Carlos, yes I will look at input network. Charles gave the hint that input offset currents may be an issue with this chip (it has no differential input, and is bipolar). So, a current injection/sinking will be required too...reminds me of your AD815 and the issue is essentially the same.

Jam, what do you mean with "a buffer after the pot"?
There is one, the internal diamond buffer of AD844. It could be replaced by an external one though, as PMA explained.

If it could be done as I have drawn, the potentiometer is in fact "isolated" from the input as well as from the output. It essentially only loads the TZ node, which is no longer a high impedance node and therefore gain of the op amp is reduced. When the TZ node is at 0 Ohms to ground, the output voltage is zero...



Tidbits to chew upon ...

Last edited by AudioLapDance; 26th January 2013 at 03:14 AM.
  Reply With Quote
Old 26th January 2013, 03:16 AM   #174
diyAudio Member
 
Join Date: Jun 2003
Location: Hamilton, was Ottawa (Canada)
George, are you familiar with any simulator? (Cool if you aren't)
  Reply With Quote
Old 26th January 2013, 03:59 AM   #175
diyAudio Member
 
Join Date: Jun 2003
Location: Hamilton, was Ottawa (Canada)
And here's a fuller schematic of Pedja's 1541 AD844 IV with buffer:

Pedja 1541 NOS AD844IV wBUFFER.png

And George, note the simple buffer! It's the classic dual 170, start with something like that instead of trying to fix/upgrade that internet crap buffer.

I like this one (classic Curl buffer?):

carlosfm FET buffer.jpeg

Tweeking R3 you can get very low DC offset (20 ohm good trim pot) . And I just got some SOIC LSK389!

Last edited by AudioLapDance; 26th January 2013 at 04:21 AM.
  Reply With Quote
Old 26th January 2013, 07:14 AM   #176
diyAudio Member
 
georgehifi's Avatar
 
Join Date: Sep 2003
Location: Manly Australia (Jewel of the Pacific)
Send a message via MSN to georgehifi
Quote:
Originally Posted by AudioLapDance View Post
And here's a fuller schematic of Pedja's 1541 AD844 IV with buffer: He is using pin 6 for the output which is the internal buffer and he has it capacitor coupled which I don't need

Attachment 326415

And George, note the simple buffer! It's the classic dual 170, start with something like that instead of trying to fix/upgrade that internet crap buffer.

I like this one (classic Curl buffer?):

Attachment 326418

Tweeking R3 you can get very low DC offset (20 ohm good trim pot) . And I just got some SOIC LSK389!
I use abrax alitos circuit for nulling the dc offset and it hold perfect to .1mV Cheers George
__________________
Avatar : Production Lightspeed Attenuator
www.lightspeedattenuator.com
  Reply With Quote
Old 26th January 2013, 10:54 AM   #177
diyAudio Member
 
georgehifi's Avatar
 
Join Date: Sep 2003
Location: Manly Australia (Jewel of the Pacific)
Send a message via MSN to georgehifi
I see now I was looking at the wrong side. But what I have now is there, I'm just trying to get the most out of it with making something from scratch.

I am believing though that the TZ resistor size that control gain is having an effect on the bass, to me less gain (1.5k resistor) = tighter punchier bass. More gain (3k resistor) softer fatter bass.
It as though if you ask for more gain from the I/V stage your also getting more 2HD which loosens the bass. Or the extra gain from the I/V stage is making the buffer give more 2HD.

Cheers George
__________________
Avatar : Production Lightspeed Attenuator
www.lightspeedattenuator.com
  Reply With Quote
Old 27th January 2013, 07:29 AM   #178
diyAudio Member
 
georgehifi's Avatar
 
Join Date: Sep 2003
Location: Manly Australia (Jewel of the Pacific)
Send a message via MSN to georgehifi
What do you guys think of this, it's a discrete clone of the AD844 with a different buffer, but it looks to me like the distortion rises below 2khz and what is the y axis measurments?
Cheers George
Attached Images
File Type: jpg ad844_knockoff.jpg (492.2 KB, 325 views)
__________________
Avatar : Production Lightspeed Attenuator
www.lightspeedattenuator.com
  Reply With Quote
Old 27th January 2013, 08:11 AM   #179
diyAudio Member
 
s3tup's Avatar
 
Join Date: Dec 2004
Location: Israel
Y is in milli/micro percent. The top THD chart is for I/V + buffer, the bottom one is for I/V alone. There is some distortion cancellation happening, at least in the sim thus allowing us to get a better 2nd/3rd of the circuit. As i look at it... the first stage performs rather close to original 844 so it's doubtful whether the discrete will benefit from anything but low costness. The "double dot" bars are for 20c and 70c temperatures to see the temperature stability. The 20c is the lower "dot".

The THD could go lower on 15v/18v rails. A bit of current would do the same - but i have no clue how much current won't overheat the stuff. I have thermal compound ordered to cover the PCB + heatsinks + thermal matching of transistor pairs/whole circuit.
I hope to get away without degeneration resistors as i don't like 'em - they take too much space on the PCB - long traces, lots of EMI pickup.

Measured at 1kHz sine, 4mA swing (tda1541).

There should be a servo to keep everything in place and DC-coupled.
__________________
The missing link between lead and gold in alchemist's world was BS and commerce.
  Reply With Quote
Old 27th January 2013, 08:29 AM   #180
diyAudio Member
 
georgehifi's Avatar
 
Join Date: Sep 2003
Location: Manly Australia (Jewel of the Pacific)
Send a message via MSN to georgehifi
Thanks s3tup, they are good distortion figures still for zero feedback. Can you tell me what happens at low frequency (20hz-1000hz) if you did pot this, as it seems the distortion is rising quite quick to 2khz, if it contiunes to rise it could get quite high at low frequencies.

Cheers George
__________________
Avatar : Production Lightspeed Attenuator
www.lightspeedattenuator.com
  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
AD844 as a common base stage in the I/V converter Pedja Digital Source 42 25th March 2013 07:56 AM
WTB AD844 Opamps apassgear Swap Meet 6 31st January 2007 02:57 PM


New To Site? Need Help?

All times are GMT. The time now is 01:16 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