I maybe wrong, but because Charles Hanson used 8 x AD844's in one of his design, maybe he discovered the definate improvments of stacking the I/V stage of the AD844 as well, as well as using it with no feedback as we are?
Quotes from Charles Hansen designer of the Ayre CDP's
"Those are not op-amps. An op-amp (by definition) operates by using feedback. Our 7-series products use Analog Devices AD844's (designed by the great, but very stubborn, Barrie Gilbert) in the analog signal path, but with no feedback. So while the part was designed to be used as a current-feedback op-amp, we don't use it that way. Just one reason that they are in the "Class A" rating in Stereophile."
Quote from Jan Didden
"Yes, I agree, the AD844 are one of the best kept secrets in audio. I used them in my error correction amp as precision gain-of-one-without-feedback.
You can set the gain with a single resistor and get an open loop buffer to boot.
You remember the AD846? Almost the same except that the max lin current for the transconductance output was 10mA in stead of 1mA for the '844. Discontinued, of course.
Barrie sort of confided that he had a much better part on the shelf but unless there was sufficient demand, they would probably not produce it.
If I only had a 100 grand!"
Cheers George
Quotes from Charles Hansen designer of the Ayre CDP's
"Those are not op-amps. An op-amp (by definition) operates by using feedback. Our 7-series products use Analog Devices AD844's (designed by the great, but very stubborn, Barrie Gilbert) in the analog signal path, but with no feedback. So while the part was designed to be used as a current-feedback op-amp, we don't use it that way. Just one reason that they are in the "Class A" rating in Stereophile."
Quote from Jan Didden
"Yes, I agree, the AD844 are one of the best kept secrets in audio. I used them in my error correction amp as precision gain-of-one-without-feedback.
You can set the gain with a single resistor and get an open loop buffer to boot.
You remember the AD846? Almost the same except that the max lin current for the transconductance output was 10mA in stead of 1mA for the '844. Discontinued, of course.
Barrie sort of confided that he had a much better part on the shelf but unless there was sufficient demand, they would probably not produce it.
If I only had a 100 grand!"
Cheers George
Last edited:
Both are problematic, but in different ways.
I have BOTH here, and work on BOTH.
I don't have to like Black or White - I can like BOTH. This is not a war between two systems, one right and the other wrong.
To say that only one has 'jump factor' just means you have heard one side sounding any good.
I do agree (I always try to agree as I am an agreeable person and try to find common ground) that most D-S systems are plain boring - no jump factor - because they are BADLY done - and turned out like factories turns out potato crisps with fake/false chicken and barbecue type flavours by the zillions. Horrible chemically flavoured food.
But make fresh chips (with freshly caught barbecued snapper) - oh boy, you are in heaven. I just LOVE fresh chips. Alas, my waistline does not.
I have two NOS-DACs here, both built here, they don't even have any post-DAC circuit, only passive, no silicon at all, no transistors, no fets, no opamps... 8 stacked 1543s run much higher voltage and with heatsink that runs very warm indeed. I AC couple the output and I/V is a grounded resistor - with LOTS of AC current from that 8 stack. Again, an imperfect I/V too, but is there anything perfect in digital.
Went to Elson's yesterday and had to listen as he ranted (again) about all the problems he had cutting LPs back in his RCA days. It's an imperfect world. LOL
Cheers, Joe
Last edited:
See below: This works with differential current DACs.
No values are shown, but the first two resistors on "+" and "-" must be very low value and a small fraction of the nominal 50 Ohm input impedance of "-" and the gain set on Pin 5.
The filtering can be done in a number of ways, the caps shown represents the options. The output RC filter is an optional Ben Duncan type - it should be the highest pole IMO.
Works very well with Burr-Brown 1792, 1794, 1796 DACs.
Cheers, Joe
No values are shown, but the first two resistors on "+" and "-" must be very low value and a small fraction of the nominal 50 Ohm input impedance of "-" and the gain set on Pin 5.
The filtering can be done in a number of ways, the caps shown represents the options. The output RC filter is an optional Ben Duncan type - it should be the highest pole IMO.
Works very well with Burr-Brown 1792, 1794, 1796 DACs.
Cheers, Joe
Attachments
Here's the post about the AD844
http://www.diyaudio.com/forums/part...plifier-ad830-lt6552-tda856x.html#post3392798
a good overview of the 844 and the transimpedance node (pin 5).
Cheers
http://www.diyaudio.com/forums/part...plifier-ad830-lt6552-tda856x.html#post3392798
a good overview of the 844 and the transimpedance node (pin 5).
Cheers
Last edited:
As George said, it is that Pin 5 that makes things possible. One of the earliest people to use the 844 was Ben Duncan, and it reminds me how old the 844 is, but he never used Pin 5 and used it as a current feedback I/V "virtual earth" - so that shows how far back it goes in its I/V history.
Lars Clausen also used it, but as part of a phono stage. The 844 is quite low noise too - especially if you have a spectral analyser like ClioFW and select the best samples.
It's great to see a good oldie come back to life.
Cheers, Joe
AD846
Hi all, Really good information. The first chip I really liked was the AD846. It was a huge upgrade from plain opamps back in the old days. Sadly it was maybe 12 bit resolution. I still have a pair and may use them as a moving coil phono preamp first stage. Just waiting for a worthy place to use them. Will place an order for 844's next week. Dave
Hi all, Really good information. The first chip I really liked was the AD846. It was a huge upgrade from plain opamps back in the old days. Sadly it was maybe 12 bit resolution. I still have a pair and may use them as a moving coil phono preamp first stage. Just waiting for a worthy place to use them. Will place an order for 844's next week. Dave
That's awesome!!1! Please point him in the direction of my post:
http://www.diyaudio.com/forums/part...-feedback-amplifier-ad830-lt6552-tda856x.html
I think he'd get a big flashback out of the summary and Abraxalito is having as much fun with the TDA8561/6 as you are with pin5 of the AD844 (well, almost as much fun...)
Last edited:
Jeff thanks for that, in among all the links I found an interesting post by Jocko Homo. We all know that he thinks (and may well have) the best sounding I/V stage that nobody has ever seen, except in simpler circuit he gave to us and said this is 70% of the good one. Well in this short post directed at Walt Jung, they are discussing the AD844 and Jocko came out with this.
"When used in the open-loop manner, you have basically created the I/V stage I made, but in chip form. Jocko"
Could it be that the famed Jocko I/V stage is a discrete form of the AD844 run in open loop form?
Cheers George
"When used in the open-loop manner, you have basically created the I/V stage I made, but in chip form. Jocko"
Could it be that the famed Jocko I/V stage is a discrete form of the AD844 run in open loop form?
Cheers George
Isn't that the wildest thread! I couldn't believe it when I found it!
From an old thread (2004!) with ... Holy Heavy Hitters Batman! Charles Hansen, John Curl, Walt Jung, ... and the old gang! Jocko, PMA, CarlosFM, Pedja, Elso
DIYHiFi.org • View topic - "Op-amps unsuitable for audio DAC I/V" - myth?
I'm into direct quotes (link=web), do you remember where you saw Jocko's comment?
His IV stage?:
http://www.diyaudio.com/forums/digital-source/6121-easy-build-i-v-stage.html
At post 7 Elso says:
Hi Jocko,
Nice to see you back. I thought you had left us.
I see a common base amplifier between to current sources or should I say between current source and current sink?
My problem is I don't like common base amplifiers....
All the best.
Jocko:
If you carefully adjust the current, it will work.
Yes, it is a common-base. Not an RF engineer, eh? I am, so I do.
Jocko
In post 42 he provides an update:
Why don't you guys ever believe me???
I built this thing......made a few changes........
The load resistors are 3 K, so that you actually get some gain out of it. The resistors for the current mirror were changed to 5 K so that the collector voltage was decent. Raised the rails to +/- 24 V.
With 1.2 mA (peak.....which is what a '1704 puts out) the distortion is.........drum roll, please........
- 40 dB.
Just one problem. That is when you measure it from the collector to ground. When measured between the collectors, it is -23 dB.
So there.
Jocko
PS See how cool links are!
PSS Speaking of quotes:
http://www.diyaudio.com/forums/digital-source/6121-easy-build-i-v-stage-8.html#post79525
Doh!
Last edited:
Hmm!!. Very interesting thread. I have used AD844 with pin 5 as output I/V and built in buffer for many years. May be it is time for some improvement according to the information posted by georgehifi. My Dac is a DIY old school PCM63 multibit DAC which I intend to keep. Transport slaved to master clock in the DAC. I agree: keep it coming guys...
dacen double stack your AD844 I/V stage, and keep using just one of the buffer stages, you will be very surprised.
PS: I have now settled on a 4.7kohm and 470pf for the TZ gain set and filter, this gives me all the output at the rca's I need, using the 1.2mA of the PCM1704, and has a dead flat FR to 20khz and -3db at 72khz. I noticed that with and without the 470pf filtercap that the dc offset changes a couple of mV and needs to be re-adjusted, it may be nice if the AD844 nulling circuit will work when only one buffer is used and stacked i/v's, and whether you have to stack pins 1 and 8 as well???
Cheers George
PS: I have now settled on a 4.7kohm and 470pf for the TZ gain set and filter, this gives me all the output at the rca's I need, using the 1.2mA of the PCM1704, and has a dead flat FR to 20khz and -3db at 72khz. I noticed that with and without the 470pf filtercap that the dc offset changes a couple of mV and needs to be re-adjusted, it may be nice if the AD844 nulling circuit will work when only one buffer is used and stacked i/v's, and whether you have to stack pins 1 and 8 as well???
Cheers George
Last edited:
Time to get some Sabre DACs then....
I am sure there will be caches of them around, you just end up paying more.
Cheers, Joe
I beg to differ, no SD or Hyper Stream no matter how they are tweaked playing Red Book can give the rhythm and drive and jump factor that that correctly implemented R2R Multibit can, that I've heard.
This is also backed up and explained why by heavier hitters than most here on the other forum, that still use multibit in their models and designs. Why would guys like Thorsen Losetch still use a very hard to find discontinued TDA1541 at 20 x $ of the Sabre chip in his flagship player the AMR CD-77 that is touted to be the best consumer made Red Book players ever, when he could of saved and used the Sabre 9018 and got sacd "hirez" playability as well
To me I admit the Sabre done right is smooth and detailed, but it's got no clout, even the designer of the ES9018 on a conference that was video'ed that I posted the link to on SNA stated, "that the 9018 had come the closest to the sound of the best multibit than any single bit or Delta Sigma dacs could."
Cheers George
This is also backed up and explained why by heavier hitters than most here on the other forum, that still use multibit in their models and designs. Why would guys like Thorsen Losetch still use a very hard to find discontinued TDA1541 at 20 x $ of the Sabre chip in his flagship player the AMR CD-77 that is touted to be the best consumer made Red Book players ever, when he could of saved and used the Sabre 9018 and got sacd "hirez" playability as well
To me I admit the Sabre done right is smooth and detailed, but it's got no clout, even the designer of the ES9018 on a conference that was video'ed that I posted the link to on SNA stated, "that the 9018 had come the closest to the sound of the best multibit than any single bit or Delta Sigma dacs could."
Cheers George
