Hi, do you have any schematic for this compensation circuit? also, what is your favorite digital filter instead of that crappy 7220 ? thanks
Pin 2 cap
As I understand it. The cap at pin 2 slows the slew response of the AD844, it also works against some DAC noises (high frequencies) produced in the conversion process. For the most part it can be ignored as a first order filter component. The reason is the cap across the TZ resistor has a greater effect on the response working across the I/V resistor. Experimenting is recommended as I understand your using the AD1862. Other dac implementations may behave differently.
Sine/X compensation
Pedja had a compensation circuit in his AYA DAC. It is just a simple booster circuit with an inductor and capacitor as I recall. The idea was to boost the high frequencies for a bit smoother high end. With NOS there is a pronounced roll off as frequencies rise. Do you need the AYA dac PDF? Digital filter... I will need to get my cousins permission as he discovered the filter back in 1989/1990. He created a corrected datasheet. I have at least 3 DAC's running with this filter. I will ask him today when I see him. The part is obsolete although readily available even today. Maybe I will start a new thread with the information as I am sure we should not high jack George's thread here.
Hi George, Just assembled my first Pedja DDNF discrete circuit board today. I had a small batch of circuit boards made. After adjustment and 30 Minutes burn in I must say I am surprised it beats my other board. The one Oliver dvb projekt built. Have no idea why that is. My opinion is it also easily beats my triple stack AD844. Plan now is to build one for my PCM1704 DAC. I am using a BUF03 behind the DDNF board so I could use the best quality 0.1 uF cap I could find. So it is kind of in the same configuration as the AD844 more or less. I know how you feel about caps in the signal path.
just for an update.
Did you try the PCM1704 yet? Are you still thinking that the DDNF is better than the stacked AD844s as i/v and did you overcome the dc drift?
PCM1704 with a DDNF stage
No. I had my knee replaced on March 22. So I am still getting used to walking again. It looks like I will be moving around well enough by the end of April to get busy again on the DC coupled DDNF board. Speaking for the TDA1541 version I posted a picture of with caps, yes. I believe it is better then the triple stack. Even with the coupling caps. I listen to it nearly every day. I will tackle that drift problem later this month. I need a better meter. I am holding out for a bench DMM with at least 5 1/2 digits.
As I understand it. The cap at pin 2 slows the slew response of the AD844, it also works against some DAC noises (high frequencies) produced in the conversion process. For the most part it can be ignored as a first order filter component. The reason is the cap across the TZ resistor has a greater effect on the response working across the I/V resistor. Experimenting is recommended as I understand your using the AD1862. Other dac implementations may behave differently.
I have put in 5,7nf at pin 2 and it works nicely.
Nelson Pass's D1 has 10nF in similar place. The impedence of that mosfet stage is differrent so is the cap value. I remember some where in this thread someone simulated that ad844 has an input impedance of about 50 ohm. but I think it may vary according to frequency. My DAC is tda1541 so 3 times the value suggested by Pedja is logical. I believe Pedja did the measurement
I don't have a scope.For other DAC with a DF this cap may have little effect because the hf noise of OS dac is much higher frequency.
Some people also commented in this thread that they lost pin point imaging with multiple ad844 stacked. I believe they did not change this cap value and therefore the sound sufferred by HF noise being too strong.
Knee
It is a slow process. My therapy guy said I am a week ahead of most people at this point. I dumped the walker over a week ago. No cane or anything. Tuesday is 3 weeks out from surgery. Need the right one done at some point too. Going to wait at least a year before doing that one. To much surgery in the last 6 months.
It seems that Charles Hansen of Ayre had the same stacking approach with the hybrid PCM1738, in the CX-7.
"]"The PCM1738's analog outputs are in the form of current and appear to be fed directly to Analog Devices AD844 chips. This is a high-speed (2000V/µs!), low-settling-time bipolar op-amp optimized for current/voltage applications. Eight AD844s are used for the direct-coupled analog output stages. Although there are no coupling capacitors, no DC servo circuits are used; Ayre's Charles Hansen feels that this seat-of-the-pants approach optimizes low-frequency quality. Ayre also claims that the circuitry is "zero-feedback." Not only is there no loop voltage negative feedback, there are not even current feedback loops around the op-amps. When the player is powered up, two arrays of green LEDs, presumably used for biasing, lend the interior a festive appearance."
Read more at Ayre CX-7 CD player | Stereophile.com
Cheers George
"]"The PCM1738's analog outputs are in the form of current and appear to be fed directly to Analog Devices AD844 chips. This is a high-speed (2000V/µs!), low-settling-time bipolar op-amp optimized for current/voltage applications. Eight AD844s are used for the direct-coupled analog output stages. Although there are no coupling capacitors, no DC servo circuits are used; Ayre's Charles Hansen feels that this seat-of-the-pants approach optimizes low-frequency quality. Ayre also claims that the circuitry is "zero-feedback." Not only is there no loop voltage negative feedback, there are not even current feedback loops around the op-amps. When the player is powered up, two arrays of green LEDs, presumably used for biasing, lend the interior a festive appearance."
Read more at Ayre CX-7 CD player | Stereophile.com
Cheers George
George, according your experiments you find it better without JFET DC servo ?
I'm wondering if anyone had also tried some op861 stacked ?
I have also a curious question has I have no technical background; Could it be possible to use instead some SIC transistor for I/V ? Like the few ones used by Pass or Kaneda in their actual amp designs ? (I'm asking if Kaneda hasn't already a PCM1792 with SIC conversion or buffer maybe ?)
I'm wondering if anyone had also tried some op861 stacked ?
I have also a curious question has I have no technical background; Could it be possible to use instead some SIC transistor for I/V ? Like the few ones used by Pass or Kaneda in their actual amp designs ? (I'm asking if Kaneda hasn't already a PCM1792 with SIC conversion or buffer maybe ?)
I use a PCM1704, with this there is no need, I just use the pin 1 and 8 null pins with a trim pot on one only of the stacked 844's, that gives me a steady dc offset of +- 1mV, regardless of having it's own buffer or an external buffer setup.
Cheers George
Re-read three posts back http://www.diyaudio.com/forums/digital-source/227677-using-ad844-i-v-75.html#post4680776 three words after trimpot.
Cheers George
I've heard of CX-7 before. People thought the secret was the CD-ROM mechanism. Apparently, it's not like that.
Hi
I just want to share that I tried a few different types of caps for HF filtering with the ad844 i/v including silver mica, philips polystyrene, wima fkp. After all I found that c0g capacitor sound the most natural. silver mica sounds aggressive on top end. polystyrene emphasis upper midrange. wima fpk is a bit tight. Just my experience with this particular i/v stage.
I just want to share that I tried a few different types of caps for HF filtering with the ad844 i/v including silver mica, philips polystyrene, wima fkp. After all I found that c0g capacitor sound the most natural. silver mica sounds aggressive on top end. polystyrene emphasis upper midrange. wima fpk is a bit tight. Just my experience with this particular i/v stage.