Sorry for being off topic. This is the output stage of my ak4396.
I think changing the load is a bit trouble some because I will have to change the value of filter caps around the opamp.
I previously used 100nf pps cap for bypass but i have some biger electrolytic further away (a few hundred uF). The sound is cleaner and more dynamic with x7r caps as bypass compared to pps caps
Last edited:
Yes you would have to also change the filter caps. But when I did this mod on WM8740 (similar chip) the improvement in bass dynamics was dramatic. Ideally you'd change all the 1k resistors to about 39k and then C46 goes down to 82pF, 1n goes to 22pF. You'll want to change the opamp to a JFET type too or the noise will go up considerably, I use ISL28210.
Actually I changed the bypass caps for the dac and the opamp at the same time so the 'blow away' effect may be a combination of the effects of the x7r caps on the opamp as well as the DAC chip.
Thanks for the tip. I will try at some point as I need to find some good enough resistors and filter cap with these values.
ISL28210 is a good choice here. 9fA/sqrt(HZ) current noise is ideal for 39k input resistor (only generate 351pV/sqrt(HZ) noise). Thanks for the tip.
Hi canvas (Poting),
Thanks for your feedback
... I will look for some 10 uF TDKs (or similar) in place of the electrolytics I have been listening to. Hmmm... maybe you know already but Samuel Groner some time ago recommended that C0Gs be higher than a 100 Volt types - apparently distortion is lowest in these types. I've been listening to some different 100 nF C0Gs (
) and to my ears the Kemet types were the most harmonic sounding closely followed by TDKs.
Cheers,
Jesper
Thanks for your feedback
... I will look for some 10 uF TDKs (or similar) in place of the electrolytics I have been listening to. Hmmm... maybe you know already but Samuel Groner some time ago recommended that C0Gs be higher than a 100 Volt types - apparently distortion is lowest in these types. I've been listening to some different 100 nF C0Gs (
) and to my ears the Kemet types were the most harmonic sounding closely followed by TDKs.Cheers,
Jesper
AD844 gerber
Remember to triple stack as in post #1907
http://www.diyaudio.com/forums/digital-line-level/227677-using-ad844-i-v-191.html#post4922758
Gerbers attached.
Cheers
Simon
Feel free to use my gerbers if you like (or buy some from me). Have received my boards and they work well.
Remember to triple stack as in post #1907
http://www.diyaudio.com/forums/digital-line-level/227677-using-ad844-i-v-191.html#post4922758
Gerbers attached.
Cheers
Simon
Last edited:
Well, I am kind of late to the party. No, I don't know about this, but the 100nF C0Gs at hand are rated 50V and 450V, both in different size. Now, I put 10uf X7R (with 100nF across both rails) on DAC chips, I/V and buffer. The resolution on LF & Mid improves greatly. Be sure to use Rhopoint GG102 as I/V resistor. The sound stage becomes 3D!
Poting
Lightning...
A UPS would be nice to have. I have an audio club DIY power conditioner here. In fact I was still in vacation mode when the lightning struck. The AC power was isolated at the time. It came in the cable companies coax. My mistake was not having disconnected one ethernet cable to a Blu ray player. So it came in via my internet connection. Live and learn I guess. Who would expect a thunderstorm in Minnesota in late November?
A UPS would be nice to have. I have an audio club DIY power conditioner here. In fact I was still in vacation mode when the lightning struck. The AC power was isolated at the time. It came in the cable companies coax. My mistake was not having disconnected one ethernet cable to a Blu ray player. So it came in via my internet connection. Live and learn I guess. Who would expect a thunderstorm in Minnesota in late November?
Sounds great. Like a typical r2r dac. Certainly very involving and lots of micro details with extra balls. Have only done a little bit of comparing with my Pdx dac. Will do some more critical listening tommorow against an Audio Gd sigma delta dac.
Buffer amplifier that beats OPA627
Hi guys, On the advice of an ASM member of my local audio club, I started looking at other options to the OPA627. So I have had the OPA1641's in both the PCM1704 dac and the parallel TDA1541A S1 crown dac version. As a buffer for Pedja's DDNF it is clearly better. More transparent, a touch more bass and a sharply focused sound stage. Holographic. Extra levels of detail. Only down side I see is there is no nulling option with this chip. So hanging one off the TZ of a AD844 you likely will have no adjustment of offset. Maybe a DC servo could be used? I am using this chip as a unity gain device and have tried it as a gain =2 and gain = 1.5. It works very well. I have a pair of discrete op amps on order to give that a try as well. I am also going to try the OPA1611's in lower impedance buffer positions. Today I experimented with the parallel TDA1541A S1 crown dac using the DDNF into a 7th order GIC filter with the OPA1641's in the final buffer stage. It is the best headphone sound to date I have heard. I am using a Schiit Magne 2 Uber headamp and Audio Technica ATH M50 Studio monitors. Impressive. I thought I'd have to get a pair of planar headphones to get this sort of resolution.
Hi guys, On the advice of an ASM member of my local audio club, I started looking at other options to the OPA627. So I have had the OPA1641's in both the PCM1704 dac and the parallel TDA1541A S1 crown dac version. As a buffer for Pedja's DDNF it is clearly better. More transparent, a touch more bass and a sharply focused sound stage. Holographic. Extra levels of detail. Only down side I see is there is no nulling option with this chip. So hanging one off the TZ of a AD844 you likely will have no adjustment of offset. Maybe a DC servo could be used? I am using this chip as a unity gain device and have tried it as a gain =2 and gain = 1.5. It works very well. I have a pair of discrete op amps on order to give that a try as well. I am also going to try the OPA1611's in lower impedance buffer positions. Today I experimented with the parallel TDA1541A S1 crown dac using the DDNF into a 7th order GIC filter with the OPA1641's in the final buffer stage. It is the best headphone sound to date I have heard. I am using a Schiit Magne 2 Uber headamp and Audio Technica ATH M50 Studio monitors. Impressive. I thought I'd have to get a pair of planar headphones to get this sort of resolution.
Hi guys, On the advice of an ASM member of my local audio club, I started looking at other options to the OPA627. So I have had the OPA1641's in both the PCM1704 dac and the parallel TDA1541A S1 crown dac version. As a buffer for Pedja's DDNF it is clearly better. More transparent, a touch more bass and a sharply focused sound stage. Holographic. Extra levels of detail. Only down side I see is there is no nulling option with this chip. So hanging one off the TZ of a AD844 you likely will have no adjustment of offset. Maybe a DC servo could be used? I am using this chip as a unity gain device and have tried it as a gain =2 and gain = 1.5. It works very well. I have a pair of discrete op amps on order to give that a try as well. I am also going to try the OPA1611's in lower impedance buffer positions. Today I experimented with the parallel TDA1541A S1 crown dac using the DDNF into a 7th order GIC filter with the OPA1641's in the final buffer stage. It is the best headphone sound to date I have heard. I am using a Schiit Magne 2 Uber headamp and Audio Technica ATH M50 Studio monitors. Impressive. I thought I'd have to get a pair of planar headphones to get this sort of resolution.
Have you tried an LME49710. I hear they are meant to pretty good as well?
Output is wired as Post#1907. http://www.diyaudio.com/forums/digital-line-level/227677-using-ad844-i-v-191.html#post4922758
I have got a NOS AD1862 boards from diyhnk which use a dual opamp for output. I find the Ad844 to opamp method to be very ballsy if you know what I mean. It gives the music plenty of energy and drive. But then part of that could be the R2R Dac chip at play?
I have got a NOS AD1862 boards from diyhnk which use a dual opamp for output. I find the Ad844 to opamp method to be very ballsy if you know what I mean. It gives the music plenty of energy and drive. But then part of that could be the R2R Dac chip at play?