analog_sa said:
How so? The D1 is an I/V stage and the Zhaolu has voltage outputs. Some of Pass differential preamps seem much more viable with some simple passive filtering at the input.
A Zapfilter clone will be completely diy
nevermind you're right. yes, a zapfilter clone is the way to go, especially since the zapfilter is a proven solution and even the newest stuff like LM4562 falls far short of the dynamics and openness of it. does anyone know how much money a DIY clone would save? here in USA, the Zapfilter is $215 including some premium parts like Silver MICA capacitors and a 25VA toroid transformer. But the PSU parts look cheap to build and uses Jamicon capacitors, with Shocty diodes. I bet it can be built for pretty cheap. It's also fine to put the Zapfilter and the PSU very close, so maybe in the DIY implementation, everything can be combined onto a single board. Both the PSU and the Zapfilter produces loads of heat though. Especially the two 5W 100ohm resistors on the PSU get super hot and I'm wondering whether 10W resistors should be used instead. A layout for better cooling would be an improvement. It introduced so much heat in my Zhaolu that I had to drill extra holes in the casing for better cooling.
another upgrade for the Zhaolu is to change the PSU capacitors to Blackgates. I use the standard polar Blackgates, but the nonpolar Blackgates are better, as good as polypropelenes in measurements. If you are using the Zapfilter, upgrade the Jamicon capacitors on its PSU to Blackgate NX ones. They will be bigger so you have to find room for them.
there is also the matter of shielding. Using shielding paper makes a startling difference. Shield above and below the boards, and in between the PSUs and the rest. I use the ERS shielding paper, and it gives tighter bass and clearer treble. The effect is easily noticeable. Digital components like DACs and Class D amps really benefit from shielding. The nongrounded stuff is better than the grounded stuff for shielding. They operate on different principles. But it's conductive so beware what it touches or you can get a dead short and there goes your Zhaolu up in flames (the ERS paper is flammable, but you can use silicon carbide instead if that scares you).
there is also the matter of shielding. Using shielding paper makes a startling difference. Shield above and below the boards, and in between the PSUs and the rest. I use the ERS shielding paper, and it gives tighter bass and clearer treble. The effect is easily noticeable. Digital components like DACs and Class D amps really benefit from shielding. The nongrounded stuff is better than the grounded stuff for shielding. They operate on different principles. But it's conductive so beware what it touches or you can get a dead short and there goes your Zhaolu up in flames (the ERS paper is flammable, but you can use silicon carbide instead if that scares you).
Given a lot of help and some volume I would think we could at least chop the cost in half with my solution.. Clone isn't quite the word, the idea is the same, the solution is currently intended to be specific to the Zhaolu Dac, however it would work with any dac having a similar voltage output configuration. The topology is somewhat different than the ZF in order to avoid running afoul of copyright law and charges of plagiarism, etc.
If anyone is interested in stepping up to the plate and doing the proto pcb layout I would be happy to collaborate with said individual.. hint.. hint..
Just PM me if you are serious.. Note that the design will be all SMD.
Kevin
If anyone is interested in stepping up to the plate and doing the proto pcb layout I would be happy to collaborate with said individual.. hint.. hint..
Just PM me if you are serious.. Note that the design will be all SMD.
Kevin
Congrats!
Kevin,
Congratulations on the new job. It was an initial shock to me that a person of your qualifications was looking for a different position. I hope the new job will offer you many good opportunities.
I am also interested in your development of the discrete approach on the Zhaolu.
Dale,
Is that you from the old "Harmonic Discord" days? Do you still have that AKSA gear and preamp?
Hope you are enjoying the great music down there in Austin, and if you like the "old" alternative folk of the early 70's, I recommend you catch the dvd, "Gram Parsons-Fallen Angel". It figures it would take a young German filmmaker, with the support of the BBC, to make the first documentary on a musician who influenced the Byrds, the Stones, the Eagles, and country music and died in '73 at the age of 26. The small interview bits with Keith Richard was worth the price of admission.
Next up, the BBC documentary on the good Captain Beefheart.
Lyndon
(Hey, I couldn't find the "Trout Mask Replica" Smiley!)
Kevin,
Congratulations on the new job. It was an initial shock to me that a person of your qualifications was looking for a different position. I hope the new job will offer you many good opportunities.
I am also interested in your development of the discrete approach on the Zhaolu.
Dale,
Is that you from the old "Harmonic Discord" days? Do you still have that AKSA gear and preamp?
Hope you are enjoying the great music down there in Austin, and if you like the "old" alternative folk of the early 70's, I recommend you catch the dvd, "Gram Parsons-Fallen Angel". It figures it would take a young German filmmaker, with the support of the BBC, to make the first documentary on a musician who influenced the Byrds, the Stones, the Eagles, and country music and died in '73 at the age of 26. The small interview bits with Keith Richard was worth the price of admission.
Next up, the BBC documentary on the good Captain Beefheart.
Lyndon
(Hey, I couldn't find the "Trout Mask Replica" Smiley!)
I measured the offset again with the input caps on the filter shorted out ( waited 1 hour to settle down ).
LT 1057 -44.8 mV -55.0 mV
OPA2134 -41.6 mV -55.6 mV
LM4562 -44.2 mV -56.0 mV
So they are all pretty much the same. That is expected as the source of the problem is the DAC . I guess nothing can be done about it there (?).
Sound wise from better to worse ( not 'worse' really)
LM4562
LT1057
OPA2134
Of these the LT1057 seems to be pretty close to the LM4562 in some areas.
I still haven't got my discrete filter done. To many other problems cropped up.
Cheers.
LT 1057 -44.8 mV -55.0 mV
OPA2134 -41.6 mV -55.6 mV
LM4562 -44.2 mV -56.0 mV
So they are all pretty much the same. That is expected as the source of the problem is the DAC . I guess nothing can be done about it there (?).
Sound wise from better to worse ( not 'worse' really)
LM4562
LT1057
OPA2134
Of these the LT1057 seems to be pretty close to the LM4562 in some areas.
I still haven't got my discrete filter done. To many other problems cropped up.
Cheers.
Ashok
The 50mV offset is in the 1852 datasheet. As Kevin previously noted you can't get rid of it in the differential stage without upsetting balance and distortion. You can use the second opamp in the package, currently connected as a follower to compensate the offset. Not quite sure if an extra opamp is better or worse than a coupling cap though.
The 50mV offset is in the 1852 datasheet. As Kevin previously noted you can't get rid of it in the differential stage without upsetting balance and distortion. You can use the second opamp in the package, currently connected as a follower to compensate the offset. Not quite sure if an extra opamp is better or worse than a coupling cap though.
And i'm already confused
The 8416 in the Zhaolu uses a 11.289MHz crystal to get the Mclock output running even if there is no lock. Looks like the given circuit will provide 11.289 master clock irrespective of whether the 8416 is getting any input or not.
What i'm not sure about is if this master clock will be usable for the 1852. Page 8 of the datasheet says:
"Master clock should be synchronized with L/RCLK but
phase relation between master clock and L/RCLK is not critical."
I fail to distinguish the difference between "should be" and "not critical".
Maybe someone, who actually understands digital can explain?
The 8416 in the Zhaolu uses a 11.289MHz crystal to get the Mclock output running even if there is no lock. Looks like the given circuit will provide 11.289 master clock irrespective of whether the 8416 is getting any input or not.
What i'm not sure about is if this master clock will be usable for the 1852. Page 8 of the datasheet says:
"Master clock should be synchronized with L/RCLK but
phase relation between master clock and L/RCLK is not critical."
I fail to distinguish the difference between "should be" and "not critical".
Maybe someone, who actually understands digital can explain?
caps improvement ?
Would some Nichicon FG 100µF do a good job on my D2.5a analog board instead of the small black electrolytic ones ?
what could improve the sound on this board ? I have 2 LM4562 on and don't want to invest in a zapfilter.
I will remove the DC blocking caps ans wamy place some better RCAs.
Are resistors good enough or should they be changed ?
Do green caps (value 1000pF, 0,01µF, 0,1µF written on the board) and yellow caps (value 47pF according to the board) have to be replaced ?
What about using PIO caps (like cheap K40y9) in // of the chemical caps on the power board (to act as a filter) ? Or Mica, polystyrene (LCR FSC caps)... ?
Regards.
Would some Nichicon FG 100µF do a good job on my D2.5a analog board instead of the small black electrolytic ones ?
what could improve the sound on this board ? I have 2 LM4562 on and don't want to invest in a zapfilter.
I will remove the DC blocking caps ans wamy place some better RCAs.
Are resistors good enough or should they be changed ?
Do green caps (value 1000pF, 0,01µF, 0,1µF written on the board) and yellow caps (value 47pF according to the board) have to be replaced ?
What about using PIO caps (like cheap K40y9) in // of the chemical caps on the power board (to act as a filter) ? Or Mica, polystyrene (LCR FSC caps)... ?
Regards.
The existing film caps are good quality ones, (mine has wima) and it is probable that replacing them with other types is not going to result in a significant improvement.
Many seem to prefer the 47uF input caps removed (jumpered) I installed a set of non-polar BG and thought that there was little if any improvement in sound. I don't like the idea of dc offsets at the output of the dac, so if I were to remove the input caps I would add a very good coupling cap at the output.
I am working on a plug in all discrete replacement for the analog board, but have not made much progress lately due to my new job.. I'll keep people posted when I start to work on this again.
Many seem to prefer the 47uF input caps removed (jumpered) I installed a set of non-polar BG and thought that there was little if any improvement in sound. I don't like the idea of dc offsets at the output of the dac, so if I were to remove the input caps I would add a very good coupling cap at the output.
I am working on a plug in all discrete replacement for the analog board, but have not made much progress lately due to my new job.. I'll keep people posted when I start to work on this again.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.