You can replace this OP with AD825 , but it is not clear to me why they use this stage as attenuator ad output buffer. Try change resistor values to get some gain. The sound will become more spacious, and transparent.
Next you must change regulators and caps in power supply , those L78XX are bad solution.
Next you must change regulators and caps in power supply , those L78XX are bad solution.
Thanks guys, AD emailed me back and said yes the AD825 is unity gain stable, they forgot to mention it in their data sheets.
And yes Peranders the two NPN transistors are the switch to deck for the de-emphasis circuit, but i was talking about the output opamps, if you look at the return feedback resistor for the feedback it's 330ohm and the one to deck is 680ohm. 330/680 is .485 times gain, less than unity, so i hope the AD825 is stable at less than unity gain.
Cheers George
And yes Peranders the two NPN transistors are the switch to deck for the de-emphasis circuit, but i was talking about the output opamps, if you look at the return feedback resistor for the feedback it's 330ohm and the one to deck is 680ohm. 330/680 is .485 times gain, less than unity, so i hope the AD825 is stable at less than unity gain.
Cheers George
BTW guys has anyone tried (or heard)to run the output straight from the CS4328 dac as it has zero dc offset and can give out 5v of audio signal, i suppose that's why Roksan use a negative gain buffer, as to keep to the red book 2v output.
I've see some schematics on running it straight out with a simple 1st order filter of 51ohm in series and a 10nf to deck. Which in my calculations gives a low pass filter of 312khz at 3db.
Cheers George
Whoops made a mistake with the gain of the opamp, you should add the resistor that goes to deck to the one from the out put then divide by the one that goes to deck.
680ohms+330ohms=1010ohms, divided by 680ohms =1.48xgain.
That's got it (i think)
I've see some schematics on running it straight out with a simple 1st order filter of 51ohm in series and a 10nf to deck. Which in my calculations gives a low pass filter of 312khz at 3db.
Cheers George
Whoops made a mistake with the gain of the opamp, you should add the resistor that goes to deck to the one from the out put then divide by the one that goes to deck.
680ohms+330ohms=1010ohms, divided by 680ohms =1.48xgain.
That's got it (i think)
In fact the best results with CS4328 DAC and similar are obtained by running the DAC output straight to RCA's. This was suggested somewhere on the net - discussion on buffering different DAC types. (hope CS4328 is multi bit Vout DAC!!!)
Back to buffers for a moment:
I have tried using two halfs of OP275 in parallel with excellent results. One half just was not good enough to drive highish capacitive load of my interconnects, but two together made huge difference. I believe AD825 would benefit as well.
Extreme_Boky
Back to buffers for a moment:
I have tried using two halfs of OP275 in parallel with excellent results. One half just was not good enough to drive highish capacitive load of my interconnects, but two together made huge difference. I believe AD825 would benefit as well.
Extreme_Boky
Hi georgehifi,
I have just checked the gif file you posted and you should definitely try feeding the Vout straight to RCA connectors. Use the filtering cap and series resistor as well (not sure if that mosfet output could copy with accidental shorts to GND).
825 is single smd and you could try stacking one on top of the other for parallel buffering, if you decide that "straight - out from DAC" is not satisfying option (it should be!)
You could also try excellent OP275, or AD826 (might be too revealing and too fast some some...)
Cheers,
Extreme_Boky
I have just checked the gif file you posted and you should definitely try feeding the Vout straight to RCA connectors. Use the filtering cap and series resistor as well (not sure if that mosfet output could copy with accidental shorts to GND).
825 is single smd and you could try stacking one on top of the other for parallel buffering, if you decide that "straight - out from DAC" is not satisfying option (it should be!)
You could also try excellent OP275, or AD826 (might be too revealing and too fast some some...)
Cheers,
Extreme_Boky
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.