You'll want to add filtering in between somewhere for best performance, and the amp is a strange place to put that, but not it of the question...
I'm assembling a 9038 with Mercury and have a few general questions.
I am configuring it with fixed output because I want SE outputs to a tube headphone amp (I actually need 2 SE outputs).
I also have active monitors that connect via XLR, and need volume control.
So,
1) Is there any disadvantage to connecting an analog pot for XLR as opposed to using the digital volume?
2) I need dual SE outputs. Is there any issue wiring to 2 at the same time or should I add a rotary switch between outputs?
3) For XLR outputs, is it as straight-forward as wiring the 3 outputs, or do I need to ground anything to chassis?
Thanks for the help
I am configuring it with fixed output because I want SE outputs to a tube headphone amp (I actually need 2 SE outputs).
I also have active monitors that connect via XLR, and need volume control.
So,
1) Is there any disadvantage to connecting an analog pot for XLR as opposed to using the digital volume?
2) I need dual SE outputs. Is there any issue wiring to 2 at the same time or should I add a rotary switch between outputs?
3) For XLR outputs, is it as straight-forward as wiring the 3 outputs, or do I need to ground anything to chassis?
Thanks for the help
? on Mercury output filter with 9038/28 and DSD 256 input
Hi Russ, I have not yet seen the ESS 9038/28 data sheet, so I do not know if it addresses my question or not. The question is, do the requirements for the I/V stage's output filter relax some when the input rate to the ESS 9038/28 is very high?
In my case, I am only running DSD 256 into the DAC, at this rate would it make sense to relax the analog output filter of the Mercury I/V? Or, to put in another way, does the ESS 9028 move output noise to higher frequencies when in putting DSD 256?
Hi Russ, I have not yet seen the ESS 9038/28 data sheet, so I do not know if it addresses my question or not. The question is, do the requirements for the I/V stage's output filter relax some when the input rate to the ESS 9038/28 is very high?
In my case, I am only running DSD 256 into the DAC, at this rate would it make sense to relax the analog output filter of the Mercury I/V? Or, to put in another way, does the ESS 9028 move output noise to higher frequencies when in putting DSD 256?
1) Yes. A simple analog pot will have much higher noise than the digital volume control. In addition, you will need a precision pot so that the + and - poles track very closely. It will need to be a four-pole pot as well, which can be hard to find. Something like this would work: 652-PTD904-1020KA103
2) No issue connecting two at the same time, as long as the inout impedance is reasonable (>10K).
3) Yes. Generally there is no need to ground the XLR connectors to the chassis.
Last edited:
In my experience - leave the filters alone (they are not very aggressive to start with).
You could raise the corner freq (because of the filter topology this is not as simple as it sounds) - but that would also change the overall cct phase margin - I wouldn't advise that.Cheers!
Russ
I believe it would work fine - but I must warn that I have not tested it.
I have used IVY for COD and it works really well!
My one concern with using Mercury for COD is I am not sure how that DAc chip would do with the self leveling input bias. It may be that that chip needs the input to sit at or close to GND - not to center at the DAC outputs... I have not tested that scenario and I am not setup to test it at the moment - so if you want to try it it would have to be at your own risk. That said - I am inclined to think it would work fine. I would use something like 390R to start - increasing that value if the output is not high enough for you.
Cheers!
Russ
I’ve played around with different I/V stages for the Burr Brown 179X series DAC chips, and while the Mercury should work, this DAC chip family is designed to output into a low impedance virtual ground. As the I/V stage input moves away from gound the sound quality begins to degrade. I found that as a general rule of thumb the I/V input needs to stay within 50mV of ground with an input impedance off less than 10 ohms to get best performance.
Terry
Hi Terry!
Nice to see you around!
Yeah input impedance will be very near zero up to the Mhz region on the Merc
I suspect we might get a bit better results with a fully symmetrical front end like that of the mercury?
The input will basically just float to the mid point of the current bias of the DAC.
I would not try it with a conventional I/V stage (dual trans-impedance stages) - but I suspect the super symmetrical nature of this circuit might fare better.
In any case - I can't say - I have not tried it. It would be fun to give it a go and measure it. BTW it would be pretty trivial to reference the Merc's common mode input to GND for testing purposes.
Cheers!
Russ
Nice to see you around!
Yeah input impedance will be very near zero up to the Mhz region on the Merc
I suspect we might get a bit better results with a fully symmetrical front end like that of the mercury?
The input will basically just float to the mid point of the current bias of the DAC.
I would not try it with a conventional I/V stage (dual trans-impedance stages) - but I suspect the super symmetrical nature of this circuit might fare better.
In any case - I can't say - I have not tried it. It would be fun to give it a go and measure it. BTW it would be pretty trivial to reference the Merc's common mode input to GND for testing purposes.
Cheers!
Russ
Hi Terry!
Nice to see you around!
Yeah input impedance will be very near zero up to the Mhz region on the Merc
...
The input will basically just float to the mid point of the current bias of the DAC
Russ, it’s been awhile, but I’ve been ghosting around, and recently starting to play back in DIY again. Nice to see I’m still remembered.
Floating at the bias midpoint could work here. I see what you mean now.
I hope to make a venture into Buffalo / Mercury territory in the future, but right now I don’t have the time or cash flow to make that a reality. Big props for what you have accomplished with Twisted Pear! You’ve gone waaaay beyond the humble beginnings of the designs I had a finger in. Well done indeed!
Terry
A few questions before I try this myself:
1) What is the correct wiring for a 4 pin XLR socket from the balanced outputs?
2) Can I connect both 4 pin XLR for headphone and dual XLR's for active monitors?
3) I have already installed the resistors for 4V output... do I need to change this?
I plan on driving the above mentioned headphones.
Thanks for the help!