Regarding noise modulation, seems like in principle it should be measurable by way of FFT. Sometimes it is measured that way. One problem is that we don't have limits of audibility data for it. ESS says they have proprietary research on its audibility, but they haven't published it. IMHO A research effort to get to an estimated audibility limit for noise modulation would be substantial, especially so if it turns out to depend in some way on type of noise (understanding that a 'limit of audibility' is not a hard limit; rather it is a statistical estimate of an average limit for a population)
Last edited:
IMHO there would have to be substantial grant money at stake and a dean/chair that would accept the line of research as academically germane.
EDIT: By the way, there is no evidence I am aware of that noise modulation is perceived by humans in the same way that, say, fixed level white (or frequency shaped) noise is.
EDIT: By the way, there is no evidence I am aware of that noise modulation is perceived by humans in the same way that, say, fixed level white (or frequency shaped) noise is.
Last edited:
I can think of one forum member who didn't bother to measure AVCC power supply current depending on signal in order to design a proper power supply for a recent ES9038PRO project. Otherwise we would have been told about the measurements.
I can also think of a member which has no idea what he's talking about when it comes to the AVCC ES9038PRO power supplies; a close look at the board would reveal a lot about the choices made.
Recommending an off board AVCC power supply, of whatever topology, is the worst advice you tried to pass on; much simpler, any DAC board, Chinese or from the US, that does not have the power supplies regulators and distribution on board should be declared an engineering abomination rip off, disregarding the DAC chip model.
I believe a study on the reproduction habits of the Hippocampus Guttulatus, AKA the long-snouted seahorse, has much more chances to be financed by any govt. or private entity.
I can think of one forum member who didn't bother to measure AVCC power supply current depending on signal in order to design a proper power supply for a recent ES9038PRO project. Otherwise we would have been told about the measurements.
LOL
If you use an IV that keeps the output voltage constant, the joke is, it draws constant current.
It does put out a huge amount of HF noise though.
IMHO for this dac you could use 560pf or 470pf. If you used 470pf then you could put another small cap in parallel with it to make the total capacitance closer to 510pf. It will change the filter corner frequency if the cap is the wrong value, but for the amount of change you ask about there will probably not be much audible effect.
It is probably more important that you use appropriate caps, say, polypropylene film, or else C0G/NPO rated for at least 50v or 100v, for any caps in the audio path, and or for filter caps after the AVCC dividers (that filter the offset voltage for the I/V opamp non-inverting inputs).
Regarding the schematic you posted, it looks about right except there is only one AVCC divider for each channel, not one divider for each opamp in one channel. For one channel, both I/V opamps share the same AVCC divider (the 10k and 6.3k resistor voltage dividers).
It is probably more important that you use appropriate caps, say, polypropylene film, or else C0G/NPO rated for at least 50v or 100v, for any caps in the audio path, and or for filter caps after the AVCC dividers (that filter the offset voltage for the I/V opamp non-inverting inputs).
Regarding the schematic you posted, it looks about right except there is only one AVCC divider for each channel, not one divider for each opamp in one channel. For one channel, both I/V opamps share the same AVCC divider (the 10k and 6.3k resistor voltage dividers).
Last edited:
Default is 2-channel. It would use all the default settings shown in the data sheet. That's not good though, since DPLL_Bandwidth should be set to the lowest stable setting for best sound quality.
Also, the diyinhk board is designed for stereo. It parallels 4 dac chip output channels into each I/V opamp inverting input.
Also, the diyinhk board is designed for stereo. It parallels 4 dac chip output channels into each I/V opamp inverting input.
An Arduino can work, or any other MCU you know how to program. One should probably use an I2C isolator chip between the MCU and the dac, but so long as care is taken with the interfacing an isolator is not required. The I2C pins the dac board should have pullup resistors added that go to a digital 3.3v supply on the board (not AVCC power). That way there is no pullup voltage when the dac is powered off. The dac I2C pins are rated as 5v tolerant so long as the dac is powered on. That means is can be risky to apply 5v to them when the dac is powered off. Probably safest to use a 3.3v MCU, unless an isolator used in which case it doesn't matter.
Last edited:
