I have a couple of Soekris dacs and an ancient but overbuilt in any respect ESS9018. All my dacs benefit from some form of fifo or reclocking, often combined. The 1862 as well.
Fwiw, i find the Soekris dacs intolerably dull with the NOS filter, much worse than the 1862.
The only digital filter i have at hand is a DF1704 and one of these days will try it out with the AD dac.
And please don't call me funny names 🙂
Damit the Soekris was the one that disapointed me the most despite profiting of a bench of better polymer cap at the rigth area. I sold it fastly after few months. And whatever the fifo, all the front ends mods could be heard. It was not a good dac according my expectations. I haven't heard the further iterrations though.
Btw is DF1704 better than PMD100 ?
Btw is DF1704 better than PMD100 ?
Not sure if this can be very productive as everyone's system and expectations differ.
Tonal balance is generally good, bass resolution and delineation is almost spectacular. Drums too. Strings otoh, especially violins leave me wanting.
Something that is present with other dacs and most definitely with analogue is lacking here. Playing higher than 44/48kHz bit rates does not really improve it and neither does software oversampling.
I should really connect a hardware oversampling filter, just for elimination purposes, and then look elsewhere for a culprit.
Btw, i only recall two attempts on this board to address NOS issues in the analogue domain:
sinx/x analogue filtering as discussed per example here https://www.diyaudio.com/community/threads/how-to-design-sinx-x-filter-for-tda1541a.6331/
and of course Abraxalito's amazing Zanden clone https://www.diyaudio.com/community/threads/dark-led-passive-filter-i-v-stage-for-nos-dacs.395208/
The idea of upconversion to high rate dsd is a little mystifying. Apparently some extremely clued in people support it, so it cannot be bunk.
But my personal experience with HQPlayer and a reasonably high end commercial dsd dac was not positive. Indeed, very smooth, maybe even too much. 🙂
According to knowledgeable people the 1704 is pretty bad, but i don't have a PMD100.
The Soekris took a lot of work and patience, horrid out of the box.
Thanks @zoom777, @rehanabid, and @diyiggy for your tips, they were very helpful. Below are some pics of how I have the wavio connected to the TDA1541. I did have it mounted underneath the miro power supply, as there was ample room there. I took it out for ease of access to test further, after playing with it for a bit I was able to get sound at the outputs of the dac it's very distorted, music sounds like it's clipping badly. I am wondering if it's because the waveIO is not properly mounted, I will have the opportunity to test further tomorrow night and mount it on standoffs on chassis.
the waveIO does have ufl pads, though the TDA1541 dac does not have corresponding GND ufl pad. I can try to purchase and mount some ufl pads on the TDA1541 to see if I have better luck there. Is this the correct part? https://www.electromaker.io/shop/pr...UR53uQTFY2egUiKuPkiGkbo4gJSISXJMXUaPcnOhnx6ug
the waveIO does have ufl pads, though the TDA1541 dac does not have corresponding GND ufl pad. I can try to purchase and mount some ufl pads on the TDA1541 to see if I have better luck there. Is this the correct part? https://www.electromaker.io/shop/pr...UR53uQTFY2egUiKuPkiGkbo4gJSISXJMXUaPcnOhnx6ug
each ufl pads are connected to ground. Only one side of an uf-l pad is the signal input, the three others are going to ground (so you have to know to solder it)
Does the W . I/O already worked with another dac ? Try any VDC from an isolated reg from 3.5V to 5V to check if it is not a gnd loop problemo.
Have you tried to feed the W. I//O main power plug with a clean 5V in spite of feeding it from the USB (read the datasheet for the jumpers and re read the datasheet).
there is only one size, so the link is ok. you need spares too and more cables than the three needed cause it is fragile.
Does the W . I/O already worked with another dac ? Try any VDC from an isolated reg from 3.5V to 5V to check if it is not a gnd loop problemo.
Have you tried to feed the W. I//O main power plug with a clean 5V in spite of feeding it from the USB (read the datasheet for the jumpers and re read the datasheet).
there is only one size, so the link is ok. you need spares too and more cables than the three needed cause it is fragile.
If anyone needs, I have a few boards left for Miro's DACs, just pay postage. If someone wants me to mount the SMD parts for them, let me know I'll get them ordered. Also some PSU boards that might help.
2 x AD1865
1 X AD1862
2 X PCM1704
1 X PCM63
2 x AD1865
1 X AD1862
2 X PCM1704
1 X PCM63
I had not verified operation of the waveio with another dac. This is where I had hoped to first to put it to use.
So I did try with another power supply supplying the waveio with 5v to isolated v+ input. With x2 from miro psu connected to dac x2, but miro psu turned off, the led for x2 was coming on and I was measuring some 3v across positive and gnd of x2. I have since disconnected x2 from psu to x2 or dac. Last I looked at audio settings the sampling bitrate was set to 32 bits. Will try asio drivers tomorrow.
@arteom set the sample rate to 24bits instead of 32, I did share my experience with bit rates and different DACs (I think AD1862 and TDA1541).
I experienced this kind of noise from the DAC (AD1862) when the bit rate was set higher.
Connect Wave to supply at X2 both +5v and the ground.
Also check a proper connection between WaveIO I2S pins and I2S connector on the DAC with a multimeter, my guesstimate is that you should get "proper sound" when you set the sample rate to 24 bit and lower.
I experienced this kind of noise from the DAC (AD1862) when the bit rate was set higher.
Connect Wave to supply at X2 both +5v and the ground.
Also check a proper connection between WaveIO I2S pins and I2S connector on the DAC with a multimeter, my guesstimate is that you should get "proper sound" when you set the sample rate to 24 bit and lower.
@analog_sa ,
I am asking a genuine and sincere question - in what way is Abraxalito's product similar to Zanden ?
@zoom777
The similarity is limited to the use of a multistage passive LC filter as the only filtering mechanism in a NOS dac. It is clearly not exactly a clone.
The similarity is limited to the use of a multistage passive LC filter as the only filtering mechanism in a NOS dac. It is clearly not exactly a clone.
Attachments
@analog_sa , what's that tube and (output?) caps on the picture on the left? Had some correspondence with Abraxalito, to add a tube output seems not feasible, given the current amplification...
The Zanden filter has rather a different concept to mine. In brief, its a multiple band-stop design to create nulls at and around multiples of the sample frequency with the aim of eliminating phase shift. Whereas Dark LED uses a Chebyshev low-pass with its attendant phase shift.
Zanden patent
Zanden patent
Thanks for the clarification. So, neither addresses the hf roll-off, right?
A couple of dbs @20kHz may not sound like a big deal, but the roll-off is already measurable at 10kHz, so covers a significant range and may easily be audible.
My guess is both the ultrasonic images and the roll-off may cause issues in some setups. An amplifier suffering from slew rate induced distortion may produce audible IMD when receiving unfiltered ultrasound and a system than has been tuned to avoid brightness may sound dull due to the roll-off.