Hi DYIyers!
I being browsing this great forum for a while and found a lot of great info and tips. Now I came up with question which strikes me after careful testing of MUSE DAC from ebay build on DIR9001 + 4x TDA1543 chips in parallel.
Initially I found it sharp sounding, somewhat interesting. I didn't use it for few month. Than I took it for some comparison test on new home made gear and found sound harsh and unnatural.
Than I took my TEST CD made in Holland, connected DAC to my Marantz CD6000 OSE by SPDIF and Toslink cables and checked output with oscilloscope. What I saw shocked me to the core. How people claim Hi-End sound from it? Or may be it failed some how? I like some opinion before I start digging PCB. For now I just checked all electrolytic caps, for shorts e.t.c. and looks no problem there. No difference between SPDIF and Toslink. The other portable DAC from FiiO I have works fine.
See pics taken with 1, 10, 20 kHz sinus test signal. CH1 (yellow) is from DAC, CH2 (blue) from Marantz analog out.
Thank for viewing!
Genick.
I being browsing this great forum for a while and found a lot of great info and tips. Now I came up with question which strikes me after careful testing of MUSE DAC from ebay build on DIR9001 + 4x TDA1543 chips in parallel.
Initially I found it sharp sounding, somewhat interesting. I didn't use it for few month. Than I took it for some comparison test on new home made gear and found sound harsh and unnatural.
Than I took my TEST CD made in Holland, connected DAC to my Marantz CD6000 OSE by SPDIF and Toslink cables and checked output with oscilloscope. What I saw shocked me to the core. How people claim Hi-End sound from it? Or may be it failed some how? I like some opinion before I start digging PCB. For now I just checked all electrolytic caps, for shorts e.t.c. and looks no problem there. No difference between SPDIF and Toslink. The other portable DAC from FiiO I have works fine.
See pics taken with 1, 10, 20 kHz sinus test signal. CH1 (yellow) is from DAC, CH2 (blue) from Marantz analog out.
Thank for viewing!
Genick.
Attachments
I've had one of these - what you're looking at is the effect of no filtering of image products. All unfiltered NOS DACs have this. The first image which shows clipping at both top and bottom (worse at the top) is indicative of a too-large valued I/V resistor. You can fix this up easily by doing the math based on the PSU voltage of the DAC chips and then fitting a pair of reduced value resistors.
Fixing up the other two isn't quite as simple - you'd need to install a very steep filter to get some semblence of a sinewave output at 20kHz. My own NOS DACs have a 5th order filter (3 caps, 2 inductors) and I get a decent sinewave out at 10kHz. The wobble starts to appear beyond this but its still recognisably a sine until at least 15kHz.
The long and short is - no, there's nothing wrong with your DAC other than the I/V resistors being too high a value.
Fixing up the other two isn't quite as simple - you'd need to install a very steep filter to get some semblence of a sinewave output at 20kHz. My own NOS DACs have a 5th order filter (3 caps, 2 inductors) and I get a decent sinewave out at 10kHz. The wobble starts to appear beyond this but its still recognisably a sine until at least 15kHz.
The long and short is - no, there's nothing wrong with your DAC other than the I/V resistors being too high a value.
spectrum comparison
Thanks abraxalito !
Honestly I was expecting something like 1 kHz on 15 kHz and even more. But looks like DAC works as 2-3 Bit resolution after 8 kHz. That is creating a lot of HF harmonics which are extended up to 600 kHz on sinus test signal 20 kHz !!! Considering I have amplifiers with up to 200 kHz range, all that "beauty" going out and even not filtered by speakers crossovers. Not even mention full range speaker don't have crossover and output up to 40 kHz - 6dB. The only filters left are my ears and brain. Now I asking again: what we actually hearing from CD after 5-8 order filters or / and oversampling?
As for the small clipping you pointed out, I fixed it by slightly reducing load resistors 2x680 ohm -> 560 and one 390 -> 320.
Since I've invested some time already, here some spectrum screenshots below.
Two taken from Marantz CD analog out on 1 kHz and 20 kHz sinus, and other two from MUSE DAC on the same frequencies. Would be interesting to see same test on highly polished NOS DAC which is have better resolution in 8 -20 kHz range.
Enjoy.
Thanks abraxalito !
Honestly I was expecting something like 1 kHz on 15 kHz and even more. But looks like DAC works as 2-3 Bit resolution after 8 kHz. That is creating a lot of HF harmonics which are extended up to 600 kHz on sinus test signal 20 kHz !!! Considering I have amplifiers with up to 200 kHz range, all that "beauty" going out and even not filtered by speakers crossovers. Not even mention full range speaker don't have crossover and output up to 40 kHz - 6dB. The only filters left are my ears and brain. Now I asking again: what we actually hearing from CD after 5-8 order filters or / and oversampling?
As for the small clipping you pointed out, I fixed it by slightly reducing load resistors 2x680 ohm -> 560 and one 390 -> 320.
Since I've invested some time already, here some spectrum screenshots below.
Two taken from Marantz CD analog out on 1 kHz and 20 kHz sinus, and other two from MUSE DAC on the same frequencies. Would be interesting to see same test on highly polished NOS DAC which is have better resolution in 8 -20 kHz range.
Enjoy.
Attachments
- Status
- Not open for further replies.