I have received a LPF board. Looks great.
Wanted to test it but it took me a long time to find the J1 to J4 markings.
A more readable marking on the other side with in and out would have helped a lot🔎
Wanted to test it but it took me a long time to find the J1 to J4 markings.
A more readable marking on the other side with in and out would have helped a lot🔎
today one German guy bought a bit strange set of devices, I still don't realize his config. Cosmos ADC 1pcs, Cosmos APU 2pcs, so far is ok, maybe vinyl reaper enthusiast but.. Cosmos LPF 2pcs and all required cables. 2pcs LPF is strange to me 😉
ah, maybe but how do you realize the full cables set for that? Sunny asked me to think what is the full set of cables for such list of devices.
The LPF filters can be set in series to get steeper slope and more attunation of harmonics.
So that is a use case
So that is a use case
Tried the LPF yesterday with REW
Just with ASUS U7 -> LPF ->ADCiso. Balanced. That is L+R out, R inverted
The ASUS has about -105 dBFS THD up to 1 k, then it rises. Maybe -90dBFS @ 20k
With LPF it benefits from
So a very good solution to get a old/mediocre DAC "up to medium speed"
Just with ASUS U7 -> LPF ->ADCiso. Balanced. That is L+R out, R inverted
The ASUS has about -105 dBFS THD up to 1 k, then it rises. Maybe -90dBFS @ 20k
With LPF it benefits from
- Even harmonics cancelled about 50 dB on my card
- 3rd harmonic attunated about -18 dB
- The output of the LPF can drive the ADC-iso on 640R setting.
So a very good solution to get a old/mediocre DAC "up to medium speed"
torgeirs, I think you can replace Asus with Meizu HiFi DAC nearly for free to get a lot better result:
It works fine with unbalanced cheap DACs(I saw it for $13) too, but one channel is inverted to emulate balanced output.
I used the 1.7 Volt setting on the ADC. Guess I am more interested in closer to 0 dBu values. That is around 1 volt out. Not 9 V ;-)
With a SMSL 39pro I get about -132dB 3rd with same setting on ADC. Not much difference with or without LPF. Maybe 3 to 5 dB better with LPF
That is OK by me
With a SMSL 39pro I get about -132dB 3rd with same setting on ADC. Not much difference with or without LPF. Maybe 3 to 5 dB better with LPF
That is OK by me
Hopefully I will have the time to meaure a ADAU solution with a cheap cirrus codec soon. The ADAU can generate its own sin wave. And a 3rd order compensation sin wave can be made. I will post the results if they are usable.
Yes quite nice.
ADAU1452 generating sin wave and compensating 3rd harmonic with a x^3 addition at about -80 dB
The DAC is a Cirrus logic codec about 10 years old. Analog output filter removed from DAC board.
Still 1,7V setting on ADC. 2 channels used on codec/DAC, one inverted.
ADAU1452 generating sin wave and compensating 3rd harmonic with a x^3 addition at about -80 dB
The DAC is a Cirrus logic codec about 10 years old. Analog output filter removed from DAC board.
Still 1,7V setting on ADC. 2 channels used on codec/DAC, one inverted.
Last edited:
Is your ADC marked as "THD+" on the bottom? H3 -148db isn't an average performance for Cosmos ADCiso, if I see such a unit I mark it "THD"(all harmonics -140db) or "THD+"(-145db and lower). The worst unit has H3 -133-134db, average about -136-138db, H2 is always <-140db. I measure/calibrate ADCs at +27C and 768k. 768k makes the ADC chip warmer and harmonics lower.
I see no marking outside the box. I will come back with more measurements to show more low distortion 😉
Regarding the Bal/Unbal stage. I get just 18dB of rejection when I try in two in phase sin wave. LPF part has about 50 on my setup.
So something has to be wrong. Measured the resistors and found the answer? Wrong values at some of the resistors?
They match very well, by the way, the ones with same value, that is.
There is one more trick, the differential opamp stage uses two different values of the resistors(510/510 and 180/180) to match Zin of In+ and In-. It reduces H3.