• The Vendor's Bazaar forum is for commercial offers and transactions. Only unmoderated members can post here.

    diyAudio provides this forum for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members. Use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

Cosmos APU a notch+LNA $70 to outperform APx555b for $30,000

Jump to Latest
#762
I tried that but third harmonic get worse.
Switching clock changes nothing.

CLK0.jpg


CLK1.jpg
 
#764
IVX said:
And pls don't use the Crosscorrelation of REW.
Click to expand...
Why? Is there an error in the algorithm or something that makes its output incorrect? (It does happen) Very low harmonic measurements are really difficult to validate. On the Shibasoku I had to add a marker at a known levlel (-120 dB) to have some confidence it was a real measurement. Even more difficult to do with REW/ARTA etc. If there is any frequency drift between source and FFT the harmonics just melt away, looks great but is just a lie.
 
#765
If he has Cosmos APU, it is better not to use anything else. The first time I tested croscor was very promising, with close to APU results. The second try wasn't so nice, >1db of THD+N difference vs APU(unfortunately I don't remember if the same 9039S DAC was used or different but APU shows -125.8db both). So, just in case, APU is enough and the result is reliable and stable.
PS: bare ADC anyway better to use with croscor function, at least if you measure the THD+N or DR.
 
#768
So... Potentially, this device would also be useful as a kind of substitute for the APU when making 19 + 20 KHz IMD measurements.

Also, maybe a better way to set the compensation adjustments on the ADC would be to use something like a Victor oscillator into the LPF, then adjust the tweaks for the ADC. That way, no dueling compensation or cancellation between the DAC and ADC. Once the ADC is set properly, then tweak the DAC settings.
 
#769
I have added to the board 2 channels diff2se stages, and pretty flexible commutation logic with 10 poles DIP-switch. Left se-output may be fed:
1) from the output of LPF
2) before LPF i.e. from the Left balanced input
3) from the Right balanced input
4) from the Left but with gain = 2
5) from the Left and Right balanced input

The 1st mode is obvious, kind of wiping out all harmonics and noise over 1K, however, it is too difficult to get the same clean sine with SE stage, unlike Diff. I didn't see THD@1K <-140db and THD+N < -125db.
2nd mode just converts 9039S balanced2unbalanced outputs with the gain =1, THD@1K <-140db and THD+N < -125db. The highest harmonic is H2 -145db.
5rth mode I hope would be useful to mix L+R DAC's outputs to get 3db lower noise(in theory) or 19k+20k, 60+7000, and so on IMD tests.
 
  • Like
Reactions: YashN
#778
IVX said:
5rth mode I hope would be useful to mix L+R DAC's outputs to get 3db lower noise(in theory) or 19k+20k, 60+7000, and so on IMD tests.
Click to expand...
For IM the best results come from mixing well isolated oscillators with really good resistors. The active circuit will limit the performance. The quality of the oscillators is not that critical as long as they are well isolated.
 
You must log in or register to reply here.
Top Bottom