Low-distortion Audio-range Oscillator

[1audio]

Are you measuring using the differential input? it makes a difference in both distortion and noise/hum artifacts.

 

[bozidarms]

Hi Demian,
i am using Neutrik cinch to XLR ad.
Yes i know fahr from optimal, but another osc. is measured unter same circumstances.
BTW I posted yet a measurement from VOsc. now on batteries.
In principle the same harmonic constellation as with Silentsw.

https://www.diyaudio.com/forums/attachment.php?attachmentid=796071&stc=1&d=1574184965

 

[1audio]

I would propose that you measure the connections in the adapter. There are numerous ways to connect them internally. Typically pin 3 is left unconnected inside which will compromise measurements.

I would connect pins 2 and 3 to the RCA and bring pin 1 out on a separate wire for a ground connection to the source chassis.

 

[bozidarms]

TNX Demian,
i'll give it a try.

 

[JensH]

I have seen good results with cables made as shown at the top of page 5 in this application note: https://www.hypex.nl/img/upload/doc/an_wp/AN_Legacy_pin_1_problems.pdf

 

[HpW]

BTW I posted yet a measurement from VOsc. now on batteries.
In principle the same harmonic constellation as with Silentsw.

Is your Victor LDO BOXED? Otherwise do so 😀

Hp

 

[diyralf]

Is your Victor LDO BOXED? Otherwise do so 😀

Hp

A housing has only effects on the noise, not on the distortions.

 

[scott wurcer]

A housing has only effects on the noise, not on the distortions.

Unless it's ferro-magnetic.

 

[Cobalt1]

A housing has only effects on the noise, not on the distortions.

it is not so.
distortion is slightly reduced

 

[diyralf]

it is not so.
distortion is slightly reduced

q.e.d. I see reduced noise, but almoust the same amount of distortion.

 

[Cobalt1]

q.e.d. I see reduced noise, but almoust the same amount of distortion.

distortion is always reduced, sometimes a little more, sometimes a little less.

 

[Vovk Z]

q.e.d. I see reduced noise, but almoust the same amount of distortion.

Low-order harmonics are almost the same, but noise floor is lower - it counts noise as high-order harmonics. So lower noise floor - lower THD.

 

[jan.didden]

I agree. Subsequent measurements are always slightly different anyway. The difference is too small to draw any conclusions from it.

Jan

 

[bozidarms]

I have followed Demian's and Jens instructions and have made a new cable + separate grounding.
Interference pattern is maybe a bit quieter now, but THD and harmonic intensity is more or less the same.
On 3,16V range and 1V signal THD and harmonic intensity remains the same
-123dB for THD and 2H.

https://www.diyaudio.com/forums/attachment.php?attachmentid=796233&stc=1&d=1574249877
https://www.diyaudio.com/forums/attachment.php?attachmentid=796234&stc=1&d=1574249877


Only on 10V input range and 1V signal, pattern become improved, THD was -126dB now is -132, 2H was -128dB now is -136dB (I have made averages time longer, so that could be the reason for better result)


https://www.diyaudio.com/forums/attachment.php?attachmentid=796235&stc=1&d=1574249995
https://www.diyaudio.com/forums/attachment.php?attachmentid=796236&stc=1&d=1574249995

It seems that I have still problem with to many interferences, but also with osc. itself-harmonic content is simply to large.

 

[jan.didden]

Are you sure the power supply setup is correct? Can you measure the supply voltage directly at the opamp pins on the back of the board?

Also, in the last case H2 is -136dB but the fundamental is -6dB so H2 is -130dB below fundamental or -130dBc.

Jan

 

[phofman]

A ground loop present in the measurement setup?

 

[bozidarms]

On both opamp:
pin 4 -15, pin8 +15V

 

[Vovk Z]

About -90 dB 50 Hz noise - it's too much. Something noisy is near the oscillator, or near tbe cable, or something else.
-110 is ok, -120 is good.

 

[diyralf]

I suspect that is the problem with the power transformer in the RTX.

 

[Vovk Z]

10 kHz Victor's oscillator

Here is an exaple of one of my old spectrums, 10 kHz original Victor's osc, unbalance cable of cause (1.5 or 2.0 meters of RG58 or RG59 coaxial, I don't remember exact type, about 6 mm diameter), battery powered, aluminium 1.5 mm case: