1kHz Reference source

[scott wurcer]

https://www.diyaudio.com/forums/equ...rtion-audio-range-oscillator-post5936075.html

Yes, I used an external PS as Jan did.

 

[Rja4000]

https://www.diyaudio.com/forums/equ...rtion-audio-range-oscillator-post5936075.html

Yes, I used an external PS as Jan did.

"auto-tracking active notch" ?
That seems interesting

 

[PaulD]

Yes that one is photo-resistor based, Victor's is FET based. That article references the one we did in 1994 which was Blackmer VCA based.

I quoted Jan's measurements of Victor's oscillator above which added nothing to the floor of his AP.

Doh! Sorry for the noise...

 

[Ralf Stocker]

Very interested: https://viccc42.wixsite.com/uld-audio
Maybe beats AP !?

 

[SIY]

I have one. Lower distortion than the AP.

 

[chris719]

I have a few also, newer and Rev 1. They are even better after the OPA1656 replaced the LME49720. He was also able to make me one at 100 kHz for testing a non-audio ADC, with some compromise in the performance of course, but far cleaner than even a decent arb generator.

 

[dominikz]

Very interested: https://viccc42.wixsite.com/uld-audio
Maybe beats AP !?

I got one of these recently - really happy with the performance! Most definitely a huge overkill for my hobbyist needs, though - but a fun project nevertheless.
Anyway, I'm powering mine from an isolated multi-output guitar pedal power supply, connecting 4 x 9V outputs in series and using series resistors to get the voltage down to the required 35V. Didn't get the matching Twin-T from Victor sadly, so I decided to build one myself.

Here's some photos and measurements.

Oscillator board I received from Victor:

Inside the Twin-T notch + final devices built into boxes and with labels printed:

Have to say it was a bit tedious to match the Twin-T notch filter components to get a nice, deep notch - especially for balanced use!

As far as measurements go, I first recorded the notch frequency response, so I could use it as a calibration curve in REW to compensate for the filter response:

Here's zoomed to the 1kHz area, to better see the filter depth:

As you can see, the channels are not perfectly identical, but still fairly close and good enough for me

And this is the oscillator performance when running through the notch filter (with notch calibration curve applied in REW, 1M-point spectrum using Blackman-Harris 7 window and 32 averages)

Basically the distortion components are ~160dB down from the fundamental and buried in the noise floor.

I'm using the oscillator mainly to try and isolate the distortion components of my soundcard ADC from those generated by any device I'm measuring - since I don't really have calibrated equipment, this was a nice and not too expensive way to get some point of reference.

 

[chris719]

Looks cute.

I designed and built a 5V in Micro USB to 35V out galvanically isolated boost converter to use with these oscillators. I am not exactly happy with the performance and I'm not sure it can live right next to the oscillator in the same box yet without showing up in the spectrum. If I get more time to work on it and can improve the performance I will post the files.

I wish there were a composite / buffered version that could drive more demanding loads without much distortion increase.

 

[Ralf Stocker]

Nice stuff! Which sound card do you use and which software?

 

[dominikz]

Nice stuff! Which sound card do you use and which software?

Thanks! I'm using the first gen RME Babyface (Silver Edition), had it since 2011.
As far as measurement SW goes, I'm using the typical hobbyist stuff - mosty REW. Previously I used RMAA a lot, and now I started using ARTA again for a few things (e.g THD vs level plot).
But have to say, if I was doing measurements more seriously, I'd probably invest in some kind of specialized rig like QA401 - doing it manually like I am can be quite time consuming (and frustrating)

 

[pizzigri]

Looks cute.

I designed and built a 5V in Micro USB to 35V out galvanically isolated boost converter to use with these oscillators. I am not exactly happy with the performance and I'm not sure it can live right next to the oscillator in the same box yet without showing up in the spectrum. If I get more time to work on it and can improve the performance I will post the files.

I wish there were a composite / buffered version that could drive more demanding loads without much distortion increase.

I purchased Viktor's 1khz oscillator too... I'm actually very interested in a nice and clean PS, and cant wait to see your solution.
Franco

 

[KT88]

Accidentally came by this thread now. I have a similar 1KHz oscillator from Victor, I've created a small PCB with 5V input to power it.
I've posted this a couple of years ago on my blog here:
http://tolisdiy.com/2019/03/11/low-thd-oscillator-power-supply-and-pcbs-as-case-panels/
If this is of use to anyone, feel free to use/modify as you see fit. The Gerbers are available for download at the end of that page as well.

 

[Rja4000]

Accidentally came by this thread now. I have a similar 1KHz oscillator from Victor, I've created a small PCB with 5V input to power it.
I've posted this a couple of years ago on my blog here:
http://tolisdiy.com/2019/03/11/low-thd-oscillator-power-supply-and-pcbs-as-case-panels/
If this is of use to anyone, feel free to use/modify as you see fit. The Gerbers are available for download at the end of that page as well.

I had a look to your page before purchasing Victor's 1kHz oscillator.
I feed mine with a set of 9V cells.
Distortion is pretty low, but noise is higher, in my case.

 

[KT88]

I had a look to your page before purchasing Victor's 1kHz oscillator.
I feed mine with a set of 9V cells.
Distortion is pretty low, but noise is higher, in my case.

The measurements shown on that page are when its feeding an EMU 0404 USB directly, so I would not use these as the oscillator performance, but rather a measure of the EMU itself.

When I've measured distortion via a notch the oscillator performance were far better than this (~140dB below the 1KHz tone).

 

[TheReasonIsSound]

I got one of these recently - really happy with the performance! Most definitely a huge overkill for my hobbyist needs, though - but a fun project nevertheless.
Anyway, I'm powering mine from an isolated multi-output guitar pedal power supply, connecting 4 x 9V outputs in series and using series resistors to get the voltage down to the required 35V. Didn't get the matching Twin-T from Victor sadly, so I decided to build one myself.

Here's some photos and measurements.

Oscillator board I received from Victor:
View attachment 121655

Inside the Twin-T notch + final devices built into boxes and with labels printed:
View attachment 121658
Have to say it was a bit tedious to match the Twin-T notch filter components to get a nice, deep notch - especially for balanced use!

As far as measurements go, I first recorded the notch frequency response, so I could use it as a calibration curve in REW to compensate for the filter response:
View attachment 121659
Here's zoomed to the 1kHz area, to better see the filter depth:
View attachment 121661
As you can see, the channels are not perfectly identical, but still fairly close and good enough for me

And this is the oscillator performance when running through the notch filter (with notch calibration curve applied in REW, 1M-point spectrum using Blackman-Harris 7 window and 32 averages)
View attachment 121662
Basically the distortion components are ~160dB down from the fundamental and buried in the noise floor.

I'm using the oscillator mainly to try and isolate the distortion components of my soundcard ADC from those generated by any device I'm measuring - since I don't really have calibrated equipment, this was a nice and not too expensive way to get some point of reference.

Hi,

Sorry for jumping in but I really need to ask:
How did you make the balanced connection out of the Victor Oscillator?
Do you have a schematic?

I have Victor Oscillator too, but Im not sure how the circuit works, so Im not sure how to create the balanced connection out of it.

Thank you in advance!
Dodo