How low do you want to measure? -160 exists off-the-shelf. I saw 4-5 ShibaSoku 725D for sale on ebay at low price. First time in years. Probably all gone now. But, it is crammed full of electronics to do the job. Way beyond DIY IMO.
The Panasonic goes to -140 and is avail for about 1200-1500 USD frequently. Its internal osc measures at -135-ish. Does IM also.
Just as we did with the HP339A, they can be improved with opamp upgrades and selected parts changes.
The osc are generally worse than the analyzer capability.
.... As another route to get to your goals.
David's variable freq gen and a -100dbv ref cal tone:
-141-149dBvTHx-RNMarsh
Last edited:
If we can get david's schematic and make pcb available for it (with profit to his family), we might be able to tweek it lower or use it as-is for many.
Here it is being tested by the Panasonic. It couldnt accurately measure it, so I went to the ShibaSoku and QA401 on the monitor port.
THx-RNMarsh
Here it is being tested by the Panasonic. It couldnt accurately measure it, so I went to the ShibaSoku and QA401 on the monitor port.
THx-RNMarsh
Attachments
Last edited:
That makes sense, it would mean they solved the problem.
My approach would be one of the new 20bit "R2R" DAC's with a de-glitcher and maybe an error correction cycle. I think one of these could approach -130dB so divided down as a lock in signal there should be no problem. SRC has a precision oscillator that does -120dB and exact frequency done this way. I remember being disappointed in the far out of band spurs. This is a lot of complexity to add to an instrument when you already have another solution.
Last edited:
I've been able to find 4 files: "dlb SVO V5 Sheet 1.pdf" and similar sheets 2-4.
I can post here if that's helpful and not in violation of copyright issues.
This article might shed some light, in thinking about it if you have a single frequency oscillator you might want a simple pitch shifter.
Pitch Shifting Using The Fourier Transform | Stephan Bernsee's Blog
In the test arrangement were you using the usual trick of feeding back 1/2 of the drain-source voltage to the gate? Under some conditions and with some JFETs, I have found that the exact value for optimum low distortion is not always exactly 0.5, so in my distortion analyzer I have a trimmer that can adjust that ratio slightly. Another question would be how the threshold voltage, IDSS and Rds-on of the Siliconix devices to the ones from eBay.
Cheers,
Bob
I was actually testing them as a switch. The Boonton uses a lot for switching both in the oscillator and the attenuator (and the notch filter). The J108's showed much less distortion in the later Boontons so I thought I would sort for that. It seems Rds on is not a good enough parameter. In this case I tied the gate to the source through a 5K resistor to emulate the circuit. I tried swapping source and drain and the distortion was slightly higher. As a switch I could add the feedback resistor but there are 30 or more transistors so that's not too practical. I still might do some experiments.
I'll make some more formal measurements and post.
A very low distortion audio range oscillator can be easily constructed using a Raspberry Pi and a Pi DAC hat like the IQAudIO DAC. Install Raspbian and then use the command line to generate whatever frequency you may wish. You may also use a graphical interface like VLC and an audio file with the required frequency.
Not sure about that...here is data for the Pro DAC:
"Class leading audio; 114db SNR, and -94db THD"
Not really a very low distorsion generator....
"Class leading audio; 114db SNR, and -94db THD"
Not really a very low distorsion generator....
Sorry this is not possible. Every DAC on earth has more THD than a well designed analog oscillator.
If they can't measure the distortion at -140dB they need better test equipment...
