Looking for ultra low distortion 1KHz sine generator

[dimkasta]

I have an HP 3312A but it has high harmonics, even after calibration

Anyone knows of a cheap-ish generator that can do less than -120DB (0.0001%) THD+noise at 1KHz?
I am open to DIY too.

Thanks a lot for your time

[cat87]

Unfortunately I can't say much about the commercial availability of these sort of generators, but I can point you to something DIY.

There's an article written by Cordell, on how to build a THD analyzer. The signal generator for it is something along the lines of 0.0003 % or better (that's what he claims) Youd could make some tweaks to it (better op-amps, though the 5534 is already pretty good) and maybe get to your desired -120 dB THD.

This is the article:  http://www.cordellaudio.com/instrumentation/thd_analyzer.shtml

I think you could try doing some I-Q DDS generators.... I think these also have some low THD, but don't take my word for it, I might be wrong on this one.

Hope it helps.

[ebclr]

http://www.ebay.com/itm/Ultra-low-distortion-0-00001-1kHz-sine-generator-assembled-and-tested-PCB-/222449792287?hash=item33cb0a751f:m:mET82r_nyfTTV3JMQmFCAKg

[Karel]

Just take a reasonably simple oscillator and add some low-pass filters.
In order to get the highest possible purity, the last stages of the filters should be passive
in order to remove the distortion from the output buffer and/or active filter.

[agdr]

Yep, +1 for Victor's oscillator in the eBay link above.  It is the real deal.     His oscillator has been pretty thoroughly reviewed over the last couple of years in (the middle of) this massive audio oscillator thread over at diyaudio.com where I usually post:

http://www.diyaudio.com/forums/equipment-tools/205304-low-distortion-audio-range-oscillator.html

Several posts from Victor himself in that thread.  Folks over there have checked his oscillator with high quality audio distortion analyzers from Audio Precision, Panasonic, and Shiba Soku and his low THD claim checks out.  Victor has even made a small improvement about a year ago and lowered the THD even a bit more.  If you contact him he can custom-make other frequencies too.  I have a 1KHz, 10KHz, and 11KHz, the last two being handy for IMD tests.

I have a DIY power supply & case project for it over there too which can hold up to 4 Victor boards.     His board takes 35V in, you can just use a lab supply which is what a lot of folks do. His board uses kind of a unique power supply setup, forming its own ground between the two TL431's.  Important to note that the Victor board ground will be different than the supply ground feeding it.

[dimkasta]

Thanks a lot

I already had Victor's oscillator on my map. I read about it a while back on diyaudio.
Along with the one by Frex (http://www.diyaudio.com/forums/equipment-tools/287892-eosc10kv3-lt-an67-10khz-oscillator-new-updated-version.html#post4637153)

I was just wondering if there is something usable that is commercially available, just to avoid any issues with construction quality etc.
But I guess audio is a restricted market and such devices tend to be expensive

[agdr]

Victor's is about the best commercial bang for the buck that I'm aware of.  There are some other threads over at diyaudio where folks have made modifications to older HP oscillators to improve their THD.

This fellow, tomchr, has the latest model Audio Precision tester, apparently $20K or so:

http://www.diyaudio.com/forums/headphone-systems/300491-neurochrome-hp-1-ultra-high-end-headphone-amp.html#post4918641

Here he notes that the oscillator in the AP only goes down to -112dB THD while Victor's goes all the way down to -140dB THD.  So he has a Victor oscillator in addition to the ultra-picey AP.  Here is the post tomchar is talking about, some possible minor artifacts in Victor's board:

http://www.diyaudio.com/forums/headphone-systems/300491-neurochrome-hp-1-ultra-high-end-headphone-amp.html#post4918630

I'll be curious what other replies you get!  They may be some other good stuff out their I haven't heard of.

[RGK]

Some additional ideas in this recent thread:

https://www.eevblog.com/forum/beginners/high-quality-sine-wave/msg1162949/#msg1162949

BK

[nfmax]

Some additional ideas in this recent thread:

https://www.eevblog.com/forum/beginners/high-quality-sine-wave/msg1162949/#msg1162949

BK

Just updated! https://www.eevblog.com/forum/beginners/high-quality-sine-wave/msg1171576/#msg1171576

[RGK]

I think I'm the other fellow evb149 mentions.

I'm actually going a few different routes just because

JanasCard PCB + notch filter PCB = $33
Frex PCB = $20
Victor built/tested = $52
agdr PS board = priceless

BK

[f5r5e5d]

also from the diyAudio thread - several <$200 "prosumer" soundcards can do better than -120 dB THD @ 1 KHz

"the trick" being using their optimum amplitude range -3 to -10 dB fs, which then requires long records/averaging to reach down into the noise floor of some of them

and if you want high resolution distortion measurement to -140 dB, a notch filter with gain lets you measure harmonics of the notch frequency with cheap usb PC soundcards or even better motherboard sound chipsets if you are willing to dodge noise, spurs, average long datasets


for under $1K there are pro DAC or ADC+DAC using ESS latest 9038 DAC that hit both distortion and noise numbers with room to spare - should be able to go even lower by resistively summing multiple channels