Entry level signals generator

[ginetto61]

Hi i would start with some measurements on line level buffer and preamp
So the max Vout from the units under test would be 3-4 Volts ( i have two low noise sound cards to try on the receiving end just to start)
But i would like to buy a standalone signals generator
I wonder which would be the price for something decent (i.e. low THD)
I see some units but i cannot understand their real performance

https://www.amazon.it/YIWENG-Generatore-arbitrarie-campionamento-memorizzazione/dp/B0CP58XNVG/ref=sr_1_57?crid=30DKIZFY2IUG2&dib=eyJ2IjoiMSJ9.Yz-ADEB7520kupg1GfSit1pm0QbliAuqk1XafbbUKOAVN8-gqXbMerZ8MiUVA7NZ_Jc5Z3SugzYtCN-G1zyoGFiK89UDOuuZbfCZtGKxw52vLy2XYb_cv1gACfePQAdr5VyOngxqZvjZlwqx9M4Fdwi63VqLNLkEo2XrY5KhEBSCihnHdyLOkJ4RIUHfeozRND_ZNmIB6etexmZ1KWLf1BkmWmdCKnlkGzpkpa3lM9NjdNRzpkLuF4l6nWA1gWxG5UsedpskOdz3XrmZoSq16SfWOkLmAJVig0hHrg46l5Y.mAQ7ZClWtKKu-uUk5nHDhDpiJ8RIDQwsmC4nRccuElM&dib_tag=se&keywords=generatore+di+segnali&qid=1719653548&refinements=p_36:6800-&sprefix=generatore+di+se,aps,257&sr=8-57&th=1

https://www.amazon.it/Generatore-Contatore-Controllo-Sinequenza-Ac100-240V/dp/B0CSDZ3VT8/ref=sr_1_92?crid=30DKIZFY2IUG2&dib=eyJ2IjoiMSJ9.eb34Ik6wXn-CoZU6FrlBGS-X1U2l_ecDWcGBaMnreFP-Zc7B4KXSewv1zZWEpcEsYpogH4yM8VAS2VFWymXuRXdYbOTf1af1xGG3IAbJ-Aa9ueyPI_f2wlTjxDak4BnFdZIHcmlFTm-fhUnK1JY-i4X7nq5Am1C1kueIOGYMpv0VV4ZANDpjKPA5SsNG4a_WfOxCuszLq15QdG6-hz5phJ7ifggL_bvxHSiK7cf8miRyMsdRLKFsphoXrG5chshcouM5HnJaCmM8BJjdINyBH7XgZd1MeRZD4-VqjnT0VB4.rvEk96AKTC2mT27RApA0b1SMlyJJoz_xbGAPxgLrmGY&dib_tag=se&keywords=generatore+di+segnali&qid=1719653615&refinements=p_36:6800-&sprefix=generatore+di+se,aps,257&sr=8-92

Would these cheap units clearly a better option than a soundcard plus SW ? i am mostly interested in sine and square waves
Thank you very much to all

 

[Mooly]

My thoughts...

Any cheap signal generator that is essentially a function generator like these will not deliver low THD. The sine is artificially synthesised and distortion can be quite high.

What you really need (imo) is an old fashioned Wien Bridge type generator for low distortion audio work and also a Function Generator for general purpose work giving sine/square/triangle output. Audio generators have gone out of fashion though and I don't see any decent new ones.

This is a typical function generator:
https://cpc.farnell.com/gw-instek/sfg-1003/function-generator-1ch-dds-3mhz/dp/IN07959

but the distortion is quoted in dBc which I suspect is a way of covering up how bad it is (in relative terms). I found an online calculator to convert db to percent. dBc I assume is C weighting and I'll bet that is done to make results look even better than they otherwise are. This is not a bad example, its typical.

Data sheet for it:
https://www.farnell.com/datasheets/1691990.pdf

Over the audio range that generator is -55dBc which for -55dB is 0.18%

This type of thing is much better distortion wise (not that I'm recommending this although it may be fine)

https://www.ebay.co.uk/itm/17640832...Qu6t8ivfo0wLznDExtv642SSGzzP|tkp:BFBM0LDAtoxk

 

[ginetto61]

Good morning Sir ! thank you very much indeed for your always kind and precious advice
I will buy one immediately I like a lot also the rotary knob Very very nice
I am quite sure the THD from the generator will surpass the one from the circuit under test
But i need one signal generator standalone The price is very very good
Thanks again

 

[Mooly]

👍 hope you like it 🙂

I use one of these 😱

https://www.diyaudio.com/community/threads/belco-sig-gen-age-info-etc.126467/

And one of these:

The sine produced on the function generator is not totally pure as it is synthesised from the basic triangular waveform the generator produces by using diode shaping. Although it looks OK on a scope you can actually hear that its not quite pure when compared to a true sine, and if you look really carefully on a scope you can see a very tiny glitch 'like a join' at the top and bottom of the sine. Its microscopic but its there and you can hear it.

 

[bucks bunny]

With your soundcards and REW you already have a low THD and very stable tone generator.
No need to buy that old stuff.
Furthermore buy a decent used notebook like lenovo X220i3 for about 100€
that you can use exclusively as your audio measurement station.

 

[MarcelvdG]

dBc normally means decibel with respect to the carrier. It's a quite normal term for RF engineers. -55 dBc is about 0.18 %, as Mooly already wrote.

 

[Mooly]

Thanks Marcel 👍 I could only think of SPL weighting curves tbh.

 

[ginetto61]

With your soundcards and REW you already have a low THD and very stable tone generator.
No need to buy that old stuff.
Furthermore buy a decent used notebook like lenovo X220i3 for about 100€
that you can use exclusively as your audio measurement station.

Hi sorry for the belated reply Thanks a lot for the advice I tried an Audient sound card this one
https://www.thomann.de/it/audient_id14_mkii.htm
but when the SW performs a preliminary check it tells me that the input gain is too low even with the pot at max
I will take a screen shot when i will be back home I am visiting parents
I have already a refurbished Lenovo and it is not the problem The sound card is Its gain is not enough

 

[NickKUK]

I just have a little Siglent SDG1032X that’s been unlocked to 64MHz etc.

It’s not going be the perfect audio analyser for analysis but it:

• can do audio, harmonics and combined signal. Although it has a noise level and THD above my DIY ADC and will limit analysis
• can do 24MHz for clock signals
• can push 20Vpp output and sub 1Vpp
• has waveforms including duty time
• connects to the Siglent 1104X-E to give me bode plots etc
• can be accessed/controlled by ethernet or USB.
• small ish

In short very useful but for analysis you probably want a dedicated low THD oscillator.

 

[ginetto61]

Hi ! thanks I see it is a 350 euro unit I am just torn by the fact that the signal generator THD could be much higher than the one of the units under tests
Like measuring a tape measure with spans It does make little sense
It seems that very good generators are very expensive There are some kits but with limited functionality
like just 1kHz sine generators

 

[NickKUK]

Yes the signal is 14bit on the SDG and the THD/noise floor seems close to a 1998 CD player.. so as I've indicated, it's useful but not the low noise/THD source that some of the specialist/analogue sources are.

 

[tomchr]

The HP 3312A is a solid function generator. For low distortion I'd use a sound card.

Tom

 

[torgeirs]

An thinking out of the box solution is letting a ADAU 1452 and sigma studio being the signal generator
https://www.diyaudio.com/community/threads/low-cost-adau1452-china-board.309680/post-7714361
Spdif or i2s output
Or sheap board with caps removed and eksternal bal to unbal converter

 

[gerrittube]

An GW Instek GAG-810 gives good results (because it’s analog…).

I have one and it’s good for THD around 0.2%. I’m using an external DIY frequency counter to get exact frequency measurements.

Regards, Gerrit

 

[ginetto61]

good morning just to say that after thinking about it I have quite changed my mind 🙂
And that is, even a barely decent signal generator could still be absolutely usable
You just need to save the own distortion spectrum of the generator and compare it point by point with the distortion spectrum after passing through the device to be tested
If the device is completely transparent the difference should be zero It should add nothing to the generator own THD and noise
I don't know if it is technically easy to make this comparison though
In this case even a cheap digital would be a valid testing tool

 

[torgeirs]

If your scope can show a difference signal between the gain adjusted signal and the device its clearly visible

 

[adason]

Why not to use REW or RightMark audio as signal generators and thd measurements. Both offer spectrum analyzers and lots more. All you need is laptop and soundcard.

 

[ginetto61]

If your scope can show a difference signal between the gain adjusted signal and the device its clearly visible

Hi thanks a lot for the very valuable advice This situation makes the receiving device quality more important than the signal generator
The very good amp will add nothing to the original THD and Noise ... only some gain

 

[ginetto61]

Why not to use REW ...

hi thanks a lot I tried but i could not make it work with my soundcard A problem of gain not enough
I have to try it again but now i am far from home I will do it again and upload some screens with the messages from the SW
I would be very interested to know about usb sound cards already tested with REW or RightMark with great results
I am only interested in testing signals up to 3-4 Vrms
Power amps will follow in the future

 

[adason]

I would not see any issue with that signal level, most soundards have sensitivity volume adjustment. If you need to measure much higher signals, you simply use two resistor pad to reduce the signal. Two resistors do not affect your thd measurements, only attenuate signal.
There is whole thread dedicated how to measure with rew, i believe x has started it. I find RightMark so intuitive that no manual is required.