Signal generator recommendation

[BYacey]

Most function generators don't have a very pure sine, there's usually distortion at the peaks because it is derived from a triangle wave shape. I have a Wavetek 180 that works nicely for quick tests and sweeping circuits, but a pure sine is not to be had.

My other generator for more demanding measurements is a HP 201B, in fact here's one on Ebay cheap - http://www.ebay.com/itm/HP-Audio-Oscillator-Model-201C-/230884065977?pt=LH_DefaultDomain_0&hash=item35c1c342b9

 

[Brian Roth]

All that I can say is that a function generator is quite useful.  One of my 'specialties' is working with analog tape machines and desks.

With the tape machines, the ultra-fast settling time and flatness for levels, and a decent square wave output for desk go/no-go greatly reduces my time at a client's studio.

btw..try pushing a 1 kHz square wave into many A/D D/A converter combos and looking at a 'scope at the output....

A decent desk can handle that...an analog recorder much less so.

so...we all need to 'pony up' for $10K or $20K for an Audio Precision!

Bri

 

[buildafriend]

All that I can say is that a function generator is quite useful.  One of my 'specialties' is working with analog tape machines and desks.

With the tape machines, the ultra-fast settling time and flatness for levels, and a decent square wave output for desk go/no-go greatly reduces my time at a client's studio.

btw..try pushing a 1 kHz square wave into many A/D D/A converter combos and looking at a 'scope at the output....

A decent desk can handle that...an analog recorder much less so.

so...we all need to 'pony up' for $10K or $20K for an Audio Precision!

Bri

I would love to run that test.. square wave into and out of my converter with a scope.. I bet my tools suck way too much to do that accurately. It must be a blessing to have such great tools.

 

[JohnRoberts]

Google "Gibb's phenomenon" before reading too much into square wave tests through digital paths. Square waves often contain above band content, and how that gets removed can lead to interesting looking waveforms.

Square waves through analog paths can look bad in more mundane ways...

JR

 

[buildafriend]

Thanks JR!! I'll read up!!

 

[buildafriend]

ahh this speaks of fourier series! I have done a few of these calculations.

Things will never be perfect but we can get them to where they are "good enough".. dare i say.

you're pointing out the "somewhat spike looking part" at the end of each sharp slope between the up cycle and down cycle??

My professor made us learn this on a graphing calculator so that we could see that each square wave has less and less oscillations within it's self.

 

[bruno2000]

These are very nice signal generators.  This one is a little expensive but "make offer"

http://www.ebay.com/itm/HEATHKIT-SG-18A-SINE-SQUARE-WAVE-GENERATOR-LOOKS-GOOD-TESTED-/281021082540?pt=LH_DefaultDomain_0&hash=item416e296bac

disclaimer:  I have NOTHING to do with this auction.

Best,
Bruno2000