Signal Generator Advice Needed

[abpea]

I'm looking for a signal generator to use with my scope to help me troubleshoot amps and other audio gear. Would this signal generator be a good one to use for audio purposes:

Ebay Signal Generator

There are a ton of these for sale on Ebay but I'm not sure if it's what I need or not.

Thanks for your help -
Bruce

 

[AndrewT]

Hi,
that would do for much of our amateur audio testing.
The <2% sinewave distortion is a little higher than some claim, but most do not get much lower than 1%.
Check for output and load impedance. It can vary from 50r to 2k.
Probably the lower impedances allow more flexibility. But you must load the output correctly.

I am toying with the idea of building a low distortion fixed frequency sinewave generator (<0.01% or <0.001%??) for distortion testing, but don't yet know if that is sensible.

If you can find an affordable one with some kind of input to gate the signal and for AM/FM modulation and/or sweeping, then even better, once you learn how to use these extra facilities, (I don't).

And that bench DMM 20,000 count looks pretty good for $99.
Mine is only slightly better and it cost me double that second hand.

 

[d3imlay]

The item on ebay is a function generator. It likely starts with a square wave and has a sine shaping circuit. hence the high distortion.

Linear Technology had an app note in the 80's on designing a low distortion generator. It's an excellent article and even talks about the original designers concepts. The designers names were Hewlett & Packard and this was there first product. I built up and updated version from the article that uses 1 opamp, a light bulb and a couple resistors/caps. Works great.

If you want something already built, get an HP 204 off of ebay, usually dirt cheap. The 200 is the original HP tubed oscillator.

 

[jackinnj]

The item on ebay is a function generator. It likely starts with a square wave and has a sine shaping circuit. hence the high distortion.

Linear Technology had an app note in the 80's on designing a low distortion generator. It's an excellent article and even talks about the original designers concepts. The designers names were Hewlett & Packard and this was there first product. I built up and updated version from the article that uses 1 opamp, a light bulb and a couple resistors/caps. Works great.

If you want something already built, get an HP 204 off of ebay, usually dirt cheap. The 200 is the original HP tubed oscillator.

I believe that the Linear Tech application note 43 "Bridge Circuits" is the item to which you refer.

You can get a Heath IG18 on EBay for around $20. There is meter amplifier circuit on the web which eliminates the ugly rectifier circuit which Heath used -- with this in place the device will yield THD in the low 0.01%'s. If you better the power supply you can knock this down a bit.

Bob Cordell has the article which he wrote in 1980 on his website -- if you want to roll your own state variable oscillator he describes will do as well as the Tektronix SG505 -- it's pretty much the same circuit.

Lastly, don't knock the square wave oscillator -- you can use the Linear Technology or Maxim switched capacitor low pass filters to get performance in the 0.001%'s Here's a PDF from Maxim:

http://pdfserv.maxim-ic.com/en/an/AN2081.pdf

 

[Speedskater]

The Bob Cordell and the Linear Technology circuits are both great designs, but both are very difficult to DIY as some of the parts and switches are hard to source. What I would use for troubleshooting is a battery powered "CD" player. You can download or create any test tone you need.

 

[Conrad Hoffman]

That's really basic, and since they included the readouts, I have to wonder how much quality went into the actual circuitry. OTOH, the price is reasonable. I had a similar Fordham or something, but it also had sweep capabilities. Sold it and now use a used Wavetek. There are many used Waveteks for not much more money, that have more features, wider bandwidth, and better performance. The portable CD player idea also has a lot of merit, as you can use the PC to record any waveforms you want, including arbitrary waveforms. The downside is you're limited to 20khz, and the associated rise and fall times- pretty slow for circuit work.