Super-Freq: Oscillators

[TrentO]

I seek opinions on using standard through-hole clock oscillators to generate 'test-patterns,' if you will, that I can pump through my oscilloscopes, and my eventual o-scope project...

COTS frequency generators appear to be freeking expensive for anything above 5Mhz or so. I do have my eye on the Rigol DS1022 20Mhz for approximately $700US, but I have a feeling I'm going to outgrow that limit fairly quickly. The 120Mhz unit is about $4000US-- I don't think I would use it enough to justify that kind of purchase. However, paging through my favorite Digi-Key catalog, I see that there are myriad crystal and/or oscillator offerings-- could I simply provide the requisite +5v / +3.3v to the OSC can and use the output for testing? I could simply buy 10-20 OSC's in xMhz steps to get to my 120Mhz requirement! Are there limitations in doing so? What kind of circuit (apart from power) would I have to build in order to support this?

And finally, if I wanted to amplify these signals (and change the amplitude displayed on my o-scopes,) could someone recommend a circuit that I could build to do so? I would assume that an op-amp or two would be sufficient....

As for my O-scope project, I'm looking to build a three-chip DSO out of a PIC32, AD9283 ADC, and FIFO-- for no other reason than to learn about microcontroller development, ADC's and FIFO's.! But step one, in my opinion, would be to have the capability to produce a signal of known quantities/qualities.

Any and all assistance, comments and opinions are sincerely appreciated!

-Trent

[GeekGirl]

However, paging through my favorite Digi-Key catalog, I see that there are myriad crystal and/or oscillator offerings-- could I simply provide the requisite +5v / +3.3v to the OSC can and use the output for testing? I could simply buy 10-20 OSC's in xMhz steps to get to my 120Mhz requirement! Are there limitations in doing so? What kind of circuit (apart from power) would I have to build in order to support this?

Well, lets treat this as a learning exercise, first step is to get the datasheet, there will be a link on the page , read it well, it will have some "typical" circuits this will do EXACTLY what you want, only thing is it will be a square wave only from the can. A simple external circuit will allow you to turn it into a sine or triangle wave

And finally, if I wanted to amplify these signals (and change the amplitude displayed on my o-scopes,) could someone recommend a circuit that I could build to do so? I would assume that an op-amp or two would be sufficient....

You would need to find an opamp that will give the required output voltage, and have a reasonable bandwidth. 

As for my O-scope project, I'm looking to build a three-chip DSO out of a PIC32, AD9283 ADC, and FIFO-- for no other reason than to learn about microcontroller development, ADC's and FIFO's.! But step one, in my opinion, would be to have the capability to produce a signal of known quantities/qualities.

Excellent idea , cheap graphic LCD (or use a computer for waveform display), a decent ADC with a PRECISION REFERENCE supply, do not make the mistake of thinking that the +5v out from the linear reg is good enough for the vref of an ADC.  Something like an LM4040 is a good choice.

If you brainstorm, draw up a schematic (if you post it here, please make it an image from screen cap or PDF, so everyone can view it we can give you pointers

[Mastro Gippo]

DIY oscope? Did I mention http://www.seeedstudio.com/depot/digital-storage-oscilloscope-diy-kit-with-panels-p-515.html ? 

[TrentO]

Mastro Gippo--

I have seen this before, and I would have tried it, had it been based on the PIC architecture... I decided about two months ago to take the PIC route and not the Atmel route, in (re-) learning microcontrollers. I'm not looking to build an Oscilloscope for anything practical-- just to learn more about the practice of A to D conversion and perhaps driving a graphical LCD display.

-Trent

[TrentO]

Andrew,

Awesome response. I think I'll take your suggestions on as a project by itself. Are there any books on HF circuit design that you would recommend? (NOT that I want to build a frequency generator from scratch!)

Thank you very much for your opinions on the DG-1022-- I never really considered the PC-driven aspect of such a device-- I didn't think I needed the arbitrary-waveform aspect... but now that you bring it up, its probably worth considering-- I might have a need for it down-the-road.

-Trent

[Mastro Gippo]

Mastro Gippo--

I have seen this before, and I would have tried it, had it been based on the PIC architecture... I decided about two months ago to take the PIC route and not the Atmel route, in (re-) learning microcontrollers. I'm not looking to build an Oscilloscope for anything practical-- just to learn more about the practice of A to D conversion and perhaps driving a graphical LCD display.

-Trent

I'm a pic guy too, but that thing is a nice way to learn something new! Also, it's a basic oscope for very little money! I think I'm going to get it just for the cool factor, I don't think I'll ever use it