how arbitray waveforms are generated

[rstofer]

It's common to think about repetitive waveforms but that doesn't have to be the case for an 'arbitrary' waveform.  The waveform will repeat but the experiment may be over before it does.
In the case of the Siglent SDG2082X, the device supports sample memory from 8 points up to 8 megapoints and sample rates from 1 uSa/s up to 75MSa/s in what they call TrueArb mode.  In DDS mode the sample rate is 300 MSa/s.  And then they claim a sample rate of 1.2GSa/s using 4x interpolation.
TrueArb mode is point by point, every output voltage comes from sample memory.

DDS will generate most common waveforms and most AWGs have a large menu of built-in waveforms.  TrueArb is different in that there are discrete samples for every time tick.

There is some very useful iinformation in the SDG2000 datasheet:
https://mediacdn.eu/mage/media/downloads/SDG2000X_DataSheet.pdf

About page 6, they talk about the EasyWave software which allows for manual drawing, line-drawing, equation-drawing, coordinate-drawing and so forth.

User Manual:
https://mediacdn.eu/mage/media/downloads/SDG2000X_UserManual.pdf

Starting on page 40, there is a list of built-in functions extending for 4 pages.  There are a LOT of built-ins.

[nfmax]

AWGs produce periodic signals, at least the few I've played with did. The shape of a cycle is arbitrary but the cycle plays repeatedly as the selected frequency. I don't actually know what people typically use them for, other than playing around I never found a need for anything other than the usual sine, square and triangle waves.

Not necessarily. You can run in 'triggered burst' mode where an external pulse causes the waveform to play, either just once or for a programmed number of repeats. Recently, I have been using this to play shaped tone bursts, designed using Matlab to have a specified bandwidth, duration, and autocorrelation function, into an ultrasonic transducer. This is what ARBs really shine at.

The waveforms are only a few hundred points long. I'm using an Agilent 33522B which lets you set the waveform sample rate to whatever you want (50 Msps in my case) while interpolating up the actual, fixed DAC sample rate of 250 Msps using a digital filter in the internal FPGA.

[rstofer]

At some point, generating the waveform is like a side issue to all of the surrounding features.  A true AWG can do a lot of really neat things.
On the low end, I built a Halloween animation that was simply a skeleton rocking in a chair while snoring.  The sound effect was derived from a .wav file that I ripped from somewhere and played by a PIC with an R-2R DAC.  It worked really well.  It's been a long time, I don't remember how I got from .wav to discrete samples but it was easy, whatever is was.

[james_s]

A .wav file is nothing more than discrete samples with a header so it's not hard to do.

[joeqsmith]

Fun times.   I constructed an engine simulator and needed to have different waveforms for the various sensors.  I wanted to be able to control the frequency as well as AM and FM modulate so I used an FPGA to do everything.

I like analog computing and simulating the suspension of a vehicle (at least one front tire) is a classic problem since it is essentially a mass-spring-damper problem and those are well understood.  But what if the ground under the tire contains a chuck-hole?  I have toyed with the idea of using an AWG to generate the surface.  I could do this quite easily with the Analog Discovery 2.  I could even model those nefarious speed bumps!

Other than sheer laziness, I don't know why I haven't done it yet.

AWGs could be a lot of fun to play with but I'm just guessing that the frequency is going to be a whole lot less than 400 MHz.  For my suspension project, even the 12 MHz bandwidth of the Analog Discover 2 is FAR more than required.  Nothing on the chassis is working in microsecond time scales.  If it does, I guess we move to Euler's Method.

I need sub 1 MHz for the simulator.   I will collect data off the real setup and feed that into the simulator.   It was a fun and useful project.  Video showing the basics.   
https://youtu.be/q_89qoFMivg?list=PLZSS2ajxhiQBvWvqMVLdRQMjGofKpQUJr&t=2202 

It's common to think about repetitive waveforms but that doesn't have to be the case for an 'arbitrary' waveform.  The waveform will repeat but the experiment may be over before it does.

True.  Running a test right now doing just that.   

[rstofer]

A .wav file is nothing more than discrete samples with a header so it's not hard to do.

Something along the lines of 'dump the header' and 'keep the samples'.  It's been about 15 years and I'm getting old...

[ant17]

hi guys thanks so much for all input every body has contributed. i feel like i have  been flooded with information i am really going to have to take my time in reading all carefully so thanks so much to all who have contributed to this discussion thanks after i have read it and understood it i will post any questions i have thanks again

[rstofer]

I need sub 1 MHz for the simulator.   I will collect data off the real setup and feed that into the simulator.   It was a fun and useful project.  Video showing the basics.   
https://youtu.be/q_89qoFMivg?list=PLZSS2ajxhiQBvWvqMVLdRQMjGofKpQUJr&t=2202 

Interesting project and very well done!

[ant17]

hi guys thanks for all your input cany anyone suggest fpga development board to get experimenting i was thinking a zynq 7000 series fpga development board or other suggested developments with most room to experiment with arbitrary waves

[james_s]

It doesn't really matter, any FPGA will get you started, you can go a long way before you start pushing the performance envelope.

How much experience do you have with FPGA development? To say the learning curve is steep would be an understatement.

[ant17]

i have had experience with with a lattice semiconuductor PLD development board it was just a programmable logic device.programmed with hardware description language made at a basic electronics school, about 20 years ago not a fpga so i am a novice.

[james_s]

Well that's more than most. Either way first step is start learning either VHDL or Verilog, the FPGA choice isn't too important until you have a working prototype.

[rstofer]

You might check in over at NandLand.com  The tutorials are excellent and he offers a reasonably priced FPGA board for his projects.
Notice that there are tutorials in both VHDL and Verilog.

[ant17]

thank you