Checking Out Digilent's Analog Discovery - Workbench Wednesdays 76

Okay, I had a few minutes to check out the latest beta. This is a very exciting update for the AD3!

I have an AD3 and an R&S MXO scope in the picture. The MXO's waveform generator goes into the AD3's Scope channel 1. And then the AD3's Wavegen goes to the MXO's Channel 1. (I'm only using the MXO to make it easier to keep straight where the ins and outs are going!)

In Waveforms, I set up an FIR filter of Channel 1 and then output it on the Wavegen.

So, the MXO generates a 5 MHz square wave, and the Analog Discovery 3 outputs a digitally filtered copy of that signal!

It is cool to see that working, and it is also nice to see you added FIR Frequency options for the AD3.I had the impression you couldn't do frequency filters with an FIR. To me, that makes it seem less critical to add IIR However, I am not developing digital filters, so many someone doing that kind of work would like to be able to mess with the taps in real-time?

tldr; A feature I really wanted to see in the AD3 is now available (and works as expected!)

So now the wavegen outputs the FIR from the scope?

That would be great news. I didn't get into in the video, but that was one of my largest disappointments.

It is the only thing I could think that would make the FIR even remotely interesting since you could change the coefficients in near-real time with a real signal.

Regarding a dedicated configuration with FIR, I think it would depend on what trade-offs were involved. But having FIRs available that run in real-time is a great option. I use it on my bench scope all the time (specifically low-pass).

Hello,

The Wavegen output of Scope Filter input is fixed in the latest version:
https://forum.digilent.com/topic/8908-waveforms-beta-download/

I got notified about this issue yesterday.
The newer version also brings several new features.

The ADP3X50 implements both FIR and IIR.
The AD3 having smaller FPGA only contains FIR. If you think it is useful, I could try to create dedicated configuration with IIR.

Convenience of being able to disconnect the tool from the breadboard really. If you have an IDC to separate Dupont pins then it's either up to 30 pins to connect and disconnect or you have a IDC to Dupont pin breakout harness dangling from each breadboard. 

I've used the Adafruit T-Cobbler breadboard breakout with the R-Pi before, so already familiar with that sort of set-up.  

You can always expand the breadboard if you run out of space. (A number of the Analog Devices tutorial circuits are built on a half-sized breadboard anyway so using a full breadboard with the mini breakout PCB leaves room to spare.)   

There is a 'Studio' version of the Analog Discovery that takes a double breadboard module (canvas), but gets expensive if you want to work with different circuits.

https://digilent.com/shop/analog-discovery-studio-a-portable-circuits-laboratory-for-every-student/

https://digilent.com/shop/breadboard-canvas/

Aimed at classes of students I guess where each student gets their own canvas to build their circuits on and then it can be dropped into the Studio unit once ready.

The ADALM2000 requires you to disconnect the circuit under test at start up for calibration purposes, so a quick and reliable means of disconnection and reconnection from the breadboard is useful.

I imagine the breadboard adapter is a result of the designer focusing more on how to make the PCB layout easier and cleaner than how it would get used. But I could see how some confusion could happen.

I wonder how useful the (smaller) adapter really is though. That would burn up a lot of breadboard space and then you would still need to run jumper wires. But I guess you could build a breadboard circuit and quickly plug the adapter into it.

I was about to make a T-Cobbler type breadboard adapter for the ADALM2000 when I noticed the Analog Discovery 2 one had one with the same pin allocation.

However it is a bit annoying that they have decided to reverse the pinout from the device itself.

If you are connecting to the device header directly then the positive pins, W1, and Digital 0-7 are all on the top row, whereas on the breadboard breakout for some reason they end up on the bottom row. 

Given the silk screen text, then perhaps they use their breadboard vertically.

So I might go ahead and make one anyway as have been meaning to give KiCAD a try. I would have gone for thru-headers and added some additional headers to line up with the breadboard top and bottom power rails.

The other protoboard/breadboard adapter matches the device pinout by keeping the positive pins, W1, and Digital 0-7 on the top row of the header. 

A small breadboard is supplied with this one if you don't intend to use it as proto board. 

Good catch. I am not aware of any intentional collaboration. But, I could definitely see how ADI might want to utilize the pseudo-standard. 

Looks like they even copied the color code! 

I noticed that the pinout is the same as that of the Analog Devices ADALM2000 module. The BNC breakout module is also very similar. 

Is there some collaboration going on here in an attempt to standardise educational lab kit ? 

Nice overview of the device.

I will keep this unit in mind if I need those capabilities.

Yup. Totally agree!

Students having the Analog Discovery (2 or 3) make me super jealous. I would have loved to have such a device when I was in school!