Raspberry Pi Waveforms

Post your experiences of using the Raspberry Pi with the DrDAQ
Post Reply
Site Admin
Site Admin
Posts: 12
Joined: Tue Feb 08, 2011 4:04 pm

Raspberry Pi Waveforms

Post by Pete » Wed May 30, 2012 3:00 pm

So, a few people in the office have started to receive their Raspberry Pi's. Normally when we see new gadgets around the office we take it apart but there's no case on the Pi so we thought we'd 'scope' it out instead.

I reached for a 6403 and started to probe around, there's some pretty interesting stuff that you wouldn't normally think to look at so I grabbed some;

Ethernet, connected at 100Mb
Ethernet Capture on the Raspberry Pi
Persistence mode can show some really interesting stuff, I figured this would give a nice capture but had to think a little bit about why there's so little happening around the trigger. This is down to 100BaseT physical encoding which is done in MLT-3 (an interesting read) which means with a simple trigger you would never observe a negative 'lobe' in the period after the trigger and never a positive 'lobe' before. Also something that I'd forgotten is that 100Meg Ethernet is transmitted at 125Mhz! (I'll leave that for you guys to discover)

We had Debian loaded which and we were controlling with a USB mouse and keyboard, so figured that was a good next stop. Since I had the 6403 connected it seemed a good idea to just use that, althoughI think any of our USB scopes would easily capture these (forgot how slow USB1.0 was!) but it is nice to have deep memory.
USB Capture on the Pi
When a USB device is connected you get start of frame pulses, which are the regular pulses you get every mS, then you can see that when data needs transmitting it's inserted. in reality with our advanced triggers you could capture 'data-only' frames but when you have such large memory you can be a little lazy because you can zoom in so much. I decided to capture both D+ and D- so I can show that USB is actually semi-differential?!......
this is just zoomed in on the previous waveform, I'll upload the file below, this only uses 12.5Msample of the memory and we go upto 1Gig!

Thought I'd look for other cool persistence waves and saw the composite video.......
I like the way you can see the color burst as the most common and the video data after, someone should fill me in on the other two bits of information, I assume there's a HSync or something? (don't normally think about video stuff)

At this point I thought I'd change scope (oscilloscope), I figured a 4262 would be cool to measure the sound output given it's 16Bit resolution. On the Raspberry Pi there's PWM outputs which have an RC arrangement before the output, thought it would be nice to see both sides of the filter. Unfortunately, we couldn't get the sound going which was massively disappointing, so expect an update here when we've poked it into life.

Since I had the 4262 plugged in I thought I'd capture one more simple wave before I went home. This is a trick I often do to measure currents, since there are normally '0' ohm links or low resistance inductors inline with most power supplies and the 4262 has nice small ranges it means you can measure currents really easily, I just probed across F3 on the R-Pi board, I didn't measure the resistance but you can do and then make a maths channel to turn this into current really easily.

Maybe we should get out a 2205-MSO and get down with some SD card transfers?!

Site Admin
Site Admin
Posts: 184
Joined: Thu May 25, 2006 8:02 am

Re: Raspberry Pi Waveforms

Post by alan » Tue Jul 03, 2012 9:17 pm

Over at code and life they have been benchmarking the Raspberry Pi GPIO Speed as a way of testing how efficient different languages / compilers are - a simple task for a PicoScope 2000.
http://codeandlife.com/2012/07/03/bench ... pio-speed/

Post Reply