Jump to content

Pocket oscilloscope as waveform monitor


Recommended Posts

  • Premium Member

Hi,

 

A while ago, I posited the idea that one of those little pocket oscilloscopes could be used as a waveform monitor. So, I bought one, since they're not too expensive and I wanted one for other reasons (truly, I am a geek).

 

post-29-0-45824300-1339416891.jpg

 

 

Cute, eh?

 

Anyway, as a WFM, the short answer is: sort of, not really.

 

Below is an image captured using the scope's inbuilt screen grab feature. At this point, it's reading the Y output of a Blackmagic card that's displaying a 1080p25 signal with some picture on it.

 

post-29-0-84125300-1339416957.png

 

 

As we can see, the sync pulse and one line's worth of luminance information is clearly visible, and it does work to some extent as a single-line waveform monitor. What it doesn't do is to integrate all the lines of the image together, in the way an old-style CRT scope (and old style waveform monitors) do. The rate at which it displays each line's worth of information is limited by the refresh rate on its little LCD and you therefore don't get the integration effect that's intrinsic to CRT scopes.

 

So, you can't really use it as a waveform monitor. There's an outside chance I may be able to modify the software to make that possible, but I'll have to look into it.

 

P

Link to comment
Share on other sites

Hi,

 

A while ago, I posited the idea that one of those little pocket oscilloscopes could be used as a waveform monitor. So, I bought one, since they're not too expensive and I wanted one for other reasons (truly, I am a geek).

 

post-29-0-45824300-1339416891.jpg

 

 

Cute, eh?

 

Anyway, as a WFM, the short answer is: sort of, not really.

 

Below is an image captured using the scope's inbuilt screen grab feature. At this point, it's reading the Y output of a Blackmagic card that's displaying a 1080p25 signal with some picture on it.

 

post-29-0-84125300-1339416957.png

 

 

As we can see, the sync pulse and one line's worth of luminance information is clearly visible, and it does work to some extent as a single-line waveform monitor. What it doesn't do is to integrate all the lines of the image together, in the way an old-style CRT scope (and old style waveform monitors) do. The rate at which it displays each line's worth of information is limited by the refresh rate on its little LCD and you therefore don't get the integration effect that's intrinsic to CRT scopes.

 

So, you can't really use it as a waveform monitor. There's an outside chance I may be able to modify the software to make that possible, but I'll have to look into it.

 

P

To be honest though at least it gives you a rough approximation of exposure, which is better than trusting a monitor with a 100x10^200 contrast ratio where even the sun looks correctly exposed.

Link to comment
Share on other sites

  • Premium Member

Well, sort of. The problem is that it doesn't settle on any one particular line.

 

In the image, the trigger threshold is the dotted white line marked T toward the bottom of the display, which is triggering on the positive-going edge of the negative part of the line sync, which is therefore aligned with the vertical orange line. The scope will then take 4096 samples (the size of its sample memory), display those samples, then immediately trigger again whenever it sees another similar edge. This means you have no control over what line you're seeing, which rolls through the image as the sampling period of the scope and the line syncs go through various phase relationships. The result of this is that the line you see there ripples and wobbles about as you get various lines displayed.

 

This does, sort of, give you information about the exposure of the whole frame, if you watch it for a couple of seconds, but it's a bit of a stretch to really call it a waveform monitor, and it's difficult to associate exposure peaks and troughs with the parts of the image that are causing them.

 

I have done some research on the hardware and software of the scope and there is a little scripting engine available for it which makes new software development quite easy. It would, I believe, be possible to create an application that would operate as a single-line waveform monitor with selectable per-line readout. It would also be possible to create a histogram display, and show other characteristics of the signal. The problem with creating a true per-frame waveform display is that it requires creating a composite of all the lines in the frame overlaid and summed together for intensity, which is a fairly complex operation on a bitmap image that the hardware in the scope is probably not up to doing.

 

Serious nerd stuff:

 

The hardware is not unlike a very primitive cellphone with some extra bits and pieces. It has a little ARM Cortex M3 as its main processor, plus a SiliconBlue ICE65L FPGA to handle driving graphics to the display and some of the mathematical heavy lifting. The processor alone is probably capable of producing the required display, especially if you could tolerate a 10-15fps update and not capturing every single line. The FPGA is well capable of capturing and compiling every line at a resolution that would drive the display to its full horizontal resolution of 400 pixels, although you'd have to write some moderately complex code in a low-level language to make the FPGA do it.

 

The problem in either case is the paucity of available RAM. There is only 36kB of RAM on the device that's available for applications and this is not enough to build a buffer for the 240x400 TFT, even in 16 shades of grey or something like that which would probably be usable. The only memory in the device that's capable of storing the entire display is actually the RAM on the TFT module, and while it's possible to retrieve values from that RAM, it's very slow to do so and this would probably clobber the performance unacceptably.

 

You could possibly do it if you were willing to tolerate a very low resolution display such as 100x200, which might still be marginally useful as a rough idea of relative exposure. Probably some monitors in "overlay" mode, when they give you the scopes in a border under the image, don't do much better than this.

 

You could almost certainly do a max/min display, where there would be two lines, one at the maximum luminance and one at the minimum.

 

There are other possible workarounds, but it's not looking good for a traditional WFM.

 

P

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...