Forums

How to zoom on waveform on oscilloscope ?

Started by janii June 22, 2014
I'm using Yokogawa DL1540L Scope to measure watchdog timeout.

CH2 = 5V/div,  CH4 = 5V/div,  50ms/div

Single shot, rising edge trigger set on CH2.  CH2 is gpio port.  In
software I set this port to logic low right before I service the watchdog. 
In scope I see a positive spike, which is actually logic high and then
after some short time, logic low.  I want clear display that zooms this
spike so it clearly show logic high and then logic low.  

I tried to use buttons on scope to zoom but no luck.  

Is there any way I can post my waveform so question can be answered.

If my time scale is set lower than 50ms/div, then I would see the pulse,
logic high for short duration and then logic low.  I can't see lower time
scale because then I won't catch watchdog reset on CH4.	   
					
---------------------------------------		
Posted through http://www.EmbeddedRelated.com
> How to zoom on waveform on oscilloscope ?
Thats highly 'scope specific (especially when you use anything else than an old-style CRT device), so I can only present you with generics. I get the feeling that the "zoom" you want is actually ment to be "horizontal zoom". And that should be easy as pie: just change the current 50ms/div to something lower. As the DL1540L seems to be a digital 'scope (assuming here its a memory scope) thats pretty-much all you need to do. Ofcourse, that could push the second (watchdog is serviced) signal change outof the screen (depending on where the current cursor is positioned it could also push the first signal change outof the screen). To scroll thru the whole, origional signal you should be able to use the "horizontal position" knob. If you want the 'scope to focus on (only store the time around) the the second, low-going edge (ignoring the first, the one you triggered on) you could see if your scope has got something like a trigger hold-off timer setting (it simply delays the moment of when your scope will actually store the signal). And a remark: When you say "spike" I hope you ment "edge". If not your 'scope might still be in AC mode. Regards, Rudy Wieser (using a Tectronics TDS 210 memory scope) -- Origional message: janii <85588@embeddedrelated> schreef in berichtnieuws LuednTWO07lZ7TrOnZ2dnUVZ_qydnZ2d@giganews.com...
> I'm using Yokogawa DL1540L Scope to measure watchdog timeout. > > CH2 = 5V/div, CH4 = 5V/div, 50ms/div > > Single shot, rising edge trigger set on CH2. CH2 is gpio port. In > software I set this port to logic low right before I service the watchdog. > In scope I see a positive spike, which is actually logic high and then > after some short time, logic low. I want clear display that zooms this > spike so it clearly show logic high and then logic low. > > I tried to use buttons on scope to zoom but no luck. > > Is there any way I can post my waveform so question can be answered. > > If my time scale is set lower than 50ms/div, then I would see the pulse, > logic high for short duration and then logic low. I can't see lower time > scale because then I won't catch watchdog reset on CH4. > > --------------------------------------- > Posted through http://www.EmbeddedRelated.com
On 23/06/14 01:11, janii wrote:
> I'm using Yokogawa DL1540L Scope to measure watchdog timeout. > > CH2 = 5V/div, CH4 = 5V/div, 50ms/div > > Single shot, rising edge trigger set on CH2. CH2 is gpio port. In > software I set this port to logic low right before I service the watchdog. > In scope I see a positive spike, which is actually logic high and then > after some short time, logic low. I want clear display that zooms this > spike so it clearly show logic high and then logic low. > > I tried to use buttons on scope to zoom but no luck. > > Is there any way I can post my waveform so question can be answered. > > If my time scale is set lower than 50ms/div, then I would see the pulse, > logic high for short duration and then logic low. I can't see lower time > scale because then I won't catch watchdog reset on CH4.
Details depend on your scope, but the general principle is - set trigger threshold to somewhere between logic low and logic high, e.g. for a 3.3V CMOS signal, 1.5V would be appropriate - set trigger direction to +ve to see a positive pulse - set holdoff or delay time to zero - start scope running: you should see the positive pulse at the left of the display - reduce horizontal timescale until you see what you want
On 23/06/2014 01:11, janii wrote:
> I'm using Yokogawa DL1540L Scope to measure watchdog timeout. > > CH2 = 5V/div, CH4 = 5V/div, 50ms/div > > Single shot, rising edge trigger set on CH2. CH2 is gpio port. In > software I set this port to logic low right before I service the watchdog. > In scope I see a positive spike, which is actually logic high and then > after some short time, logic low. I want clear display that zooms this > spike so it clearly show logic high and then logic low. > > I tried to use buttons on scope to zoom but no luck. > > Is there any way I can post my waveform so question can be answered. > > If my time scale is set lower than 50ms/div, then I would see the pulse, > logic high for short duration and then logic low. I can't see lower time > scale because then I won't catch watchdog reset on CH4. > > --------------------------------------- > Posted through http://www.EmbeddedRelated.com >
The DL1540 has a Zoom button! Press this and select MainZoom. You will have the trace as you see it at the moment at the top and then a zoomed version underneath. You can zoom in and out and move the zoom window lef and right to see what you need. You say the "spike" is short time, how short? Experiment with just triggering on this and zooming the timebase to measure it's width. It may be that it's so short with respect to the overall time you want to see that the 'scope can't take many samples of the "spike" so you may not be able to see a clear display. The only way to solve this realy is to get a 'scope with a higher sampling rate. Maybe you should consider a different way of trying to measure the watchdog timeout. By your description it sounds like you are servicing the watchdog once and then waiting for it to timeout and I assume the servicing of the wachdog is one of the first things you do in the code, hence the short spike. Instead of this wait some longer time time before servicing the watchdog so that the time the GPIO stays high is longer. You should be able to achieve this by disabling the watchdog initially and only enable and service it some time after program start. I must admit I don't get what the significance of this short pulse duration is if the reset is being monitored on another channel. Isn't the watchdog timeout the time from when the pin goes low to the time the reset is pulled?