LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

How do I overlap waveforms?

I am capturing a number of uniformly spaced waveforms using a NI PCI-6269 board with the help of its analog triggering feature. I can then produce these waveforms on an excel spreadsheet. The specification I have now been given is to capture these waveforms in a way that they overlap one another. It was explained to me that any faults in the system I'm measuring would show up when the waveforms become offset.

Can anyone assist.

SimonC
0 Kudos
Message 1 of 11
(3,559 Views)
Can you explain what you mean by overlapping the waveforms?
Randall Pursley
0 Kudos
Message 2 of 11
(3,544 Views)
Randall,

At present I have a wave graph.vi, in Labview, and an excel spreadsheet graph that presents, for example 5 waveforms, one after the other. I have been asked to set Labview so that the 5 waveforms are presented in the excel spreadsheet graph on top of one another. If there is a fault then one of the waveforms would be offset from the other 4.
0 Kudos
Message 3 of 11
(3,535 Views)
this strikes me as more of an excel question if I understand you correctly. You want to take 5 waveforms that you collected and overlay them on one graph. Is this correct?

What I can see is the rub, is that they are not necessarily phase aligned. May that assumption is wrong as well because your trigger condition would always be the same.

If my assumption is correct, excel can do this, but it's a bit of a pain (my way is at least). I'm assuming that you have five xy arrays of data. I start by plotting the first one and then adding data series include the rest of them.

Sheldon
Technical geek, engineer, research scientist, biodegradable...
0 Kudos
Message 4 of 11
(3,528 Views)
Sheldon,

Thanks for your reply. I also thought I could manipulate excel but realised the pain and thought Labview could ease this pain for me.

Unfortunately I don't have five xy arrays of data. ( 5 was an example)I have one sensor that detects paper passing over it and 1000s of pieces of uniform paper will be passing over this sensor daily. I have a triggering condition that triggers on a rising slope. ( I have not quite mastered the Analog window yet but I'm working on it) When paper passes over the sensor Labview records the waveforms created and I can plot this on Excel. Due to the position of the sensor the paper will pass over it at a certain time and this is recorded as evenly time spaced waveforms. If the paper gets delayed in the mechanism then this time period will be altered.

I would like to find a way of being able to switch the triggering on and off automatically which will result in the waveform for each piece of paper appearing to overlap one another. If there is a fault in the mechanism or torn paper etc the waveform will appear offset.

In the longer term, the idea is to raise a flag in real time when an anomoly arises. However, that is not the immediate objective.

I am thinking of using some sort of timed pulses to switch the trigger on or off or look at software triggering but would the excel handle the switching on and off of the trigger?
0 Kudos
Message 5 of 11
(3,520 Views)
I think what you want to do is to record the time each time you receive your trigger signal and subtract the new time from the previous time to get the difference. This is all the info you really need to determine whether there has been a delay. The time is referenced in your waveform file by t0.

If you do need to plot the waveforms then take each waveform data set that is acquired and modify the t0 by subtracting the previous t0 from current t0. This will give you a relative time relationship between waveforms (i.e. if there is normally 1 second between sets then the new t0 should always be 1 second unless there is a delay then that waveform will start at a different time).
Randall Pursley
0 Kudos
Message 6 of 11
(3,508 Views)
I think I understand what you are trying to do a bit better now. Is this correct:

You have a sensor looking at pieces of paper zipping by under it. The sensor is being continuously sampled and the output data looks like a square wave of sorts, where the high level is the paper and the low level is absence of paper?

It appears that you want to trigger sampling at the start of a sheet of paper each time, so rather than have a long pulse train you want individual "records" where each record is a single sheet.

I see two methods to do this, the first is to use the analog trigger and trigger on a DC level. This will start sampling at the beginning of each sheet. You can set the sample length to be longer than you'll want and ignore the end of the data set. The second method is to continuously sample like I'm assuming above, and you will "find" the sheets in software from the pulse train. This finding process is easier than it may seem assuming that there are clear transitions and reasonably defined paper and no-paper states.

Let us know if I'm on track and we can work from there.

Sheldon
Technical geek, engineer, research scientist, biodegradable...
0 Kudos
Message 7 of 11
(3,506 Views)
Randall/Sheldon,

Thanks for ideas, it going to take me a couple of days to try out both ideas, i'll get back to you once I've given them a go.

Cheers,

SimonC
0 Kudos
Message 8 of 11
(3,489 Views)

Hi, did you resolve this? I would like to create something very similar using Labview. I would be interested to know about overlapping of waveforms, but I would like to use an SQL server to store all the data and labview to pull it out and analyse it. Thanks 

0 Kudos
Message 9 of 11
(2,501 Views)

Don't hold your breath for an answer as SimonC hasn't logged into the forums since that last message over 14 years ago.

0 Kudos
Message 10 of 11
(2,489 Views)