SCOPE Soft Front Panel

[DFGray]

To change the number of points acquired, select Edit->Device Configuration.  Change Record Length on the dialog which comes up.

I finally understand what you want to do with the soft front panel.  Unfortunately, it does not support that feature.  You want an "auto" trigger mode.  In "auto" trigger mode, the oscilloscope triggers if there is something to trigger from.  If no trigger occurs after a short timeout, the scope triggers anyway.  This triggering mode is not native to any of the current NI oscilloscopes, but could be implemented in software.  That has not been done in the NI-SCOPE Soft Front Panel (I will call it SFP from now on).  However, you can probably get the information you want, just not the way you are used to doing it (my apologies).

There are two main controls which set how the triggering on the SFP works - Type and Mode.  Let's start with Mode.  Mode determines if the SFP triggers continually or only once.  In "Norm" mode, the SFP draws another trace every time it gets a new trigger.  In "SGL" mode, the SFP draws a trace the first time it get a trigger, then waits until the user hits the go or pause button to start waiting for another trigger.  Type determines what the SFP looks for in the signal to trigger.  There are two categories (but just one list) - Immed. and everything else.  With Immed. (immediate) triggering, the scope device does not look at the signal at all, it just takes a set of data immediately.  The other trigger types, Edge, Hyster., etc. all look for a particular pattern in the waveform (e.g. rising edge) and take a waveform when they detect that pattern.

So, it appears that you have intermittent data.  I would leave triggering on Hyster. and set the record length long enough so I could see that the record is indeed going to zero.  I will freely admit that this is a poor second to a real "auto" mode, but it is the best the SFP can do.  Alternately, you can use Immed., but you will probably miss your data most of the time, since the scope is not taking continuous data, but snapshots.  If you want truly continuous data, you will need to write your own code.

[wang]

Hi DFGRAY:

Thank you very much for your advice. The following is what I did following your suggestions. Would you please take a look:

1) I tried to set Hyster mode, level 2, hystersis 2.2-2.6, rising slope adn set record length to 100000,1000000,10000000,16776960.

    After I set my record length to 100000. It did not work. It was a zero volt. According to your previous mail, you said the record is set to

    long enough so I could see that the record is indeed going to zero. I don't understand why I need to set a very high record length becasue

    the definition of record length request the minimun no of point the digitalizer acqurire to the specified channel. If the record length is

    set to very high, that means the digitalizer acquire the minimum number of point to the specified channel increases. Why did you say

    waveform is indeed goint to zero? Am I doing anything wroing with it?

2) Then I fellow your instruction in your previous mail to try setting the time/division. or increasing the number of samples. I could not

     find how to set the number of samples. Could you tell where I can set the number of sampes.

3)  I tried to adjust the time/division setting. I started to use 10 microsecond/div and then increased the value. I use the default value of

     20 MHz, The result did not work. It did not indeed going to zero. It did not act like an "Auto mode" of the oscilloscope.

     After setting into 10ms/division, the waveform will look like the diagram of what I attached. 

4) Is there anymore setting I did not do? Or you can conclude that due to no "Auto Mode" on SFP, that is why I can not get

     a wavform like what I attached, except I write a program for SFP.

5) When you say writing a program, do you mean writing a program for SFP? I want to be sure because I need to do my job

6) If I don't write a program, the result is the second diagram of my attached file, right? I need to make sure.

7) Does SPF has peak to peak acquistion fuction like oscilloscope?

😎 I know the bandwidth can think of the internet bandwidth. The more you have, the more data you can carry. How about the

    20Mhz bandwidth and 100MHz bandwidth here. What is the difference and when should I use 100MHz bandwidth.

[DFGray]

Take the questions one by one...

1. The reason you want to increase the record length is to make sure you are getting your entire signal. If the record length is too short, you will only get a portion of the signal. If you only get a portion of the signal, you cannot tell if it is returning to zero or not. If you are getting your entire signal pulse, there is no reason to increase the record length any further.
2. A sample is one point in a record. Therefore the number of samples is the same as the record length. You have successfully done this.
3. The SFP does not have an AUTO mode. You cannot make it act like one. As mentioned in my previous e-mail, you can get the information you need, just not as easily.
4. You can easily get the waveform like the data you attached if you use SGL mode and an appropriate trigger level. It just will not be a continuous display. You will only get a trace when there is a signal. You will not get a straight line at zero because there is no trigger associated with this. Make sure you set the record length and reference position correctly to get your entire waveform.
5. When I say write a program, I mean do it with LabVIEW, CVI, Visual C/C++ or some other environment which supports NI-SCOPE. If you are in a hurry, use the one you are most familiar with. The SFP does not have a programmable interface. It was written in LabVIEW using NI-SCOPE for the data acquisition and measurement functions.
6. That is correct. You know that the SFP can export data in a variety of formats. See the help file for details.
7. I am unsure what you mean by "peak to peak" acquisition function. At first guess, I would say no. You can use the measurements to find the value of the maximum in your waveform. You can also find the period (time) between two pulses, if your acquisition is long enough to have two pulses.
8. All analog devices have a frequency response curve. With oscilloscopes, the bandwidth usually refers to the point at which input signals are attenuated 3dB from their actual values. The native bandwidth of the 5112 is 100MHz, so 100MHz signals will be attenuated by about 50%, 200MHz signals by a lot more, 50MHz signals almost no attenuation. With 20MHz bandwidth, this -3dB point is moved to 20MHz. The 20MHz bandwidth is sometimes referred to as a noise filter. Use it if you have no frequency components in your signal over 20MHz and you want to reduce noise.

[wang]

Dear DFGray:

Thanks!

Is it very hard to write a prgram in LabVIEW for "Auto trigger mode"?

It seems it is very easy to see the new trigger when coming. Am I right?

I am thinking why NI not making this "Auto mode feature". Is it becasue it is not very useful? Could you answer to me pls. Thanks

What is the difference between AC and DC couple in SFP? Is AC for measuring analog signal and DC couple for measuring digital and

analog waveform? Are they the same under channel function and trigger type fuction?

In order to verify my signal use 20Mhz bandwidth, the best way is to look at the ampilute of the voltage to see if it is  correct? Am I correct?

I need to be sure.

[wang]

Dear DFGrary:

I am sorry, I forgot to ask you the "Peak to Peak" which is refer to the acquision button on the o-scope. After you press the

acquision button, you can select "Sample" mode or "Peak to Peak" function. I read the manual of the o-scope about "Peak to

Peak". Peak to Peak stand for tracing a thin waveform rather than "Sample" mode. It is more precise.

The reason I ask is because I need to acquire the waveform like the file a attached. If I don't use Peak to Peak on the oscilloscipe,

I could not get a nice waveform like the file attached. After I use, peak to peak, I can get it.

Does SPF set Peak to Peak?

Also, what is the meaning of hard ground of NI 5112? Does it means it will drag current from other ground to hard ground? Why use diff?

[DFGray]

Answers for first post

From what you have said, it sounds like the scope is either finding peaks using a subsampling technique, or, more likely, sampling at its maximum frequency and giving you the maximum value in a given time bin.  You can get the same effect by increasing your acquisition rate and record length, keeping your acquisition time constant.  You will have a lot more points to deal with, but the 5112 can handle somewhere between 4MS to 32MS, depending on the version you have.  The SFP has no trouble with these numbers, although it will slow down as you acquire more data (it has to decimate it for display).  Don't forget that you can save your data.  Don't use the ASCII format (.LVM) for a large data size.  Performance will be horrible.  Use one of the binary types.

Most oscilloscopes have a hard ground (the outer shell on the BNC connector).  This ground will force almost anything to zero volts, although a large current signal connected to it will probably damage the scope.  This is the only real ground on most scopes.  Some scopes have a ground input.  This connects the input through a low value resistor to ground.  It is used for calibration.  Differential mode is used when you must measure a signal floating on an unknown level, such as the voltage across a random resistor in a circuit.  It is also used to reduce common mode noise (noise on both the signal and ground) for sensitive measurements.  DMMs and intrumentation amplifiers (as well as NI DAQ boards) typically have differential inputs.

Answers for second post

It is not very difficult to write a program in LabVIEW to implement auto trigger mode.  You can easily poll the NI-SCOPE driver to determine if the scope device has received a trigger.  If a trigger has not occurred for a set length of time (200ms?), the scope device can them be software triggered.  I don't plan NI policy, so I can't answer why the SFP doesn't have auto triggering.  We have had many requests for it and recognize it is a problem.

AC coupling is essentially a high pass filter at about 10Hz (frequency depends on the probe attached to the input).  It is normally used for analog signals to remove a DC offset from the signal.  See the tutorial AC and CD Coupling. for more details.  Coupling choices exist for both the trigger circuit and the analog inputs.  This can get somewhat complex, depending upon how the circuit is implemented, and I cannot remember the details for the 5112.  Coupling on the analog inputs always effects the signal you acquire.  Coupling on the trigger effects how the trigger level is determined.  If you use the trigger input, the coupling is simple - it is what you set.  If you use an analog input for the trigger source, the actual trigger coupling will depend on the scope model and both the analog and trigger coupling.  See your scope manual for details.  The SFP takes care of all these details for you and always shows you the correct coupling choices.

The easiest way to see if you can use 20MHz bandwidth is to switch it on and off while your signal is showing on the scope (running, not paused).  If the signal changes appreciably (other than noise going away), you should not use 20MHz.  The gold standard is to take a signal, save it, then run a Fourier power spectrum on it.  If you have appreciable signal components over 20MHz, don't use it.

[wang]

Hi DFGray:

Thanks for your great help. You are very experience person and I don't want to bother you a lot. But I can not think of any way to solve my problem. I wish you can help me out. Don't worry, I will handle the small problem by myself.

By following the way to told to do, after that I found the first thing I need to slove right now is to make sure my hardware is working

correctly before I tried using SPF to output the waveform.

I have two DAQ boards. One is 6260 with CB-68LP connector boards. The other is NI5112. I also have one test fixture hardware

with oscilloscope ouput. Last last, you told to try using differential mode for CB-68LP and 6250 to remove the hard ground of

NI 5112. It was successfully and the hardware is working correctly. But this time, I put 5 digital output lines with 1 DGND from 6250 to

the test fixture 5 channels. It will then affect my 6250 signals on test panel from positive voltage to some uncorrect negative voltage.

Why? Should I tried putting two bias resitors on different mode? I am not very sure. Could you help me out? Thanks!

[DFGray]

It appears your digital and analog grounds are sufficiently different to cause problems.  You probably still have the original ground loop that caused you to use differential mode for the DAQ board.  Make sure your DAQ board and 5112 are using the same ground.  Are they in the same PXI chassis or computer?  If not, are the different computer/chassis connected to the same power connector?  If not, do the power connectors they are connected to have the same ground?  If this is not the problem I would need a current wiring diagram to debug further.  Good luck!

[nyborn]

Hi DFGray:

Thanks for your replying. I need to tell you that my 6250 board and 5112  are in the same computer. But I don't know if you remeber my

test fixture has two input. One is current source and the other go to the NI5110 scope. The reason I chose to use the differential mode

is because I had a hard ground on the BNC cable. I also have a CB-68LP terminal block which I remeber if I am not wrong that it switch

the gound to different wire. Am I correct? Pls verify me here.

I got the correct voltage of 0.51V in differential mode. Then I tried to put digital ground into DGround on my terminal block. It cause my

differential voltage to drop from positvie value of 0.51 volt to -0.8V.

Attached is my digram. Could you help me to look at what wrong with it?

By the way,  The reson I use differential mode is because of the hard ground. When I use differential mode, does it mean I have two different wire (seperate) so, I can only measure it in differentoal mode? Becasue my RSE in 5112 wire has been seperated by terminal

block of 6250 RSE mode. Am I right? If not, could you please expain to me. I don't understand here. Thanks!

[DFGray]

I missed the fact that you are attempting to measure current with the 6250.  The 6250 measures electric potential (voltage), not current (amperage).  From your diagram, it appears you are using the 6250 to measure the current flowing from your device under test to ground.  You can do this by converting the current to a voltage with a resistor - which you are sort of doing at the moment.  Using your diagram, I would guess you would use the +/-0.1V range of the 6250 and a small value resistor (0.1Ohm, will depend on what current you are getting) placed between the red and black binding posts.  Use RSE for the 6250 and attach the positive side to the red binding post, the negative side to the black.  Adjust the 6250 gain and resistor value as necessary to get good readings.  Find your current with Ohms law.  Keep the resistor as low as you can so you disturb your measurement as little as possible.