GPIB timing problem

[Tweety]

Hello,

I have the following problem.

I need to measure the rise time between an input and output voltage signal. The input signal will be created with two power supplies connected in series. (Agilent E3633A)

The power supplies are connected to the GPIB interface Agilent 82357A. The problem is that I couldn't put it on at the same time. Thereby I have a time offset on the input signal and there are steps in the signal.

We have a software based on Excel (VB programming) that is running well. With this software we don't have these problems.

That looks like that LabVIEW and VISA need more time to process the data.

Is there an other way to manage this problem?

Exist anywhere settings to improve the timing process of GPIB?

Tweety

[SFK]

---

Tweety wrote:

I need to measure the rise time between an input and output voltage signal. The input signal will be created with two power supplies connected in series. (Agilent E3633A)

The power supplies are connected to the GPIB interface Agilent 82357A. The problem is that I couldn't put it on at the same time. Thereby I have a time offset on the input signal and there are steps in the signal.

We have a software based on Excel (VB programming) that is running well. With this software we don't have these problems.

That looks like that LabVIEW and VISA need more time to process the data.

Is there an other way to manage this problem?

Exist anywhere settings to improve the timing process of GPIB?

---

Hello Tweety,

I can only guess that you are referring to Agilent's IntuiLink software that you are using in Excel.

Naturally, their communication software will use the most efficient communication settings and parameters available, but as I do not have access to the internals of their power supplies and communication protocols, there is no way for me to judge the performance optimization potential in LabVIEW. When controlling both instruments through MS-Excel - are you setting both voltages with separate function calls, or are you using some kind of "broadcast" function that sets all connected power supplies to the same voltage?

You wrote that you have to insert a delay between the setting of the output voltages of the two instruments. May I ask you how you have implemented the communication? Did you implement one or two (parallel) the VISA sessions? Are you using the E3633A LabVIEW driver provided on http://sine.ni.com/apps/utf8/niid_web_display.download_page?p_id_guid=0475216F9FCA5335E0440003BA7CCD..., or did you write the whole communication routines yourself?

Furthermore, I would like to know if you can see both power supplies listed in MAX in the GPIB tree, and if you have given them individual aliasses.

Best regards,
Sebastian

[Tweety]

Hello Sebastian,

I don't know exactly how it works with Excel, but I found out that the software is using the VISA32.dll and the SCPI-commands.

For the setting of the power supply outputs exist seperate function calls that will be called successive. And the power supplies will be set to different voltages.

In LabVIEW I'm using the provided drivers from the NI archiv.

At the beginning I initialize all GPIB devices so that every device have their own VISA session. I use the VISA address (e.g. GPIB0::11:INSTR) as identifier that is shown in MAX and not any aliasses.

SFK wrote:

You wrote that you have to insert a delay between the setting of the output voltages of the two instruments.

Maybe I misunderstand you but I won't a delay between the settings of the output voltages of the instruments, the delay is my problem.

The delay causes the steps within my input signal. I need to switch both instruments on at the same time without a delay.

I've solved the problem now with another hardware solution, because I have not enough time to investigate this problem.

But I'm very interested in the solution of this problem.

Thank you for the support so far.

Best regards,

Tweety

[SFK]

---

The delay causes the steps within my input signal. I need to switch both instruments on at the same time without a delay.

---

Hello Tweety,

switching without any delay is fundamentally not possible if using two separate commands to address the instruments, as the GPIB commands must be send one after another. You can though use the trigger function of the instruments to prepare all settings to a point where both instruments are ready to change their outputs almost at once - see attached example:

Best regards,
Sebastian

Message Edited by SFK on 04-30-2008 05:22 PM

Message Edited by SFK on 04-30-2008 05:24 PM

Message Edited by SFK on 04-30-2008 05:24 PM

[Tweety]

Hi Sebastian,

thank you for the example.
Okay I know that it's not possible to switch both devices at exactly the same time. But there must be differences between the handling of GPIB and VISA with LV and VB.
I attached two pictures. One picture shows the measured signal when I'm using your example. (Input voltage LV.jpg)
The other picture shows the measured signal by using VisualBasic. (Input voltage VB.jpg)
How I described, Visual Basic sends the commands seperate one after another. But if you can see in the pictures, the VB version is much faster.
The risetime between Input and Output signal is around 11-12ms. And if you look at the timebase of the LV picture, then you know my problem.


regards
Tweety

[SFK]

Hi Tweety,


comparing the two screenshots you prodided is quite hard, as they display completely different voltage levels for initial and target voltage.

The 'LabVIEW-diagram' does include multiple steps, whereas the 'VB-diagram' shows only one voltage step.

On the other hand, I do think I can now fully understand your problem: the issue is not the time it takes to transmit the new settings to the instruments, but the somehow decreased slew rate of your output voltage signals. Looking at the graphs you sent, it seems as the voltage increase is "damped" by a capacitance - so there must be some current flowing. I cannot guess the actual amount, but as the example I sent you limits the output currents to 1 Ampère in the beginning, you might want to increase the allowed currents to the instruments' maximums...

Best regards,
Sebastian