Data transmission from a HP 4194A Impedance Analyzer is extremely low since updating to LabVIEW 7.1

[martinheinz]

We had developed a complete series of measurement programms running on Hewlett Packard HP 4194A impedance analyzer. These applications run on a 150 MHz Pentium PC with Windows 98, ISA-Bus HPIB-Card and LabView 5.1. We also had other applications for other measurement devices. After updating to LabView 7.1, PCI-BUS HPIB-Card, and 1 GHz Pentium PC all applications run much faster, but in the application for the HP 4194A impedance analyzer only sending commands for measurement was faster. The impedance analyzer runs the measurement - and than: extremly long duration for sending the measurement curves to the PC. The duration for sending the complete impedance and phase curve was very very much longer than on the 150 MHz PC with ISA-Bus HPIB card. Since we have no idea for the reason any hint would be helpful.

[Brian_Beal]

When you say slow, how slow do you mean? Seconds? Tens of seconds? Be aware that this instrument doesn't like fast communications. There are several posts about using old 488 equipment with Pentium processors and PCI GPIB cards.

You should be getting your A and B array data over in milliseconds, not seconds.

[martinheinz]

---

@brian Beal wrote:
When you say slow, how slow do you mean? Seconds? Tens of seconds? Be aware that this instrument doesn't like fast communications. There are several posts about using old 488 equipment with Pentium processors and PCI GPIB cards.

You should be getting your A and B array data over in milliseconds, not seconds.

---

When measurements are completed the impedance analyzer gives a beep. Than - with the old equipment (150 MHz Pentium PC with ISA-Bus HPIB-Card, Windows 98, 64 MB RAM, LabVIEW 5.1) the two data fields was sended to the PC within about 1 second.
With the new equipment (1,1 GHz Pentium PC with 256 MB RAM, PCI-Bus HPIB-Card, Windows XP professional (optimized for performance) and LabVIEW 7.1 programs - all programs was created by the corresponding application builder) the transmission of the two data fields has a duration of over 8 (!) seconds.

[Brian_Beal]

Are you running a program in the HP or are you sending it individual commands? In other words, is the PC just triggering the start of the program and watching the status byte to see when the measurements are done?

[martinheinz]

Measurement parameters were send to the impedance analyzer. Than the analyser makes the measurement with the built-in options. On the display of the analyzer there can be seen that each measurement (e.g. equivalent circuit, impedance and phase curve)is properly executed. This takes the normal time. Than, if the measurement data has to be transmitted to the PC, the transmission takes extremly long time (more than 8 seconds).

[Brian_Beal]

You need to find out how your Labview program determines that the HP is done with the measurements. Can you post the VI that your executable was created from?

[martinheinz]

If necessary I can post the VI`s. Where (e-Mail-adress?) have I to send it?
I will search for the items you pointed to. The question is: The same VI has run perfectly with the slow PC system. Why the fast system could make trouble?

[Brian_Beal]

The fast system is too fast for the 4194A. Delays(about 50ms) need to be added after each read and write statement.

[martinheinz]

Meanwhile I had analyzed the complete application. I put break ponits to each program step. Thus I could determine the command where the time delay was caused. The reason for the time delay is NOT the HP 4194A impedance analyzer but (this is very surprising for me) the storage of the data to the hard disc when using the vi "write characters to file.vi". Within a for-loop each of the about 400 data values of the A anf B field (first impedance, than phase) were changed into string format and than written to a file using the "write characters to file.vi" VI. Before writing the data to the file all data were read out from the HP 4194A into local variables (fields). This reading out of data from HP 4194A takes less than a second. Storing the A field to the file takes about 3 seconds, storing the B field to the file takes about 5 secondes although the field dimension is equal. In the old PC system this storage was much faster.

[Sumiteee]

Hi Martin,

I m a student of Munich University of Applied sciences. In our lab we have a Hewlett Packard HP 4194A Impedance/ Gain-Phase Analyzer. It would be very kind of you if we could get the Labview program for this equipment. Is it possible to get it?

Yours faithfully Sumit