AnsweredAssumed Answered

HP 436A Matlab programming issues

Question asked by mattdeibel on Feb 22, 2011
Latest reply on Feb 26, 2011 by mattdeibel

Like • Show 0 Likes0
Comment • 4

We are trying to communicate to the HP 436A over IEEE 488 using MATLAB. We want to be able to set the units the power is measured in, and read out the power measurements. We have 2 simple lines of code:

        fprintf(power1_obj, '9D+T'); % controls power meter - sets it to read in dB and puts it in trigger mode
        power_1 = str2double(fscanf(power1_obj)); % read the power *** this line does not work

the first line works, we are able to change D to A or other ascii characters to change the units of the power meter. However we are unable to read the power out. I think it has something to do with the "eol" command sent by the power meter (I read a couple posts about this but not too sure what it is or how to look for it) because when we run the program the power meter portion always times out. Any suggestions on how to fix this?

I am fairly non experienced in this area so can if you know how to help me please try to break down the problem into its simplest form. Thanks for your time.

Matt Deibel

No one else has this question

Outcomes

StevenLee

Feb 23, 2011 9:00 AM

Hi Matt,

I'm too familiar with Matlab but what might be a good way to break it down is to try sending SCPI command equivalents to the instrument. You can do this by using the Agilent I/O Library Suites. Any drivers or functions in Matlab that you use eventually breaks down to low level SCPI commands so if the commands work by itself, you can monitor the data coming back via the BUS using the I/O monitor and see whether the Matlab program is parsing/interpreting the data in the right way.

Steven

Actions

mattdeibel Feb 23, 2011 10:42 AM

Thanks for your response, we will be sure to try this. Unfortunately we will not be able to return to the lab until next Monday and our time in this particular lab is relatively limited, so we are trying to prepare as much as possible for our next visit.

In the mean time we are trying to do as much research as possible to ensure we have all the information we can find in order to either figure out if this is feasible or not.

In our research we noticed the manual for the 436A power meter dates back to 1979 when the programming guide was updated. This means the communication used to be done via IEEE 488.1 protocol with the power meter. And we also found that SCPI was released in 1990. So with this in mind, we want to make sure before continuing that we will be able to receive output from the 436A power meter if programmed correctly. In other words, is our task possible? Do we need to make sure to use any special cable to connect the power meter to the computer? And will MATLAB be able to communicate and receive over these old standards the power meter was designed for?

This kind of programming is not in my teams area of study so we are having trouble trying to crank through this, sorry for all the questions.

Thanks for your time,

Matt Deibel

Actions

RayH Feb 25, 2011 4:08 PM

The 436A always outputs a string a 12 characters, followed by the ascii <CR><LF> (cariage return, line feed) characters. The crlf is intended as the eos (or eol) sequence

Some interfaces (NI) are set to expect a GPIB hardware line, EOI (end-or-identify) to be asserted as the EOS "character". The 436A never uses this line - so if the software is waiting for it, there will be a timeout.

are you using the matlab instrument control toolbox? There is some info on setting EOS characters here:
http://www.mathworks.com/help/toolbox/i ... 29308.html

The info in the 436A manual talks about formats used in 1970's era computers, and is just confusing. See the attached 1 page explanation of the 436A GPIB commands. If it is not shown on this page, the 436A does not support it - in other words, "*IDN?" is NOT recognized and should not be used, there is NO status byte, NO serial poll, no SRQ interrupts, etc.

FYI - 436A Introduction Date: 01 FEB 1975

good-luck!

Attachments

436Aprogramming.pdf104.0 KB

Actions

mattdeibel Feb 26, 2011 5:55 PM

Thank you for all your help. Were going to give this one last try (and also post over on the MATLAB forums if anybody has done this/knows how to this). Thanks again!

Actions

4 Replies

StevenLee Feb 23, 2011 9:00 AM
Mark Correct
Correct Answer
Hi Matt,

I'm too familiar with Matlab but what might be a good way to break it down is to try sending SCPI command equivalents to the instrument. You can do this by using the Agilent I/O Library Suites. Any drivers or functions in Matlab that you use eventually breaks down to low level SCPI commands so if the commands work by itself, you can monitor the data coming back via the BUS using the I/O monitor and see whether the Matlab program is parsing/interpreting the data in the right way.

Steven
Like • Show 0 Likes0
Actions
mattdeibel Feb 23, 2011 10:42 AM
Mark Correct
Correct Answer
Thanks for your response, we will be sure to try this. Unfortunately we will not be able to return to the lab until next Monday and our time in this particular lab is relatively limited, so we are trying to prepare as much as possible for our next visit.

In the mean time we are trying to do as much research as possible to ensure we have all the information we can find in order to either figure out if this is feasible or not.

In our research we noticed the manual for the 436A power meter dates back to 1979 when the programming guide was updated. This means the communication used to be done via IEEE 488.1 protocol with the power meter. And we also found that SCPI was released in 1990. So with this in mind, we want to make sure before continuing that we will be able to receive output from the 436A power meter if programmed correctly. In other words, is our task possible? Do we need to make sure to use any special cable to connect the power meter to the computer? And will MATLAB be able to communicate and receive over these old standards the power meter was designed for?

This kind of programming is not in my teams area of study so we are having trouble trying to crank through this, sorry for all the questions.

Thanks for your time,

Matt Deibel
Like • Show 0 Likes0
Actions
RayH Feb 25, 2011 4:08 PM
Mark Correct
Correct Answer
The 436A always outputs a string a 12 characters, followed by the ascii <CR><LF> (cariage return, line feed) characters. The crlf is intended as the eos (or eol) sequence

Some interfaces (NI) are set to expect a GPIB hardware line, EOI (end-or-identify) to be asserted as the EOS "character". The 436A never uses this line - so if the software is waiting for it, there will be a timeout.

are you using the matlab instrument control toolbox? There is some info on setting EOS characters here:
http://www.mathworks.com/help/toolbox/i ... 29308.html

The info in the 436A manual talks about formats used in 1970's era computers, and is just confusing. See the attached 1 page explanation of the 436A GPIB commands. If it is not shown on this page, the 436A does not support it - in other words, "*IDN?" is NOT recognized and should not be used, there is NO status byte, NO serial poll, no SRQ interrupts, etc.

FYI - 436A Introduction Date: 01 FEB 1975

good-luck!
Attachments
- 436Aprogramming.pdf104.0 KB
Like • Show 0 Likes0
Actions
mattdeibel Feb 26, 2011 5:55 PM
Mark Correct
Correct Answer
Thank you for all your help. Were going to give this one last try (and also post over on the MATLAB forums if anybody has done this/knows how to this). Thanks again!
Like • Show 0 Likes0
Actions

HP 436A Matlab programming issues

Outcomes

Attachments

Related Content

Recommended Content