HP 8903A Audio analyzer and Galvant industries USB-GPIB interface

Status
Not open for further replies.
Hi All

I am looking for people that have successfully used the Galvant Industries USB-GPIB interface with a HP 8903A audio analyzer.

It is not possible to use pmillets code and a python coded project I forgot the name of.

I hope someone else is out there that have had success with this interface 🙂

EDIT: I found the python project and that it was updated since the version I have on my PC, I will try the new version later today.

https://github.com/cosmonaut/hp8903
 
Last edited:
Here is a short thread from a while ago with a bit of information. It is not directed at your exact USB-GPIB adapter, but still informative. http://www.diyaudio.com/forums/equipment-tools/242429-why-gpib-converters-so-expensive.html

I ended up buying a generic HP style USB-GPIB adapter and it works perfect with my 8903b. Not sure if you downloaded Pete's files recently. If you have been holding onto those files for a while, he has updated 8903 files with some bug fixes.
 

Attachments

  • 8903.png
    8903.png
    92.8 KB · Views: 386
Hello Nicholas

I tried to find an email address to send you an email about it, but I only ended up trying to send you a message on linkedin.

I just wanted to say thank you for the effort and give you some details on a few minor bugs, if you wanted it. But here goes.

If I try to run a THD+n plot right after start of program, standard settings, the program will halt with an error in the python command line after a few samples. If I however run a frequency plot first, I can run the THD+n afterwards without problems... a little weird.

A simple future function would be comma separated list export of results, the graphs are not that smooth with few steps per decade 🙂

I changed maximum decade step to 100000 instead of 1000, to get a better resolution (1.2Hz steps or so) at a more narrow bandwidth, it can however not take more than a 2kHz span at that sample rate without halting.

If possible, I would like to be able to define the step resolution instead of steps per decade, something does not add up as with 100000 steps giving a 1.2Hz resolution. Maybe it is a calculation error. For some tests I would like 1Hz or 10Hz resolution.

But overall I am satisfied, your program makes it possible to use the Galvant adapter and get some results good enough to adjust the amplifiers 🙂
 
I just wanted to say thank you for the effort and give you some details on a few minor bugs, if you wanted it. But here goes.

No problem, glad it helps you out some. Feedback is the only way it will improve, so I welcome it. I've updated the issues list on the github page for the software with what you've written here, thanks!.

If I try to run a THD+n plot right after start of program, standard settings, the program will halt with an error in the python command line after a few samples. If I however run a frequency plot first, I can run the THD+n afterwards without problems... a little weird.

Yuck. Can you paste the error?

A simple future function would be comma separated list export of results, the graphs are not that smooth with few steps per decade

This should already exist. After you run a sequence try File->Save Data.

I changed maximum decade step to 100000 instead of 1000, to get a better resolution (1.2Hz steps or so) at a more narrow bandwidth, it can however not take more than a 2kHz span at that sample rate without halting.

Any error or does it just freeze? I've been hacking on an hp54506 gpib interface with the galvant device lately and found some bugs in my interface code... these fixes may correct the issues in hp8903, we'll see.

If possible, I would like to be able to define the step resolution instead of steps per decade, something does not add up as with 100000 steps giving a 1.2Hz resolution. Maybe it is a calculation error. For some tests I would like 1Hz or 10Hz resolution.

I should be able to add this. Currently it is designed to put an even number of data points in each decade of log space (because I have used it to scan 20Hz-20kHz in 99% of my use cases). It should not be a problem to add a mode where frequency points are defined by a single resolution value.

Hopefully I'll get a chance to hack on it again in the near future and kill some bugs...
 
Hello!

how challanging is to add VISA support to your pythono software? I own a visa-compatible Agilent GPIB-to-USB interface for controlling my lab equipment via VISA and python under win7.

Nevertheless, i use linux for my personal work. So i really apreciate your linux effort 😉.

Best regards!
 
Hello everyone,

So..I'm trying to run cosmonaut python code on Windows 10 and the software doesn't want to run. I have tried python 3.7 and 2.7. On 3.7 I was able to install all the required modules and on 2.7 I can't seem to install the GObjects packadge.

In both cases the software doesn't open and when opening with the IDLE editor I can see the software gets stuck on this line:
Traceback (most recent call last):
File "C:\Users\L&C Edinburgh\Desktop\hp8903-master(1)\hp8903-master\hp8903.py", line 3, in <module>
from gi.repository import Gtk, PyGObject
ImportError: No module named gi.repository


I have the GI USB.

Any help would be much appriciated.
 
Status
Not open for further replies.