Instrument Control (GPIB, Serial, VISA, IVI)

cancel
Showing results for 
Search instead for 
Did you mean: 

EABO(6)

Hi All,

 

I am using NATIONAL INSTRUMENTS NI PCI-GPIB WITH NI-488.2 778032-01 (system controller) to communicate with two tester. 

Here is the problem, the bytes i write sometimes is not match with the ibcntl and cause EABO(6). Please see the attached picture from NI trace.

ibwrt(UD1, "S:O:C?" ,6) but the ibcntl is only 0x0001, where ibcntl indicates the number of bytes that were successfully transferred during the last I/O operation.

Is this hardware problem or my software problem? Please help.

0 Kudos
Message 1 of 4
(3,896 Views)

Well, EABO is of course "Error ABOrted" and the Status "C100" breaks down to Err 0x80+Timeout 0x40+CIC 0x01 

 

It seams to "Mostly Work" since there are plenty of good transfers.  This usually means that you are operating on the edge with the PHY layer.  Check:

  • cabling for tightness and correct length
  • System for number of devices
  • system for percent of devices on bus that are powered on
  • System for ground loops or noise

Depending on if you are HS or not the PHY layer requirements change


"Should be" isn't "Is" -Jay
0 Kudos
Message 2 of 4
(3,857 Views)

Hi Jeff,

Thanks a lot for your reply.

There is 1 system controller (my PC) and 2 tester. I am using 1x4meter(763507B-01) and 1x2meter(763507B-02) GPIB cable, NI PCI-GPIB connect 4meter gpib cable to tester1 and from tester1 connect 2meter gpib cable to tester2. Is that fine for the application?

The both tester is connected to ground.

How can i verify the "system for percent of devices on bus that are powered on"

0 Kudos
Message 3 of 4
(3,846 Views)

A quick look at the standards

For HS GPIB you need a load per meter of cable with all devices powered

For reliable Low Speed you need an average length between devices <2Meters.  with at least 2/3rds of the devices powered.


"Should be" isn't "Is" -Jay
0 Kudos
Message 4 of 4
(3,833 Views)