time-nuts@lists.febo.com

Discussion of precise time and frequency measurement

View all threads

Re: [time-nuts] GPIB, Proloigix, cables

HM
Hal Murray
Sun, Jan 27, 2013 8:09 PM

Connectivity to the prologix from host isn't the problem.    The challenge
is on the gpib side. It was almost working, so I was wondering if there was
a systematic way to test.

How about a simple utility/hack that read/checks the device ID strings?

Plan 1 would be to scan all possible device numbers and print what it finds.

Plan 2 would be to read a file with number/string pairs and verify that the
device at that number returns the expected ID string.

If the file is extended to include a command on each line, you have the
structure for being able to write test scripts.  I'm thinking of commands
like sleep, write, read...

--
These are my opinions.  I hate spam.

jimlux@earthlink.net said: > Connectivity to the prologix from host isn't the problem. The challenge > is on the gpib side. It was almost working, so I was wondering if there was > a systematic way to test. How about a simple utility/hack that read/checks the device ID strings? Plan 1 would be to scan all possible device numbers and print what it finds. Plan 2 would be to read a file with number/string pairs and verify that the device at that number returns the expected ID string. If the file is extended to include a command on each line, you have the structure for being able to write test scripts. I'm thinking of commands like sleep, write, read... -- These are my opinions. I hate spam.
EP
Ed Palmer
Sun, Jan 27, 2013 8:15 PM

I don't know if it would work on the Prologix adapter, but when you
install the drivers for the NI controller, they include a program called
'Measurement & Automation Explorer' that includes the ability to scan
the bus, eavesdrop on the bus, send low-level commands and get
responses, etc.

Ed

On 1/27/2013 2:09 PM, Hal Murray wrote:

Connectivity to the prologix from host isn't the problem.    The challenge
is on the gpib side. It was almost working, so I was wondering if there was
a systematic way to test.

How about a simple utility/hack that read/checks the device ID strings?

Plan 1 would be to scan all possible device numbers and print what it finds.

Plan 2 would be to read a file with number/string pairs and verify that the
device at that number returns the expected ID string.

If the file is extended to include a command on each line, you have the
structure for being able to write test scripts.  I'm thinking of commands
like sleep, write, read...

I don't know if it would work on the Prologix adapter, but when you install the drivers for the NI controller, they include a program called 'Measurement & Automation Explorer' that includes the ability to scan the bus, eavesdrop on the bus, send low-level commands and get responses, etc. Ed On 1/27/2013 2:09 PM, Hal Murray wrote: > jimlux@earthlink.net said: >> Connectivity to the prologix from host isn't the problem. The challenge >> is on the gpib side. It was almost working, so I was wondering if there was >> a systematic way to test. > How about a simple utility/hack that read/checks the device ID strings? > > Plan 1 would be to scan all possible device numbers and print what it finds. > > Plan 2 would be to read a file with number/string pairs and verify that the > device at that number returns the expected ID string. > > If the file is extended to include a command on each line, you have the > structure for being able to write test scripts. I'm thinking of commands > like sleep, write, read...