9914 - MAX

[sma]

I have a device designed around the 9914 ASIC, it hase been used successfully for several years however when trying to detect the device in MAX it detects the Listen command but does not receive any information.  It appears max simply sends the listen command to detect if something is on the bus at that address.  Can somone provide a method of detecting this operation to enable the interface to correctly handle this scenario.  Note that the unit was used in an ATE which did not perform the listen activity described so I am confident that the implemented design correctly trnsfers data.

[Meekshow]

Hi sma,

Could you be more specific on what exactly you are trying to implement? What exactly is the command you are trying to use and what is your expected return?

[sma]

The device I have implemented sits between a gpib CIC and an instrument to convert the sent SCPI commands to CILL (command translator) and correctly performs the necassary data transfers.  At this point I am attempting to repurpose the unit for an ATE which uses Labwindows which performs the same initialisation/instrument identification functions as done by NI Max. The initialisation sequence causes the unit to listen for data which is not being sent and as a result locks up the implemented interface.

The original unit was used in an ATE which did not perform the initialisation sequence and operated for a period of 5 years without fault.

What I am after is a detailed explaination of what NI Max is doing during the instrument identification process to assist in isolating the cause.

[Dennis_Knutson]

Turn on I/O Trace before you select the Scan for Instruments to capture the low level GPIB commands that are being used. This assumes you are using an NI controller.

[sma]

Managed to capture the commands being sent as suggested.

All is fine until the ibclr(UD0) command is sent.  The status value for this function = 0xC100.

My interface detects an IFC interrupt as a result of this command, at this point I cause a software reset to be performed however this leaves the 9914 unable to recieve and data.  I have also just ignored the IFC interrupt with the same results.

What do I actually need to do to correctly handle the IFC received?

[Dennis_Knutson]

It's been a long, long time since I've done any instrument programming at this level. I found

http://digital.ni.com/public.nsf/allkb/17133B14615BE5DE86256E7000688005

http://digital.ni.com/public.nsf/allkb/DDF4FAB478F48492862571EA00093B20

[sma]

Further investigation shows that following receipt of the IFC the 9914 generates four interrupts.  Checking the contents of ISR[0] a value of 0x0 is returned.  Investigating the cause of these a BO interrupt is high indicating that the device is now a talker with no data in the data out register.

How do I respond/configure the device to accept the GPIB write to the device which follows these four interrupts.