Racal Dana 1992 Gpib not connecting

[Deathwish]

Ok so I have spent the best part of half a day or more trying to get this damned thing to talk to my PC via the gpib card. No go.

Took a look on the time nuts mailing list to read that there is a jumper set sk4 or some such that has to be changed to get it to speak natively , anyone got any idea as to where that is on board as I cant see it in my schematics.

Has anyone got one of thee to talk to a gpib controller card ?.

[edpalmer42]

I don't think that all the GPIB cards have the jumper.  If yours does, it's near IC13, right at the edge of the board toward the front of the counter.

What GPIB card and software are you using in your PC?  Can you talk to any other GPIB devices?

Ed

[Deathwish]

I have both benchview and labview on the pc with a agilent 52350b PCI card, it talks ot my 34401A so I know it works.

[MarkL]

I have a 1992 and, yes, the GPIB interface works.  However, it's not SCPI.  Don't know if that matters with {lab,bench}view.

SK4 is the socket designation on the main board for the ribbon cable to the GPIB card.  There is a jumper next to it, set as shown on mine.  I'm not sure what the jumper does.  It's probably described in the service manual.

My GPIB address switch is set to all down, except position 4 which is up (DDDUDD).  That sets address 4 and disables talk-only mode (normal mode).

[Deathwish]

thanks will take a look in a while and see if it is in my unit.

[edpalmer42]

Sorry, I'm not familiar with either BenchVue or Labview.  I see that there's a Labview driver for the 1992.  Have you installed it?

I use a National Instruments GPIB board with their free IEEE488 software and a freeware program called EZGPIB to access it.  It's basically a Pascal interpreter with some GPIB extensions.

I remember back when I was trying to get my 1992 working on GPIB, there was an issue with the EOI line.  Apparently, the 1992 doesn't use it.  Make sure that your program isn't expecting it.  The author of EZGPIB was quite surprised by the 1992's odd behaviour, but he made changes to his program to accomodate it.  I would think that the Labview driver would take care of that.

Ed

[MarkL]

Good point, Ed.  In linux-gpib, I have "set-reos" enabled with a note I put in years ago that it's required for the 1992.

Here's what it does:

Enables the termination of reads on reception of the end-of-string byte for descriptors returned by ibfind(). Same as setting the REOS bit in a ibeos() call.

EOS is set to 0x0a.

Perhaps you need to set up the equivalent.

[Deathwish]

Sorry it took so long to get back to you, not a very well bunny I am afraid - heart problems

Anyway this is the board in mine, the SK4 header is on the top Gpib board, the other jumper is as yours I believe on the bottom board by the cable connector.

[Deathwish]

typical i shove the bloody following into google search racal gpib board 401820 and get this

https://www.febo.com/pipermail/time-nuts/2008-March/030301.html
which leads me to this which I couldn't find earlier

http://millimeterwave.free.fr/RACALDANA-1991-1992-IM.pdf

and there it is as large as life !

[MarkL]

Interesting.  My GPIB board looks quite different from the layout in the link you posted, and mine doesn't have any jumpers.  It's labeled with 19-1146/2, with the ending "2" written in marker.  Probably the rev. number.

And for more confusion, it looks like Racal re-uses connector designations from board to board.  For example, there's an SK4 on the main board for the GPIB ribbon which then goes to SK1 on the GPIB board.  On the GPIB board, SK4 is your jumper block.

Clear as mud.

[Deathwish]

looks like they used your board in the 2101 microwave counter too