[time-nuts] GPIB interfaces these days

Adrian Godwin artgodwin at gmail.com
Fri Nov 2 07:39:04 EDT 2018


How important is ATN in a typical time-nuts usage ?

I can see it being important in a complex ATE setup where some instruments
are automatically providing data to a schedule and need to be serviced, but
in my understanding the time-nuts case is often capturing a stream of data
from a single TIC. Are there other common configurations that need ATN
handling ?

Although GPIB is (obviously) a bus, it seems to me that modern usage could
be handled more easily using multiple USB interfaces, each connected to
only one instrument. This means the interface can be much cheaper than an
82357a or similar, because it doesn't need bus buffers capable of driving a
full load, and you can use lightweight USB cabling instead of the heavy
GPIB cable.

Getting further off-topic (so please followup off-list if this is of
interest), I'm surprised that the 82357a appears to contain both an FPGA
and a cypress FX2 USB device. The cypress device has a crude (8051)
processor but also a programmable DMA engine that ought to be capable of
doing the source-acceptor handshake on its own. Why does the HP interface
need the FPGA too ?

You can buy FX2 dev boards (sold to be used as clones of Salae and other
USB logic analysers) for a few pounds each. Of course, this doesn't allow
instrument-to-instrument transfers but this is probably mitigated by having
a fast PC and multiple busses - a star rather than a bus topology.



On Fri, Nov 2, 2018 at 9:43 AM Forrest Christian (List Account) <
lists at packetflux.com> wrote:

> I use a national instruments PCI-GPIB card in a Windows 10 PC, works just
> fine.   Usually can find them <$100 on eBay.
>
> I've also used a HP/Agilent 82357A (or B) which does USB-GPIB for those
> cases when you need it for a laptop or something else without a pci or pcie
> slot.
>
> I understand the USB ones in particular are prone to being counterfeited,
> but evidently most of the counterfeit ones work even though they're not
> original HP/Agilent.
>
> There are various other options out there, for instance prologix (and
> possibly others) make a GPIB-ETHERNET converter which will convert GPIB
> instruments to network instruments.
>
>
>
> On Fri, Nov 2, 2018 at 2:05 AM Rex <rexa at sonic.net> wrote:
>
> > So I've got some test equipment devices (mostly HP) with GPIB (or
> > actually HPIB) connectors. Also a few others as non-HP stuff.
> >
> > Mostly I have talked to them with a NI GPIB card in a PCMCIA slot in a
> > laptop. Works great but the small notebook PC I have with a PCMIA slot
> > is from the early 2000's and I'm worrying what if it dies. It is running
> > XP but usually not on the internet.
> >
> > I also have a couple very early aluminum case Prologix USB interfaces
> > that I haven't tried to use in 10 years. I think I remember hearing
> > these early ones had some issues, and I'd have to dig to re-learn how to
> > talk to them.
> >
> > So I haven't looked at GPIB interface devices in a long time but I'm
> > getting a bit paranoid about the good NI PCMCIA card in a very old PC.
> >
> > I don't remember seeing much discussion about this lately.
> >
> > Is there anything new I should look at. I would have thought there might
> > be something with Arduino or maybe Ras Pi by now, possibly needing some
> > interfacing hardware, but I'm not aware of anything.
> >
> > So, any advice from the group?  Old or new. Are my very old Prologix
> > interfaces still worth looking at?
> >
> > -Rex
> >
> >
> >
> > _______________________________________________
> > time-nuts mailing list -- time-nuts at lists.febo.com
> > To unsubscribe, go to
> > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> > and follow the instructions there.
> >
>
>
> --
> *Forrest Christian* *CEO**, PacketFlux Technologies, Inc.*
> Tel: 406-449-3345 | Address: 3577 Countryside Road, Helena, MT 59602
> forrestc at imach.com | http://www.packetflux.com
> <http://www.linkedin.com/in/fwchristian>  <http://facebook.com/packetflux>
> <http://twitter.com/@packetflux>
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe, go to
> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.
>


More information about the time-nuts mailing list