simulate VME master

[zou]

Try to test a VME slave (UUT) with no backplane/crate.
It is possible to read/write data from/to UUT with NI-HSDIO?

George

[Jason W]

Hey George,

NI-HSDIO is a driver that is used with our HSDIO cards.  Are you using one of our HSDIO cards to read/write to your VME device?  Can you provide a little more detail about your current setup?

Thanks,

Jason W
Applications Engineer
National Instruments

[zou]

Yes, I am going to use NI HSDIO card, PXI-6542.  What else you want to know?

George

[Raajit_Lall]

Hello George,

Since you do not have a backplane/crate, how are you actually communicating between your VME device and the HSDIO card?  Since, these have two different communication protocols, you would probably need to have a third party/customer cable to be able to communicate between the two.

Regards,

Raajit L

[tbd]

Dear George,

      I've been following your post with some interest because of familiarity with what you want to do, having done something similar with the DIO32HS.  We used the DIO32HS to test a device designed for a VME shaped backplane - though, I don't think the DUT followed VME write/read protocols.  Anyway, the device was tested without backplane, just a 3' home-made cable - and bouncing DIO pulses needed to be "conditioned".   My job was to reverse-engineer/rewrite an old (DOS) test in LabVIEW; I got a lot of practice using a logic analyzer to dissect bus "write" cycles.  Also, in the late 90s, we wired a VXI controller directly to a VME device.  Cabling was very short, yet it wasn't as trivial as you might think considering the controller was VXI.  Not that any of this will help you!..

I expect what you'd like to hear is - "Yeah, you can do that, no problem - here's the code!" - hey, it could (still) happen!  But until then...

Try finding a copy of "The System Engineer's Handbook - A Guide to building VMEbus and VXIbus systems".  See "VMEbus Specifications, Chapter 2, VMEbus DATA TRANSFER BUS" describes bus-cycles, signals and timing.   I think this project is very possible - but I wouldn't even attempt it without a logic-analyzer!. 

Cheers.

[zou]

---

@Raajit L wrote:
Hello George,

Since you do not have a backplane/crate, how are you actually communicating between your VME device and the HSDIO card?  Since, these have two different communication protocols, you would probably need to have a third party/customer cable to be able to communicate between the two.

Regards,

Raajit L

---

Raajit,

I am going to make a customized cable to connect the front of HSDIO card and the back of the VME device.

Thanks for reply.

George Zou

http://gtoolbox.yeah.net

[zou]

---

@tbd wrote:

Dear George,

      I've been following your post with some interest because of familiarity with what you want to do, having done something similar with the DIO32HS.  We used the DIO32HS to test a device designed for a VME shaped backplane - though, I don't think the DUT followed VME write/read protocols.  Anyway, the device was tested without backplane, just a 3' home-made cable - and bouncing DIO pulses needed to be "conditioned".   My job was to reverse-engineer/rewrite an old (DOS) test in LabVIEW; I got a lot of practice using a logic analyzer to dissect bus "write" cycles.  Also, in the late 90s, we wired a VXI controller directly to a VME device.  Cabling was very short, yet it wasn't as trivial as you might think considering the controller was VXI.  Not that any of this will help you!..

I expect what you'd like to hear is - "Yeah, you can do that, no problem - here's the code!" - hey, it could (still) happen!  But until then...

Try finding a copy of "The System Engineer's Handbook - A Guide to building VMEbus and VXIbus systems".  See "VMEbus Specifications, Chapter 2, VMEbus DATA TRANSFER BUS" describes bus-cycles, signals and timing.   I think this project is very possible - but I wouldn't even attempt it without a logic-analyzer!. 

Cheers.

---

Thanks.  I found read/write cycle info on vita.  I'll tell our HW engineer to looking for a logic-analyzer.

George