tnt 5002 interface

[theotigan]

Mr Collin

My Board type is PCI-GPIB (488 2) , maybe i must make some extra configuration before i try to communicate whit the tnt 5002 interface??

Thanks

[Collin M]

I think we should focus on getting the GPIB signals to match at the device and controller before anything else.

When you read the BSR at the device you get 0x5C which indicates DAV, NRFD, EOI, and SRQ are asserted. When you do an iblines at the host you only see SRQ and REN asserted. This does not make sense. The BSR register simply reflects the state of the GPIB signals on the bus. iblines does the same thing. These should always match.

Can you verify the connections between the TNT5002 and GPIB connector? Is it possible that some of the signals are swapped?

Is it possible to make a pdf or scan a hard copy of your schematic?

Another test is to do the following:

Initiliaze the PCI-GPIB and device. Do not make the PCI-GPIB a controller. You can control each GPIB signal through the BCR regsiter in the TNT5002. Set each of the following bits by writing a '1' to the corresponding bit in the TNT5002 BCR, the do an iblines on the PCI-GPIB. You should see a correlatioin. Try this with EOI, SRQ, NRFD, NDAC, and DAV.

For example, when you write 0x08 to the BCR in the TNT5002 you force the EOI signal to assert. When you do an iblines on the PCI-GPIB you should see the EOI signal is asserted.

[theotigan]

I do the test

also

 Write(0x7C, BCR)

I read Read(BSR) = 0x7C

and iblines                                   EOI ATN SRQ REN IFC NFRD NDAC DAV

                                                      0       0      1       1       0      0         1         0

[Collin M]

Can you try setting each bit in the BCR individually so that only 1 bit in the BCR is set at any given time?

[theotigan]

i will test evry bit and reply whit the reponse ..

thanks

[theotigan]

Line test the tnt 5002 is controller

here are the result.....

SRQ_line   Write(0x04, BCR)//  Read(BSR) =   0x04   // iblines   SRQ=1, REN=1 NDAC=1

EOI_line   Write(0x08, BCR)//  Read(BSR) =0x0C      // iblines  SRQ=1 REN=1 NDAC=1

NRFD_line   Write(0x10, BCR)//  Read(BSR) =  0x0C    // iblines  SRQ=1 REN=1 NDAC=1

NDAC_line   Write(0x20, BCR)//  Read(BSR) = 0x24     // iblines  SRQ=1  REN=1  NDAC=1

DAV_line   Write(0x40, BCR)//  Read(BSR) =   0x44   // iblines  SRQ=1 REN=1 NDAC=1

[theotigan]

Mr Collin

I whill scan the Schematic  and send it letter...

my scanner is broken down,

I must do it by  a friend  letter ...

[theotigan]

Mr Collin

My Board type is PCI-GPIB (488 2) , maybe i must make some extra configuration before i try to communicate whit the tnt 5002 interface??

Thanks

[Collin M]

You do not need to do any special configuration to the PCI-GPIB to communicate to the TNT5002.

I think there is a problem is with the GPIB connector or GPIB cable.

In the TNT5002 if pon in cleared then writing a '1' to any bit in the BCR will assert the corresponding GPIB signal and will be detected by other devices on the bus.

For example, when you write 0x08 to the BCR the EOI signal will assert (as long as pon is cleared). This should be detectable at the PCI-GPIB when you do the iblines command. I think your GPIB connection to the TNT5002 is incorrect. For example, you might have the SRQ signal on the cable connected to the EOI signal on the TNT5002.

We need to focus on the having both the TNT5002 and PCI-GPIB detect the same state for the GPIB signals before anything else is possible.

Here is another test. Disconnect the GPIB cable from the PCI-GPIB and do an iblines. You should not see the SRQ signal asserted. Then connect the GPIB cable so the PCI-GPIB and TNT5002 are connected, but hold the TNT5002 in reset (either hardware reset or keep pon asserted). Then do another iblines at the host. It is important that the TNT5002 is in reset when you do the iblines at the TNT5002.

[theotigan]

here are the result of the test

iblines whitout cable

Ren=1 NDAC=1

iblines whit cable

SRQ =1 REN=1 NDAC=1