D510MO

PCW wrote:

A base thread is not needed with Mesa cards (effectively the base thread runs on the FPGA card)

Thanks for the explanation.

I'm thinking that I will follow the hardware plan from the MV35/40 thread.
Does this seem reasonable?

1x D510MO
1 x 5i23 (not a lot more than the 5i20, and a more spacious FPGA) @ $229
1x 7i33TA @ $79
2x 7i64 @ $198
1x 7i44 RJ45 breakout @ $69

I've got the D510MO running just fine for some time.

However, and that's a big however, I seem to have no output from the built-in parallel port. I've got a second pci parallel port card plugged in, and it works fine.

The HAL is set to use 378 for the 0 card, and 1008 for the 1 card.

I'm thinking that the port address is wrong for the zero card, but can't find a way to find the 0 port address.

Any others?

Tom

You can enter
lspci -v
from a terminal

Good reading here...
wiki.linuxcnc.org/cgi-bin/emcinfo.pl?NetMos

Did you try 0x378 as the address?


Rick G

Oops,
lspci -v
does not show my on board pport either.
Never actually tried to find it that way before because, Well it has always been where I expected it to be 0x378

Rick G

andypugh wrote:

TarHeelTom wrote:

I'm thinking that the port address is wrong for the zero card, but can't find a way to find the 0 port address.

Mine is at 0x378, as shown by
cat /proc/ioports
I wonder if the problem is that you are giving an address in decimal, not hex?
Alternatively, it might be disabled in the BIOS.
You ought to be able to access it on address 0 and the PCI one on address 1, EMC2 uses parport_pc to find cards by index as well as by address.
Oddly, I don't see my p-port with lspci -v, but I do with cat /proc/ioports I have also never actually used it, so don't know if it works, it should be easy enough to find out…

I've used the lspci command which Rick suggested, and it finds the second port, but not the first port. I'll try the cat /proc/ioports next.

And although I typed 378, I am using 0x378 and 0x1008.

Just had a horrifying thought last night.

On the D510MO board the parport header has a key at one end. And the red strip on the ribbon cable from the header plug to the db25 connector is at the key end of the header plug.

However, in looking at the pinout in the literature for the board, it shows pin 0 and pin 1 being at the far end of the header from the key. My first project today is to remove the printer cable from parport 0 and use a meter to see if the header cable is wired correctly, or wired backwards.

The second test will be to just not use parport 0 for the moment, and see if parport 1 will create motion.

But for the moment I'd be willing to bet a beer that the db25 connector is pinning out backwards.

Tom

I would guess the key is pin 26 on the header (since the DB25 only has 25 pins vs the headers 26)
so that would make pin 1 on the opposite end from the key

PCW wrote:

I would guess the key is pin 26 on the header (since the DB25 only has 25 pins vs the headers 26)
so that would make pin 1 on the opposite end from the key

Kinda maybe sorta. The header is two rows. The key IS pin 26, but pin 1 is on the opposite row on the opposite end.

Page 34 of the D510MO ProductGuide01 shows this very clearly.

I just can't imagine why the red stripe on the ribbon cable is on pin 26, instead of pin 1.

Tom

Sure pin 26 is a long diagonal from pin 1
the red stripe is always a pin 1 marker
Both on DB25s and 26 pin headers

It should be easy to verify the pinout by checking if the DB25s
pins 18 through 25 are connected to ground