Pico Universal Stepper Voltage questions

I am stepping (no pun intended) through brining up a linuxcnc - unknown pci parallel port - cable - Universal Stepper - Geiko 203v drivers system. There is a 12 V supply connected to the USC board P2-1 and P2-2. The regulated voltage on P2-3 and P2-4 reads +5 V. The documentation says the E-Stop logic can be bypassed for testing by connecting P5-15 to P5-EG. The voltage on P5-15 reads 0 and the voltage on P5-EG reads -6.5 relative to the ground voltages on P2. This is not what I expected. Are these values correct?
(I am surprised to find any negative voltages on the board. )
Am I correct in reading the pins on P5? There are 8 pins on P5. Only 7 oif them are labeled. EG is the second from the end?
What is the E+5 pin? Should this be +5V?
Is is safe to connect P5-15 to P5-EG (second from the end)?
Thank you, Dan

dansawyer wrote:

I am stepping (no pun intended) through brining up a linuxcnc - unknown pci parallel port - cable - Universal Stepper - Geiko 203v drivers system. There is a 12 V supply connected to the USC board P2-1 and P2-2. The regulated voltage on P2-3 and P2-4 reads +5 V. The documentation says the E-Stop logic can be bypassed for testing by connecting P5-15 to P5-EG. The voltage on P5-15 reads 0 and the voltage on P5-EG reads -6.5 relative to the ground voltages on P2. This is not what I expected. Are these values correct?
(I am surprised to find any negative voltages on the board. )

Yes. All the terminals on P5 and P4 are isolated from the main power/ground on P2. I'm not sure why you are reading any solid voltage there. You should see about +6 V from P5-15 to P5-EG.

Am I correct in reading the pins on P5? There are 8 pins on P5. Only 7 oif them are labeled. EG is the second from the end?
What is the E+5 pin? Should this be +5V?
Is is safe to connect P5-15 to P5-EG (second from the end)?

E+5 is an external isolated +5 V, relative to EG. But, it is from an unregulated DC-DC converter, so usually looks more like 6 V. it has very limited current capacity, so don't use it unless you need to power a low-current sensor, for instance.
Yes, connect P5-15 to EG (second from the end) to indicate there is NOT an E-stop condition. Green LED D5 (marked Estop OK) should light up. This will enable you to come out of E-stop.

Jon

Good News: Thank you for the reply. Yes, after connecting 15 to EG the green led came on.
Bad News: The comtest is failing at 100%
I have a USB to parallel adaptor. Can I replace the PCI card with a USB card for testing?
(I understand USB does not work well under production.)

dansawyer wrote:

Good News: Thank you for the reply. Yes, after connecting 15 to EG the green led came on.
Bad News: The comtest is failing at 100%
I have a USB to parallel adaptor. Can I replace the PCI card with a USB card for testing?
(I understand USB does not work well under production.)

The USB-parallel adaptor won't work, it barely works for some printers. We need fast bidirectional EPP-mode communication for the Pico Systems devices. You must use a compatible parallel port. Some motherboards have a built-in parallel port, otherwise you can use a PCI pr PCIe plug in card.

But, if you have an existing parallel port in there, you may need to use a command to set the port to EPP mode.
We have the pcisetup program at www.pico-systems.com/codes/pcisetup.tgz

which will set an on-motherboard port to the EPP mode. If you use a PCI or PCIe port, there is some tricky address computation required to give the program the right number.

Some parports need this mode setting, some don't, it is kind of hit and miss. LinuxCNC knows how to deal with it so this mode setting issue does not happen.

You must also use a cable designed for IEEE-1284 (EPP mode), generic 25-pin cables do not have the right twisted pairs of the control signals.

Jon

Is this correct or is there a remaining issue?

btw: ./univstepdiags 378 commtest is still failing with 100% error.

./pcisetup 378
wrote 0x80 to 0x77a

./univstepdiags 378 bus
io addr = 378
parport addr 0x378
Bus Map
Board Addr Type Ver.
0 Unknown f
1 Encoder f
2 DAC-16 f
3 DIO f
4 Univ. Stepper f
6 Unknown 6f
7 Unknown 7f
8 Unknown 8f
9 Unknown 9f
a Unknown af
b Unknown bf
c Unknown cf
d Unknown df
e Unknown ef
f No Board ff

dansawyer wrote:

Is this correct or is there a remaining issue?


./pcisetup 378
wrote 0x80 to 0x77a

./univstepdiags 378 bus
io addr = 378
parport addr 0x378
Bus Map
Board Addr Type Ver.
0 Unknown f
1 Encoder f
2 DAC-16 f
3 DIO f
4 Univ. Stepper f
6 Unknown 6f
7 Unknown 7f
8 Unknown 8f
9 Unknown 9f
a Unknown af
b Unknown bf
c Unknown cf
d Unknown df
e Unknown ef
f No Board ff

The above shows that there is no communication between the computer and the USC board. Are you sure you have the right port address? 0x378 has to be an on-motherboard parallel port. But, you said something earlier about a PCI card.
If you are using a PCI plug-in parport card, you need to use lspci -v to find the port address(es) that it has been assigned.

Jon

I thought the fact that the './univstepdiags 378 bus' command is now showing '4 Univ. Stepper f' was a sign that it had identifed the USC board. Am I mistaken? How else would it know about the 'Univ Stepper'?
BTW: it only started reporting this after I ran the command './pcisetup 378'.

Yes I do have a PCI board.

BTW:cat /proc/ioports |grep par
0378-037a : parport0
037b-037f : parport0

dansawyer wrote:

I thought the fact that the './univstepdiags 378 bus' command is now showing '4 Univ. Stepper f' was a sign that it had identifed the USC board.

No. 4f is the ID code of the USC board. Notice the pattern of the boards the program seems to be reporting count up from 04 up to ff. These are the addresses the parallel port is sending to interrogate the board. They are left over on the cable for a few microseconds and that is what is being read back. A momentary phantom bit of data on the parallel port cable. If you unplug the USC you will get the same pattern.

Also, the USC can only appear at address 0 or 2.

BTW: it only started reporting this after I ran the command './pcisetup 378'.

Ah, this is a good sign, it means the port is probably in the right mode, now,

Yes I do have a PCI board.

BTW:cat /proc/ioports |grep par
0378-037a : parport0
037b-037f : parport0

[/quote]
Is that all? Because, normal PCI enumeration requires all PCI devices to be at addresses 0x1000 and above.
So, I really think your 0x378 port is directly on the motherboard. It is possible this port does not have a DB25 connector, but just a 26-pin header on the board. You would need a cable to adapt to the DB-25.

Jon

dansawyer wrote:

BTW:cat /proc/ioports |grep par

Oh, the above command may be removiing an unidentified port. It does NOT find the PCI par ports on some of my systems. I would rather do :

lspci -v | more

and scan through it for parallel port cards.

Jon