Error,

OH, DIO serial # 0019 did report mostly OK tests, it had a few missed values. That is interesting, that 3 boards have no response, but one board mostly responds correctly. Even stranger!

Jon

Your telling me! They are all set to address 6 and in the same slot. When power is applied a red light on the encoder 2 card blinks once and the I/O all has green lights.

When diotest is halted there are random red lights on.

When diotest is halted there are random red lights on.

OK, this all sounds right, then. There may be some noise issues on the parport bus, but it seems to be mostly working. I think, (as others have said) that there is some kind of issue with the lube pump relay that is sparking so brightly in the video. I'd just disconnect that for now, and see if the rest of the system comes up and works properly. If so, then you will need to find out why that relay sparks and fix it, possibly with arc suppressors on the contacts.

Jon

I have already disabled the contact that was sparking. I have also posted about that and that if I comment out the drive ready signal in the code it will not do any of that.

The drive ready signal has been cycling so fast that the machine would home out. That contact is no longer sparking and the machine seems to be locked into E-stop all the time now.

I am going to put everything back in its original positions and see what works or doesn't at this time.

I put the cards back to where they were when I started working on this machine, with the exception of the DI/O cards. I am not sure they are in their original spots but the machine turns on and runs at this time. The motion has alot of vibration in it while jogging and homing. The spindle is also jerky when rotating. The single DAC is velocity control for the axes and spindle.

Well, that's some bit of progress, but it obviously is not right. I think what you need to do now is bring up Halscope and see if the encoder velocity has spikes in it. That would indicate that LinuxCNC is not reading the encoder counts correctly. It could indicate noise in the encoder signals or communication errors between the PPMC and the computer. If the encoders are being read properly, then it would indicate the DAC is not getting updated reliably. (That's a condition that I've never seen.)

Jon

It is back to loading Linux CNC but will not start the machine. When I boot the computer I go to terminal and type "sudo ./ppmcdiags 378 bus" and get DAC-16 on every board address.

If I type "sudo ./pcisetup 378" and then the "sudo ./ppmcdiags 378 bus" it shows me the cards in the addressed locations.

It is back to loading Linux CNC but will not start the machine. When I boot the computer I go to terminal and type "sudo ./ppmcdiags 378 bus" and get DAC-16 on every board address.

If I type "sudo ./pcisetup 378" and then the "sudo ./ppmcdiags 378 bus" it shows me the cards in the addressed locations.

Yes, the ppmc driver built into LinuxCNC has the smarts to set the parallel port to EPP mode. The diagnostic doesn't have this feature built in, so you have to use the pcisetup program (or others) to do that, on certain parallel ports.

You should then run the diagnostic program with the commtest option and see if it can doo 100,000 test cycles without any errors reported. If not, then some detective work needs to be done to find out why. (It may be necessary to disconnect encoders from the encoder board, if the encoders dither it can fool the test into reporting errors.)

Jon

Just an update.

I had a bad DAC and spare DAC. A contactor was noisy. It is running at tis time. Thanks for the help!