MONSTER - A learning experience... pyVCP machine

I'm glad you found it helpful. The graphics are all created using Inkscape which is a free graphics program. In the image folder you will see a sub folder SVGA which has all the raw graphics files so that you can modify them and create your own buttons and icons.

The machine was very stable... then today... just a couple of hours ago, I turned it on and the Y-Axis is jerking and having issues, the Z-Axis ERR signal is blinking... and the two digital inputs are showing constant ON mode. I have no idea what the heck happened. Power surge? Component failure? So now, this weekend instead of working on some projects I really was looking forward to doing... I get to disconnect wires and check components one-by-one.

I'm really starting to wish I had large steppers instead of servos... hopefully I'll find an easy fix... but it just doesn't seem like it. I may have to opto-isolate everything.

Ah the fun of building your own machine eh?

Askjerry wrote: I'm really starting to wish I had large steppers instead of servos.

You don't really. Small steppers are quite nice and it's hard to argue between them and servos, but big steppers don't really scale. They get disproportionally slower and hotter.

Try to find a 750W stepper motor.

If you've been rewiring, it may be a loose connection. I had similar on one of my mills. After a week of a "jittery" Y slide, it turned out to be a loose +Ve wire at the motor... Whew! (Need to update the spiral-pocket file... I uploaded the wrong one!

I have two computers... Mongo (test machine) and Monster (mill actual) I swapped the parallel card between the two machines and the DIG0 and DIG1 input error swapped with the card... so I have a KNOWN bad parallel port #2 card.

I am heading to the store to pick up a new ($15 USD) card. I am also going to evaluate my Gecko reset circuit... currently it applies 5vdc to the three ERR pins of the X/Y/Z gecko-drives. (diode separation.) I'm wondering if the interface I build could have an issue.

I had an issue with the magnetic home sensors a week or so ago... thought I patched that one...

while I'm at it... I may just make a new interface board... set up some opto-isolators and/or relay dry contacts. Then I don't have to worry about ground loops, etc.

It is a pain in the back-side... however, it would likely result in a more stable machine. :dry:

Hopefully the Geckodrives didn't sustain any damage. I'll check the output signals with my O-scope here in the office to be sure the PC is sending proper signals. And, I suppose I'll verify that the 40v DC supply isn't noisy... oh fun. :S

Like the post says... "A Learning Experience"... :blink:

The full history is here: askjerry.info/monster_mill_machine.shtml

Further examination determined that whatever happened took out two Geckodrives and a parallel port.

I'm really not sure what happened... I'm thinking the Geckodrive somehow got a voltage spike that triggered the internal microprocessor to go stupid. Internally to the Geckodrive are four (4) MOSFETS... a high power transistor capable of driving about 20 amps of power.



By switching them in a proper configuration...

You can go Forward. Or reverse.

Unfortunately... I think the microprocessor went insane and switched one side... or the other... or both... to the on position...

... or... ... or Both.

Which would cause a direct short to ground... effectively blowing out the drivers almost instantly.

I'm going to order replacement MOSFETS and replacement logic opto-couplers... and hopefully I can resolve this issue.

Not happy right now... not happy at all.

$10 (USD) Replacement parts to fix the Geckos (if successful.) or.. $140 (USD) to replace each one. ($280 total)

Send them back to Maris he will fix them. Which drive are you using?

JT