about my 48" X 48" X 5" Techno/ISEL woodworking router build

I have uploaded my LCNC configs to my GitHub ( github.com/travis-farmer/linuxcnc ) if anyone is interested. unfortunately, i did not save a backup before my recent changes, and i have not tested the recent changes, yet. but i will try and remember to keep it up to date, for troubleshooting things. so there may be tons of errors, at least until i am able to test it, and make the corrections. though if anyone finds blatant errors, feel free to let me know how to correct them.

there is, i hope, config values for two spindles, if i copy-and-pasted correctly. spindle 0 is controlled via a HY VFD over Modbus, and spindle 1 is a PWM controlled laser for cutting and burning. hopefully i have configured those somewhat correctly.

~Travis

worked on some config changes today to get my HY VFD to communicate over RS485. first i had to add the current user to the "dialout" group to be able to init communication.
i ended up commenting out the PWM for Laser configs, as apparently i do not know how to add a second spindle, or maybe i was doing it wrong. i will attack that when i know more.
anyway, i pushed the latest configs to the Github above.

I am waiting on delivery of more wire to finish connecting things up, so wiring progress is slow.
major todo's are:
finish power supply connections.
install and connect stepper drivers.
connect home switches.
connect relay to control Z-axis stepper brake.

then i should be able to see some first moves. i really should lube the machine axis again too. been sitting awhile.
after i get in some first moves, i need to fine tune some parameters regarding stepper direction, homing direction, tie back in my tool probe and 3D probe.

when i go to make first moves, i will try to remember to drag out my video camera, and pop a video on my YouTube. could either be boring (if everything goes right the first time), or entertaining when it all goes to heck.

~Travis

i connected the machine up today with temporary power, and signal wires, and didn't have anything worth posting the video of. in fact, i deleted the video...
X-axis: very slow, but responds to LCNC control, though direction was inverted (fixed by inverting direction in HAL).
Y-axis: nothing.
z-axis: nothing.

observations: the 5V logic to 24V logic opto converters i bought off eBay have diagnostic LEDs.

LCNC machine OFF, stepper amp enable signals

---

X: input LED HIGH, output HIGH
Y: same as above
Z: same as above

LCNC machine ON, stepper amp enable signals

---

X: in LOW, out LOW
Y: in LOW, out HIGH (about 1.79V)
Z: in LOW, out HIGH (same volts)

those are just examples of similar activity across the different signals. my stepper drivers are active when enable is low. my interpretation is that the Mesa 7i80HD-16 is not outputting true LOW signals (not tested with meter), so the opto level converters output high all the time. or, and this is more likely, the opto level converters are defective. i checked the stepper drivers inputs for current leakage through the internal optos, and found none, as i suspected. so the drivers are not inducing voltage. i don't know where the power could be coming from. i used shielded twisted-pair cable (CAT5e) for the stepper driver signals, with the shield drain tied to the shielding on the control board. the dark colors were used for the signals, and the light colors tied to ground.
admittedly, the control board shielding is not yet tied to anything, so it could be inducing something on the signal wires...

if i can't figure out the issue (i stopped for the day due to a headache) sometime tomorrow, i may eliminate the 24V level converters for the stepper drivers, and connect directly from the Mesa card to the stepper drivers. the inputs have optos already, and the signal wire length is short (18 - 20 inches), so with the shielded twisted-pair signal wires, there shouldn't be too much error. first i will try grounding the shielding, and test the wiring some more, but it is getting to the point where i want this machine usable again.

just a little frustrated, is all.

~Travis

ok, i unplugged the enable, and to my surprise, the LED on the voltage level shifters went out. i unplugged all the signal wires, and all the LEDs went out, or rather, they responded to the output of the 7i80HD-16. so, something about the drivers is driving the LEDs on, though this should not be, as the drivers inputs are just LEDs, with a weak current limit resistor set for 5V input. i added resistors for 24V... oh, wait. i added 3.3Kohms, should be 2Kohms... i can't imagine it would be that big a difference, but maybe. waiting for my scope/meter to recharge, so i will head back out in a few minutes to see if changing the 3.3K to 2K solves everything.
i suspect it won't change anything... but what do i know? (less than i thought, apparently)

as far as "set the machine up properly", this is a little vague. but then, my opening statement could have been misinterpreted. the power wires are temporarily installed, but the signal wires were done proper, i assume. i have since grounded the shielding on the control board, and it didn't have any change, but it is done.

i have some sort of plan for moving forward. i will change out one current limiting resistor, and test for effect. if it solves, i will change out them all.
i think my biggest brain-fart is that i disregarded the other end of the twisted-pair light colored wires. one end is tied to ground, the other is floating, with a separate ground... and yes, i see the flaw in this (as i said, brain-fart). so that will have to be fixed. i think i was regarding it as a shield, but as a twisted-pair, it needs current flow to be effective. sometimes my mind gets doing stuff without thinking about exactly what needs to be done, correctly.

but anyway, when i unplugged the enable signal, the other axis were able to move, but it also got me thinking, and more critical of my wiring job. once the signals are straightened out and working, i will solidify my power wires.

as far as non-stepper signals between the control box and machine, i am going to mount a junction box under the machine (side accessible) with a DB25 breakout board, and another in the control box. that way i can use a parallel port cable to route signals to and from the machine, and be able to disconnect the cable easily when i move the machine from the trailer to the to-be-built shed. it will also tidy up the wires quite a bit too.

~Travis

We have motion on all 3 axis!!!
i eliminated the 24V step-up logic opto converters for only the stepper signals. so the stepper drivers are receiving 5V from the 7i80HD. other signals, when connected, will use opto converters, so the homing switches, and probing will be at 24V. i honestly wasn't expecting it to move all three, so i forgot to video the moment.
at present, i don't have the homing switches connected, so i setup the machine INI for instant homing. the signal wires are attached, but not immobilized yet (to keep wires from breaking).

so, with the wiring in a known moving state, i can focus on solidifying power. when the DB25 breakout boards arrive (generic, for custom wire-planning), i can start to clean up the non-stepper wiring. i rigged in the power for the brake on the Z-axis, but that will be routed through the DB25 cable. i just needed it powered to move the Z-axis.

one thing i noticed, my USB XHC (wired) pendant functions, but will not jog, so i will need to look into that. machine is pretty slow, but i can work on that when the wiring and all sensors are complete.

~Travis

Quick update, i received the DB25 parallel printer cable breakout boards, so i was able to wire in the homing switches correctly, and was able to home the machine! the machine is still deathly slow, and i think the INI has misconfigured STEP_SCALE values, and i posted about that in a more appropriate board here.
anyway, it was great to see it getting closer to functioning correctly, and i have more work planed for tomorrow.

i have abandoned the idea of shielding the control board, as the more wires i add, the less convenient it will be to enclose it. i will however try to shield the rest of the enclosure, and place the stepper drivers in some shielding to protect the control board integrity. perhaps i am just getting too lazy, or getting burnt out from weekend after weekend of wiring.

~Travis

today was filled with wonder and excitement. excitement about seeing some of the little details getting completed. and a lot of wondering about how other details are going to turn out.

i worked on trying to tidy up some wiring, so it looks better. going through a lot of zip-ties, and such.
the enclosure is getting a little cramped with hardware. i think the solution is to move the power converters out of the enclosure, and into a different one. then i can use buss bars to distribute power inside the enclosure. should be fairly straight forward. so far, the only thing using 5Vdc is the Mesa board, and the 5Vdc <==> 24Vdc logic level opto converters. relays use 12Vdc, but not much. 24Vdc for a lot of stuff. and the stepper drivers have their own power supply anyway.
one idea i have been toying with, is that with the relays making and breaking contact, and causing power spikes, i want to shield the ones in the control enclosure, and use them to switch 24Vdc that is routed out to an enclosure under the machine where i will have 24Vdc coil relays that will switch higher powered items, like the dust-collection system. but the 24Vdc for those relays will have an isolated power supply. this is due to an observation of the LEDs on the logic level opto converters blinking dimly when the relay breaks contact.

though what i say, and what i end up doing all depends on my mindset at the time. best i can do is outline my plans, and hope for the best.

as for the laser component, i think i will end up just using two independent configurations. that way it will be simpler. i can copy the basics from one to the other, and just make the needed modifications. i think i will find that i make more use of the spindle than the laser anyway.

something else i am toying around with, is to move the stepper drivers from the control enclosure to an enclosure on the machine. perhaps with the above mentioned relays. just an idea though, as i doubt i will try it. may just be too much complexity.

I can't seem to get my spindle to turn... the VFD responds to forward and reverse, but will not turn up. i suspect something is not connected in HAL, but i can't seem to figure out what. my configs are all here: github.com/travis-farmer/linuxcnc (currently up to date with my machine).
i still have to figure out why my XHC pendant will not jog, but that is for another day. don't want to overload my weak mind

~Travis

spent a little time going over the configs, after work today, and i got my spindle working again! took it up to 20000 rpm, just to celebrate, and it ran nice and smooth. though the general sound it made was terrifying. sounded like it was ready to take flight.
in any case, it works now. i had it in HAL, but it wasn't connected to anything in HAL, so it just didn't turn up. all i could do is select forward or reverse.

now i just have to go through the configs and see why my XHC LHB pendant won't jog...

~Travis