[RCaffin]

Hi Washuu

In 99% of cases that means the Dir command is not reaching the driver. Pin mis-assignment, broken wire, loose screw connection, blown Dir input to motor driver, inadequate drive signal on Dir line, whatever. Multimeter time.

Cheers
Roger

[dharmic]

First thing that springs to mind is maybe the homing switch polarity is wrong in the software or is connected to the wrong terminal so it's in a triggered state. If you have the hard limit setting (or whatever it's called) activated, the motor won't drive past that spot. Easy enough to check: jump into the diagnostic page on Mach3 and have a look at the inputs. If my guess is right, lights will come on when you push the switch on X and Y and go out when you release them but, for the Z, the light will be on until you push the switch instead.

:edit: between RCaffin's response and mine: what happens when you try to drive up on the Z? Does it just not move, or does it drive down as well?

[Washuu]

First thing that springs to mind is maybe the homing switch polarity is wrong in the software or is connected to the wrong terminal so it's in a triggered state. If you have the hard limit setting (or whatever it's called) activated, the motor won't drive past that spot. Easy enough to check: jump into the diagnostic page on Mach3 and have a look at the inputs. If my guess is right, lights will come on when you push the switch on X and Y and go out when you release them but, for the Z, the light will be on until you push the switch instead.

:edit: between RCaffin's response and mine: what happens when you try to drive up on the Z? Does it just not move, or does it drive down as well?

I should have mentioned that, both up and down both go down. I will go test the connections.

[dharmic]

Then yep, dir pin.

It's pretty hard for it to go wrong with their loom all set up with locking connectors both ends of the cables between the bob and drivers. If you're stuffing around inside the black box it's REALLY easy to knock the dip switches on the drivers with cables etc, worth checking they're the same on each driver. Also check the pin and polarity of the Z DIR pin as Roger mentioned just in case it's not set right. But, if you're grabbing the config from OmioCNC and you haven't dicked around with the settings they should be right...

[Washuu]

Then yep, dir pin.

It's pretty hard for it to go wrong with their loom all set up with locking connectors both ends of the cables between the bob and drivers. If you're stuffing around inside the black box it's REALLY easy to knock the dip switches on the drivers with cables etc, worth checking they're the same on each driver. Also check the pin and polarity of the Z DIR pin as Roger mentioned just in case it's not set right. But, if you're grabbing the config from OmioCNC and you haven't dicked around with the settings they should be right...

This is a brand new setup I just finished assembling. Since I had the spare A-axis cable I tossed that on the Z-axis and that confirms it is not a cable issue. The configuration is the one provided by OmioCNC besides me enabling the limit switches.

I am taking the cover off the control box now to check the DIP switches.

[dharmic]

It won't be the cable from the box to the stepper.

It might be the cable from the BOB to the stepper driver inside the black box. Unlikely, but possible.
It might be the DIP switches on the Z driver. Unlikely - especially if you've not had the box open yet.
It might be the direction pin's active low setting in Mach3. Or the pin's selection.

Try using a voltmeter on the direction pin + and - inputs on the Z driver with the BOB cable connected. Heading up should be different to heading down. If not, you have either the wrong active low setting or pin selected in the config in Mach3 or a fault in cable or BOB.

[Washuu]

The Z-axis cabling and motor driver are fine. I moved the connector cable from the motion controller card from the Y-axis motor controller to the Z-axis motor controller and was able to make it move up and down by using the Y-axis jog. Additionally, hooking the Y-axis motor controller up to the Z-axis connector on the motor controller card results in the same error. So this seems like a software configuration issue so far.(Or maybe a busted controller card.)

[dharmic]

Excellent - you're getting somewhere

[Washuu]

Excellent - you're getting somewhere

Yep! Thanks by the way. DIR+/DIR- on the Y-axis switches between 0.5 and 4.5 volts when switching directions. The Z-axis is stuck at 0.4 volts.

Port and Pin configuration: http://i.imgur.com/yj7zivP.png

[dharmic]

Hmm.

Those settings are the same as mine. Doesn't bode well for your BOB unfortunately.

You might be able to assign the Z DIR pin to a spare output pin on the BOB but it'll mean cable surgery.

[Washuu]

Hmm.

Those settings are the same as mine. Doesn't bode well for your BOB unfortunately.

You might be able to assign the Z DIR pin to a spare output pin on the BOB but it'll mean cable surgery.

Thanks. I will pull the MCC(I guess that is what you mean by BOB?) out of the box tomorrow night and put it under my oscilloscope. The worst case of losing one spare output pin that I would not have used anyway to repair the Z-axis is better than shipping the control box back to China for replacement.

[dharmic]

I wouldn't call it a motion controller card - it ain't doing any motion control, it's letting Mach3 do all that. It's just busting out from USB to dir/step/analog spindle with optos. Nitpicking, I know. But yes: that thar board what the USB plugs into.

Scope? Nice when you have the tools to look at this stuff properly.

[Washuu]

I wouldn't call it a motion controller card - it ain't doing any motion control, it's letting Mach3 do all that. It's just busting out from USB to dir/step/analog spindle with optos. Nitpicking, I know. But yes: that thar board what the USB plugs into.

Scope? Nice when you have the tools to look at this stuff properly.

It was the best name I could think for it.

I bought the oscilloscope finally a few months ago when I started working on hacking CAN bus for cars. It is proving very useful in lots of diagnostics.

[Mmpie]

Had to have a major reshuffle at work, I've ended up moving all my stuff into one area and building an enclosure to contain the noise and mess this thing makes. Not used the machine since the first day I got it up and running properly, not had the time. Having to extend my cabling for it to reach from my cabinet so now seems a good time to plug in the shielded cables and the earth. Just had a look on ebay for 4 core wire, christ on a bike is it expensive! Seemingly only available in 30m reels. Lots of different wire diameters, what kind of AWG should I be looking at? For the motor/spindle cables and the earth. I've plenty of old cable lying around to make an earth from, just wondering which to use.

Also, does anyone know how many amps this thing draws? The circuit on that room can only handle 60A, I think an upgrade is going to be needed. It will have to handle the Omio, a 3D printer, two PC's, a hoover or compressor and a few LED's and PC fans hooked up a standalone PC PSU I'm using to ventilate all dust and moisture straight outside.

Will this stuff do? It looks like 4 core to me? Quite thin wire though. 305M RJ45 4 Pair Shielded FTP Cat5e Ethernet LAN Network Patch Outdoor Cable | eBay

And just to be sure, I'm hooking an earth cable up from anywhere on the Omio itself, to somewhere in/on the black control box? Can I run all my earth/shield points to what I assume is the earth point on the PSU? The blue wire which comes straight from the power cable to the 'N' (I'm guessing for neutral). Should I also change the spindle wire to a 4 core and earth it, or just use 3 core as the earth cable from the frame should do the same thing?

Finally, what are the plastic connectors called? There's three different connectors, the PC fan looking ones that go to the motors, the round plastic ones that plug in to the box and the metal one on the spindle (I've not opened that yet to see if its reusable or not).




Sorry for all the questions, I should have paid more attention to the lessons on electrics back when I was at school!

[Mmpie]

Ignore that last post, been doing some research at home and learned my error mistaking negative for ground. I know now the one next to it with the funny lines is the earth point. So I know that the PSU is not earthed, the VFD is not earthed, apparently (I read) the emi filter wants earthing which obviously isn't earthed. Nothing is earthed. I'm guessing now that the third unused terminal on the power cable socket in the control box (pictured in the other post) is the ground terminal?

I can't get my head around this, the more I read the more different things I see and I haven't a clue what to do, no one (that I can find) has made a picture or video guide for electrical idiots such as myself who haven't a clue what they're doing. I see people talking about running ground wires directly to the wall, are they just stripping the end off a wire and shoving it straight in the third socket in the wall?

Also, on shielding, I have a better grasp of this now after reading about it. The bit that's confusing me is how to connect the shield to ground. Will it have a 5th wire coming out the side, which I'll have to route to my ground point on its own, perhaps by drilling a new hole in the box? Or should I replace my 4 pin plugs on the box for 5 pin plugs, connect the shield drain wire to one of the plugs and then attach that terminal to ground?

Perhaps I should just pay someone to do it for me :drowning:

[dharmic]

Electrically and earthwise, these are the same nightmare as most of the fleaBay 6040 machines.

There's ideal and what you can get away with and, being a combination of practical and a tightwad, I opt for the latter.

Noise-wise the biggest issue by far on these is EMI generated by the spindle which can induce currents in signal lines high enough to false trigger things. Safety wise the biggest drama is that the frame isn't earthed so, if a coil shorts to the case in the spindle motor, the whole frame is live and will earth through you to ground. Inductive coupling can also cause the motor (and thus the frame earth) to climb to the point where the USB connection stops working properly which was happening to me over about a 5 hour run time.

Ground the frame for safety's sake. Get a suitable bit of earth wire - I just stripped out the earth line from a bit of house mains cable. Crimp an eye onto one end. Pull out one of the frame bolts (who cares which one) on the base, re-install it with the eye behind it. Other end of that wire into your mains earth. One day I'll put a mains plug on the end of my earth wire but, for the moment, it's just jammed into the earth pin on a power board. This got rid of my 5 hour buildup issues too.

Shielding. You're doing this because you're extending cable anyway? Suggestion: don't bother. Extend the cables with far, far cheaper (and easier to source in short quantity) unshielded cable first. I had to stretch some of my cables to fit my black box on a shelf under the machine, too. Didn't shield, no issues so far. Don't let your wallet burn fixing a problem that doesn't exist.

IF it starts showing issues, fix it then: you haven't blown too much money on unshielded cable. The big one to shield is the spindle. Running shield through connectors ideally means getting metal connectors and crimping their body clamps onto the shielding so the shield remains complete and unbroken around the connection. Not sure I'd bother with the steppers, to be honest. They're just not that energetic on a machine this size running at these low voltages.

Relax. Take a breath. Keep it simple

[RCaffin]

Noise-wise the biggest issue by far on these is EMI generated by the spindle which can induce currents in signal lines high enough to false trigger things.
RECKON! That is the killer. Screening is a start; going further can be difficult even for an experienced electronics engineer. Often much experiment is needed. It is an arcane black art.

Safety wise the biggest drama is that the frame isn't earthed so, if a coil shorts to the case in the spindle motor, the whole frame is live and will earth through you to ground.
Yep, in spades. Been there (between a live frame and earth) before. The electrician who did the work was incompetent and had not tested his own work. He was TOLD.

Here in Australia it is compulsory to earth equipment unless it is double-insulated. A CNC is NOT double insulated. Our mains leads for such gear ALWAYS have a solid earth wire, and you connect that wire to the frame first. Our power sockets all have an earth pin. Do not fool around with this stuff! Get help!

Screening signal cables: this is not as effective as screening the VFD and spindle motor. I agree that you could start by running unscreened cables for signals AT FIRST. That said, screened cable is NOT expensive and can be really worth while. Hunt around local shops: you should be able to find someone who can sell you a few metres. Note that running all UNscreened pairs flush against earthed metal surfaces is moderately effective anyhow.

Cheers
Roger

[dharmic]

I'll say it again: Earth the frame for safety. But I wouldn't bother with shielded cables on these until you have to, most of the time they work fine without it.

But, when looking at EMI and signal issues, something else to consider is how EMI gets into the signal path to false trigger things.

To get onto the signal line it needs to run parallel for a reasonable length of run. And the impedance at the controller end needs to be high enough that the current will induce sufficient voltage to flip the input state on the board. This might happen on the home switch lines due to the run along the cable chain, but it looks like the BOB's input impedance is low enough that it tolerates significant noise on these lines and so you're pretty sweet without shielding.

The other place it might cause problems is between the BOB and the stepper drivers. However, these lines are short and do not run parallel to the spindle power line inside the black box. Again, this should be fine.

The more likely (on these machines) source of noise issues is the floating ground voltage being driven up to (or past) the voltage that should flip the state of an input. If this happens you WILL get issues. For me it was because of a double insulated laptop running over USB to the BOB on a floating ground machine. The difference between "ground" voltage for the machine vs the ground voltage for the laptop exceeded the capacity of the USB electronics to buffer and, when this happened, switching the spindle on with an M3 would kill the USB connection.

Grounding the frame of the machine meant no more of these shenanigans and fixed all my noise woes.

[Washuu]

An update on the Z-axis that will only move in one direction. I was going to just temporary repurpose the A-axis output on the motion controller card. However, reassigning the pins actually does nothing. Setting them all to 0 still allows everything to work. So reassigning a spare pin most likely will not work either. I will just have to wait for Golden Week to be over and contact OmioCNC directly.

[dharmic]

An update on the Z-axis that will only move in one direction. I was going to just temporary repurpose the A-axis output on the motion controller card. However, reassigning the pins actually does nothing. Setting them all to 0 still allows everything to work. So reassigning a spare pin most likely will not work either. I will just have to wait for Golden Week to be over and contact OmioCNC directly.

Sorry, I didn't even twig. The driver for the Omio's USB interface seems to completely ignore the ports and pins for the motors and spindle control - one of the reasons I switched to a cnc-drive controller and cheapo bob instead. If you're getting a pulse train on dir/step on the other axes but the dir pin on Z isn't switching, I think you might be looking at a new board. Keep us posted on how you go with Lance...