Dc310 about to hit the bricks

[Codythecreator]

Alright guys. I posted a while back that I was having issues getting my new cnc to work. The mill came with a Dc3io and a hitachi sj200 vfd. I had the mill wired up to the point I was confident enough to attempt to do a motor tune. Come to find out, a spindle fault is inhibiting me from taking any further action. After multiple attempts to contact hitachi on what the problem could be with no reply, I decided to switch vfds. Ince I switched the drives, all of a sudden my estop contactor will not close no matter what I try. I don’t know if this is an issue with the vfd faulting from incorrect wiring/parameters or if it’s a programming error The contactor at one point was clicking on and off at random times and the “DF” led was flickering so then I thought maybe it was a problem with the fiber optics so I ordered new ones. Still didn’t solve anything I’ve rewired this endure control panel 4 separate times in case I had made a mistake and have had no further luck. Neither vfd manufacturer seems to want to answer what I thought were just a few basic questions and there’s absolutely nothing on wiring schematics for the Dc3io besides the “Ajax cnc10” install manual and the revision 040914 user guide. I’ve had the mill for coming up on 4 months now and haven’t even started the spindle up so I’m to the point of ditching this centroid controller and trying something else. I’d hate to do that as I can tell the previous owner put a lot of money into this system, but nothing I can think of is fixing the problem and customer advise from vfd manufacturers has been a joke. So if anybody has some input or thoughts before I do some major rebranding in this control cabinet, Id really appreciate it. Thank you

[cncsnw]

If the DF LED is flickering, and you are sure there is nothing wrong with the optical fibers and CPU10 board, then you have either a faulty logic power supply on the DC3IO (the micro-ATX supply on the back of the frame); or you have a faulty DC3IO unit.

[Codythecreator]

I don’t believe the issue is with the optical fibers, I disconnected each one and checked to see if light made it through the( I know that’s not the correct way to test them but it at least tells you if they’re completely broken or not ) and each one appears to be fine. I ordered brand new ones just in case so they’ll be here hopefully tomorrow. And I’ve tested the logic power coming off the power supply and it’s getting +5v and +12v as it’s supposed to. Like I mentioned, everything worked at one point perfectly, I just had the wrong parameters set on vfd so it kept throwing a spindle fault which was what led me to swap the drive with one I had on another machine that isn’t cnc so it didn’t need all the logic I/Os. Could that even be the problem or could it just be coincidence to have the cpu10 crap out at the same time I switched the drive ? The only other thing I could think of, would be that the computer I’m using doesn’t have the old pci slots that the cou10 has so I had to order a pcie to pci adapter in order to connect it. It was just a cheap 10$ adapter off Amazon but it worked when I first installed it. I wonder if the issue could be related to that somehow.

[cncsnw]

I just had the wrong parameters set on vfd so it kept throwing a spindle fault which was what led me to swap the drive with one I had on another machine that isn’t cnc so it didn’t need all the logic I/Os. Could that even be the problem or could it just be coincidence to have the cpu10 crap out at the same time I switched the drive ?

If by "drive" you mean the spindle drive (a.k.a. VFD), no.

By itself, changing the spindle drive cannot cause the DF LED on the DC3IO unit to flicker.

The DC3IO, using its 5V logic power, will light the DF LED any time it is receiving valid communications from the CPU10 board, on fibers #4 and #5. The same conditions that lead to it lighting the DF LED also lead to it closing its fault relay, allowing the emergency stop contactor coil circuit to be completed.

The only other thing I could think of, would be that the computer I’m using doesn’t have the old pci slots that the cou10 has so I had to order a pcie to pci adapter in order to connect it. It was just a cheap 10$ adapter off Amazon but it worked when I first installed it. I wonder if the issue could be related to that somehow.

This might be part of the problem, but there is no way to know except to try using a motherboard with a real PCI slot (such as a Gigabyte GA-M68M-S2P).

[Codythecreator]

Ok i have a new theory that I accidentally stumbled upon today while tracing wires. i was testing the incoming 120vac conductors and made my way through the rectifier/cap and up into the servo motor input (+vm & vm gnd), and for a split secomd i forgor to change my multimeter to dc, but that was enough time for me to notice that i had 30vac running into the +vm terminal. How this is happening is anybodys guess, because when i tested from +vm to the vm gnd on the "live" side of the contactor (because contactor is latched open), i got 108vdc, which is right around where it should be. the strange thing is i also got the 30vac when testing between +vm to the machine chassis. so the only thing i can think that this could be is emi? The servo motor conductors are running right next to 120vac conductors and neither are shielded (i know they should be but i assumed since they were both "high power", that there wouldnt be enough interference to cause any problems". now that i think about it though, the ac lines are active before the dc lines would be, so by time the cpu booted up it would immediately dectect the 30vac and fault out. Or at least thats my thinking. Do you think that sounds like a eureka moment ? Or is 30vac an unrealistic amount of interference ? because theres no way i have it wired incorrectly because theres nothing besides the 24vac tap on the transformer that could produce that low of a voltage. And i know without a doubt in my mind i didnt wire something like that to the servo power circuit lol what do u think?

[Codythecreator]

oh and i forgot to mention, i am getting a "servo drive shutdown" fault almost immediately after the cpu boots every time i start the system up. i dont think i mentioned that before

[cncsnw]

What kind of rectifier assembly do you have making the 108VDC? How big is the capacitor? Your 30VAC sounds more like ripple through the rectifier, due to an insufficient capacitor.

120VAC incoming service wires, at 60Hz, carrying the few amps they would be carrying when just the computer and logic supplies are running, are probably not going to be a significant source of noise. In any event, aren't you seeing the DF LED flickering, and getting the "436 Servo drive shutdown" message, before the E-stop contactor has closed?

"436 Servo drive shutdown" means the same thing that having the DF LED out means: the servo drive is in a fault condition, because it is not receiving valid communication on the #4 and #5 fibers.

Actually, "436 Servo drive shutdown" does tell you one additional thing: the PLC component of the DC3IO, and fibers #1, #2 and #3, are functioning properly. It is the PLC that is communicating the drive fault condition to the CPU10.

You could send your DC3IO unit, or perhaps both your DC3IO unit and your CPU10 board, to Centroid for evaluation and testing. There would be a fee for that ($150 minimum?) but at least you would get an answer as to whether either of those components is at fault.

Alternately, you could gamble $50 on the problem being the computer motherboard and its lack of a PCI slot:
https://www.ebay.com/itm/224451933964?h ... SwxRFgk36E
https://www.ebay.com/itm/124886529301?h ... OSwalxhNSg~

[Codythecreator]

the led was flickering at one point, and the contactor was opening and closing erratically. although it doesn't seem to be doing it as of lately. and honestly i was trying everything in my power to avoid having to send the components in, but i will if theres just no other option. and the 120vac that i was referring to is the hot (and neutral) wires that connect the primary side of the transformer, which is in the same vicinity as the 108vdc conductors as well not sure if that makes a difference or not as far as how much interference it would cause. but the rectifier is the full wave bridge rectifier/cap from centroid Off the top of my head i wanna say the cap is 20000mfds? im used to start and run caps in single phase motors and them being 50mfds so 20000 sounded absolutely insane to me but then again ive never even seen a cnc, let alone build one before now so im not 100% familiar with everything they entail. Would it help if i posted some pictures of the cabinet and such so you could get a better idea of what of going on? Thanks for taking your the time to help by the way. Ive been stressing on this mill for months now, going through 3 or 4 different manuals, back and forth, endless hours of videos and forums, trying to make sense of this thing, so its nice to be able to ask a question and get actual feedback instead of searching for an answer and MAYBE finding one that relates to the problem haha

[cncsnw]

I just checked a standard Centroid "low power" supply on my bench: transformer, 88VAC secondary, CAPBRDLO with 12000 uFd capacitor making 120VDC.

With no load, the AC ripple on the DC output is a few millivolts.

With heavy load (9-10A), the AC ripple on the DC output is around one volt.

For what it's worth, there is a big different in design and intent between the capacitors used to start an AC motor, and the capacitors used to smooth a DC supply. In one case you are looking to keep the basic AC wave, and just introduce a little phase shift (lag) so the motor gets started in the right direction. In the other case, you are looking to flatten the AC wave entirely, absorbing the peaks and using the stored energy to fill in the valleys.

[cncsnw]

If you have 30VAC ripple on your rectifier output, then there is something wrong with the rectifier assembly.

However, that by itself is not going to induce a fault condition in the DC3IO drive section. If you are skeptical of that, simply disconnect those wires from the VM+ and VM- terminals on the DC3IO (be sure to safely tape them off). They are not needed to run the logic circuits, light the DF LED, or close the no-fault relay.