Dell Precision test run: 1. Turned on power to Blackbox, plugged in USB from box to laptop. Immediately noticed repeated clicks from the blackbox and could hear repeated "nudges" from the stepper motors receiving power. Normally, this is just a single set of clicks from the blackbox, either when it powers on initially or when it is manually reset. This time there was a set of about 4-5 repeated clicks/resets. 2. Clicking stops. I keep the USB plugged in. 3. Next, I launch the OB Control from the Dell. It comes up just fine, but I notice that there is a toolpath already loaded up in the simulation area. I have not used this toolpath for a very long time and it appears to be a facing operation of some type. I am not sure if I have used this toolpath at all - it doesnt resemble anything that I have done. 4. I attempt to use the probe cycle to set up the part and the toolpaths that were hanging up. I notice that jogging the machine into a set up position seems slightly faster than the mac. The first axis I try to set is the Y axis (using y-zero). The signal lead is connected directly to the workpiece. Initiating a probe cycle for Y axis, and the touch does not register to stop the cycle. I am not going to attempt the other axis. 5. I eyeballed the x/y/z WCS for the tool path where the pauses/stops were happening. I removed the bit, zeroed the z-axis 3" above the WCS and just let the spindle spin above the part so as not to involve any tooling. I ran the 33k lines of code at 195% feed, total run time was 01:20:00. 6. Not a single instance of pausing or random stops. If they happened at all, they would have happened while I stepped away and the program restarted on its own. 7. Also note that the probe wiring was left hooked up to the blackbox, but not attached to anything at the other end. Just hanging over the machine platform so as not to short out.
I'm not using the 16 ft USB cable (yet). I'll be sure to keep an eye on it if I do. Once this machine is running reliably, I plan to move it into its final place. I'll try to keep the distance between the blackbox and the computer as short as possible so I can use the shortest cable that I can. Not sure where the DC ground to earth would be coming from, there is no mixing of ground/earth anywhere. There are only 4 sets of wires/cables going to the machine: 1. 4-wire cable to X axis stepper motor 2. 4-wire cable to Y axis stepper motor 3. 4-wire cable to Z axis stepper motor 4. Shielded 4-wire VFD cable with drain The z-axis column is separate, and insulated from the X and Y axis. There is a ground from the spindle back to the VFD, onward to earth. I just realized there is no ground on the X/Y axis chassis and these are separate and insulated from the column/z-axis. Grounding this to the column would be as simple as attaching a strap from the X/Y baseplate to the column. However - thinking about the continuity tests I have done, there was continuity between the X/Y base plate and carriages and the column/z-axis. This means that the stepper motors themselves are the only possible explanation for the DC ground.
If you have continuity between a motor frame and any of the motor wires then that's your issue. Should be insulated. Grounding the frames is a good idea, often overlooked, but I would check your motors for shorts first. Cheers Gary
1. Just checked all 3 stepper motors for shorts or bad windings - none found. Found a good video showing how to check by using an LED. Also went by feel, then checked each pair of wires with a jumper and each had considerable resistance as they should. Even pulled out my 4th stepper (spare) and it checked out good. 2. With both stepper motors disconnected from the blackbox - the x/y assembly is 100% isolated from the column/z-axis. With all steppers disconnected, a continuity test between the x/y assembly and the column/z-axis assembly shows no continuity. 3. Connecting only the y-axis stepper to the blackbox - no continuity found between the x/y assembly and the column/z-axis assembly. 4. Connecting only the x-axis stepper to the blackbox - no continuity found between the x/y assembly and the column/z-axis assembly. 5. Connecting only the z-axis stepper to the blackbox - no continuity, as expected since the column/z-axis and x/y assembly are isolated from one another. 6. Connecting both the z-axis and y-axis steppers to the blackbox - no continuity found 7. Connecting both the z-axis and the x-axis steppers to the blackbox - no continuity found 8. Connecting all 3 steppers to the blackbox - no continuity found **All checks were performed using multiple points so as to eliminate the possibility of mill scale and other contaminates from giving a false reading. These ranged from bare steel, to raw aluminum, the steel lower spindle housing, the stud blank in the spindle, and to the steel linear rails on each axis. Since this is the case, I am wondering how I was getting continuity between both the x/y assembly and the column/z-axis assembly. Another possibility - chip build up on the ungrounded x/y assembly. Could the build up of aluminum chips cause a static charge to collect on the x/y assembly? Since the ungrounded x/y assembly is essentially free-floating, there would be no way for the static charge to dissipate to ground. Aircraft have this same issue. However - I have never received a static shock when touching (grounding) the x/y assembly. I would think this would have occurred over the countless times I’ve loaded stock, adjust items, and made any contact with this part of the machine. Wow! Sitting here trying to figure out what could be the cause…. I go over to the machine, reach for the probe wire and pick it up - the machine goes absolutely nuts! Z-axis starts cycling back and forth, the y-axis turned on and ran all the way to its limit only stopping when I hit abort. Scared the daylights out of me! Pretty obvious now that it’s related to the probe function in some way. Either the wires, the terminals at the blackbox, or something reacting to the probe function. Oddly enough, the x/y/z on the interface shows x=0, y=0, z=0 after all this. Somehow just touching the probe wire made the controller think it was supposed to go to an arbitrary 0/0/0 WCS. I didn’t touch it to anything or ground it out in any way, literally just picked it up by the insulation and went to disconnect it. Also to note - when jogging the machine on either the Dell Precision or the OB Interface, the steppers have just a bit more pep in their step. Like the acceleration was ramped up slightly. Not sure what this is from.
That would be typical of your high resistance return path discussed above. Sort out the Gnd/Earth short. Confirms my suspicion that your Mac needs Big Sur and also may need proper FTDI drivers. Consider a Windows PC though, overall just a little better Negative, on the observation - as thousands of othet users has BlackBoxes, probes and wires - but you have DC GND shorted to AC earth somewhere sufficiently well enough that you can sometimes successfully probe (not a good thing - those should be isolated not connected) That needs fixing Didn't know you also had an OpenBuilds Interface. So far you only mentioned CONTROL. Interface's capacitive buttons are very sensitive to Ground faults too - random runaways are a good sign too that your DC GND is super noisy.you touching even an insulated wire could be enough to induce some spike into DC GND because your GND reference is dirty. The Capacitive jog buttons see this random voltage or drain as a Touch and sends jog commands. Disconnect the Interface for now (safety) until you have sorted out the Grounding! Are you using our Meanwell+Powercase or some other power supply?
Not sure if it was clear, but after attempting to probe, I removed the probe lead completely and set it aside. It was not connected to anything, in any way, when I touched it and the steppers went nuts. yes, I have the OB meanwell and case
When you touched the wire (even just insulation) , with your body touching the floor, which has a path the AC earth, which is accidentally connected to the DC GND through some yet to be found short but known from the symptoms, and in turn connected to the BlackBox through to the Interface - triggered the capacitive touch buttons. All because in your setup, your DC GND is not properly isolated from Earth/machine chassis/etc
if there is a DC ground at the x/y assembly, then I should be able to wire up a simple DC circuit and use x/y assembly as a ground, correct? Just a simple battery and a lightbulb should be sufficient, then touch the ground side of the circuit to various points for testing. I’m not sure where the DC ground would be coming from, the only electrical items on the x/y axis are the two steppers that have tested good.
Probing should be: BlackBox Probe Signal > Endmill > Touchoff plate > Plate is attached to GND on Blackbox Yours is something like this... BlackBox Probe signal > Stock > Machine Chassis > AC Earth > All around the house wiring > Some yet to be found leak between AC Earth and DC GND > Back to DC GND via some unknown path > Back into BlackBox's DC GND domain If you want to do what you are trying to do, it should be: BlackBox Probe signal > Attached to Vice (Isolated from everything else electrically) BlackBox GND > Attached to Collet (Spindle body, Z Axis, rest of machine all isolated from this GND) Or BlackBox GND > Attached to Vice (Isolated from everything else electrically) BlackBox Probe Signal > Attached to Collet (Spindle body, Z Axis, rest of machine all isolated from any other GNDs (or probe will always be in triggered state) ) Until you find that - you will keep having weird issues. Process of elimination: - Multimeter in continuity mode between machine body (AC Earth) and a convenient GND on BlackBox (Endstop or Probe terminals all expose GND) - Unplug random stuff until the multimeter stops beeping to help you zero in on where the short is Use a multimeter - injecting more higher voltages and currents is going to end up killing your controllers and won't be covered under warranty - in fact all the AC coupling around it I'm surprised its all still alive
Ok, I’ll see if anything turns up. Should only be a couple minute task since there are only 3 possible sources (each stepper).
Ok, I have located the issue. Stay with me on this... I hooked up one end of the multimeter to the ground on the probe cable. I used the other end of the multimeter to test for continuity. 1. I left everything hooked up - all 3 steppers motors and the VFD. I turned on the Blackbox but did not hook up to the computer. 2. Probing for continuity had zero results, even when unplugging each item, one at a time and plugging them back in one at a time. Nothing at all. 3. I turned on the Macbook, opened OB Control, plugged in USB, connected to blackbox 4. Probing for continuity showed there was DC ground present on the column/z-axis 5. I disconnected z-axis stepper to eliminate it from items that were plugged in. 6. Continuity still present at column 7. This left only the VFD - I opened the VFD cover and disconnected the ground terminal wire that runs from the VFD to the Blackbox 8. Continuity still present at the column 9. I disconnected the remaining VFD control terminal wiring (VI, and jumper) 10 Continuity still present on column! Even though there is nothing to link the Blackbox and the column together to form a circuit. VFD and Blackbox are completely isolated from each other. 11. Disconnect USB cable from Macbook 12. Continuity no longer present.... 13. Close/restart OB Control, I plug USB cable (one that came with Blackbox) back into the Macbook, reconnect to blackbox 14. Contintuity present at column 15. Unhook usb cable, shutoff power to Blackbox, start Dell laptop, launch OB Control, plug in same USB cable, turn on Blackbox, connect to Blackbox... 16. Blackbox starts clicking repeatedly over and over, maybe 5-6 sets of the standard 3 clicks I am used to hearing when it starts up. 17. Continuity not found at column 18. Hook up all control terminal wiring between Blackbox and VFD, restart the blackbox and control software 19. Contintuity not found. So... DC ground continuity is present when the Blackbox is 100% removed and isolated from the rest of the machine by way of ALL physical wiring. Its also evident that the Macbook enables this to be present or not. Three questions - 1. How is the machine getting a wireless DC ground? 2. What on the Macbook ports is causing this issue? 3. Can this be turned off on the Macbook so it can still be used?
Fantastic, get yourself a https://www.amazon.com/gp/product/B07235PR4V/ref=ewc_pr_img_1?smid=A30QSGOJR8LMXA&psc=1 to isolate the issue The clicks are just the computer scanning ports, actuating the DTR pin - causing resets (quite normal) - you can save yourself the audible annoyance of it by removing the jumper - see the "Disable the Relay" section of docs:blackbox:jumper-relay [OpenBuilds Documentation] (As it should be for VFDs anyway, if you note the description in that section of the docs) Its not, the Mac PSU might be coupling DC to AC (failing transformer windings or by design) Power Supply more likely - macbook is fed DC from the PSU - only place around the Mac where AC is close enough to couple to it, is the PSU. On the machine frame, AC Earth probably comes off the VFD body You can try a different Mac PSU, or the USB Isolator linked above
Ok, great. The clicks don’t bother me, it’s nice to hear them so I know something is working. It’s just with the Mac it’s three short clicks and the Dell it’s a series of three short clicks 5-6 times. Makes it seem like something is shorting out or going haywire. I’ll be sure to pick up one of those USB isolators. Thank you for your help with this, I appreciate it!
Also, if I choose a different USB to USB isolated, what are the specs that I need to worry about? Thanks!
I went with the one you recommended. Was just asking to further my own knowledge in case I need another one later on down the road.
I think part of the problem will be solved once the usb isolator comes in…. But that isn’t the only issue. While I am waiting on that isolator I decided to get everything trammed back up after the probing crashes. Rather than use the MacBook. I opted to use the interface. Because I’m using a tramming gauge, there really is no need for me to have the VFD on, therefore I did not have it powered up the entire time - no EMI. 1. while tramming the machine, I manually jogged the axis back/forth in speed intervals anywhere from 1% to about 50%. No problems. 2. After tramming for about 20 minutes or so, I began to notice that Interface would not respond to any touches. 3. There were no metal chips, particles, oils, or other contaminates on the screen. In fact, I just pulled the protective screen plastic from the screen. The interface is still that new and unused. 4. I figured there might have been something on the screen and continued on. 5. Screen went unresponsive again, axis jogged on its own very briefly and stopped. 6. I continued trying to tram the machine, tried to jog again and no response from interface whatsoever. 7. Note - there was a red light for the AUX icon that seemed to be trying to light up but was not fully illuminated red. It would flicker very weak, brighten (faintly), then back off. 8. Tried to jog machine back to its temporary machine home 9. Y-axis takes off on its own, does not respond to any buttons on the Interface, does not respond to me pressing abort, carriage crashes into the end of the y-axis, y-axis stepper motor continues to try turning until I am able to cut all power with E-stop. Notes- 1. USB cable was not plugged in at all. 2. Performed continuity test to ensure previous issue was not present before operating machine, no continuity present 3. Installed (2) ferrite chokes on each cable before operating. One choke at each end, as close to each cables plug/receptacle as possible (within 1-2 inches of terminated cable end). Installed on x,y,z stepper cables, VFD command signal cable, probe cable, Interface cable. Any ideas?
Received the USB isolator today. Started laptop, turned on machine, connected isolator and USB cable to computer... OB control will connect, but then disconnects shortly after. Maybe 20-30 seconds or so. I tried both settings on the isolator, both produced the same results. Can't connect to the blackbox without being booted off.
I was able to get it working. I think. You can't plug it in to the computer, then plug the USB cable into it. Evidently the sequence matters. USB cable < USB Isolator < Computer < OB Control Launch < Connect FTDI Seems to only work in low speed mode (dip switch = 1, at 1.5 Mbps) When on this setting there is no continuity between DC ground and the machine chassis. I was having more issues with the Interface though. It becomes unresponsive at times, and the buttons seem to be getting mixed up / confused. I made sure it wasn't my fingers accidentally pressing two buttons at the same time or even hovering too close like you'd see with a touchpad. Its gone bonkers like everything else seems to be doing. Again, no EMI present, VFD completely shut down and never turned on.
Only in the middle of a desert. There is always some electrical appliance, wire in the wall, wifi router, neighbour with an electric fence, microwave oven, or other source of interference around. Sometime it might even be an old electricians mistake (had a customer whose house had L and N swopped on the sockets in his house)
Ok, fair enough. The interface should not be malfunctioning like it is. It has never done this before, The most I have ever used it was during the build phase of the machine to jog each axis out of the way. Never had any issues. The same conditions are present now when it is acting erratic and sending my y-axis into the column. I want to know that when I go to use the interface that its not going to send a shell mill into a 3 inch thick piece of steel or try to drive a 1/4" bit through the top of my stock sending shattered pieces of bit flying. It seems to me that there is some kind of a sequence with the blackbox - for example, when I was trying to get the GT series VFD to work, the blackbox has a startup sequence that runs on the relay as well as the command signal wires. When its the USB cable, its seems like if I start the blackbox and the DC ground is somehow present, it creates an operating condition in the blackbox that can only be reset by a complete shutdown, not simply removing the cause of the problem. With the Interface - I press the button a few times and it stops responding. Then it responds. Then it doesnt respond but sends my y-axis on a crash course with other machine parts. The abort button on the control system doesnt even stop it like its supposed to. Honestly, I don't want to be on this forum at all. I would much rather be out cutting steel and aluminum, making money.
So you are relying on a connection through the greased bearings of the spindle, right? Clearly this is not a good connection, causing probe overruns. Always apply a good connection to the collet nut (make sure ti is not painted or otherwise coated!) and be aware that some coatings on bits are not conductive at all. In this case using a straight, clean, pin as the probe is the better option (as Alex detailed above.) Pretty much all forms of oxidization are not conductive, or only partially so and unreliable. Of interest here is rust on steels and anodizing on aluminum (the rails are anodized). If you are probing steel, it has to be shiny clean. If you are probing aluminum, you may need to grind through the anodizing if that exists, or the probe has to push really hard on it to make a connection. This reduces accuracy and will sometimes just never make a connection.
This issue has been sorted out - at least for now. Suspected DC to AC coupling from the Macbook pro power supply, recommended to use a USB to USB isolator. I am not worried about the cleanliness of the materials being probed - its a non-issue. Thank you though. The problem that I am having now is with the interface and erratic and runaway axis movements
Still having issues with runaway axis and button issues with Interface. No VFD involved. Posting so this doesn't get lost in sidebar topics about probing.
See earlier replies about the issue. Won't have any new replies forthcoming, the root cause is still the same. Somewhere something in your setup is still noisy, leaky, or otherwise electrically nonstandard. Same process of elimination applies.
