X-50 UI and Controller issues

[RobertD (Robert D Mroczka)]

Hi,

I am having multiple issues, I have a Onefinity, woodworker X-50, with stiffy, upgrades Z-20, 2.2kw spindle and 400hz VFD, Jtech 24watt Laser, purchased 3 months ago, set up a month ago, it is ran daily 8 to 12 hours a day, it is lubricated and maintained every 2 to 3 days, it is wiped down with a rag that has 3in1 in it daily, it is blown off and or vacuumed after every production run, it is square, level and tramed. it has drag chains, extension wires purchased from Onefinity, it has twisted braided shielded cable running from the VFD and routed away from any power cable. The VFD and wiring from the spindle is not routed near the controller with at least 18" of room between the units… and for 4 weeks this was just an amazing piece of equipment with precision and accuracy unlike any other CNC I have ran in the past.

One other note, I am using Vectric Vcarve Pro 11.5 to generate my g-code

In the last week, it has developed some really strange problems

1. The pathfinder 3d display box, and the 8 buttons below the run, stop, open, download and trash have gone missing from my controller display - see attached photo
   
   

When I inspect the element in the browser, I see a styling that sets the display of the element to None
See attached photo

if I remove this style, they return, but the next time the machine powers up, that styling returns and the buttons and 3d display box is missing again

The buttons and 3D display box were there 4 days ago, I came in powered up the machine three days ago and they were gone, I did not update or alter anything, I am running the latest firmware version 1.3.0

The next issue - When I try to visit the support website from the controller software it tells me Google Chromium is now out of date, with no options to update it and prevents me from viewing the support page

The next issue - upon shut down, I briefly see a laundry list of errors displayed some of them denoted as fatal… for two days I tried to get a bug report out of the machine and the machine would either crash or just hang… today I was able to download a bug report, I cant view it, however I have attached it here
onefinity-20231123-210806.tar.bz2 (67.4 KB)

I have tried to update the firmware multiple times at this point, both via web and USB, I am not seeing any of this clear, I have tried to restore setting from when the machine was working perfectly the week prior, with nothing resolving these issues either

Now for the big problem - I will admit, I have been running this machine a great deal over the first 4 weeks of use and now that I have hit week 5 I have all kinds of pathing errors with both v-carving and the laser

In this photo you can clearly see that the alignment of the outline is in places .050 of an inch or more off of the cross hatching

and in this photo you can see the alignment is perfect and I ran over a 100 of these ornaments flawlessly the week prior

The grub screw connections have been checked and everything is tight… the wire connections and the pins in the wires are all good, when I run a Vcarve now I have anything from a step over in the X-direction to a complete misalignment of passes, and this is especially noticeable in any text with a round edge like an R or a O and it occurs with different fonts, and it is noticeable in anything with a slant in it like a V or an N and its usually and error in the X direction, but not always…

I have seen and heard the stepper motor skip

here is an example of a piece ran the second week of having the machine in operations

and it was done perfectly with no problems at all its an exact replica of the drawing I made using the exact same equipment and software I have used all along in this process, it was done with a 60 degree Vbit, 1/2" diameter, 1/4" shank, running at 200 hz (12,000 rpm) at 100 ipm

this piece was ran with all the same settings in the same walnut material

and after it ruining the first piece of material I put in the machine with very errant pathing on the O, N and R, I was able to make this piece and still spent an hour or more with needle files and sanding sticks cleaning up a pronounced step in the R, W, O, and N

about 10 days earlier I ran this

and it was absolutely flawless, using the same fonts, the same tooling, the same feeds and speeds and even the same sized lettering for the large font

Thank you in advance for any help or advice you might be able to offer, Robert

[Aiph5u (Aiph5u (not affiliated with Onefinity))]

Hey Robert,

1. the camotics.org 3D toolpath simulation is only shown when connected remotely on a computer/laptop/tablet that has enough 3D hardware capabilites and that has support for 3D graphics acceleration installed. Both does not apply to the Raspberry Pi 3B Raspbian OS on the Onefinity controller, so when you connect a HDMI monitor directly to the Onefinity Controller, it is not displayed (and should not appear).

2. You cannot update any of the packages of the ancient Raspbian Stretch installation from 2017 on the Onefinity SD card without making the CNC controller functions ceasing to work. This is because more or less every package of which you want to install a recent version relies on countless dependencies to other recent packages that you can only fulfill by a complete OS upgrade. But a OS upgrade is not possible. This is because the crucial bbserial.ko kernel module, that makes communication with the AVR mainboard with the stepper drivers possible, is not packed as a .deb package and not integrated in Dynamic Kernel Module Support (DKMS) so will not be re-compiled automatically on a kernel package upgrade, so on kernel module loading, this module will fail. So updating chrome or any of the packages that could have old security issues is not possible.

3. CNC Controller reboots into kernel panic despite being powered off because is backfed by the power from the monitor through the USB port, but cannot complete a boot with the low power that it can draw from the monitor.

@RobertD

The next issue - upon shut down, I briefly see a laundry list of errors displayed some of them denoted as fatal… for two days I tried to get a bug report out of the machine and the machine would either crash or just hang… today I was able to download a bug report, I cant view it, however I have attached it here
onefinity-20231123-210806.tar.bz2 (67.4 KB)

If you mean this:

…this means the controller is already powered down but some power is backfed through the monitor’s USB cable to the controller. The controller reboots then, but the power is not sufficient to complete the bootup.

This only happens if you have Onefinity controller v4 (the power rocker switch version). This is the issue “CNC Controller reboots into kernel panic despite being powered off because is backfed by the power from the monitor through the USB port, but cannot complete a boot with the low power that it can draw from the monitor”. See @Matticustard’s suggestion below.

If you have a Onefinity controller v5 (the green-lit power pushbutton version), this issue does not appear because it has an additional circuit with which the Raspberry Pi can turn itself off.

• Shutdown kernel panic (this is normal in Firmware 1.0.9 and below, removed in 1.1 and above)
• Controller not turning off when powering down due to backfeeding of the screens power to the controller
• Proper Shutdown on the Onefinity CNC
• @Matticustard’s suggestion for a solution
• How to connect the Touch Display (7, 8, 10.8")

We have a search function here.

[RobertD (Robert D Mroczka)]

I have the newer version controller with the push button, as I said Machine was purchased 3 months ago with firmware v1.3.0 installed

Would you have any insight as to why there is now pathing errors developing

and I promise you 5 days ago, the 3D toolpath simulation displayed on the main display connected to the controller, its not a big deal that it does not, its just curious that there was a sudden change in things when all of these other issues cropped up as well.

Thank you for clarifying why I cant update the browser, that makes sense.

Thank you for assisting, Robert

[Aiph5u (Aiph5u (not affiliated with Onefinity))]

Hey Robert,

@RobertD

I have the newer version controller with the push button,

Even if computer error messages look like chinese to ordinary users, there was always a programmer that programmed the error message and its exact wording to get a message about where the error lies, in order to be able to help the user. That is why it is wise to provide the exact error message when calling for help.

@RobertD

and I promise you 5 days ago, the 3D toolpath simulation displayed on the main display connected to the controller,

the question here is not if I believe you (I do) but if you can tell the forum how to reproduce the error.

What I can say is, the Raspbian system that is installed on the SD card inside the Onefinity controller has no 3D graphics acceleration drivers installed. It is not able to display the 3D toolpath simulation.

I don’t know if the buttons row that you marked above should appear, because at the moment I have no monitor on the Onefinity Controller, but should they appear but disappear sometimes, I would report this in the topic of the Firmware version you are using (Firmware 1.3 (Most Recent) in this case).

@RobertD

Would you have any insight as to why there is now pathing errors developing

@RobertD

In this photo you can clearly see that the alignment of the outline is in places .050 of an inch or more off of the cross hatching

You don’t say anything about how you set the workpiece zero. Are you zeroing between the two passes, and how?

If you think you lost steps, then there are these support documents for checking the possible causes one by one:

• Troubleshooting: Rail is getting hung, stuck, or out of alignment (x35/x50 machines)

and

• Z Axis Moving Unexpectedly (Cutting too shallow or too deep)

for details on the possible causes of such issues. I would check all points one by one in the first help document to see why your carriage could possibly loose steps.

You can still think of Electromagnetic interference (EMI), as e.g. mentioned here, if you have…

1. …stepper cables that are not shielded and grounded. EMI can act on a cable but a cable also emit EMI.

2. A VFD that is not enclosed in a grounded VFD Control Cabinet. A VFD is a heavy source of EMI, because the three-phase current that it creates by switching LGBTs often does not consist of nice sine waves.

3. A spindle power cable that is not properly shielded and terminated (for this, see also this important document)

Also is your machine properly rectangular (“squared”) (bar gauge) and coplanar (“not twisted”) (fishing line method) ? This however would mainly affect the Y axis movement where the Y carriages can block.

@Aiph5u

…have you ensured that your machine is accurately rectangular (“squared”) and coplanar (“not twisted”)?

Also Onefinity provides a help file which deals with many kinds of blocking of an axis:

• Troubleshooting: Rail is getting hung, stuck, or out of alignment

I would check all points one by one to see why your carriage does not move anymore in the middle of the axis.

There are three causes that are not mentioned in the document above:

1. Should your machine not be properly rectangular (“squared”), the Y movement can block.

2. And the second possible cause not mentioned in the document above is: This mysterious issue some time ago.

3. Electromagnetic interference (EMI), as e.g. mentioned here.

If you loose steps, I add here some quotings from forum search results:

@Aiph5u

…please also think of connectors and cables possibly failing. Especially the curly Z cable! It can fail by an internal cable break and make Z positioning errors before failing completely.

Also think of connectors are only tin-plated and can have bad contacts:

@Aiph5u

Also as you can read in the document “Troubleshooting: Rail is getting hung, stuck, or out of alignment (x35/x50 machines)”, there are some tips on cable and connector checking, and the Onefinity Elite uses the same type of non-shielded control cables with similar tin-plated Molex/Amphenol connectors and also with same lacking strain relief as the Standard Series, so I would have a look at some defects like these:

and also follow the tips for these:

@OnefinityCNC

Disconnect all wires on the rail that isn’t moving, from controller to motor, triple check all 4 pins inside each connector are present, straight, and leveled with each other. Then reconnect all connections, ensuring they fit very snuggly and 'snap ’ together. Here’s some pictures of a bad wire and a pin pushed in.

To check all of these cables/connectors, after having followed the tips from the manufacturer above which consists of reseating the connectors, inspecting the crimp on their pins, etc., and this leads to no success, I would thoroughly check them this way:

@Aiph5u

To check the cable, I would unplug it on the controller side and on the other end and check all four wires one by one for conductivity between both ends, by using a multimeter set to resistance (ohms). But it could be that you measure that everything is okay because the fault my appear only in certain moments, in certain positions of the cable, when you move it, etc. Usually it’s the motion on a cable that is not made for permanent motion that can make problems after a while, mostly where the individual wire is crimped into the contact, or between the contacts which unfortunately are only tin-plated which means movement and/or corrosion can wear the plating leaving the bare copper open to air which is even more prone to corrosion. In this case, a little amount of KONTAKT 60 and then KONTAKT 61 can help.

And finally think of retrofitting strain relief. The stepper cables on the side hang around freely during movement, that is not good for the connectors and the wires in them.

@Aiph5u

everytime I hear this it makes me so sad because I belong those who appreciate the existence of wood so much, and most frequent causes to ruin it with the machine usually would not even need to be. I assume you have the Standard Series machine that can show problems with the connectors, especially because of the lack of strain relief on this machine .

@Aiph5u

the Cable Connectors used on Onefinity are the weak spot on the Onefinity (not the controller), especially in combination with the lack of strain relief on this machine. As someone who has assembled and troubleshot many computers, I can’t have a good opinion of such tin-plated connector contacts. The Onefinity CNC manufacturer addresses possible connector problems in this FAQ document .

@Aiph5u

If one day you find that your Z-axis moves up and down as it pleases and ruins your workpiece and your wasteboard, you might find that it’s because the connectors can make problems (see also here), which may be caused by the fact that you did not yet retrofit strain relief to your machine , which the machine unfortunately lacks. In this case you may want to procure what is needed to make new cables yourself. The same applies if you want custom length cables which would be the case if you want to mount the controller further away.

So what I would try before risking to loose another piece of wood, is check the cables and especially its connectors, and where they are crimped (if you can look a little into them, do they come out when you move the connector on the cable) or do you measure discontinuity in resistance between both ends with a multimeter? You can use the search function to search for cable multimeter to find how others solved their issue. Unfortunately if you search for plunges workpiece you can find many other who observed their machine ruining a workpiece.

Should this be the cause of your issue, the best thing to avoid that in the future is to retrofit a serious cable management system to the machine, especially where the cables hang freely without strain relief on moving parts, as is the case for the X and Z stepper cables where they enter the left end of X axis, and the curly Z cable often fails after a longer time too.

@Aiph5u

I would check all the points in the help document “Troubleshooting: Rail is getting hung, stuck, or out of alignment” one by one.

Also note that cables not only fail on their connectors. A multimeter can help. Also be aware that neither Onefinity cable is made for permanent motion in a CNC. Sometimes replacing the curly Z cable, that can break internally, or retrofitting strain relief on the machine can help (or better retrofitting a drag chain system that runs cables specifically made for permanent motion).

It was already the choice of these Molex connectors on the Buildbotics controller that was a mistake that Onefinity inherited on their hardware fork, but it’s quite obvious that the machine/controller was also about a low price threshold. With gold-plated circular connectors as used in the industry and a proper cable management (drag chain) this happens less of course.

[RobertD (Robert D Mroczka)]

@Aiph5u

Even if computer error messages look like chinese to ordinary users, there was always a programmer that programmed the error message and its exact wording to get a message about where the error lies, in order to be able to help the user. That is why it is wise to provide the exact error message when calling for help.

I included the Bug Report Log in the original post, please reference that, if you can tell me how to open it, I am happy to tell you what the errors are… the ones that flash on the screen on shut down are there for less than 3 seconds

@Aiph5u

the question here is not if I believe you (I do) but if you can tell the forum how to reproduce the error.

What I can say is, the Raspbian system that is installed on the SD card inside the Onefinity controller has no 3D graphics acceleration drivers installed. It is not able to display the 3D toolpath simulation.

I don’t know if the buttons row that you marked above should appear, because at the moment I have no monitor on the Onefinity Controller, but should they appear but disappear sometimes, I would report this in the topic of the Firmware version you are using (Firmware 1.3 (Most Recent) in this case).

They are back… it renders the ‘3D Tool path’ just fine unless its too big in which case it tells me DATA Unavailable:

I cant reproduce the errors, at best I can only tell you the facts of what I am dealing with and when they started, and share images and bug reports (that I cant access) here and that most of this is random

I might go all day never have an issue with any of it… come in the next day and the machine is a completely different machine

as for zeroing, I set the zero with the laser by getting my focal point at the corner of the piece, space it with the supplied spacer from Jtech, setting the Offset on the machines X, Y and Z, my my origin point in Vcarve at 0,0…

My tool path for cross hatching and outlines is all one tool path… theres no need to reset the the zero… Vcarve allows for both to be made simultaneously

as for Vcarving… its in the process of making the Vcarve that it goes errant… theres no tooling change, no different tool path… lets say its starting to make the letter O… it will go in make the first pass or two, then set over to the right about .020 to .040 thou for no reason and continue on its way completing the rest of the letter

I have already checked all of the wiring, the cables and thier routing, I have been through the trouble shooting supplied here by Tech Support:

Troubleshooting: Rail is getting hung, stuck, or out of alignment (x35/x50 machines)

Rail is getting hung/stuck/grinding. A rail getting stuck/hung is typically one of two things, either a wire/connection issue or the grub screw on the ball screw isn’t tight on the side that isn’t moving. One thing to remember, stall homing adjustments do NOT affect rail movement during cutting or jogging. If the rail isn’t moving correctly during cutting or jogging, incorrect stall homing is an affect of this greater issue. Note: Any time the axis moves in the opposite direction than what yo…

and told to run this as a test file:

Having issues with your machine? Run this test file (Circle, Square, Diamond)

Having issues with your machine? We created a test file programmed using Carbide Create that we know works correctly. Please download and run this file to verify things are operating correctly. This will help narrow down any issues you may be having. Stock size: 5" x 5" x .25" Calls for 1/8th bit (upcut/downcut is fine) Design file (Carbide create v648): temp.c2d (12.6 KB) Run this Gcode file: CSDTest.nc (30.5 KB) Gcode Preview: [Test File Preview]

which seems odd as thats an awfully small test piece 5"x5" on a machine that travels 32x32, and even odder is the fact it cuts through the outside circle before ever making the internal cuts, but fortunately I used a thicker piece of material and everything stayed in place and didn’t break a bit or damage anything else.

BTW… in the future when supply a test file, and a bit to use, the feed and speed should be included as well… for this one here 1/8" end mill, 100hz/6000rpm seemed to be ideal for the feed speed that was programed into the test file which not having the software that was used to create it, I couldnt open and get the info from… I believe the machine reported 80ipm

and theres the test cut done, three were no issues

Tomorrow I have a large production run planned, Ill be sure to give the connections and wires a going over again, after I clean and lubricate the machine, and warm up the spindle… Ill report back and let you know how it goes

Thank you again very much for taking the time to assist me with this, I really appreciate it, especially over the holiday. I am sure together we can get this sorted out.

Have a good night, Robert

[RobertD (Robert D Mroczka)]

I would like to add briefly a bit of my background just so we are on the same page
Went to a trade school for high school, my trade was Electronics with a focus on AC/DC and Digital Electronics, Microprocessors, and Data Systems

I spent the next 18 or so years after high school as a custom fabricator and installer in Mobile Electronics… from air conditioning and cruise control to remote starters and custom high end audio I did it all

The next ten years or so I worked in manufacturing and ran Laser Cutting and Punch Press CNC machines (Amadas and Strippets)

I then went back to school and got two degrees in Web Design and Web Development and did that as a freelancer for the next 10 years

after taking a year off to figure out what was next for me, I went and rebuilt a 5x10’ 35 year old CNC from the ground up including the wiring, electronics, thrust angular bearings in the spindle, rebalancing of the spindle, new X and Y drives and transmissions (it was a Rack and Pinion System), installed new controller software and brought the entire machine back to an almost new state going from nearly .060 to .080" run out and lash issues to a tolerance of .004" over the entire table 60" by 120"…

Heres some pictures of that job earlier this year

and 3 months ago, I began the process of setting up my own woodworking shop, which brings me to you today.

Thanks again for your help, have a great day ~ Robert

[Aiph5u (Aiph5u (not affiliated with Onefinity))]

Hey Robert,

although I post a lot in this forum, this should not lead to the impression that I am the support. I do this in my spare time and don’t get paid for it. I am an ordinary user like you. The impression possibly comes from my capability to type with ten finger into a keyboard as fast as I can think and that I have a good bookmarks collection for the links and quotations that I am known for, so that I can always make good postings very fast. Most of my postings is copy and paste of existing material that I stored for future use.

The bug report you sent me contains, besides your config file, the bbctrl logs of the last day and the syslog of the last three days. But you should know that every time you hit ‘play’, it puts the entire config file into the log file again and again. These rotated bbctrl.log files contain tons of information. At first sight I cannot see that it says “here I shifted everything by 0.05 in”. What lacks here is a matching analysis tool, but the Buildbotics was written from scratch by one single person and this person did not provide such an analysis tool (at least I have none). Did you have a look at bbctrl.log in your bug report yourself, to make yourself an image?

I asked for the error message on display, not for the bbctrl.log and syslog of the last days. Usually people hold their photocamera on the screen and make a screen shot.

The syslog file contains what a computer does usually, nothing special, and the bbctrl.log contains tons of redundant data. It contains, however, in a more or less cryptic way, what you do. Although I already read in these files, e.g. to find my offsets after a power outage, and understand the most of what it says, I don’t own a tool to specifically analyze this log file to sort out the relevant things quickly. That would be necessary because I am a human and you don’t want me to read with my eyes what you have milled in the last days? Nobody could do that without analysis tool (and without charging money). I really suggest that you have a look at it to understand this.

Furthermore, I don’t find it probable at all that the controller shifted your second pass by 0.05 inches. Really not. The experience tends in a totally different direction. I listed the possible causes above, so I would see sense only in checking the possible causes mentioned above, one by one, with the Sherlock Holmes method: Eliminate them one by one, so in the end the answer remains.

I would say that the possible cause is rather there where I pointed you to above. If you’d zeroed or homed or had a power outage between the passes, then it would be easy. But finding the other causes is less obvious, you need to investigate them one by one.

If you still suggest that there is something wrong with the firmware version, I already told you where to post the issue.

@RobertD

the ones that flash on the screen on shut down are there for less than 3 seconds

This is the shutdown issue mentioned above, and as Onefinity told you in the support documents linked above, it can be safely ignored.

@RobertD

I included the Bug Report Log in the original post, please reference that, if you can tell me how to open it, I am happy to tell you what the errors are…

Most people open a tar.bz2 by clicking on it on their desktop. I open it with   tar -xvf filename.

@RobertD

as for zeroing, I set the zero with the laser by getting my focal point at the corner of the piece, space it with the supplied spacer from Jtech, setting the Offset on the machines X, Y and Z, my my origin point in Vcarve at 0,0…

My tool path for cross hatching and outlines is all one tool path… theres no need to reset the the zero… Vcarve allows for both to be made simultaneously

okay this is information. But if you don’t re-zero between the passes, this would lead to assume it could be an error that is made during using your CAD/CAM program.

@RobertD

I have seen and heard the stepper motor skip

If not, then you lost steps, then please follow the recommendations for removing all possible causes for loosing steps or erratic axis movements above. Checking the pins of the connectors may not always be enough when the tin-plated pins are unreliable by themselves and never meant to be used on the outside of a device and on moving machine parts, or when the machine lacks a serious cabling system, or has no strain relief on moving cables and connectors, or uses stepper cables that are not shielded, etc. etc.

PS: It is 02:17 AM in Central Europe. Good night!

[RobertD (Robert D Mroczka)]

Brother, you are doing a terrific job with the feedback, I really appreciate it and grateful to have another set of eyes on this stuff, thank you for informing me on how to open the bug report, I did not know

Ill try to capture the stuff on the shut down screen, as you said however Onefinity says disregard this

Ill do what I can tomorrow while running to capture video or images of the exact problem Im having with carving and the laser and maybe we will have better information to work with, its very hard troubleshooting something without seeing it actually happen.

there was one thing I forgot to mention… yesterday I stated I spent nearly an hour trying to get it to connect to WiFI… the reason for this, is I would get errant keystrokes filling in the password section of the WiFi… and not be touching the keyboard… what I mean was, there were key strokes filling in the password with random characters and I didnt have the keyboard up on screen, nor was I touching the physical keyboard I use (and theres brand new batteries in that as Im trying everything here)

Now this almost sounds like a EMI or RF issue, but the odd part was, the spindle and the laser were both off at the time and nothing else was powered in the shop other than the lights, and everything is on its own dedicated outlets, to dedicated breakers in the panel, the wiring in the building is new, I have had the electric company out once to repair a loose lug in the meter that was caught before the CNC was installed, I have a 240v 3phase feed coming in and dedicated 20amp 110v circuits coming out… 1 breaker… 1 outlet all rated 20amp

I like your suggesting of re-examining the molex connectors, the strain reliefs and the routing again, Ill reconfim all system grounds, the shielding on the spindle cables and the power caples and the connections on the Onefinity extenders

You sound like an extremely informed person when it comes to this stuff and again, I just want to thank you so much for your time

Have a great night man, Ill get back to you when I have more information or a resolution.
~ Robert

[RobertD (Robert D Mroczka)]

Good Morning,

I would like to first say than you for your time and help in this matter and give you a quick update.

I believe the problem is an external one and has to do with the electricity supply to the building, I do not have all the details as of the moment, I am requesting a service check.

What brings me to this conclusion is upon looking at all of the problem that cropped up unexpectedly and that they come and go, along with the fact my T-Mobile home internet router died 3 days ago and the service tech reported to me that he could see I was having connectivity issues starting a week ago Friday, the exact same day I developed CNC issues now has me believing this is a line noise/voltage issue coming in from the street.

I have gone through and rerouted the X and Z wires, I have again checked molex connectors and my experince is telling me the source of the problem is not there

Ill update you when I know more in a few days.

Thanks again, your support in this matter has been exceptional.

Have a Great Day,
Robert