AR8 Y-Axis Skips Above 6000 mm/min

[dadealeus]

I purchased my AR8 Pro machine around 2 years ago. The machine is rated at 7620 mm/min (300 in/min) rapid travel. I never got this speed (I was only getting about half of that), but figured I was doing something wrong because this was my first CNC machine.

However, my colleague recently purchased the exact same model and I am training him in its usage. I immediately noticed his machine was capable of moving *way* faster than mine (roughly twice as fast, or the advertised 300 IPM). So, it got me curious about my machine.

After doing some testing, I found that it was maxing out around 160 IPM; I timed the movement manually and calculated the movement speed of each axis and they were all topping out around 160 IPM.

I dug around in the settings and found that my machine was set to a max speed of 4000 IPM (about 157.5 IPM) - almost exactly what I was getting in my tests. So, I moved the numbers up to the rated speed of 7620.

This immediately fixed the X and Z axes, but the Y-axis skips steps on anything over about 6000 mm/min - well short of the 7620 mm/min rated travel speed.

My immediate thought is that the stepper motor is wired incorrectly, but I know just enough about electronics and stepper motors to get myself into trouble, so I haven't messed with it.

Does anyone have any ideas?

[dadealeus]

After doing a ton of reading and examining the machine, I don't think it's wired incorrectly. Most people seem to believe the skipping is due to either one of two things:

1) Not enough electrical current reaching the stepper
2) Too much friction on that axis of movement

I turned the machine off and can freely rotate the ball-screw by hand, so it's not a friction issue, which leads me to believe it may be an issue with the current that is reaching the motor - but, again, I'm not sure.

If you have an AR8 Pro, can you check the following and let me know what yours says (all you have to do to get back out of the menu without changing anything is hit the red button repeatedly):

Menu -> Machine Setup -> Voltage Setup

In that screen, all of my arrows are pointing down which, according to my reading, indicates they are all set to high voltage. Just wondering if my machine might be misconfigured.

[Gary Campbell]

Have you considered that it could be the additional mass of the longer ballscrew?

[dadealeus]

Oct 24, 2020 17:56:52 GMT -5 Gary Campbell said:

Have you considered that it could be the additional mass of the longer ballscrew?

Hey Gary,

Thanks for the response. I had certainly considered that, but it doesn't explain why the machine is rated at 300 IPM if it can't do that and my colleague's machine will 300 IPM without any issues on all 3 axes - so I'm assuming there's something wrong with my machine.

Additionally, I tried setting the acceleration down to 10 mm/s² so that the super slow acceleration would offset the additional force required to move the entire gantry to see if that was the issue, but it simply starts to skip steps above a certain RPM - which leads me to believe that something is either misconfigured, or maybe I have a bad stepper?

Current doesn't seem to be an issue because as long as I don't exceed the 6000 mm/s rate that it can hit without skipping, I can move both the X and Y axes at the same time without any loss of power (by holding the X and Y buttons on the RichAuto controller at the same time).

It probably doesn't sound like much of an issue to most people as they wouldn't normally run their machines at that speed anyway. However, I do a lot of work with light-density HDU - which this machine eats through like it's cutting through thin air.

I was able to increase my cutting speeds from the 4000 mm/min to 6000 mm/min, but I'd love to be able to squeeze the rated speed out of the machine to get what I paid for because the material I'm using can be cut at those speeds.

[Gary Campbell]

Hey Gary,



Thanks for the response. I had certainly considered that, but it doesn't explain why the machine is rated at 300 IPM if it can't do that and my colleague's machine will 300 IPM without any issues on all 3 axes - so I'm assuming there's something wrong with my machine.


Trouble shooting is the process of assuming nothing.... and everything...at the same time.  There are almost always more variables in play than we think.


Additionally, I tried setting the acceleration down to 10 mm/s² so that the super slow acceleration would offset the additional force required to move the entire gantry to see if that was the issue, but it simply starts to skip steps above a certain RPM - which leads me to believe that something is either misconfigured, or maybe I have a bad stepper?


You can swap the X & Y steppers and/or drives (one at a time) to see if there is a difference in performance between them


Current doesn't seem to be an issue because as long as I don't exceed the 6000 mm/s rate that it can hit without skipping, I can move both the X and Y axes at the same time without any loss of power (by holding the X and Y buttons on the RichAuto controller at the same time).


Im not sure your test is valid, due to great differences in mass between X and Y motion parts.  You also dont say what speed you test at.  Torque falls off on steppers quickly as RPM increases and in many cases is only 10-15% of the advertised holding torque above a couple hundred RPM.


It probably doesn't sound like much of an issue to most people as they wouldn't normally run their machines at that speed anyway. However, I do a lot of work with light-density HDU - which this machine eats through like it's cutting through thin air.


The quest for more speed seems to be universal among CNC'ers, so I can blame you for trying, but most "advertised maximums" are not the numbers you should reach for on a daily basis.  Any electro-mechanical device run continuously near its design max will fail drastically more often than one one at 60%.  That said, and all things being "new and tuned" and sized properly, 600ipm on a 10mm pitch ballscrew is reasonable with steppers and 50% higher using servos.  Of course, there are "cubic dollars" between each of these steps.

I was able to increase my cutting speeds from the 4000 mm/min to 6000 mm/min, but I'd love to be able to squeeze the rated speed out of the machine to get what I paid for because the material I'm using can be cut at those speeds.

Yes, but.... what about linear rails & bearings, ballscrew bearings and their lube schedule?  Heat in the drives & motors?  Are you sure you want to move your cutting speeds into the "not 100% reliable" realm?  

[dadealeus]

Hey Gary,

Again, thanks for your reply; I like this forum because of how constructive it is.

The current test I performed was just to rule out there not being enough current coming from my 220v outlet to the machine. If a single axis was failing at 6500 mm/s, then running that axis at 6000 m/s while also running the X axis at 7620 mm/s should have caused a failure of one of the motors from the shared current choking the machine. However, that didn't happen, so I believe my outlet is good.

Time is definitely money and I often run 3D carvings that run into the 10s of hours. A further increase from 6000 mm/s to 7620 mm/s is an additional 27% increase in cutting speed (or close to that). That can shave over 4 hours off a 20 hour job.

The increase in heat can shorten the life of any motor, but it's fairly negligible and you can buy replacement NEMA 23s for less than $30/piece. I actually have several 5 amp NEMA 23s on hand from a project I recently decommissioned. However, there's not enough physical space to install a 5 amp stepper under the machine (nor do I even know if the controller is outputting that much current). The ones that are installed are tough to find information on, but they look like a standard 3 amp NEMA 23 and probably have a max torque of around 280 oz/in. If it's as simple as swapping out the stepper, I can give that a try, but my concern is that the controller is simply not outputting enough current.

I have emailed Axiom to see what guidance they can offer. I'll update this thread once I have more information.

Thanks, again!

[dadealeus]

Oddly enough, both the Y and Z steppers appear to be 3 amp steppers with around 280 oz/in of torque, but the X axis (the one that can drive at the proper speed of 7620 mm/min), only has a 2 inch body (the gantry arms are too narrow to fit a 3" on the X-axis without it protruding from the side of the arm) - which likely puts it around 140 oz/in of torque.

So, I believe it's either a lack of current, or a failing stepper.

Again, I'll wait to hear back from Axiom and see how they suggest to proceed.

[Gary Campbell]

Have you checked to see if the drive current settings are correct for their respective motors?  It makes perfect sense that the smaller of the 2 motors would perform at a higher RPM.

[dadealeus]

I heard back from Chad and will post his reply here - a thorough response, as expected (love Axiom's support).

To summarize his reply:

The AR8 Pro that I purchased a couple years ago uses a 24v system for the steppers, whereas the new AR8s use a 48v - so they are essentially providing twice the current - which is why my friend's newer machine outperforms mine.

He also stated that the original specs were basically referring to what the motors were capable of offering, not necessarily the machine's maximum operating rate, and that they have modified the numbers to be more in line with the expected performance of the machine so as to not mislead customers.

Here is the complete reply:

The original max speeds that were listed as 300ipm were in terms of max stepper motor potential and shuttle speeds…not cutting speeds. (This has largely been revised on the more current machines to reflect a more accurate representation factoring in the weight of the machines, but may not apply to ever application).

In fact with the Pro+ series machines, it would not be recommended to cut at higher than 150ipm on average.

Most users have an average of 100ipm for most jobs…but this may very depending upon material and job setup.

*With stepper motors, as the speeds increase…the torque decreases. With the higher weight of the machine, the lower torque at those higher speeds presents a number of problems with step/position loss.

In addition, the higher the speed…the more things like lubrication and bearing condition effect the load placed on those motors.

There are internal Max Speed limits within the handheld controller (found in the AutoPro setup portion of the Menu)…adjustments can be made to these, however, this may result in a heightened risk for loss of position.

The current Pro V5 series machines have been upgraded to use a larger 48V stepper motors (increased over the origin 24V motors) allowing for greater torque levels and those higher shuttle speeds. Though depending upon material and bit selection, the cutting speeds may remain in those same comfortable ranges for most.

[dadealeus]

As an update, the machine is now skipping at 6000 mm/sec on the y-axis as well. Additionally, I'm getting a C error when I try to use the "mill plane" function in the advanced menu (which was previously working as well).

I've emailed Chad to see what they suggest and if flashing or updating the firmware on the controller might address some of these issues.

Will update the thread again when I hear back.

[dadealeus]

Awesome! Got everything fixed! Between a few emails and a phone call to Axiom, we flashed the firmware and restored factory settings (which fixed the error I was seeing in the "mill plane" function.

Additionally, at the direction of Axiom, I opened the case and checked the potentiometers that control current flowing to each stepper. The X-axis was maxed (which explains why it had no trouble), but both the Z and Y axis were not. After maxing both of those as well, I'm now able to hit the advertised speeds without issues.

So, it looks like my machine either shipped with the pots set low, or they lowered through vibration over the months of usage.

Either way, I'm back in business! Hope this helps anyone with a similar issue in the future.

[dadealeus]

Figured I'd do one final post with a video of the solution that worked for me in case anyone else has the issue.

In the case of my problem, I believe that over the months of usage, vibrations from a faulty collet nut were transferred through the frame of the machine to the controller - which slowly caused the potentiometers to reduce the current they sent to their respective steppers. I never noticed this until I recently attempted to bump up the speeds because I was still under the threshold.

-but, it explains why the problem kept slowly getting worse.

I've already placed an order for a new collet nut and I'll likely set the controller on a separate shelf rather than on the machine's frame - this should prevent the issue from occurring in the future.

Here's a short video that describes how to access and adjust the potentiometers.

Please bear in mind that this is done at your own risk. It's always good to contact Axiom first to seek their advice: