[t2krookie]

So I seem to have some inconsistent serious drift in my axis (all three). Is there not a way t determine if the cables are responsible? Measurement of voltage across a circuit or something? Id rather not have to rewire if that is not the problem. I am including a picture o the machines stock wiring for opinions on the need for rewiring.

Thanks

[123CNC]

It's a little unclear to me what you mean by drift. Are motors or axis moving when there is no power, or when there is no commanded move, or there are 'lost steps' or large errors between commanded position and actual position?

First, there are generally 2 categories to consider for locating the source of error. The mechanical system or the electrical control system. A good approach is to use a process of elimination of the usual suspects or most likely causes, trying to work through one category and trying to isolate cause and effect to single dependency.

You have not given enough information to clearly identify the 'bad actor' and he may be elusive as you have suggested with 'inconsistent' results. The inconsistent or ghost problems, being not strictly repeatable, are the more difficult or fun ones to chase down.

But looking at your pictures, and assuming you are experiencing large differences between actual motion and expected motion, on all three axis, I would suggest further troubleshooting on a single axis and pick your staring category. Both mechanical and electrical can demonstrate random or inconsistent error, but I would argue that generally one finds more bad actors in the electrical side of things for 'ghost hunting.' If it were my problem, I would start with checking ground problems first. It is unclear if your stepper motor wires have shielded cables or not. If you have a rather large machine, with long cable runs between the stepper driver and motor, you essentially have a large antenna looking for stray EMF/RF signals which can couple with your step pulse to add or subtract actual moves.If shielding is present but not grounded, try to ground at or near the drive end alone and test, if no change ground stepper end as well, if no change, ground only stepper end and not drive side. If wiring doesn't currently have shielding, make sure your spindle and spindle drive are all properly grounded and none of your line or any supply voltage wiring is running parallel to any of your stepper motor or low voltage control signals, e.g step/dir. to drive.

If you can borrow or buy a short piece of shielded cable for one axis step/dir control signals, and step motor output, test one axis control input first, both if no change.

Try to isolate and capture the ghost before doing any grand change or rewiring, especially since doing work without results is counter productive and dreaded work. Keep in mind though that catching ghosts can be difficult, but many have done it and you will be a very happy camper once you have him in the bag.

[123CNC]

I also meant to say, the low voltage side, i.e. the control signals to the stepper drive are more susceptible to noise then the motor output. That may have been unclear in my post.

Good Luck.

[t2krookie]

The axis are moving /drifting apparently on both the x and y axis during a deep cut. It seems to handle a simple one pass cut ok. (Like a shallow smiley face or road runner relief) but any deeper and the cut seems to drift to the positive on the X axis and it appears a bit on the Y axis as well. I will try one axis at a time as you suggested but from those pictures can u not see if it is shielded cable or not? I tried to include the numbering on the cable but I am unsure of what kind of wiring is optimal for this thype of setup.

[123CNC]

Unfortunately the markings on your Chinese wiring does not clearly reveal if it is shielded cable or not. The one picture of the multi-conductors revealed, your motor power wires, don't show the shield if it is there and if they may have trimmed it back with the outer insulation. The multi pin connectors you pictured often incorporate a way to ground the shield to this connector and then through it's mount or housing. It isn't externally visible, you would have to open one up to visually verify.

However, with your new description of the problem stating that it seems worse or more evident under heavy cuts, or heavy load, then I would be more inclined to look for mechanical error, or even simply over taxing your available power and torque. Your motor to screw connectors or your screw to nut (ball or acme) or even the nut to table or slide are all mechanical joints that might seem tight under no or low load but flex or give way under heavy cuts or high load. The best approach in this case is to load and unload each joint and rig a dial indicator or other means to measure the relative 'play' in the joints, or component motion relative to next component in the system.

[t2krookie]

Unfortunately the markings on your Chinese wiring does not clearly reveal if it is shielded cable or not. The one picture of the multi-conductors revealed, your motor power wires, don't show the shield if it is there and if they may have trimmed it back with the outer insulation. The multi pin connectors you pictured often incorporate a way to ground the shield to this connector and then through it's mount or housing. It isn't externally visible, you would have to open one up to visually verify.

However, with your new description of the problem stating that it seems worse or more evident under heavy cuts, or heavy load, then I would be more inclined to look for mechanical error, or even simply over taxing your available power and torque. Your motor to screw connectors or your screw to nut (ball or acme) or even the nut to table or slide are all mechanical joints that might seem tight under no or low load but flex or give way under heavy cuts or high load. The best approach in this case is to load and unload each joint and rig a dial indicator or other means to measure the relative 'play' in the joints, or component motion relative to next component in the system.

Thank you 123. I'm again embarrassed to say one or at least part of my problem was not torquing the coupling tight enough. This actually removed the majority of repeated error on the x axis. The y axis seems to be humming along just fine. Positional accuracy is spot on after a multitude of tests (just placing objects on the table in various positions and manually putting in coordinates.)
The newest rub is something that had been happening also but had not been paid attention to as I had the major drift I mentioned to occupy my initial troubleshooting efforts.




This second problem is also a boggle. It only seems to happen on the x axis, but as I manually jog with the keyboard over long distances but manual input of coordinates no more that 40 MM can also occasionally reproduce what I can only describe as the motor having a seizure. It will get moving for a bit and then make a high pitched sound at the same time it stops moving. The DRO will show the correct coordinate I was aiming for but it will be off by significant amount when I bring it back to its soft home reference site.
I had read somewhere that sometimes the POTS on the controller should be adjusted as the motor itself may not be getting sufficient amperage but I do not know enough to know how such a condition would manifest. If almost sounds like a stalling (if that makes sense).

Another thought i had was alignment of the motor. (don't know how to correct. I tightened it even i think).

yet another thought was simply the quality of the ball-screw but it seems to work perfectly sometimes thus my mind goes back to a motor issue.)

The accuracy of the movements are spot on each time it moves successfully. Its just this "seizure" activity which is my last barrier (machine wise that is. I still have other learning curves to overcome). Is there perhaps signal adjustments to the motor itself that can be an issue? Or is the power allocation a common problem with stalling motors?


Update : I did peel back the outer wrap on the wiring and it is not shielded. I actually anticipated rewiring and bought shielded cable but I just don't get why only one axis would be causing trouble.

Update #2, 6 runs of smiley face #1 all repeated flawlessly one on top of the other with just lowering Z down a bit each time and zeroing it out. Whatever else is wrong with the motor, I figured I would slow down the acceleration and THAT has eliminated the seizing of the X axis but still frustrating. The motors are supposed to be set 320 steps 3000 acceleration and speed? of 200, and 5/3 where the stepping pulses? I don't remember exactly but still upset its not operating full speed.

[123CNC]

Thank you 123. I'm again embarrassed to say one or at least part of my problem was not torquing the coupling tight enough. This actually removed the majority of repeated error on the x axis.

This second problem is also a boggle. It only seems to happen on the x axis, but as I manually jog with the keyboard over long distances but manual input of coordinates no more that 40 MM can also occasionally reproduce what I can only describe as the motor having a seizure. It will get moving for a bit and then make a high pitched sound at the same time it stops moving. The DRO will show the correct coordinate I was aiming for but it will be off by significant amount when I bring it back to its soft home reference site.
I had read somewhere that sometimes the POTS on the controller should be adjusted as the motor itself may not be getting sufficient amperage but I do not know enough to know how such a condition would manifest. If almost sounds like a stalling (if that makes sense).

Its just this "seizure" activity which is my last barrier (machine wise that is. I still have other learning curves to overcome). Is there perhaps signal adjustments to the motor itself that can be an issue? Or is the power allocation a common problem with stalling motors?

Update #2, 6 runs of smiley face #1 all repeated flawlessly one on top of the other with just lowering Z down a bit each time and zeroing it out. Whatever else is wrong with the motor, I figured I would slow down the acceleration and THAT has eliminated the seizing of the X axis but still frustrating. The motors are supposed to be set 320 steps 3000 acceleration and speed? of 200, and 5/3 where the stepping pulses? I don't remember exactly but still upset its not operating full speed.

Looks as those you caught a ghost on your own account. Your description of 'seizures' or 'stalling' and lowering acceleration has cured the bad habits strongly suggest that you are correct. Some of the nice features of steppers are a.) they run in an open loop control, and b.) they generally exhibit good low end torque, but torque decreases as speed increases. Two ways in that steppers have seizures or stalling, or lose steps, or make unhappy sounds like whining, grunting, grinding are 1) when attempting to exceed the speed/torque limits of the motor or motor and driven components or 2) when driven against a hard stop or obstacle. Sounds as though you have encountered case 1. You may or may not be able to make improvements to your x-axis components to reduce friction or reduce load, you may just be near the mechanical limits of the system. If the latter, you would need to consider mechanical changes if you have a real need or desire for higher speeds, either different pitch screws or going to servos.

Either way congratulations, your'e having fun now.

Yes, depending on your drives and your current power supply, you may have some headroom to increase your DC supply voltage to improve your power and torque output, if your are not already at the motor limits.