[Khalid]

one of ur ball screw having problem..may be some dirt hindr the balls to rotate freely...i was getting same problem and now i solved..

[cxixer]

OK, I've about had it with this oversized paperweight. I cut two squares... SQUARES... one came out perfect, the second one, beside the first on the x axis lost steps somehow on the y axis. What bugs me is that it screwed up in the same plane as where it cut correctly, the router just moved over.

What bugs me more is that there are two motors driving the y axis (where it "lost steps" today), so if there is some kind of resonance issue and missed steps due to this issue, it must be happening on both of these motors at exactly the same time, which doesn't seem right.

I've tweaked and adjusted and came up with notta, and today I wasted a half sheet of wood, which is a big no no in my shop, I don't have "scraps", everything gets used - unless of course it gets all screwed up on the cnc.

So unless there are any more ideas to fix the R&P setup, I'm going to be ordering some 1/2-10 ACME screws and some of widgetmaster's acme nuts and re-do the drive systems on this thing. It doesn't matter how "high tech" the tools are if they're not accurate. Not only can I get better resolution from acme, it will give me a chance to re-do my z axis and carriage in precision bearings, and go through everything and make sure I didn't overlook anything that could be causing this.

[Steve-Tee]

Howdy Cxixer,

Please forgive me if this question has already been asked, but is the motor current within specs?
It appears that this problem has shown up recently, without any obvious reason. If there is not enough motor current, this could explain why they are inclined to miss steps.

Are you seeing this effect when the router is not in contact with the work?
Perhaps if you were to do those squares again, but with the router clear of the work, you might get to see those missed steps.

My machine is all home made, as are the drivers and power supplies. While experimenting, I found a simple way of measuring the load on each motor.
By placing an ammeter between the motor common and its normal point of connection, you can see how hard the motor is working.
Increased labour translates to increased current, for a given speed.
This allowed me to find those binding points in my machine's slide ways.

When the current for the motors was insufficient, I would often observe stalling. This was more apparent at lower speeds, due to the lower level of actual power being supplied to the coils. At greater speeds, the voltage rises, for the same current, and the raw power level rises with it.
This translates to extra torque, over a certain portion of the motor's speed range.

After seeing what you have been going through, I don't think that resonance is an issue. It bugs many of us, but appears not to be the cause of your present concerns.

Two motors wired in parallel can stall each other when there is not quite enough current to keep them happy.
If you could take a look at the DC current level, you could find an answer there.
While at it, also inspect the filter capacitors in the power supply.
If these are losing their ability to filter out the mains hum, you could be getting a modulation effect in the DC feed to the motors.
This will cause an intermittent effect, often related to the actions of other motors in the system.

Hope this helps.


Cheers,
Steve.

[cxixer]

Howdy Cxixer,

Please forgive me if this question has already been asked, but is the motor current within specs?
It appears that this problem has shown up recently, without any obvious reason. If there is not enough motor current, this could explain why they are inclined to miss steps.

Are you seeing this effect when the router is not in contact with the work?
Perhaps if you were to do those squares again, but with the router clear of the work, you might get to see those missed steps.

My machine is all home made, as are the drivers and power supplies. While experimenting, I found a simple way of measuring the load on each motor.
By placing an ammeter between the motor common and its normal point of connection, you can see how hard the motor is working.
Increased labour translates to increased current, for a given speed.
This allowed me to find those binding points in my machine's slide ways.

When the current for the motors was insufficient, I would often observe stalling. This was more apparent at lower speeds, due to the lower level of actual power being supplied to the coils. At greater speeds, the voltage rises, for the same current, and the raw power level rises with it.
This translates to extra torque, over a certain portion of the motor's speed range.

After seeing what you have been going through, I don't think that resonance is an issue. It bugs many of us, but appears not to be the cause of your present concerns.

Two motors wired in parallel can stall each other when there is not quite enough current to keep them happy.
If you could take a look at the DC current level, you could find an answer there.
While at it, also inspect the filter capacitors in the power supply.
If these are losing their ability to filter out the mains hum, you could be getting a modulation effect in the DC feed to the motors.
This will cause an intermittent effect, often related to the actions of other motors in the system.

Hope this helps.


Cheers,
Steve.

Steve,
Thanks for the reply. I will check the current draw on the motors. One thing though, The motors are not wired together, I have a xylotex 4 axis (425oz-in) kit, each is on it's own connection. Would that make a difference? And, once I have an ampere reading, what do I compare this number to?

Please forgive my inexperience with in-depth electronics, but how do I check the filter capacitors? I'm not sure how to go about this.

[Steve-Tee]

Howdy Cxixer,

As the motors are not wired directly together, I am guessing that they have their own driver units. This will change the way they behave, often making them less likely to affect each other.

The motors on my machine are marked with their current rating. In mine, it's 2.2 amps. They'll work at higher amps, but may overheat.
At lower amps, stalling is more likely.

To checK the condition of the capacitors in the power supply, the simplest way is to take hold of the motor shaft and apply a small amount of turning force.
When the power is on, and the motor is resisting the force you are applying, you may feel a minor buzzing.
This will be in the order of 100 to 120 events per second, and may become audible.
It occurs to me that you might be using a switchmode power supply as opposed to a mains transformer and rectifier. It that's the case, the foregoing will not be of any use, due to high switching frequencies.

Instead, you will need to use a voltmeter to monitor the DC bus voltage, looking for load induced variations.
Minor variation is ok, but if you are seeing a few volts of variation, then there may be a regulation problem with the power unit.

First things first though, check the motor current.
Perhaps even try raising it slightly.

Here's hoping you can get your machine working properly soon .


Cheers,
Steve.

Steve,
Thanks for the reply. I will check the current draw on the motors. One thing though, The motors are not wired together, I have a xylotex 4 axis (425oz-in) kit, each is on it's own connection. Would that make a difference? And, once I have an ampere reading, what do I compare this number to?

Please forgive my inexperience with in-depth electronics, but how do I check the filter capacitors? I'm not sure how to go about this.

[harryn]

Hi, I was just thinking, since you (might) be having an electronics problem, perhaps re-post the question and some information on your exact setup in the electronics area of the forum.

Either here in the more general stepper motor / driver area

http://www.cnczone.com/forums/forumdisplay.php?f=193

or here in the Xylotec specific area.

http://www.cnczone.com/forums/forumdisplay.php?f=82

I would be tempted to read through the xylotec threads, and post the question in the more general stepper one, as the problem might be noise or something else than an actual xylotec specific problem.

The more detail you provide in the starting post about your exact setup (power supply, motors, computer used, break out board, pinion size, pictures, etc), the faster they will be able to help.

I am actually surprised that an axis with 2 each 400 ish oz in motors is not able to drive that setup, esp. if you have more the axis relatively easily by hand with no binding spots.

Assuming it is electronic related, perhaps run the mach III diagnostic of the parallel port signal. I don't know how to do this, but it is in the on-line manual.

[cxixer]

Hi, I was just thinking, since you (might) be having an electronics problem, perhaps re-post the question and some information on your exact setup in the electronics area of the forum.

Either here in the more general stepper motor / driver area

http://www.cnczone.com/forums/forumdisplay.php?f=193

or here in the Xylotec specific area.

http://www.cnczone.com/forums/forumdisplay.php?f=82

I would be tempted to read through the xylotec threads, and post the question in the more general stepper one, as the problem might be noise or something else than an actual xylotec specific problem.

The more detail you provide in the starting post about your exact setup (power supply, motors, computer used, break out board, pinion size, pictures, etc), the faster they will be able to help.

I am actually surprised that an axis with 2 each 400 ish oz in motors is not able to drive that setup, esp. if you have more the axis relatively easily by hand with no binding spots.

Assuming it is electronic related, perhaps run the mach III diagnostic of the parallel port signal. I don't know how to do this, but it is in the on-line manual.

I will be asking in the xylotex forum as well. I am suprised as well, as the motors are very strong and the axis they're messing up on is on precision linear bearings, very low resistance. This is another reason I'm inclined to think it is a mechanical issue.

After I tweaked the R&P, there was some improvement. So, I'm pretty convinced that's the problem. Although I will check out the electronics just to be sure everything everything's OK.

I've ordered the supplies to replace the R&P with leadscrews. I'm also ordering some precision bearings for the z axis, that way the weakest link on the machine will be the x axis fully supported skate bearing setup, which is very sturdy and no noticable variences or flat spots right now. It should make for some nice upgrades, the z axis bearings have some noticable flex in them, which is one of the reasons I don't up the ipm very much. Hopefully I'll get better performance out of this new setup!

[ger21]

Resonance is a common problem with the Xylotex and those larger motors. That could be what you're seeing.

[harryn]

Hi, I was in the garage working on my router project, and of course, thinking about this thread, as my build is not so different in concept. A couple of things came to mind.

a) A lot of the stepper motor drivers reduce the current to the motor when they are idle for very long (seconds) to keep the motors from overheating. This "feature" is fine for screw based systems, but for a direct driven pinion, it dramatically reduces the holding torque.

I don't know how to disable this, but if you have not done it, that is a place to look.

b) Resonance Dampers - I just happen to have a couple of Vextra / Oriental Motor "Clean Dampers" model D9CL-12.7. I ended up with them when I was buying some stepper motors from an electronics junk yard locally. The dampers hid the fact that the motors are 5 phase, so useless to me since my drivers are linistepper unipolar 2 phase drivers.

I measured the motor shafts, and they seem to be approx 0.312 - 0.315 in, which is a bit odd, but perhaps you can adapt them to your motors as a test.

If you want them, just send me a PM with your adr and I will put them in the mail.

c) On the chance that you are picking up static in your system (machining a non conductor like mdf), then perhaps your wiring needs a bit more protection. Just a wild idea, but perhaps cover the wire with Al tape and ground it out. Static charges are funny and can take on remarkably high potentials - really can cause havoc. It is interesting that some of your challenges seem to come as summer is coming on.

[harryn]

Hi - just curious - does this sound like your problem ?

[ame="http://www.youtube.com/watch?v=9RoOZs9LrYo"]YouTube - 20 inches per minute resonce in stepper[/ame]

If so, there is a long thread in the stepper section about dampers, and xylotec related aspects as well. Seems to be a common problem.

[Mike Stevenson]

Wow, what an amazing effect that damper has on that wild, untamed motor.

[cxixer]

UPDATES!

Actually, I've seen that youtube vid on the thread about dampers, no, that's not what my motors were doing. It is now...

I know this isn't the fix that anyone with similar problems would want to hear, but here's what I did. I went ahead and ordered enco 1/2"-10 acme rods, replaced the x, y, z, a axes all with this, and some of wigitmaster's excellent quality delrin acme nuts. Replaced the z axis drawer sliders with NSK 8" linear ball bearings. The conversion took from my last post until now to complete, but I have cut a few things already tonight with great results!

Measured specs:

ZERO measurable backlash - X axis (tested only this axis - down to .001")
Test part (6x9 speaker ring) - accurate to .020"


This is what I have so far. I need to cut a bunch of parts, so the new parts and modified parts are not yet painted. Just assembled to cut what I got behind on and I'll take it apart again at a later date to finish it - already have the paint!

Issues still to fix:

Need to tweak all axes so they don't stall. The "couplers" I have from the shafts to the screws are not perfect, and cause wobble, stalling the motors. I'll replace with proper couplers shortly. I also will make/buy some of these dampers and see if that helps at all. I forgot that my z axis use to use 1/2-13 allthread, instead of 10tpi, and copied over the 20800 steps per inch in mach3 to all axes, and they were all stalling around 30-35ipm. Oops. Well, re-calculated it using the program I wrote a while back, 16000 steps per inch, and all axes ran fine at 35ipm... no further testing was done as I started cutting, already had the gcode set up for 30ipm, so that's all I needed. The parts came out perfect by any tape measure standards and that is what I'm going for!

With these screws I'm already getting great results - but with some additions for self-adjusting anti-backlash, precision couplers, some dampers, and maybe a tweak here or there, I'll have better resolution, accuracy, and reliability than I ever had with the rack setup.

Someone remind me why I didn't choose the acme route in the first place? It would have been cheaper, too.