[jharvey407]

Alright guys, I need some help, and I'm sure this group can make short work of my problem.

In short my Y axis is stalling during rapid moves. The axis stops moving and the stepper emits a high pitch whine. It doesn't happen on every move but at what seems like random intervals. I did a bit of trouble shooting yesterday and have listed my findings and machine specs below. Let me know what you think. I need to get this problem solved and preserve the rapid speed of the machine if possible.

Machine Specs (pic attached)



The Y axis rides on V-bearings and V-rails (hardened steel) from Pacific Bearing.
The drive system is the CNC router parts belt drive (3:1 1" diameter pinion) mated to racks from Moore Gear (20 DP).
I am using a 425 oz stepper from Keling Inc. (KL23H286-20-8B) powered by their 24V power supply, the KL-4030 stepper drive with 1/8 steps and the C33 multi function router board from CNC 4PC.
I am using Mach3, SPI are approx 1500, accel 25, rapid speed 500 ipm.

Troubleshooting and testing

The first test was to confirm that the Y axis was not binding. I checked the preload on the V bearings and everything seemed fine. I then disengaged the pinion from the rack and moved the Y axis back and forth. It moves very smoothly and does not take much effort at all to get it moving.

The next test was jogging the Y axis back and forth using Mach3. The Y axis moves smoothly and quickly, but occasionally stalls while in motion. The axis stops and the stepper emits a high pitch whine.

The third test was to check for resonance. I removed the dampener from my Z axis and placed it on the Y axis motor. I jogged the Y axis with Mach3 and got the same results as before.

Test number 4 was performed with G code. I wrote a quick test program of a series of G0 moves back and forth along the Y axis using random distances Whit an M47 command to keep the program looping. The motor did not stall on it's own, but I was able to get it to stall sometimes by applying resistance to the Y axis, but not every time. I would apply pressure with my finger against the side of the Y axis, as if I were trying to stop it from moving. Some of the time it would not stop and in fact had a considerable amount of force pushing against my finger. Other times it would stall immediately with almost no pressure from my at all.

I tried varying the engagement pressure of the pinion by adjusting the turnbuckle, but it had no effect.

I do not have a torque curve for the stepper. Does anyone have one for this particular motor?

Any idea's what might be causing this?

Thanks,
J

[zool]

Have you double check the integrity of the wiring and connections between the controller board and the steppers?

Have you checked the voltages that the parallel port is putting out, both with load and no load?

Have you checked the voltages that the controller board is putting out, both with load and no load?

Have you switched the stepper motors?

Have you done a search for "stepper emits a high pitch whine" and "stepper stalls"?

Also, maybe the stepper motor thinks you are overworking it and the high pitched whine is its trying to communicate that to you.

I am having trouble setting up MACH3 and have had really great help from folks here on CNC Zone, though to no avail. I am beginning to suspect that my wife has cast a spell on my new CNC because I am not paying sufficient attention to the Honey Do list.

[jharvey407]

Have you double check the integrity of the wiring and connections between the controller board and the steppers?

Yes

Have you checked the voltages that the parallel port is putting out, both with load and no load?

Yes

Have you checked the voltages that the controller board is putting out, both with load and no load?

Yes

Have you switched the stepper motors?

No

Have you done a search for "stepper emits a high pitch whine" and "stepper stalls"?

Yes, no joy.

Also, maybe the stepper motor thinks you are overworking it and the high pitched whine is its trying to communicate that to you.

Since it happens at random intervals I don't think I am overloading the stepper.

I am having trouble setting up MACH3 and have had really great help from folks here on CNC Zone, though to no avail. I am beginning to suspect that my wife has cast a spell on my new CNC because I am not paying sufficient attention to the Honey Do list.

It sounds like you need to get your priorities straight.

[jharvey407]

The most interesting aspect of this problem is the fact that sometimes I can apply considerable force against the movement of the Y axis and it does not stall, while other times if I just barely touch it, it stalls.

[jsheerin]

I have those motors and have the torque curve loaded in a spreadsheet. With the cnc router parts r&p, you should have enough torque at that speed / acceleration to not stall.

That power supply voltage is pretty small for those motors. They'd run better on ~80V. I run mine with a G540 and a 48V supply. I'm not sure if that would be causing your problem or not, but iirc low voltage will limit how fast your motors can go. Some of the smaller motors like the ~380oz*in sizes are better for use around 40-50V. It looks like your drive is limited to 40V, so those motors would have been a better choice (it's what I wish I had, but they were just coming out when I bought my motors).

[atwooddon]

Sounds like switching the steppers between x and y (can probably just swap cables at the controller) would be the next step. If the problem follows the axis swap, then the stepper or wiring is a likely culprit, or if the problem stays with the Y axis, a mechanical issue.

Any chance this could be some sort of electrical interference?

Don

[wildwestpat]

Hi J

First a question can you start from a known point (NOT the end of travel) and run the axis from one end to the other several times and have it come back to the known point? This checks for lost steps. And can you cut a circular path that is a circle? By a circle I mean cut 370 degrees and still have no step in the outer circumference or the inner diameter where the cutter stops. If these tests are OK then I have three comments to make about possible problems that you might be having.

1. The axis could be crabbing for mechanical reasons to the extent that the power needed to shift the axis is varying with time. When the axis has stalled under your light finger pressure turn off the power to the steppers and check the torque necessary to move that axis. Compare with the torque needed when the axis is moving freely and light finger pressure has no detrimental effect. If the two torques are very different the problem is mechanical or if you have drive on both sides of the axis then it may be that one side is losing steps or you have couplings that are slipping.

2. IMHO you may be missing steps and it is good practice to have decoupling across the power inputs to the drive amplifiers and on the inputs to your breakout board. Suggest three capacitors in parallel 50mfd (voltage to be more than the supply voltage by a margin) electrolytic with 0.01mfd disc ceramic and a 1,000pF disc ceramic. All three capacitors mounted by their leads directly to the power input terminals NOT to the strip connectors or the power supply terminals. Each driver as well as your BOB should have capacitors fitted across the supply. It might also be a good idea to put two disc ceramic capacitiors across any switch inputs on the BOB. If you have a digital volt meter (DVM) check that the signals on the switches comply with the input specification for the BOB. The signals via the LPT (parallel printer socket) should be OK if you are using a desk top machine. BUT if you are trying to use a laptop then the voltage levels might not be enough for the BOB and earth loop problems could be enough to cause missing steps to only occur when the current demands are high or in a particular combination. Do the tests in the order I have indicated and let us know how you get on if you need more help.

3. You may have earth loop problems. From the picture posted I can not see how the earthing of the power unit the amplifiers and BOB has been done. I can see that you have a long lead from the power unit to a two way high current terminal strip. Each motor driver should have its leads connected direct to the power supply. The aim is to have no shared lead in any of the supplies. The BOB looks to be one of the opto-isolated variety. If so the PC should be earthed to the zero volts of the BOB and the BOB earthed to the power supply. If the power unit is not provided with terminals suitable for all the necessary wires use the thickest wire possible and move that two pole buss strip to be as close as practical to the power terminals on the power supply. For the sake of clarity by earth I mean the zero volt terminal on the power unit's DC output. and the mains earth should also be connected to that point with a single connection. I would also put disk ceramic capacitors across the switch inputs to the BOB and make sure the leads to those switches are screened and the BOB has the pull down resistors correctly connected.

Also see if anyone else offers advice but make sure you keep the thread up to date so we all know what you have tried and in what order.

Hope this helps - Regards - Pat

[zool]

Have you checked to see if the chains are binding or chain slack is building up oddly one one side or the other of the pulleys?

Except for this one small glitch, it certainly seems all your planning paid off; the half-lap gooseneck being a stellar example.

I do have MY priorities straight, it is that my wife, a clinical psychologist, has a differing opinion, which is obviously wrong. That being said, there are at times when no battle can be won irrespective of the logic applied. There are times it is better to lose the battle than jeopardize the war; all metaphoric of course.

[jharvey407]

I'm 99% sure I have the problem solved.

I'll answer all of the posts and conclude with my findings.

Thanks for the help

I have those motors and have the torque curve loaded in a spreadsheet. With the cnc router parts r&p, you should have enough torque at that speed / acceleration to not stall.

I think you're right. I have run this machine with rapids as high as 750 ipm. The machine ran fine, but the speed seemed a bit agressive for the size of the machine.

Would you send me a copy of your spreadsheet? I don't have any information on the torque etc.

That power supply voltage is pretty small for those motors. They'd run better on ~80V. I run mine with a G540 and a 48V supply. I'm not sure if that would be causing your problem or not, but iirc low voltage will limit how fast your motors can go. Some of the smaller motors like the ~380oz*in sizes are better for use around 40-50V. It looks like your drive is limited to 40V, so those motors would have been a better choice (it's what I wish I had, but they were just coming out when I bought my motors).

I've wondered about this. I'll look into a larger power supply. I'll have to check the spec on the driver I have.

Sounds like switching the steppers between x and y (can probably just swap cables at the controller) would be the next step. If the problem follows the axis swap, then the stepper or wiring is a likely culprit, or if the problem stays with the Y axis, a mechanical issue.

I've tested the Y axis for any mechanical binding and the Y axis moves like butter on the V rails. The motor swap is on the list of troubleshooting steps. I'll let you know how it works out.

Any chance this could be some sort of electrical interference?

I looked at that. The motor cables are sheilded and grounded. I removed the ground and ran a couple of tests. At first I thought the problem was solved, but it came back. So I don't think it's interference.

Hi J

First a question can you start from a known point (NOT the end of travel) and run the axis from one end to the other several times and have it come back to the known point?

I've done a lot of testing and it's not lost steps. It's a complete stall.

1. The axis could be crabbing for mechanical reasons to the extent that the power needed to shift the axis is varying with time.

It's not mechanical binding. The Y axis glides like butter.

2. IMHO you may be missing steps and it is good practice to have decoupling across the power inputs to the drive amplifiers and on the inputs to your breakout board.

I've have not heard of this, but I am going to look into it. Thanks for the info.

3. You may have earth loop problems.

I have looked at this and I don't think it's a factor. One of my main businesses is marine electronics, which are very ground sensitive, so I am well versed in ground loop issues.

Have you checked to see if the chains are binding or chain slack is building up oddly one one side or the other of the pulleys?

Actually the chain was only used during the prototype stage. I have switched to a R&P drive system, and I have checked it for binds etc. and it is running smoothly.

Except for this one small glitch, it certainly seems all your planning paid off; the half-lap gooseneck being a stellar example.

Thanks, but I can't take all of the credit. Between this forum and about 80 different blogs that I read, I learned all that I needed to assemble a nice collection of off the shelf parts into a good, light commercial CNC machine.

I do have MY priorities straight, it is that my wife, a clinical psychologist, has a differing opinion, which is obviously wrong. That being said, there are at times when no battle can be won irrespective of the logic applied. There are times it is better to lose the battle than jeopardize the war; all metaphoric of course.

I hear ya brother.

Ok, so for the moment of truth!

It seems to be the stepper motor.

I swapped drivers, with no change in performance. So I swapped the Y axis stepper with the Z axis stepper and the motor is running fine. It is has tons of power. It takes an enourmous amount of force to make it stall. In fact I bumped the rapids upto 750 ipm and it still takes a lot of force to make it stall.

So I'll send an email to Keling and see what they say about warranty. I'm not sure what their warranty policy is, but I've had the stepper for only a few months.

Thanks for all the help.

[tjskcnc]

So I'll send an email to Keling and see what they say about warranty. I'm not sure what their warranty policy is, but I've had the stepper for only a few months.

When I purchased my motors from Keling, over a year ago, one was "bad". They replaced it with no problem. They even shipped the replacement before I had the bad motor in the mail to them. I hope they are still providing excellent service. Most of the vendors discussed in this forum provide excellent customer service. That's why we use them. Let us know how they handle this.

[wildwestpat]

A simple way of checking steppers is to simply short the wires of one set of windings and see if you can turn the spindle. Repeat for the other winding. Then see if the spindle turns freely with all wire ends floating free. This test as there is no need for drive amplifiers or electricity.

Regards - Pat

[jharvey407]

A simple way of checking steppers is to simply short the wires of one set of windings and see if you can turn the spindle. Repeat for the other winding. Then see if the spindle turns freely with all wire ends floating free. This test as there is no need for drive amplifiers or electricity.

Regards - Pat

I'm not sure if I understand this. I have removed the motor, and the spindle turns freely with the wires not connected to anything. So I am assuming that is a good sign.

As far as 'short the wires of one set of windings', do you mean to connect the A+ leads to the A- leads, with out any other type of connection?

What should the result of this test be, and what does it indicate?

Thanks for the help.

[ger21]

As far as 'short the wires of one set of windings', do you mean to connect the A+ leads to the A- leads, with out any other type of connection?

Yes.

It should make the motor very hard to turn.

[jharvey407]

Yes.

It should make the motor very hard to turn.

I see.

I love this forum.

Thanks guys.

[zool]

If there is a question about the stepper motor, it seems that it is more prudent to just switch motors, than switching this wire to that terminal and visa-versa.

I have an aversion to the magic smoke being released!

[wildwestpat]

Shorting out a winding will cause any movement of the shaft to induce a current in that winding which will resist it being turned. The result is that IF shaft is next to impossible to turn then the motor magnets are up to strength AND the winding is free from breaks. This should be repeated for the other windings.

Being able to turn the shaft with all the lead ends isolated means that the rotor is free to turn i.e. the bearings and internal tolerances are probably OK.

This is not a definitive test but is a good one to apply to a suspect motor or one that you are about to purchase or as a dead on arrival test.

Also note that opening the stepper motor runs a very very high risk of destroying the permanent magnets power.

Good luck - Regards - Pat

[jharvey407]

If there is a question about the stepper motor, it seems that it is more prudent to just switch motors, than switching this wire to that terminal and visa-versa.

I have an aversion to the magic smoke being released!

The motors have already been swapped. The Z axis motor is running the Y axis just fine, indicating that the Y axis motor is bad.

The windings test is just being done to provided more information.

A new motor has been ordered and I am simply waiting for information from Keling about warranty status of the dead motor.

Thanks guys.

[jharvey407]

Well it looks like I've been chasing ghosts.

I had some free time in the shop this morning and decided to continue troubleshooting the dead stepper motor. I did the windings test first and sure enough the motor was very hard to turn with either of the windings shorted. Since the Z axis motor is currently installed in the Y axis I decided to put the suspect motor in the Z axis for testing.

By some miracle the motor has repaired itself and is working fine!

The only conclusion I can draw at this point is that it was a wiring issue and that I managed to fix it while moving the motors and drivers around for testing.

There is a new motor on the way from Keling that I will keep on standby just in case the motor isn't completely ok and dies some other time.

Thanks for the help troubleshooting.

J

[zool]

jharvey stated407:I learned all that I needed to assemble a nice collection of off the shelf parts into a good, light commercial CNC machine.

The sign of a true genius; he did not want to reinvent the wheel.