Strangely losing Steps in X axis. Help pls!

[d3workz]

Hi to all, I am new here.

I have 2 cnc machines, one is a plasma table (nearly ready but operational) and I have a CNC mill with a 2.2kW router. Both are diys and the latter one I had it for over a year now. Obviously, since I am here I have an issue which I cannot resolve as this phenomenon it is very strange to me, but before I go any further, I want to admit that I am no guru, just your average machinist here.

So my problem is that my CNC mill/router is losing steps in the X axis during a particular job. This is happening after all roughing operations are done and happening during a scalloping operation (which are all same feeds).

This is what I am running:
BOB - MACH3V2.1-L - this is some cheap Chinese board, but so far so good.
Nema 34 1204oz.in 5.6A - all axis
Stepper drivers DM860A
Power supplies 60V 5.8 Amps - Running one power supply for each Nema motor.

Here is where I need your help guys, I ve been cutting aluminum for nearly a year and never had an issue (granted, max speed I got on aluminum is under 2000m/mins)... Now I am cutting a piece of wood at 2500m/min and 275mm/s and I am loosing steps. The funny part is that I have 5mm pitch lead screws coupled with a 1:1 ratio, which makes the moments of the axis super light, I ran operations at speeds of 4000m/min and nothing happened, I have no heating issues what so ever (power supplies does not even turn on the internal fans, the nemas stay relatively cool, stepper drives stay cool) and the machine is losing steps in a scallop operation where there there is very little cutting.

The only thing that I have tested so far is I have operated the Nema motor with our the timing pulley belt and noticed that it skips steps also, meaning that I have eliminated the mechanical part.

Knowing that I normally cut wood at 3000m/min and even higher speeds, is there something different during a scallop operation? I am concerned about this, as I ran programs that are 1 hour long, running over 2500m/min with out any issues but to be honest scallops I've rarely used!! (I ve only used it twice and had no issues)...

One other strange thing is that, I could get speeds of 8500m/min with an acceleration of 800mm/s, when tuning the motors from Mach3, however at the moment I have them set at 5000m/min at 275mm/s with 2us for the step pulse and dir pulse, thus I think I am pretty conservative here!.

May be some one can shed a light on something I am missing?

Please understand that I am threading here before I go and start fiddling with my machine, because I've had no issue before! The only thing is that my normal CNCing is roughing operations, engraving and cutting contours.

Thank you guys - Let me know if you need further details. Any help would be appreciated.

Thanks

Dim

Similar Threads:

• z axis losing steps as I cut.
• Need Help!- Z axis losing steps?
• Z axis losing steps
• Z axis losing steps
• Y Axis losing steps

[d3workz]

Hello to all of you gurus again.... I was hoping that by now some would have shed some light here... but understand that me post is all over the place... and that is how I feel.

I think what I really wanted to ask is, if it possible to have stepping issue which are not coming from the nema motors and the motor drives!??

Could it be an issue with the BOB, or MACH3? Example would it be wise if I increase the STEP pulse to 10 or 15 us? Also, why is the phenomenon only in the X axis, I mean the Y is identical to the X axis... and the Y axis carries more weight for sure!!!

Also one last thing - I failed to mention that I am running x16 mircostepping (3200)... and there are no reasons why I am running this many steps. Its just the motors run very smooth and having some hefty motors coupled with a lead screw, I wasn't concerned about torque...that much.

Also if it helps, when I mentioned of losing steps, I mean that the motor would literally stall for an entire command, or even more, I think. But want I mean is that its noticeable and I am not talking about 1 or 2 steps.

Anyways guys, any information would be helpful at the stage, as I am out of options.

Dim

[cncmakers001]

Hello to all of you gurus again.... I was hoping that by now some would have shed some light here... but understand that me post is all over the place... and that is how I feel.

I think what I really wanted to ask is, if it possible to have stepping issue which are not coming from the nema motors and the motor drives!??

Could it be an issue with the BOB, or MACH3? Example would it be wise if I increase the STEP pulse to 10 or 15 us? Also, why is the phenomenon only in the X axis, I mean the Y is identical to the X axis... and the Y axis carries more weight for sure!!!

Also one last thing - I failed to mention that I am running x16 mircostepping (3200)... and there are no reasons why I am running this many steps. Its just the motors run very smooth and having some hefty motors coupled with a lead screw, I wasn't concerned about torque...that much.

Also if it helps, when I mentioned of losing steps, I mean that the motor would literally stall for an entire command, or even more, I think. But want I mean is that its noticeable and I am not talking about 1 or 2 steps.

Anyways guys, any information would be helpful at the stage, as I am out of options.

Dim

Looks like the motor is not enough torque for the job when losing steps, try not so deep cutting and see if the losing step problem will be gone or not.

[rodw]

I would increase the step timings fairly significantly until you get this working (3-5x). Often a longer step won't impact on your machines performance as other constraints apply. There is still a lot more info required to provide meaningful help. Gearing ratios. lead screw pitch, Link to stepper motor specs, weights each motor has to move.

[d3workz]

Thank you for your comments guys...

I think I found the problem... still not sure though. I think my problem lies in my gear ratios between the motor and lead screw. I have lead screws of 5mm pitch which are then geared as 1:1 to the motor. Therefore, mechanically speaking, if 1 motor revolution gives me 5mm of travel (and this is tested), if I want to run speeds of 3500mm/min, this means that I am turning the motors at 3500mm/5mm = 700 rpm!! Now today I just read in an other forum that Nema 34 motors prefer to stay in the range of 200 to 400 rpms.

If you have any thoughts on this, please feel freee to correct me, or give you 2 cents.

In the mean time, I am preparing to gear the lead screw to 1:2, which in turn I ll be cutting the motors rpm by half.

BTW - I've cut many aluminum with this machine and I can tell you that these motors do not suffer from torque issues at low speeds. Normally I cut aluminum at 800 to 1500mm/min, and I ve never lost a step. Also I ve done silly mistakes during my learning curve and ended impressing my self as to how much it takes to actually stall these motors!!! and I mean, I ve bent steel parts that are 6 mm thick with out any hesitation, until I hit the reset button !

Thanks guys, your feed back is well appreciated. If you need more information, please let me know exactly what you need.

And let me know what is your intake on the rpm theory!

Dim

[ger21]

1) I don't think the pulse width setting has any effect when using a motion controller.

2). Yes, you are most likely trying to go too fast for your motors. What is their inductance? Big motors tend to lose torque rapidly as rpm's increase. And your screws require high RPM's. If you switched to motors half the size, or even smaller, the machine would likely be much faster, and still have plenty of power.

3). Ideally, for a wood router with steppers, you want between 10-20mm of travel per motor revolution. That gives you the best balance of speed and resolution.

4) Bumping the voltage up to 80V would likely help, and give you another 25% more speed.

[d3workz]

1) I don't think the pulse width setting has any effect when using a motion controller.

2). Yes, you are most likely trying to go too fast for your motors. What is their inductance? Big motors tend to lose torque rapidly as rpm's increase. And your screws require high RPM's. If you switched to motors half the size, or even smaller, the machine would likely be much faster, and still have plenty of power.

3). Ideally, for a wood router with steppers, you want between 10-20mm of travel per motor revolution. That gives you the best balance of speed and resolution.

4) Bumping the voltage up to 80V would likely help, and give you another 25% more speed.

Thank you for your thoughts Ger21.

I am glad that you think this may be the issue as well. Before I go and spend more money on bigger supplies, I am waiting for some 48T pulleys and belts, so that I convert my 1:1 to a 1:2 system. I strongly believe, that I would gain the speed needed to cut wood more efficiently but still be able to cut aluminum as I was before. But this i ll be able to verify later on.

Please see the attached documents for my motors and stepper drivers (I managed to find them yesterday) you'll find their specifications there.

So lets say I decide to get bigger power supplies than what I have, which are 60V 5.7Amps, and lets say I get an 80V (as that is the stepper drivers maximum voltage), what amps or wattage should I consider?

Side note: Here I must add that initially I built this machine to cut metals and aluminum. In fact it has a small cutting area of only 850 mm x 500 mm, it has a fixed gantry and weight over 400 kgs. But recently I ve been getting confident with the machine, and ended up having no patients whilst cutting wood HAHA... so I ran into this issue to losing steps.

Thank you very much

Dim

[dazp1976]

Guess what.
I was using nema34 1090oz/in 5.6A motors, 60V, cheap bob, X&Y on my mill. 5mm pitch screws.
Couldn't get more than 2000mm/min even set at 7A on drivers and had some resonance. Even with everything set up pretty light (gibs etc).
I've ditched them for nema23 4A 566oz/in and it's come to life!.
Still running 60v, drivers at 4A and I've gone up to 3000mm/min no problem. DFU screw sounded a little rough for my liking so backed it off a little.

Food for thought.

[rodw]

Its never a good sign when the manufacturer does not provide torque charts. Well I could not find any

[d3workz]

Its never a good sign when the manufacturer does not provide torque charts. Well I could not find any

This is the torque chart of the Nemas I am using.

[Muzzer]

That chart's an excellent illustration of the futility of gearing a stepper motor down to "get more torque". In doing so, you often end up with almost the same final torque due to the way torque falls off so rapidly with speed. In this example, running the motor at 4000pps instead of 2000pps pretty much halves the torque, so a 2:1 reduction would achieve no obvious benefit, although you might now be able to break something expensive if you run against a hard stop. It's not the problem here though.

As DAZSP says, you may be better off simply fitting smaller motors. Closed loop steppers or one of the low cost servos might be worth looking in to, although if you went the servo route, you probably would need to introduce a reduction ratio, as they are lower torque and higher speed. With a closed loop stepper, you will at least automatically recover any "lost" steps - or be able to stop if you exceed a programmed maximum error.

[rodw]

There is nothing wrong with Longs Motors I have their smaller DM542A drivers at 25x microstepping and I'd use them again. The problem you have is a poor match between the motor and your application.
Its also a bit out for my rack and pinion drive machine according to our model.
It can get to 6100 m/sec on 0.5 second, only needs 55 volts but it wants more amps than the driver can give it (needs 6.8 amps)
But we are only driving it at 332 RPM to get to that speed.

[d3workz]

Hi rodw, thank for your comment.

Can you please elaborate a bit more about what you meant that there is a mis-match between the motor size and the application? As I am not sure what you meant.

The machine was intended to work as a milling machine, I just ended up putting a 2.2kW water cooled Router (which I do not regret).

Dim

[d3workz]

Guess what.
I was using nema34 1090oz/in 5.6A motors, 60V, cheap bob, X&Y on my mill. 5mm pitch screws.
Couldn't get more than 2000mm/min even set at 7A on drivers and had some resonance. Even with everything set up pretty light (gibs etc).
I've ditched them for nema23 4A 566oz/in and it's come to life!.
Still running 60v, drivers at 4A and I've gone up to 3000mm/min no problem. DFU screw sounded a little rough for my liking so backed it off a little.

Food for thought.

If I may ask you, just out of curiosity, why didn't you leave the nema 34 and consider a reduction on the motor side... I am sure that these Nema 34 will not run out of torque if kept under 500 rpm.?

[dazp1976]

If I may ask you, just out of curiosity, why didn't you leave the nema 34 and consider a reduction on the motor side... I am sure that these Nema 34 will not run out of torque if kept under 500 rpm.?

I did consider trying out a 1:2 drive system but it would have cost more to impliment. I prefer the cleaner look of fitting it direct drive anyway and not thatkeen on belting it all up.

[rodw]

Hi rodw, thank for your comment.

Can you please elaborate a bit more about what you meant that there is a mis-match between the motor size and the application? As I am not sure what you meant.

The machine was intended to work as a milling machine, I just ended up putting a 2.2kW water cooled Router (which I do not regret).

Dim

Maybe I should have left the word size out. I Was really echoing what others had said after actually reviewing your drives specs against an engineering model.
I don't think increasing the power supply volts will help much.
You can get precision planetary gearboxes if you don't want belts.
BUt I would look at that step pulse first. You say you have a cheap BOB so if its a bit slow switching, you could get lost steps and you do not need high frequency step pulses. why live on the edge?

[d3workz]

Phxeww... at least you thought of it!!!, because for a minute there I thought it is still something not to consider.

I understand what you mean about the cleaner look, but my case is much simpler to just change some pulleys rather than new motors. I took that into consideration when I was design the cnc chassis and as you said, I wanted a clean look and no motors sticking out. I ll attach some photos to you understand. These are when it was still under construction.

[d3workz]

Thanks again for your thoughts.

Btw yes I do recommend longmotors, I have them on the cnc mill/router what ever, and my cnc plasma (which is 1.3m x 2.5m cutting area). Love their stuff...
However, getting back you your post, I do have a super cheap BOB and I love it and recommend it. I have the MACH3V2.1-L (google search it as I have nothing else). Very good.

On the other hand, for my plasma table, I got the NVUM v2, which is much better, but what ever you do, DO NOT buy from LOLLETTE. They sent me a faulty board, which costed me €130, I had to find the fault, which was a faulty DC to DC convector module, I told them about it, they confirmed the fault and did nothing. I had the board for only 2 weeks, and they did nothing. They said I should go buy the module locally because its only $1. Now I ended up hooking a buck converter just to bypass their faulty module until I find one online. I need the HDN3-5S12A1 module but have no luck find one as all seller cannot ship to me.

Anyways, thanks again for the help so far.

[ger21]

f I may ask you, just out of curiosity, why didn't you leave the nema 34 and consider a reduction on the motor side... I am sure that these Nema 34 will not run out of torque if kept under 500 rpm.?

The issue is that you are reducing your resolution by a factor of 2. And steppers are already limited in resolution.

I would not recommend downsizing to a Nema 23. A Small Nema 34 on 60-72V will run circles around a large Nema 23, and it'll run a lot cooler, as the larger size can dissipate more heat.

[cncuser1]

Review your decision on micro stepping. I have read opinions, possibly true, that micro stepping decreases torque. If you do not get a benefit from it then try not doing it. I think that in a setup like yours you won't see any benefit in the end product.