CNC motors are slow and miss steps

[Shamanjim]

Hi guys, I have a small 600x300x75 CNC router that I made my own toolholder for so I can do alloy when i'm up to the challenge. However right now Im just doing styrofoam and MDF with a high speed laminate cutter (Hardcore makita one) in the spindle as I get my skills up. It seems no matter what i do, even going from 1/4 step to 1/2 step, I can't get any reasonably fast speed out of it without it eventually missing a massive step and ending up way over the other side of the table, way above the work piece, or something completely stupid.
If I slow down, its fine, but I shouldn't be on 600mm/minute on foam. The Z axis seems a bit tight but other then that its often the x or y axis that goes stupid.

The motors are nema 23's but maybe really dodgey ones, is there any way to boost power to those motors, increase the speed, or anything at all you can suggest, as I don't believe my controller board can take hugely bigger motors.

[BillTodd]

Steeper motors are a pain...

In general the way to increase the maximum speed of a stepper is to increase the supply voltage, but be careful not to exceed the driver ratings. Sometimes decreasing the current limit can help. A micro-stepping driver (or an automatic switching type like gecko's) will usually get the best from any motor.

Obviously, the lower the friction in the mechanical drive the better. You may also find that some mass-decoupling between motor and drive helps (i.e. a compliant coupling like a belt or rubber tube).

Are you in the right gear? Steppers prefer to run slowly, so if you are using a lead-screw, is the pitch of the screw too fine? Or, can you change the pulleys to change speed?

[Shamanjim]

Dear Billtodd,
Thank you for your reply, this has really been quite frustrating.

Dont gasp but its an acme rod thread, 12 threads per inch.

The only axis which seems quite tight is the Z axis but it is holding a bit of weight, and does seem to be the one that gives up first when trying to bring the speed up.

I use mach 3, and compared to the tutorial videos my acceleration seems really really low on the graph, should I bring it right up or is that worse.

Also when you say increase voltage is there any links for ways to do that, other then going to 1/2 step I'm not sure if I can turn the voltage up on my board, is there a pin or something on the board for it?

Thanks for taking the time to help me out here Bill,
Kind Regards,
Jim

[NC Cams]

pursuant to the F=MA equation, any time you want/try to do something faster with a given load or resistance, you need to apply more POWER.

Any time you apply more power to a slightly bound up system, you encounter MORE RESISTANCE.

I dunno what your motors are nor your driver cards but you can run into a bottleneck if motors ask for more power INSTANTANEOUSLY than than the board is capable of providing - the driver card essentially becomes the bottleneck for current as opposed to the motor, wires, P/S etc.

As speed requirements go up, you have to get a bit more involved with the engineering of the perts being integrated. Example: how do you quantify "a bit tight"? IF your residual friction torque is say 30% of your motor output torque, you have to hit the motor with a good shot of current to simply get stuff to move INITIALLY. Start, stop, reverse direction and it is easy to see how any power reserve you may have is GONE. THe more current you draw causes the controller to heat up and they then can't pass current as readily or easily. SOme may actually go into a fold back mode to self protect.

Over the time I've been a member, there have been lots of discussions about "missing steps" and the root causes thereof. MOst seem to end up being wiring and/or friction related. DO a search for missing steps search engine and even more ideas/solutions will spring forth.

I dunno the quality of your acme but that too enters the picture. To keep friction low, you have to increase the nut clearance and reduce preload. To improve accuracy and reduce direction change flats/dwells, you have to reduce nut back lash via increasing preload. Conflicting actions for to achieve the simultaneously desired conditions.

Finally, your motors may simply not be suitable/capable of high speed operation. Some of the older/early steppers were designed to simply poke a long. WIth the advent of high speed machining, low mass motors abd big driver systems sprang forth. Simply put, your stuff may simply not want or be capable of doing what you wish, especially if it is made of older/surplus parts.

Minimize bind, go to larger gage power wiring, make sure motors are not tired/worn out, typical usual stuff is probably the root cause of your ills.The same old problems do not really need all new solutions.

[BillTodd]

Originally Posted by Shamanjim Dear Billtodd, Thank you for your reply, this has really been quite frustrating. Dont gasp but its an acme rod thread, 12 threads per inch. The only axis which seems quite tight is the Z axis but it is holding a bit of weight, and does seem to be the one that gives up first when trying to bring the speed up.

Have a read through the roller nut thread Here. It's a bit long winded but there's some useful mods you can do.

I use mach 3, and compared to the tutorial videos my acceleration seems really really low on the graph, should I bring it right up or is that worse.

Start with slow acceleration, increase the rate slowly, until you find it stalls then back it down until it's reliable.

Also when you say increase voltage is there any links for ways to do that, other then going to 1/2 step I'm not sure if I can turn the voltage up on my board, is there a pin or something on the board for it?

I was speaking generally so changing voltages etc. may require replacing things. What drivers and PSU are you using?

[BillTodd]

pursuant to the F=MA equation, any time you want/try to do something faster with a given load or resistance, you need to apply more POWER.

Someone needs to explain that to a stepper motor