More Speed needed. Bigger Stepper or increase Ballscrew Pitch ?

[Joey]

I feel like I`m chasing my tail and getting no-where fast.

I am trying to increase my X & Y table speeds. At the moment I think I`m getting around 1500mm per min (60in per min) using a 4V, 2 Amp, 1.85 Nm stepper ( FL57STH56-2008B ).
If I try to increase the speed in Mach3 motor tuning the motor stalls.

If I were to get a slightly stronger motor ( FL60STH86-3008BF 4V, 3 Amp, 3Nm - 425 Oz-inch), do you think I`d be able to bump the speed up ?

or plan B
My current Ballscrew has a 5mm pitch, if I were to increase this to a 10mm pitch would my existing FL57 motor still be strong enough at the same velocity setting in Mach3 ?

My machine is non cutting has very little load.

Many thanks for any advice.

[Greolt]

Joey what voltage are you running the steppers at?

What drivers are you using?

What is the microstepping?

What kernal speed are you running in Mach?

Greg

[mxtras]

Originally Posted by Joey using a 4V, 2 Amp,

4 Volts??

I think we have found your problem.



I say increase the voltage. By a factor of 10!

I think that's where Greolt was headed....

Scott

[Joey]

Sorry I should have given more info. The 4V was the figure taken off of the motor specsheet. (see attached pic)

I am using a 24V DC x 14.6 Amp PSU
Stepmaster 4 Axis Driver, I`m not sure what actual voltage/amperage this supplies to the motors, it`s a sealed unit.

Mach3 settings are : using 35000Hz kernel speed.
Steps per = 320 Velocity=4000 Acceleration=325 G`s=0.033 ( I have been told for my application I need to pull over 1G ? )


thanks guys

[Greolt]

Joey

In your first post you said your getting 1500mm/min and in the second it's 4000mm/min. Which is it?

I'm not familiar with the Stepmaster. What microstep does it use? Is it adjustable?

In principal a bit more voltage will increase the rpm attainable.

The easiest and first thing I would try is upping the kernal speed.

My router will rapid comfortably at about 370 stepper rpm. 25000khz

If I go to 60000khz it will rapid at more like 1000 stepper rpm.

I don't need this and don't use it but it shows how much difference it can make. Costs nothing to try.

Greg

PS: 1g is instant. No acceleration slope at all. Aim for 0.1g more realistic.

And if you reset kernal speed you need to restart and retune.

[Joey]

yeah Greolt I appologize for that discrepancy, truth is I dont understand this enough.
What I did was calculate the total length of a simple X & Y path of a program and divide this by the actual time it took to run, this gave me the 1500mm/sec speed.
I have tried changing the kernel speeds in Mach3 and have settled on the 35000 as being the fastest. I will have another go at it.
The stepmaster specs state, All axis are permanently set to 1/8 step so I guess there`s no adjustment.
My application is spraying and so the faster the change of axis direction will help as any dwell causes build up at those points.

tnx mate.

[Greolt]

Originally Posted by Joey What I did was calculate the total length of a simple X & Y path of a program and divide this by the actual time it took to run, this gave me the 1500mm/sec speed.

OK if its travel speed when running a gcode, that is determined by the feedrate in the code.

G1 moves will be at the said feedrate and G0 moves will be at the rapid rate set in motor tuning.

I have tried changing the kernel speeds in Mach3 and have settled on the 35000 as being the fastest. I will have another go at it.

If you have tried various kernal speeds and found that to be the best then that's fine. I was suggesting trying that if you had not.

The stepmaster specs state, All axis are permanently set to 1/8 step so I guess there`s no adjustment.

Nothing wrong with that

My application is spraying and so the faster the change of axis direction will help as any dwell causes build up at those points.

Sounds like you need to bump up the acceleration then. Rapids may already be fast enough.

Larger motors with their greater torque may help here. I would be spending a bit of time motor tuning before doing that.

Do you have CV turned on? Sounds like you definitely want it on.

Greg

[mcpltd]

The problem is most likely that you are running the motor at 2A in series and at only 24V.
Use a better driver, higher voltage and connect the motor in paralell you will see better performance.

[Joey]

Ok thanks for that Greg.
Could I ask you about CV.
I tried CV with the default values of Distance 180.00 and Feedrate +1.0 but got very rounded corners. I`m not sure what the distance value does, any clues here please.

mcpltd I`m discussing this via email with Gary ATM. thanks

[Greolt]

Joey

Feedrate needs to be turned off.

If rounding at corners is an issue then that is because you have slow acceleration setting.

If you can not get that up higher then you probably need to look at the type of options that mcpltd is suggesting.

From what you have described, sounds like acceleration is mostly your issue.

Greg

[pastera]

At 1500mm/min with a 5mm lead, you are spinning your steppers at 5rev/sec

Do the motors make nasty sounds when they stall? Then it is probably resonance - look up the thread on DIY dampers or get a set of Geko drives.

Also, with a 24 volts supply the drives are at the edge of reaching full phase current at 5 rps. So you are loosing torque rapidly at this point. Kick up your PS voltage or run the motor phases in parallel

Running the motors in parallel with the same driver would allow you to get full torque to about 15 rps but the "full" torque would be about 63% of nameplate torque on the motor.

Aaron

[pminmo]

Since you can't supply the full 4A of the parallel winding, you would probably be best served by using half coil wiring. When using half coil you would normally set the current to the unipolar rating, but your driver is limited to 2.5A so you would set it no more than that. It doesn't seem intuitive, but the half coil arrangement will give you more top end speed even though it's actually producing less torque, try it you will be surprised.

My concern would be the ability of the driver to handle 2.5A without overheating. The tssop package that board uses is difficult to get full rated current and keep from kicking in the overtemp of the chip.