[BrooksMachining]

Using Mach 3, X3 machine.

I have Keling 5056 drivers.
(* Supply voltage up to +50VDC, current output up to 5.6A peak
* Optically isolated input signals,
* 3-state current control for less motor heating
* 16 selectable resolutions in decimal and binary
* Suitable for 4,6,8 lead motors
* Stepping on upward or downward pulse edge (selectable)
* DIP switch current setting with 8 different values
* Over-voltage and short-circuit protection)


I'm using 2 Nema 23s (KL23H86-20-8B Dual Shaft Motor 425 oz-in) for the X and Y, and a Nema 34 (1 PCS NEMA34 KL34H280-45-4A, HIGH TORQUE STEPPER MOTOR
640 oz-in) for the Z.

My power supply is 48V/7.3A.

Okay, so the steppers stall like crazy. What should I select on the amperage on my driver boards, and what should I select for the microsteps? More info on the drivers can be found here: http://kelinginc.net/KL-5056.pdf
Also which position of the switch is "1" or "0" on my driver? I'm kinda lost. I can CAD on SW and CAM on Mastercam and program Haas and Fanuc, but I am a newborn when it comes to wiring this all up.

[BrooksMachining]

Ok so I see Nema23 is rated at 2.7A, so I select 2.7A. Next I went in and changed the velocity until it wouldn't stall. I can get 32ipm, but if I try 50 it stalls.

Then I noticed the motor was getting pretty hot. Are these motors suppose to get hot fast? The microstep is on the highest number setting - I think it is a binary setting - 125?

[Crevice Reamer]

Hi Brooks. Welcome to the Zone.

I assume you have this wiring diagram?

http://kelinginc.net/KL-5056WiringDB25.pdf

Your motors will run best with 84V PSU and you are supplying 48. This is going to affect performance some.

All microstepping above 10 is wasted and just cuts down on how fast the motors turn.
set micro stepping to 8 and adjust steps per inch accordingly to about 8000 if you're using ball screws.

If you have the acceleration set too high, motors will stall early.

Steppers DO run hot, but motors will get hot fast if Amp settings are too high. Looks like the 425s should be set for 2.7A, and the N34 for 4.3A.

X3s can have really rough gibbs/ways. Have you at least smoothed up the gibbs? Are the gibbs adjusted properly and not too tight?

You may be experiencing mid band resonance. To correct that, you either have to install dampners or get Geckos.

I'm sure the option didn't exist when you bought your electronics, but nowadays for the same money you could have a g540 with 381s and a 906 N34. That would give you super-easy installation and much greater performance.

CR.

[BrooksMachining]

Hi Brooks. Welcome to the Zone.

I assume you have this wiring diagram?

http://kelinginc.net/KL-5056WiringDB25.pdf

Your motors will run best with 84V PSU and you are supplying 48. This is going to affect performance some.

All microstepping above 10 is wasted and just cuts down on how fast the motors turn.
set micro stepping to 8 and adjust steps per inch accordingly to about 8000 if you're using ball screws.

If you have the acceleration set too high, motors will stall early.

Steppers DO run hot, but motors will get hot fast if Amp settings are too high. Looks like the 425s should be set for 2.7A, and the N34 for 4.3A.

X3s can have really rough gibbs/ways. Have you at least smoothed up the gibbs? Are the gibbs adjusted properly and not too tight?

You may be experiencing mid band resonance. To correct that, you either have to install dampners or get Geckos.

I'm sure the option didn't exist when you bought your electronics, but nowadays for the same money you could have a g540 with 381s and a 906 N34. That would give you super-easy installation and much greater performance.

CR.

Yes you are correct on the wiring diagram. On the ways I smoothed them down with wd40 and a stone, and they are not tight.

Ok when you say adjust steps to around 8000, I'm lost.

My motors are 1.8 degree, and I am using ballscrews with a 5 TPI, so I was calculating 200 (for the motor) * 5 and getting 1000. And then since I am running in half-step mode, I double that to 2000.

Should I be running in half-step mode? How do you correctly calculate your steps for motor tuning? Is your microstep setting on your driver added into this formula?

Thanks and best regards,

[Crevice Reamer]

WD40 is NOT a lubricant. I don't recommend using a stone on the ways. I hope you have not screwed things up by creating an uneven surface.

How do you correctly calculate your steps for motor tuning? Is your microstep setting on your driver added into this formula?

Is your Mach set for inch or metric?

To calculate SPI: 200 x TPI X microstepping = Steps Per Inch.

So you have calculated it correctly for 1/2 step.

Maybe your axes are binding due to misalignment of ball screws and bearings. Run table to one end. Loosen BS nut. Run table to other end. tighten nut. See if that helps.

With your undervolted motors, and I suspect mid band resonance, this may be the best the mill will do. Increasing IPM may be expensive.

You could change motors to Keling 381s and a 906. Those would run much faster with only 48V. The 906 would detune to 846oz with your max 5.7A:

$099 one KL34H295-43-8A 906oz N34
$098 two KL23H2100-35-4B 381 oz N23s
-----
$197

I would estimate this will give 50 - 75 IPM.

If you want higher performance, you may have to spend $289 more and go Gecko 540.

CR.

[BrooksMachining]

WD40 is NOT a lubricant. I don't recommend using a stone on the ways. I hope you have not screwed things up by creating an uneven surface.



Is your Mach set for inch or metric?

To calculate SPI: 200 x TPI X microstepping = Steps Per Inch.

So you have calculated it correctly for 1/2 step.

Maybe your axes are binding due to misalignment of ball screws and bearings. Run table to one end. Loosen BS nut. Run table to other end. tighten nut. See if that helps.

With your undervolted motors, and I suspect mid band resonance, this may be the best the mill will do. Increasing IPM may be expensive.

You could change motors to Keling 381s and a 906. Those would run much faster with only 48V. The 906 would detune to 846oz with your max 5.7A:

$099 one KL34H295-43-8A 906oz N34
$098 two KL23H2100-35-4B 381 oz N23s
-----
$197

I would estimate this will give 50 - 75 IPM.

If you want higher performance, you may have to spend $289 more and go Gecko 540.

CR.

I checked my ways with a tenth indicator on a granite surface and the needle didnt move.

The ballscrew installation was checked very closely. The ballnuts and screws are parallel to the ways and centered within .002" of each datum surface.

I will adjust my settings as you have noted when I get home later. If I want more performance I will just get a bigger power supply.

Thanks

[Crevice Reamer]

If I want more performance I will just get a bigger power supply.

Thanks

No can do! 5056 driver can only use 50V max.

CR.

[BrooksMachining]

No can do! 5056 driver can only use 50V max.

CR.

I see said the blind man

[BrooksMachining]

Ok so on the X axis I can get 100+ IPM going one way, but it stalls on one half of the table when going in reverse. The ballnut also makes a clicking noise with every rotation of the screw. I'm guessing that the ballnut isn't aligned, but I know for sure that the ballnut was lined up right and it was tightened down very tight.

I guess I should do what the gecko man suggests and move the table to the side that it moves easy on, loosen the ballnut from the casting, move table over to side that sticks, and tighten it back up.

I can get 70ipm consistently with no stalls but I would still like to make sure everything is right.

[CNC2Z]

Hi Brooks
I have just read over your post here
The question I would have, Is it better
to go with a servo based xyz drive vs stepper?
would it have eliminated these issues that
are going on with your project?
The way Iam looking at the stepper thing
is by the time you beef up$ all of the components
/drives/ motors power supply .could servos just
have been the better/easier way?

[Crevice Reamer]

Hi Brooks. Good to see you are making progress. Seventy IPM is much better.

Hi Brooks
I have just read over your post here
The question I would have, Is it better
to go with a servo based xyz drive vs stepper?
would it have eliminated these issues that
are going on with your project?
The way Iam looking at the stepper thing
is by the time you beef up$ all of the components
/drives/ motors power supply .could servos just
have been the better/easier way?

Hi 2Z! Welcome to the Zone!

Ah. The classic steppers VS servos question. No question servos are better, but can you afford to pay for them? You can't get near to a servo system for the $500 or so that a G540 setup will cost. Check out this Iron Man commercial router, which uses a G540 and 387 motors. Isn't that acceptable performance?:

http://www.cnczone.com/forums/showthread.php?t=77409

Jeff Davis is running G540/381/906 on his X3. He is getting 180 IPM. (He might get more, but stopped testing at 180)

http://www.cnczone.com/forums/showthread.php?t=89153

Some of this may be helpful to you:

http://www.cnczone.com/forums/showthread.php?t=78713.#3

CR.

[BrooksMachining]

Yah I wish I would have waited a bit and did more research. Looks like Jeff Davis' setup is real nice.

[CNC2Z]

Hi CR Thank you for taking the time for us newbs here
I was starting to think the Knowledgeable guys have left
Here is why I was asking about servos
http://www.machmotion.com/Drives-and...26b8a110324ac0
But looking at your examples, Ironman and Jeff, show some really
good performance at a total cost of less than(1) axis price of servo pkg.
I will do more reading of your guys post. very good Info+++:cheers: