leadshine motor drives

[Ken Rychlik]

Charlie, after changing out several drivers on different machines, if you want a smoother running machine put some Leadshine drives in it. I thought Gary was just full of crap about gecko, but the difference is from noisy/grinding sound to not knowing if it's running or not. (Thanks Gary)

[Charlie.G]

Going from 4:1 to 3:1 will reduce your resolution (and torque) by 25%. Going to a 24T pinion will increase both by 20%.

Since resolution can be "what you set it at" and resolution via microstepping and available torque are often inversely proportional do you have a specific issue that you are trying to solve? I ask because more router type cutting issues can be cured by more mechanical reduction, rather than less.

Gary,
first of all I want to say thank you for your help my knowledge is limited, I don't want to change things just for the sake of changing them,
but I know technology has changed since the machine was built 10 years ago or so. that is why I am looking at each component to see if it should be updated, and if it makes sense to do so and is worth the cost.
I know that more reduction makes better cuts to an extent anyway I had 20 tooth pinions on it in the beginning but the thought was the bigger dia.
of the 30 tooth would be smoother. maybe that thought was misguided? I can easily change this do you think the 20 tooth are to small ?
when I cut aluminum you can see a bit of chatter in the cuts (haven't cut aluminum since the rebuild of the box) but before the box rebuild I found an issue with the Z slide guide wheels being a bit loose and I resolved that issue (this could be most of the chatter haven't tested it with aluminum yet) the two things left on the machine are the final drives and the dust collection system...

thank you,
Charlie

[Charlie.G]

Charlie, after changing out several drivers on different machines, if you want a smoother running machine put some Leadshine drives in it. I thought Gary was just full of crap about gecko, but the difference is from noisy/grinding sound to not knowing if it's running or not. (Thanks Gary)

Ken,
I debated this for quite some time, some like the geckos some do not, I am indifferent makes no difference to me which drives they are as long as they work, you could be right the leadshines might be better, but I opted to keep the lizards I did order new resistors from gecko for the amp rating of my motors I don't know if this made a difference but since the box rebuild things seem much smoother and quieter to me at least I still need to tune the motors which I plan on doing today if time permits.
thank you Ken for you help as well,

Charlie

[Gary Campbell]

"I know that more reduction makes better cuts to an extent anyway I had 20 tooth pinions on it in the beginning but the thought was the bigger dia.
of the 30 tooth would be smoother. maybe that thought was misguided?"

I believe that you were misguided. Most likely by an urban myth started by ShopBot when they used 20T pinions on an older model with a motor that had more torque than the brackets that held the motors could control. The brackets twisted, the pinions ran at an angle, wore out fast and due to low resolution provided choppy cuts. I use pinions down to 18T with good success.

One of the things that happens when using 1/10th microstepping is that the available torque is reduced so much that especially when cutting plastics and aluminum, the motor will "jump" back to a full, half, or quarter step. This puts a "serrated" or "faceted" edge on angular or especially round cuts. I am sure you suffer from that syndrome, did you ever notice that those "facets" line up across the whole edge of the part thru multiple passes? How is that possible?

Back in '13 I took a ShopBot controller, added mechanical reduction and eliminated the 1/10th microstepping from the drive via programming and produced virtually facet free cuts, with before and after cuts of the same geometry in 1" aluminum.

[ShawnM]

One of the things that happens when using 1/10th microstepping is that the available torque is reduced so much that especially when cutting plastics and aluminum, the motor will "jump" back to a full, half, or quarter step. This puts a "serrated" or "faceted" edge on angular or especially round cuts. I am sure you suffer from that syndrome, did you ever notice that those "facets" line up across the whole edge of the part thru multiple passes? How is that possible?

My Chinese CNC router machine that I retrofitted came with Leadshine DM542 drives and they are set to 1/10th microstepping. It has NEMA 34, 86BYGH450A, motors that direct drive ball screws. It has 2510 ball screws on the X and Y axis and a 1605 ball screw on the Z axis. I have noticed this "serrated edge" on aluminum and acetal but I can somewhat get it to go away with a thin "separate last pass". The machine has a 4 HP water cooled spindle on it.

Being that the machine is direct drive what suggestions do you have to get some torque back?

[Gary Campbell]

First to begin with, if I were to design a machine for aluminum I would use 2505 screws, just for the extra reduction. Then I would see my cut quality at full/half and quarter steps. Of course this reduces rapid speeds lower than most desire on a large table, so the 2510's are a decent compromise. 10mm pitch screws have 2.54 turns per inch, where Charlies 4:1 with 30t pinion only has ~.85. That's 3 times the mechanical resolution. With steppers I like using between 1.5 and 2.5 turns per inch, which keeps the torque up and in the "power band" of rpm. Servos more turns as the power drop is much less severe.

I would make a quick test of cutting circular part(s) at 1/10th and then at 1/4 and 1/8 microstepping and see if there is a difference. If it improves, and it usually will, it shows that you are experiencing a torque detent slip (loss of microstep) which often causes the scalloped edges.

More info here: viewtopic.php?f=63&t=3359

As far as adding torque to your machine you could do any of the following: Use a drive that accepts higher input voltage, use 900 or 1200ozin motors, or use a closed loop drive which in most cases will make sure 100% of drive power is applied if there is slippage beyond your set threshold.

[Charlie.G]

I believe that you were misguided. Most likely by an urban myth started by ShopBot when they used 20T pinions on an older model with a motor that had more torque than the brackets that held the motors could control. The brackets twisted, the pinions ran at an angle, wore out fast and due to low resolution provided choppy cuts. I use pinions down to 18T with good success.

One of the things that happens when using 1/10th microstepping is that the available torque is reduced so much that especially when cutting plastics and aluminum, the motor will "jump" back to a full, half, or quarter step. This puts a "serrated" or "faceted" edge on angular or especially round cuts. I am sure you suffer from that syndrome, did you ever notice that those "facets" line up across the whole edge of the part thru multiple passes? How is that possible?

Back in '13 I took a ShopBot controller, added mechanical reduction and eliminated the 1/10th microstepping from the drive via programming and produced virtually facet free cuts, with before and after cuts of the same geometry in 1" aluminum.

Gary,
I will order some 20 tooth pinions and see what happens, the idea of the dia of the pinion being to small came from the mechmate site.
which more than likely was influenced by shopbot. I have looked at some mechanical planetary reductions which you have talked about before
but the shaft size on the input and output are different than what I have on my motors, I realize that the output is not a huge deal I can get the pinions
to match that but the input is 14 mm my motor shafts are .5
thank you,
Charlie

[Gary Campbell]

Most planetary's have a 14mm input clamp. Most that sell them have bushings to accommodate .25, 8mm, .375, 10mm and 1/2" shafts

[ShawnM]

First to begin with, if I were to design a machine for aluminum I would use 2505 screws, just for the extra reduction. Then I would see my cut quality at full/half and quarter steps. Of course this reduces rapid speeds lower than most desire on a large table, so the 2510's are a decent compromise. 10mm pitch screws have 2.54 turns per inch, where Charlies 4:1 with 30t pinion only has ~.85. That's 3 times the mechanical resolution. With steppers I like using between 1.5 and 2.5 turns per inch, which keeps the torque up and in the "power band" of rpm. Servos more turns as the power drop is much less severe.

I would make a quick test of cutting circular part(s) at 1/10th and then at 1/4 and 1/8 microstepping and see if there is a difference. If it improves, and it usually will, it shows that you are experiencing a torque detent slip (loss of microstep) which often causes the scalloped edges.

More info here: viewtopic.php?f=63&t=3359

As far as adding torque to your machine you could do any of the following: Use a drive that accepts higher input voltage, use 900 or 1200ozin motors, or use a closed loop drive which in most cases will make sure 100% of drive power is applied if there is slippage beyond your set threshold.

Thanks again for your input Gary. I'm not really worried about speed as the table is only 2x3, I'd rather have more torque as I cut mainly aluminum and hard plastics with it (sometimes carbon fiber) so my IPM speeds are never that high. I'll experiment with lowering the microstepping and see what happens. This is my second CNC machine but my first that I retrofitted with Acorn. It runs really well and I dont want to sink a lot more money into this one as I now find myself needing a larger table, 4x4. This machine was my "learning curve" and the next one will get even better stuff and I'm thinking AC closed loop stepper drives for it. I'll consult with the group when that time comes.

[Ken Rychlik]

I have a question for Gary about Charlie's mechmate. Do you think he might benefit more from changing to linear rails, than swapping motors and or drivers? Ii wonder if the gear chatter is being allowed more by the v rollers, that would be imparted to the cut with linear.