Re: gecko 201/210 opinions
Posted by
caudlet
on 2004-12-18 17:39:34 UTC
--- In CAD_CAM_EDM_DRO@yahoogroups.com, Hal Eckhart <hal@c...> wrote:
The step multiplier will only be of use if you get in this
condition: Your design needs more pulses per second from the
computer than it is capable of to go the speed you want. ALl it does
is take the pulses from the computer and multiply them by 2, 5 or 10.
Lets say you wanted to whip your plasma along at 300 IPM and with the
gearing you are using you do the math and find out that equates to
65,000 pulses per second (don't forget that a microstepping drive
needs 8 to 10 times (depending on the drive) the number of pulses
that a full step unit does. It's pretty easy to find yourself boxed
into a corner because of drive ratios and things like belt reduction
and lead screws. Most of the computer driven solutions top out at
40,000 to 45,000 pulses per second. You either have to re-engineer
your design or go with a step multiplier. Do not despair. A 210 is
just a 201 witht he piggyback multiplier card added. You have the
option to buy and add the card later if needed.
Now about microstepping. Steppers have issues with mechanical
resonance that can cause lost steps at speeds far below the
theoretical max of the motors. The "jump" from one pole to the next.
On most motors there are 200 detents per rev. Half stepping applies
currents to two windings at once to create a phanthom pole halfway
between. Quarter step creates 2 such poles in between. What this
does is smooth out the jumps and make things not so ragged with the
resulting vibration and resonance problems. As you approach 8X
microstepping you start to see the motor behave closer to a
conventional motor. The Gecko is a 10X microstepper. The only down
side of this feature is you have to send 10 times and many pulses to
move the motor the same distance. Thus a motor that normally takes
200 pulses now takes 2000...no big deal unless you start to have
other factors involved like a belt reduction or leadscrew ratio.
Lets say you are direct coupled to a lead screw with the motor that
has a pitch of 10TPI (.1). You have to spin the motor with
microstepping 2000 * 10 (20,000) to move one inch. If you want to
move 300 IPM you have to move 5 Inches per second and you need to
send 20,000 * 5 pulses per second = 100,000 pulses per
second....woops, can't get there. Either you change out the
leadscrew or don't go with microstepping. Plug in the step
multiplier and set it to X5 and bingo now you only have to send the
20,000 per second and you are home free.
The advantages that microstepping bring far outweigh any issues with
pulses. With the step multiplier you still get the advantage and
smoothness of microstepping.
Run your numbers for your machine with the target speeds and the
microstepping and see if you can stay below the max pulse per second
of of 45,000.
> Hi folks,G201 and G210
>
> Can anyone explain what the practical difference between the Gecko
> is? I realize there's some sort of step multiplier on the 210, butwhy should I
> care? The intended use is on a relatively low-accuracy plasmatable. I am toying
> with the idea of upgrading to Geckos with the Pico Systems USCcontrolled with
> EMC. It's just not clear to me what "Microstep-smooth Full and Half-step" will
> do for me.<http://www.casaforge.com>
>
> Thanks in advance for any light you can shed on this.
>
> Hal Eckhart - Casa Forge - Minneapolis MN -
The step multiplier will only be of use if you get in this
condition: Your design needs more pulses per second from the
computer than it is capable of to go the speed you want. ALl it does
is take the pulses from the computer and multiply them by 2, 5 or 10.
Lets say you wanted to whip your plasma along at 300 IPM and with the
gearing you are using you do the math and find out that equates to
65,000 pulses per second (don't forget that a microstepping drive
needs 8 to 10 times (depending on the drive) the number of pulses
that a full step unit does. It's pretty easy to find yourself boxed
into a corner because of drive ratios and things like belt reduction
and lead screws. Most of the computer driven solutions top out at
40,000 to 45,000 pulses per second. You either have to re-engineer
your design or go with a step multiplier. Do not despair. A 210 is
just a 201 witht he piggyback multiplier card added. You have the
option to buy and add the card later if needed.
Now about microstepping. Steppers have issues with mechanical
resonance that can cause lost steps at speeds far below the
theoretical max of the motors. The "jump" from one pole to the next.
On most motors there are 200 detents per rev. Half stepping applies
currents to two windings at once to create a phanthom pole halfway
between. Quarter step creates 2 such poles in between. What this
does is smooth out the jumps and make things not so ragged with the
resulting vibration and resonance problems. As you approach 8X
microstepping you start to see the motor behave closer to a
conventional motor. The Gecko is a 10X microstepper. The only down
side of this feature is you have to send 10 times and many pulses to
move the motor the same distance. Thus a motor that normally takes
200 pulses now takes 2000...no big deal unless you start to have
other factors involved like a belt reduction or leadscrew ratio.
Lets say you are direct coupled to a lead screw with the motor that
has a pitch of 10TPI (.1). You have to spin the motor with
microstepping 2000 * 10 (20,000) to move one inch. If you want to
move 300 IPM you have to move 5 Inches per second and you need to
send 20,000 * 5 pulses per second = 100,000 pulses per
second....woops, can't get there. Either you change out the
leadscrew or don't go with microstepping. Plug in the step
multiplier and set it to X5 and bingo now you only have to send the
20,000 per second and you are home free.
The advantages that microstepping bring far outweigh any issues with
pulses. With the step multiplier you still get the advantage and
smoothness of microstepping.
Run your numbers for your machine with the target speeds and the
microstepping and see if you can stay below the max pulse per second
of of 45,000.
Discussion Thread
sargossa_99
2004-12-17 06:58:00 UTC
plasma accuracy
turbulatordude
2004-12-17 11:19:50 UTC
Re: plasma accuracy
Graham Stabler
2004-12-17 13:35:47 UTC
Re: plasma accuracy
V FONG
2004-12-17 14:33:45 UTC
Re: [CAD_CAM_EDM_DRO] Re: plasma accuracy
turbulatordude
2004-12-17 15:22:18 UTC
Re: plasma accuracy
Graham Stabler
2004-12-17 15:50:52 UTC
Re: plasma accuracy
John Heritage
2004-12-18 07:30:52 UTC
Re: [CAD_CAM_EDM_DRO] Re: plasma accuracy
Graham Stabler
2004-12-18 08:35:04 UTC
Re: plasma accuracy
caudlet
2004-12-18 08:35:55 UTC
Re: plasma accuracy
Graham Stabler
2004-12-18 09:12:58 UTC
Re: plasma accuracy
R Rogers
2004-12-18 09:38:47 UTC
Re: [CAD_CAM_EDM_DRO] Re: plasma accuracy
turbulatordude
2004-12-18 10:46:15 UTC
Re: plasma accuracy
caudlet
2004-12-18 13:55:31 UTC
Re: plasma accuracy
metlmunchr
2004-12-18 15:14:59 UTC
Re: plasma accuracy
Hal Eckhart
2004-12-18 15:42:28 UTC
gecko 201/210 opinions
Raymond Heckert
2004-12-18 17:24:32 UTC
Re: [CAD_CAM_EDM_DRO] plasma accuracy
Raymond Heckert
2004-12-18 17:24:33 UTC
Re: [CAD_CAM_EDM_DRO] Re: plasma accuracy
caudlet
2004-12-18 17:39:34 UTC
Re: gecko 201/210 opinions
Hal Eckhart
2004-12-18 18:31:52 UTC
Re: [CAD_CAM_EDM_DRO] Re: gecko 201/210 opinions
Hal Eckhart
2004-12-18 18:53:42 UTC
Re: [CAD_CAM_EDM_DRO] plasma accuracy
Jon Elson
2004-12-18 20:40:43 UTC
Re: [CAD_CAM_EDM_DRO] gecko 201/210 opinions
turbulatordude
2004-12-18 21:40:12 UTC
Re: plasma accuracy
Randy Brown
2004-12-19 08:03:26 UTC
RE: [CAD_CAM_EDM_DRO] Re: plasma accuracy
caudlet
2004-12-19 13:12:59 UTC
Re: gecko 201/210 opinions
m0nkey0ne
2004-12-20 16:22:25 UTC
Re: plasma accuracy
R Rogers
2004-12-20 17:40:36 UTC
Re: [CAD_CAM_EDM_DRO] Re: plasma accuracy
turbulatordude
2004-12-20 23:26:48 UTC
Re: plasma accuracy
Randy Wilson
2004-12-21 05:42:05 UTC
Re: [CAD_CAM_EDM_DRO] Re: plasma accuracy
Bruce Pigeon
2004-12-21 09:40:37 UTC
Re: [CAD_CAM_EDM_DRO] Re: plasma accuracy
turbulatordude
2004-12-21 09:46:24 UTC
Re: plasma accuracy