Discussion:
Leadshine ACS806 - Large f-error at higher speeds - what is acceptable?
(too old to reply)
Curtis Dutton
2014-04-30 18:37:34 UTC
Permalink
I recently installed an ACS806 brushless drive along with a BLM57130
leadshine motor. The motor is a 130W motor with a 4000ppr encoder.

The controller is a step and direction type, with the encoder feedback
passed back through to linuxcnc via a mesa 5i25 and a 7i86s.. So I'm using
the drive to run the PID loop but having linuxcnc watch my f-error.

I've replaced a 500oz/in stepper motor with it on a medium sized cnc wood
shaping router.


I have tuned it up and it seems to be tuned well, and has a low following
error (< 0.01mm) at about 400mm/min and below.

The problem is that it seems to rapidly increase from there. 0.1mm around
3000mm/min and 1.5mm at around 12000mm/min.


Does anyone have experience with these drives? I've tried to utilize feed
forward but it doesn't seem to make a difference.



Being that this is my first servo drive, what would be a typical f-error
for various types of machinery?

Say for instance what would the following error be on a 1000in/min high
dollar cnc vmc.

What do people with home shop mills see?


Thanks,
Curtis
Jon Elson
2014-05-01 02:34:59 UTC
Permalink
Post by Curtis Dutton
I have tuned it up and it seems to be tuned well, and has a low following
error (< 0.01mm) at about 400mm/min and below.
The problem is that it seems to rapidly increase from there. 0.1mm around
3000mm/min and 1.5mm at around 12000mm/min.
12 m/min? That error is only .0125 of the travel at that speed.
Post by Curtis Dutton
What do people with home shop mills see?
Certainly no home shop mills are running at 12 m/min!
Probably no
home shop routers are going that fast, either. My mill is
limited to
less than 2 m/min, and I get an error of less than .1mm at
that speed.

Jon
Peter C. Wallace
2014-05-01 02:41:36 UTC
Permalink
Date: Wed, 30 Apr 2014 14:37:34 -0400
Reply-To: "Enhanced Machine Controller (EMC)"
Subject: [Emc-users] Leadshine ACS806 - Large f-error at higher speeds - what
is acceptable?
I recently installed an ACS806 brushless drive along with a BLM57130
leadshine motor. The motor is a 130W motor with a 4000ppr encoder.
The controller is a step and direction type, with the encoder feedback
passed back through to linuxcnc via a mesa 5i25 and a 7i86s.. So I'm using
the drive to run the PID loop but having linuxcnc watch my f-error.
I've replaced a 500oz/in stepper motor with it on a medium sized cnc wood
shaping router.
I have tuned it up and it seems to be tuned well, and has a low following
error (< 0.01mm) at about 400mm/min and below.
The problem is that it seems to rapidly increase from there. 0.1mm around
3000mm/min and 1.5mm at around 12000mm/min.
Not sure why this would happen but what tuning options are available
for the drives PID loop?

A suitable amount of I term should be able to pull this in.

Another option is to close the position PID loop in linuxcnc
by setting the stepgen to velocity mode and adding a PID loop
with just P, FF1 and a bit of I and getting the feedback position
from the encoders
Does anyone have experience with these drives? I've tried to utilize feed
forward but it doesn't seem to make a difference.
Being that this is my first servo drive, what would be a typical f-error
for various types of machinery?
Say for instance what would the following error be on a 1000in/min high
dollar cnc vmc.
What do people with home shop mills see?
Thanks,
Curtis
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos. Get
unparalleled scalability from the best Selenium testing platform available.
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.
Kirk Wallace
2014-05-01 03:19:18 UTC
Permalink
Post by Curtis Dutton
I recently installed an ACS806 brushless drive along with a BLM57130
leadshine motor. The motor is a 130W motor with a 4000ppr encoder.
This is just my opinion, but to me 130 Watts is way too small,
especially with a router where you might need extra feed rate when
cutting wood. My mill has 350 Watt steppers but it doesn't need to go
very fast, and doesn't. I would feel better with 500 Watt, 90 Volt motors.
Post by Curtis Dutton
The controller is a step and direction type, with the encoder feedback
passed back through to linuxcnc via a mesa 5i25 and a 7i86s.. So I'm using
the drive to run the PID loop but having linuxcnc watch my f-error.
It would be interesting to see what kind of load the drive is seeing. An
advantage of having the PID loop in LinuxCNC is that you can use
HALscope to monitor the PID parameters to see if the output command is
getting saturated or hitting the load limit. It might be useful to see
if you can get the drive to tell you how hard it is working.

Hmmm... I just looked at the drive information:
http://www.americanmotiontech.com/Products/ProductDetail.aspx?category=4&model=ACS806

I would max out the drive voltage to the specified max. of 80 Volts if
you haven't done so. It also looks like there is a serial port on the
drive which you might monitor with LinuxCNC or with the tuning software
that comes with the drive.
--
Kirk Wallace
http://www.wallacecompany.com/machine_shop/
http://www.wallacecompany.com/E45/
andy pugh
2014-05-01 11:22:38 UTC
Permalink
Post by Curtis Dutton
The motor is a 130W motor
1.5mm at around 12000mm/min.
That seems like a very small motor, and a very fast travel.

Is it possible that the motor is simply running out of steam?
--
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto
Jon Elson
2014-05-01 14:44:06 UTC
Permalink
Post by andy pugh
Post by Curtis Dutton
The motor is a 130W motor
1.5mm at around 12000mm/min.
That seems like a very small motor, and a very fast travel.
Is it possible that the motor is simply running out of steam?
Generally when the drive runs out of available voltage
the following error very suddenly grows without bound.
So, you can be at 500mm/min with error of .01mm,
and then at 550mm/min the error rises continuously
because full voltage applied to the motor is only
giving you 520mm/min, to give an example.

So, having the following error increase only a modest bit
at higher speeds may indicate the drive just has a
constant time lag in the internal loop. Or, it may
be a torque limit, where the 130 W motor is nearly
maxed out on current driving the axis at 12 m/min.
Those conditions might cause a bounded error
that increases roughly proportionally to velocity.

Jon
Curtis Dutton
2014-05-01 15:34:12 UTC
Permalink
Ok this makes sense. Thanks all for your explanations. I guess I just
wasn't sure what was reasonable behavior and what wasnt.

So if the motor is rated for 36v, and the drive is rated for 80 volts max.
How much voltage can I get away with delivering to the drives without
damaging equipment?


Thanks,
Curtis
Post by Jon Elson
Post by andy pugh
Post by Curtis Dutton
The motor is a 130W motor
1.5mm at around 12000mm/min.
That seems like a very small motor, and a very fast travel.
Is it possible that the motor is simply running out of steam?
Generally when the drive runs out of available voltage
the following error very suddenly grows without bound.
So, you can be at 500mm/min with error of .01mm,
and then at 550mm/min the error rises continuously
because full voltage applied to the motor is only
giving you 520mm/min, to give an example.
So, having the following error increase only a modest bit
at higher speeds may indicate the drive just has a
constant time lag in the internal loop. Or, it may
be a torque limit, where the 130 W motor is nearly
maxed out on current driving the axis at 12 m/min.
Those conditions might cause a bounded error
that increases roughly proportionally to velocity.
Jon
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos. Get
unparalleled scalability from the best Selenium testing platform available.
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Jon Elson
2014-05-01 16:47:52 UTC
Permalink
Post by Curtis Dutton
Ok this makes sense. Thanks all for your explanations. I guess I just
wasn't sure what was reasonable behavior and what wasnt.
So if the motor is rated for 36v, and the drive is rated for 80 volts max.
How much voltage can I get away with delivering to the drives without
damaging equipment?
Most of these motor ratings are figures by what voltage can be
applied from a very strong power supply with the motor at
standstill, and not cause damage to the motor. A servo amp
will be current limited, and will only apply whatever average
voltage is required to close the servo loop. So, you should
be able to apply quite a bit more voltage to the servo amp
than the motor's rating. However, you shouldn't NEED to
provide much greater voltage. But, the PWM process and
the limits on duty cycle may require 10 - 20% greater
voltage if you are running the motor at the rated speed.
At 12 m/min, maybe you are spinning it that fast.

Jon
Gene Heskett
2014-05-01 18:26:19 UTC
Permalink
Post by Curtis Dutton
Ok this makes sense. Thanks all for your explanations. I guess I just
wasn't sure what was reasonable behavior and what wasnt.
So if the motor is rated for 36v, and the drive is rated for 80 volts
max. How much voltage can I get away with delivering to the drives
without damaging equipment?
Thanks,
Curtis
I, as an electronics type, would look at it from the motors rated currant
viewpoint regardless of the family of motor.

The motor more than likely has permanent magnet fields, and allowing more
currant than about 1.25x the nameplate rating (based on my reading on the
subject but I don't have an URL's to offer) gets you into a magnetic
territory where the field magnets can be damaged by reducing their magnetic
strength, and its an instant and permanent effect.

The same effect applies to steppers, usually at currants above 1.25x
nameplate.

Applying an 80 volt supply to a 32 volt rated motor seems like it would be,
if not currant limited in the driver, playing with fire. I would have to
assume they said that assuming a condition where it could spin freely,
letting its counter EMF control the current and therefore the resultant
magnetic field.

This isn't normally a concern with steppers because the 10 to 30x over
voltage is just normal standard operating voltage for them. The drivers
chopper limits on the currant are many times more important to the long
term health of the motors. I see no reason not to apply much the same
thinking to PM field servo motors. Any difference is in where the magnets
are, the steppers magnet is the rotor, where a brushed servo has the magnet
in its stator. But its still the strongest magnet we know how to build in
production quantities.

Now, in servo's I'll have to plead the big dummy because in brushless, hall
effect commutated motors (BLDC?), it seem like a 3rd phase of drive to what
is basically a 3 phase wound stepper motor frame assembly, meaning the
rotor is the PM, would this not also apply?

Or do they have something even more complex for the "BLDC" format? I am
not using them, so I've not spent a lot of time researching how they are
built.

My understanding is quite incomplete for those, and is not clarified a bit
by having so many available mappings in the BLDC driver. I suspect the
reason for that boils down to a profound lack of a standard way of marking
the motors leads as to phase & polarity, making the builder try every
combination until he hits the right one that just happens to be correct for
the wire hookup sequence he used?

Is there a URL to read that would help me understand that Jon?
Post by Curtis Dutton
Post by Jon Elson
Post by andy pugh
Post by Curtis Dutton
The motor is a 130W motor
1.5mm at around 12000mm/min.
That seems like a very small motor, and a very fast travel.
Is it possible that the motor is simply running out of steam?
Generally when the drive runs out of available voltage
the following error very suddenly grows without bound.
So, you can be at 500mm/min with error of .01mm,
and then at 550mm/min the error rises continuously
because full voltage applied to the motor is only
giving you 520mm/min, to give an example.
So, having the following error increase only a modest bit
at higher speeds may indicate the drive just has a
constant time lag in the internal loop. Or, it may
be a torque limit, where the 130 W motor is nearly
maxed out on current driving the axis at 12 m/min.
Those conditions might cause a bounded error
that increases roughly proportionally to velocity.
Jon
----------------------------------------------------------------------
-------- "Accelerate Dev Cycles with Automated Cross-Browser Testing -
For FREE Instantly run your Selenium tests across 300+ browser/OS
combos. Get unparalleled scalability from the best Selenium testing
platform available. Simple to use. Nothing to install. Get started
now for free." http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
------------------------------------------------------------------------
------ "Accelerate Dev Cycles with Automated Cross-Browser Testing - For
FREE Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform
available. Simple to use. Nothing to install. Get started now for
free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Cheers, Gene
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS
Curtis Dutton
2014-05-01 22:56:23 UTC
Permalink
Thanks to everybody for the information. If anything it helped me
understand the nature of servos better as I've been sitting here tweaking
the machine.

I tried 2 experiments to see if I could get better results.

#1. Give it 80Volts: I hooked the driver up to an 80v PS. The drive clearly
had more power, however the following error is exactly the same behavior.

#2. Run the motor with no load: I disconnected the motor drive belt.
Re-tuned the servo and then started running it in linux cnc. Net result...
exactly the same behavior.

The higher the velocity, the more the error increases, and it appears to be
completely linear.


So it has something to do with some sort of lag in the servo driver. I'll
be emailing leadshine with my fingers crossed hoping they can help.

For what the servo is driving, seat of the pants says that the servo has
more torque then the old stepper did, so the machine should behave better
than it used to.

Thanks!
Curt
Post by Gene Heskett
Post by Curtis Dutton
Ok this makes sense. Thanks all for your explanations. I guess I just
wasn't sure what was reasonable behavior and what wasnt.
So if the motor is rated for 36v, and the drive is rated for 80 volts
max. How much voltage can I get away with delivering to the drives
without damaging equipment?
Thanks,
Curtis
I, as an electronics type, would look at it from the motors rated currant
viewpoint regardless of the family of motor.
The motor more than likely has permanent magnet fields, and allowing more
currant than about 1.25x the nameplate rating (based on my reading on the
subject but I don't have an URL's to offer) gets you into a magnetic
territory where the field magnets can be damaged by reducing their magnetic
strength, and its an instant and permanent effect.
The same effect applies to steppers, usually at currants above 1.25x
nameplate.
Applying an 80 volt supply to a 32 volt rated motor seems like it would be,
if not currant limited in the driver, playing with fire. I would have to
assume they said that assuming a condition where it could spin freely,
letting its counter EMF control the current and therefore the resultant
magnetic field.
This isn't normally a concern with steppers because the 10 to 30x over
voltage is just normal standard operating voltage for them. The drivers
chopper limits on the currant are many times more important to the long
term health of the motors. I see no reason not to apply much the same
thinking to PM field servo motors. Any difference is in where the magnets
are, the steppers magnet is the rotor, where a brushed servo has the magnet
in its stator. But its still the strongest magnet we know how to build in
production quantities.
Now, in servo's I'll have to plead the big dummy because in brushless, hall
effect commutated motors (BLDC?), it seem like a 3rd phase of drive to what
is basically a 3 phase wound stepper motor frame assembly, meaning the
rotor is the PM, would this not also apply?
Or do they have something even more complex for the "BLDC" format? I am
not using them, so I've not spent a lot of time researching how they are
built.
My understanding is quite incomplete for those, and is not clarified a bit
by having so many available mappings in the BLDC driver. I suspect the
reason for that boils down to a profound lack of a standard way of marking
the motors leads as to phase & polarity, making the builder try every
combination until he hits the right one that just happens to be correct for
the wire hookup sequence he used?
Is there a URL to read that would help me understand that Jon?
Post by Curtis Dutton
Post by Jon Elson
Post by andy pugh
Post by Curtis Dutton
The motor is a 130W motor
1.5mm at around 12000mm/min.
That seems like a very small motor, and a very fast travel.
Is it possible that the motor is simply running out of steam?
Generally when the drive runs out of available voltage
the following error very suddenly grows without bound.
So, you can be at 500mm/min with error of .01mm,
and then at 550mm/min the error rises continuously
because full voltage applied to the motor is only
giving you 520mm/min, to give an example.
So, having the following error increase only a modest bit
at higher speeds may indicate the drive just has a
constant time lag in the internal loop. Or, it may
be a torque limit, where the 130 W motor is nearly
maxed out on current driving the axis at 12 m/min.
Those conditions might cause a bounded error
that increases roughly proportionally to velocity.
Jon
----------------------------------------------------------------------
-------- "Accelerate Dev Cycles with Automated Cross-Browser Testing -
For FREE Instantly run your Selenium tests across 300+ browser/OS
combos. Get unparalleled scalability from the best Selenium testing
platform available. Simple to use. Nothing to install. Get started
now for free." http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
------------------------------------------------------------------------
------ "Accelerate Dev Cycles with Automated Cross-Browser Testing - For
FREE Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform
available. Simple to use. Nothing to install. Get started now for
free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Cheers, Gene
--
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos. Get
unparalleled scalability from the best Selenium testing platform available.
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Peter C. Wallace
2014-05-01 23:18:29 UTC
Permalink
Date: Thu, 1 May 2014 18:56:23 -0400
Reply-To: "Enhanced Machine Controller (EMC)"
Subject: Re: [Emc-users] Leadshine ACS806 - Large f-error at higher speeds -
what is acceptable?
Thanks to everybody for the information. If anything it helped me
understand the nature of servos better as I've been sitting here tweaking
the machine.
I tried 2 experiments to see if I could get better results.
#1. Give it 80Volts: I hooked the driver up to an 80v PS. The drive clearly
had more power, however the following error is exactly the same behavior.
#2. Run the motor with no load: I disconnected the motor drive belt.
Re-tuned the servo and then started running it in linux cnc. Net result...
exactly the same behavior.
The higher the velocity, the more the error increases, and it appears to be
completely linear.
Can you add any I term to the drives position control loop?
Enough I term should pull the following error to an average of 0
on a long slew
So it has something to do with some sort of lag in the servo driver. I'll
be emailing leadshine with my fingers crossed hoping they can help.
For what the servo is driving, seat of the pants says that the servo has
more torque then the old stepper did, so the machine should behave better
than it used to.
Thanks!
Curt
Post by Gene Heskett
Post by Curtis Dutton
Ok this makes sense. Thanks all for your explanations. I guess I just
wasn't sure what was reasonable behavior and what wasnt.
So if the motor is rated for 36v, and the drive is rated for 80 volts
max. How much voltage can I get away with delivering to the drives
without damaging equipment?
Thanks,
Curtis
I, as an electronics type, would look at it from the motors rated currant
viewpoint regardless of the family of motor.
The motor more than likely has permanent magnet fields, and allowing more
currant than about 1.25x the nameplate rating (based on my reading on the
subject but I don't have an URL's to offer) gets you into a magnetic
territory where the field magnets can be damaged by reducing their magnetic
strength, and its an instant and permanent effect.
The same effect applies to steppers, usually at currants above 1.25x
nameplate.
Applying an 80 volt supply to a 32 volt rated motor seems like it would be,
if not currant limited in the driver, playing with fire. I would have to
assume they said that assuming a condition where it could spin freely,
letting its counter EMF control the current and therefore the resultant
magnetic field.
This isn't normally a concern with steppers because the 10 to 30x over
voltage is just normal standard operating voltage for them. The drivers
chopper limits on the currant are many times more important to the long
term health of the motors. I see no reason not to apply much the same
thinking to PM field servo motors. Any difference is in where the magnets
are, the steppers magnet is the rotor, where a brushed servo has the magnet
in its stator. But its still the strongest magnet we know how to build in
production quantities.
Now, in servo's I'll have to plead the big dummy because in brushless, hall
effect commutated motors (BLDC?), it seem like a 3rd phase of drive to what
is basically a 3 phase wound stepper motor frame assembly, meaning the
rotor is the PM, would this not also apply?
Or do they have something even more complex for the "BLDC" format? I am
not using them, so I've not spent a lot of time researching how they are
built.
My understanding is quite incomplete for those, and is not clarified a bit
by having so many available mappings in the BLDC driver. I suspect the
reason for that boils down to a profound lack of a standard way of marking
the motors leads as to phase & polarity, making the builder try every
combination until he hits the right one that just happens to be correct for
the wire hookup sequence he used?
Is there a URL to read that would help me understand that Jon?
Post by Curtis Dutton
Post by Jon Elson
Post by andy pugh
Post by Curtis Dutton
The motor is a 130W motor
1.5mm at around 12000mm/min.
That seems like a very small motor, and a very fast travel.
Is it possible that the motor is simply running out of steam?
Generally when the drive runs out of available voltage
the following error very suddenly grows without bound.
So, you can be at 500mm/min with error of .01mm,
and then at 550mm/min the error rises continuously
because full voltage applied to the motor is only
giving you 520mm/min, to give an example.
So, having the following error increase only a modest bit
at higher speeds may indicate the drive just has a
constant time lag in the internal loop. Or, it may
be a torque limit, where the 130 W motor is nearly
maxed out on current driving the axis at 12 m/min.
Those conditions might cause a bounded error
that increases roughly proportionally to velocity.
Jon
----------------------------------------------------------------------
-------- "Accelerate Dev Cycles with Automated Cross-Browser Testing -
For FREE Instantly run your Selenium tests across 300+ browser/OS
combos. Get unparalleled scalability from the best Selenium testing
platform available. Simple to use. Nothing to install. Get started
now for free." http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
------------------------------------------------------------------------
------ "Accelerate Dev Cycles with Automated Cross-Browser Testing - For
FREE Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform
available. Simple to use. Nothing to install. Get started now for
free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Cheers, Gene
--
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos. Get
unparalleled scalability from the best Selenium testing platform available.
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos. Get
unparalleled scalability from the best Selenium testing platform available.
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.
Curtis Dutton
2014-05-02 01:40:59 UTC
Permalink
So the actual tuning software does not provide this parameter for high
speed adjustment. It has both low speed and high speed settings.


However there is a parameter set which can be downloaded and saved and the
high speed Ki is in there. I will alter the saved file and re-upload it to
try different values of Ki.
Date: Thu, 1 May 2014 18:56:23 -0400
Reply-To: "Enhanced Machine Controller (EMC)"
Subject: Re: [Emc-users] Leadshine ACS806 - Large f-error at higher
speeds -
what is acceptable?
Thanks to everybody for the information. If anything it helped me
understand the nature of servos better as I've been sitting here tweaking
the machine.
I tried 2 experiments to see if I could get better results.
#1. Give it 80Volts: I hooked the driver up to an 80v PS. The drive
clearly
had more power, however the following error is exactly the same behavior.
#2. Run the motor with no load: I disconnected the motor drive belt.
Re-tuned the servo and then started running it in linux cnc. Net
result...
exactly the same behavior.
The higher the velocity, the more the error increases, and it appears to
be
completely linear.
Can you add any I term to the drives position control loop?
Enough I term should pull the following error to an average of 0
on a long slew
So it has something to do with some sort of lag in the servo driver. I'll
be emailing leadshine with my fingers crossed hoping they can help.
For what the servo is driving, seat of the pants says that the servo has
more torque then the old stepper did, so the machine should behave better
than it used to.
Thanks!
Curt
Post by Gene Heskett
Post by Curtis Dutton
Ok this makes sense. Thanks all for your explanations. I guess I just
wasn't sure what was reasonable behavior and what wasnt.
So if the motor is rated for 36v, and the drive is rated for 80 volts
max. How much voltage can I get away with delivering to the drives
without damaging equipment?
Thanks,
Curtis
I, as an electronics type, would look at it from the motors rated
currant
Post by Gene Heskett
viewpoint regardless of the family of motor.
The motor more than likely has permanent magnet fields, and allowing
more
Post by Gene Heskett
currant than about 1.25x the nameplate rating (based on my reading on
the
Post by Gene Heskett
subject but I don't have an URL's to offer) gets you into a magnetic
territory where the field magnets can be damaged by reducing their
magnetic
Post by Gene Heskett
strength, and its an instant and permanent effect.
The same effect applies to steppers, usually at currants above 1.25x
nameplate.
Applying an 80 volt supply to a 32 volt rated motor seems like it would
be,
Post by Gene Heskett
if not currant limited in the driver, playing with fire. I would have
to
Post by Gene Heskett
assume they said that assuming a condition where it could spin freely,
letting its counter EMF control the current and therefore the resultant
magnetic field.
This isn't normally a concern with steppers because the 10 to 30x over
voltage is just normal standard operating voltage for them. The drivers
chopper limits on the currant are many times more important to the long
term health of the motors. I see no reason not to apply much the same
thinking to PM field servo motors. Any difference is in where the
magnets
Post by Gene Heskett
are, the steppers magnet is the rotor, where a brushed servo has the
magnet
Post by Gene Heskett
in its stator. But its still the strongest magnet we know how to build
in
Post by Gene Heskett
production quantities.
Now, in servo's I'll have to plead the big dummy because in brushless,
hall
Post by Gene Heskett
effect commutated motors (BLDC?), it seem like a 3rd phase of drive to
what
Post by Gene Heskett
is basically a 3 phase wound stepper motor frame assembly, meaning the
rotor is the PM, would this not also apply?
Or do they have something even more complex for the "BLDC" format? I am
not using them, so I've not spent a lot of time researching how they are
built.
My understanding is quite incomplete for those, and is not clarified a
bit
Post by Gene Heskett
by having so many available mappings in the BLDC driver. I suspect the
reason for that boils down to a profound lack of a standard way of
marking
Post by Gene Heskett
the motors leads as to phase & polarity, making the builder try every
combination until he hits the right one that just happens to be correct
for
Post by Gene Heskett
the wire hookup sequence he used?
Is there a URL to read that would help me understand that Jon?
Post by Curtis Dutton
Post by Jon Elson
Post by andy pugh
Post by Curtis Dutton
The motor is a 130W motor
1.5mm at around 12000mm/min.
That seems like a very small motor, and a very fast travel.
Is it possible that the motor is simply running out of steam?
Generally when the drive runs out of available voltage
the following error very suddenly grows without bound.
So, you can be at 500mm/min with error of .01mm,
and then at 550mm/min the error rises continuously
because full voltage applied to the motor is only
giving you 520mm/min, to give an example.
So, having the following error increase only a modest bit
at higher speeds may indicate the drive just has a
constant time lag in the internal loop. Or, it may
be a torque limit, where the 130 W motor is nearly
maxed out on current driving the axis at 12 m/min.
Those conditions might cause a bounded error
that increases roughly proportionally to velocity.
Jon
----------------------------------------------------------------------
-------- "Accelerate Dev Cycles with Automated Cross-Browser Testing -
For FREE Instantly run your Selenium tests across 300+ browser/OS
combos. Get unparalleled scalability from the best Selenium testing
platform available. Simple to use. Nothing to install. Get started
now for free." http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
------------------------------------------------------------------------
Post by Gene Heskett
Post by Curtis Dutton
------ "Accelerate Dev Cycles with Automated Cross-Browser Testing -
For
Post by Gene Heskett
Post by Curtis Dutton
FREE Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform
available. Simple to use. Nothing to install. Get started now for
free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Cheers, Gene
--
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS
------------------------------------------------------------------------------
Post by Gene Heskett
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos. Get
unparalleled scalability from the best Selenium testing platform
available.
Post by Gene Heskett
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos. Get
unparalleled scalability from the best Selenium testing platform
available.
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Peter Wallace
Mesa Electronics
(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos. Get
unparalleled scalability from the best Selenium testing platform available.
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Kirk Wallace
2014-05-02 02:25:41 UTC
Permalink
On 05/01/2014 03:56 PM, Curtis Dutton wrote:

... snip
Post by Curtis Dutton
#2. Run the motor with no load: I disconnected the motor drive belt.
Re-tuned the servo and then started running it in linux cnc. Net result...
exactly the same behavior.
... snip

I think you should tune the drive with the load that it will normally
see -- with the motor connected to the lead screw, table and such.
--
Kirk Wallace
http://www.wallacecompany.com/machine_shop/
http://www.wallacecompany.com/E45/
Steve Blackmore
2014-05-02 07:17:12 UTC
Permalink
Post by Curtis Dutton
Thanks to everybody for the information. If anything it helped me
understand the nature of servos better as I've been sitting here tweaking
the machine.
So it has something to do with some sort of lag in the servo driver. I'll
be emailing leadshine with my fingers crossed hoping they can help.
Here are the Specs

Position following error : +/-1 count
Velocity accuracy: +/-2rpm
Maximum acceleration speed (No Load) : 80 rpm/ms2
Input frequency up to 600 kHz
Maximum speed : 4000 rpm
Allowable low speed reaches1 rpm
Positioning accuracy :+/-1 count
Suitable for 18 - 80 VDC AC servo

Are you within them?

Have you used the available tuning software using the serial connection
to the drive? If not you need to to see what's going on and tune it
properly. You will need to adjust the Kp, Kd and Vp parameters. All
explained in the software manual.

Steve Blackmore
--
Curtis Dutton
2014-05-02 22:36:22 UTC
Permalink
Today I received a second AC806 drive in the mail. It is a V1 ACS806.

The V1 is simpler to tune. It only has a current loop Kp and Ki, as well as
position loop Kp, Ki, and Kd.

I tried that on the same motor that the AVS806V2 is having trouble with.
Following error is now down to under 20 counts at full velocity, about
1400rpm, and down to under 5 counts or so at low velocities. Which is well
within my needs.

So by process of elimination it is some form of setting on the AVS806V2
which is the problem. (which has more parameters, etc... involved with it)

I'll keep troubleshooting the V2, but being that an older, basically
identical drive is performing properly, that helps me eliminate motor/ps
sizing concerns.
Post by Steve Blackmore
Post by Curtis Dutton
Thanks to everybody for the information. If anything it helped me
understand the nature of servos better as I've been sitting here tweaking
the machine.
So it has something to do with some sort of lag in the servo driver. I'll
be emailing leadshine with my fingers crossed hoping they can help.
Here are the Specs
Position following error : +/-1 count
Velocity accuracy: +/-2rpm
Maximum acceleration speed (No Load) : 80 rpm/ms2
Input frequency up to 600 kHz
Maximum speed : 4000 rpm
Allowable low speed reaches1 rpm
Positioning accuracy :+/-1 count
Suitable for 18 - 80 VDC AC servo
Are you within them?
Have you used the available tuning software using the serial connection
to the drive? If not you need to to see what's going on and tune it
properly. You will need to adjust the Kp, Kd and Vp parameters. All
explained in the software manual.
Steve Blackmore
--
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos. Get
unparalleled scalability from the best Selenium testing platform available.
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Emc-users mailing list
https://lists.sourceforge.net/lists/listinfo/emc-users
Loading...