Re: [Emc-users] What Would You Suggest?

[John Dammeyer]

Sigh...  A web link or two would be more helpful than just a comment.

> From: Robin Szemeti via Emc-users [mailto:emc-users@lists.sourceforge.net]
> 
> Sigh ...
> 
> " Perhaps you can find some and post those that do show flat torque up to a
> 'corner speed'?"  ... all ... every single one,  when connected to a
> current limited driver.
> 
> On Sun, 6 Feb 2022 at 00:45, John Dammeyer  wrote:
> 
> > No.  The motors are designed to handle N amperes although they get quite
> > warm that should be a 24/7 rating. Because they get so warm many drivers
> > back the current off when the motors have been idle for a period of time.
> >
> > Let's take a step backwards with a long explanation.   When I first sized
> > the motor for the knee mill I tested how much torque was required to
> > overcome 'static' friction by attaching a bar and adding weight until it
> > started turning.  The 'kinetic' friction is always less so as a rule of
> > thumb using twice that experimentally determined torque value was a good
> > starting point.
> >
> > But you can also use math if you know the weight of the table you are
> > lifting.  I didn't but if I guess at 325 lbs with a 1" lead 4 TPI ACME
> > leadscrew with Bronze nut
> > https://daycounter.com/Calculators/Lead-Screw-Force-Torque-Calculator.phtml
> > I get pretty close to the value I determined experimentally which is 600
> > oz-in
> >
> > So choosing a 600 oz-in motor for roughly 600 oz-in torque determined
> > experimentally looked like it would work.  Then looking at the torque curve
> > I determined that to get that 600 oz-in torque at the desired speed would
> > require a 3:1 reduction belt drive to get the 600 oz-in at target speed.
> >  I used a Gecko Drive with a 60VDC power supply and a motor with 6.5A,
> > 2.2mH rating.  The calculator here
> > https://daycounter.com/Calculators/Stepper-Motor-Calculator.phtml
> > shows max 10.5 RPS or 630 RPM and therefore with 3:1 reduction I'd be
> > looking at 210 RPM into the lead screw drive.
> >
> > With a 4 TPI drive that means 52 ipm of table motion.  In reality the best
> > I got was 15 ipm before the motor locked up.   Click on the torque curve
> > and at 630RPM it's about 1.5NM which is 33% of the 4.5NM rating (640 oz-in).
> >
> > https://www.automationtechnologiesinc.com/products-page/nema-34/nema-34-640-oz-in-stepper-motor-kl34h280-45-8b-dual-
> shaft
> >
> > Long story short is that I updated to a 1200 oz-in motor and had reliable
> > motion with a Gecko Drive and 60VDC at 25 ipm from LCNC or MACH3.  This
> > moved the torque up but not really the top speed since now this motor has
> > 6mH rather than 2.2mH with a 6A drive rating.  I might have been better to
> > have changed the reduction belt drive to 2:1 or less.
> >
> > Back to that pesky power rating.  At 60V the motor calculator suggests
> > it's drawing 380W.  Change the voltage to 120VDC and with the same current
> > we get twice the RPM and twice the power 750W.  So then theoretically I
> > should be able to get 50 ipm reliably if I had a driver that could handle
> > 120VDC and if the motor winding insulation was rated for that voltage.
> >
> > Instead I ball-parked an AC Servo motor, 750W, 3000 RPM, 3.5NM (495 oz-in)
> > and max 2000 RPM as the solution.  This drive runs off 220VAC and is rated
> > at 3A with a peak torque of 10.5NM.  With 3:1 and that 495 oz-in torque
> > value all the way up to 2000 RPM (667 shaft RPM or 167 ipm) I can
> > comfortably run this motor at 150 ipm without it ever faulting.  That's 6x
> > the stepper motor speed.
> >
> > What that suggests is what everyone already knows is that stepper torque
> > falls off badly at higher RPM even with a higher voltage but they are
> > really good at low speeds.  Change things to get more torque and if the
> > inductance goes up the torque at low RPM is there but not at the high RPM.
> > Unless your driver and motor can handle a correspondingly higher voltage.
> >
> > But look around and most drivers for steppers are in the 24V to 80V range
> > so with steppers,  motor power in watts means nothing if you can't get the
> > voltage up.   Now if you go 1:1 then 0.25"/200 results in a resolution of
> > 0.00125".  Torque falls off to 50% with half stepping and micro-stepping
> > results in a 70.7% reduction in max torque.  And Micro-stepping also can
> > require as many as 4 steps before static torque is overcome and the motor
> > shaft turning.  That's just due to the nature of the current waveform so
> > for accuracy you must design for full step resolution.
> >
> > John
> >
> >
> > > -Original Message-
> > > From: Thaddeus Waldner [mailto:thadw...@gmail.com]
> > > Sent: February-05-22 12:40 PM
> > > To: Enhanced Machine Controller (EMC)
> > > Subject: Re: [Emc-users] What Would You Suggest?
> > >
> > > I believe the flat part of the curve is defined by the motor thermal
> > limits. Is this correct?
> > >
> > > > On Feb 5, 2022, at 11:44 AM, John Dammeyer 
> > wrote:
> > > >
> > > > ?All f

Re: [Emc-users] What Would You Suggest?

[John Dammeyer]

The link I posted for stepper motors might give you a better idea.  Also 
there's this data sheet which I've found has a good description on how 
micro-stepping works.   How the current in one phase is set with respect to the 
other for each micro-step.  
https://www.ti.com/lit/ds/symlink/lmd18245.pdf

>From the perspective of current draw think about what an inductor does to 
>current flow.   If you apply a voltage across a resistor the current starts 
>flowing pretty well instantly.  When you put a voltage across an inductor it 
>takes time for the current to reach maximum.

So what is maximum?  Well if the current rating of the motor is say 3A and the 
winding resistance is 1 ohm then Ohms law say the motor voltage to create 3A 
must be 3V.   But it doesn't happen immediately because the inductance slows it 
down.  And since it's not at 3A right away the motor doesn't develop full 
torque right away.

One way to get the current to max faster is to use a higher voltage.  Let's use 
30V instead.  30V/1 Ohm is 30A.  Oops.  Magic smoke just came out of the motor. 
 So what he drivers do is sense the current and when it reaches 3A turn off the 
voltage.  When it drops below 3A which also takes time because of the stored 
energy in the coil, the controller switches on the voltage again.  It's called 
chopping.

Now instead of say taking 2.5% of the step pulse to reach max current it takes 
0.25%  so max torque is reached and held for much longer.   Look at page 23 of 
the data sheet I linked and you can see from the table that the one big feature 
about stepper motors is the winding current changes direction every step.  When 
you start micro-stepping current through one winding is 100% while the other is 
0%  So half the torque if you use the T=Amp x Turns rule and reversing the 
current also takes time.

Again, 30V for a 3V coil or even 60V for a 3V coil means virtually zero time to 
reach full current.  Let me take a little side trip here.  If you've ever 
played with slot cars or small dc motors and increased the voltage to make them 
turn faster you have, as a kid, learned the simple rule.  More voltage makes 
the motor spin faster.  But why?

With brushed DC motors as the windings that are not powered move through the 
magnetic field, just like a generator a voltage is induced in the windings.
Simply put this induced voltage equals the applied voltage at the speed created 
by the applied voltage.  So if it turns 1000 RPM at 6V and you apply 12V it may 
will reach 2000 RPM (for example) but will go no faster.  Change it to 18V and 
now we get 3000 RPM.  How does relate to stepper motors?

To change the direction of the current through the winding the controller has 
to exceed this generated voltage caused by the motor turning through the 
magnets.  When the induced voltage reaches 3V below the applied voltage; 27V 
induced and 30V applied the time for the current to reach that 3A is back up to 
2.5mS.  Trouble is the step pulses are 1.25mS long.  So again with lots of 
simplification (and a linear viewpoint), the current only reaches 1.5A before 
it's time to change the direction of the current again.  That means only 50% of 
the desired torque.  It's actually less because the actual math will talk about 
the area under that triangular waveform and as Robin mentioned, voltage * 
current = power and remains steady.  But it doesn't help us for the torque.

Double the voltage, the time to change direction of current decreases, there're 
more current in the windings for a longer time and the torque increases. And 
yes,  2x voltage * current under the curve equals a doubling of power.  But 
again, we don't care.  Unless we know that within a given step period that the 
'average' current has increased we don't have more torque.   So indirectly 
power might well remain the same but generally the one _fixed_ parameter in a 
stepper drive system is the power supply voltage.  We don't change that.  So 
the average current through the windings during a specific step determines the 
motor torque.

Power as a number just isn't important once you've bought the power supply and 
the stepper driver that can set 1A to 8A for different motors.   Now you choose 
the motor that has a torque verses rpm that matches your system.   There are 
very few, if any stepper drivers that use 200V.  Most are max 80V or so.

So where is the flat part of the graph?  At a very low speed if the stepper 
driver voltage is 3V.  Much longer and at a higher speed with 24V and really a 
lot longer at 80V.  But unlike DC motors, it's the reversing of the current 
through the windings every step that is the Achilles heel of the stepper motor 
architecture.   So although the power part of the curve might look flat the 
drivers don't keep the average current the same for faster step rates by 
allowing the winding current to increase above the rated value so the area 
under the curve remains the same regardless of step pulse width.  And that

Re: [Emc-users] What Would You Suggest?

[Thaddeus Waldner]

> On Feb 5, 2022, at 6:48 PM, John Dammeyer  wrote:
> 
> No.  The motors are designed to handle N amperes although they get quite warm 
> that should be a 24/7 rating. Because they get so warm many drivers back the 
> current off when the motors have been idle for a period of time. 

Yes, N amperes.  Is that not the flat part of the graph? 

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] What Would You Suggest?

[Robin Szemeti via Emc-users]

Sigh ...

" Perhaps you can find some and post those that do show flat torque up to a
'corner speed'?"  ... all ... every single one,  when connected to a
current limited driver.

On Sun, 6 Feb 2022 at 00:45, John Dammeyer  wrote:

> No.  The motors are designed to handle N amperes although they get quite
> warm that should be a 24/7 rating. Because they get so warm many drivers
> back the current off when the motors have been idle for a period of time.
>
> Let's take a step backwards with a long explanation.   When I first sized
> the motor for the knee mill I tested how much torque was required to
> overcome 'static' friction by attaching a bar and adding weight until it
> started turning.  The 'kinetic' friction is always less so as a rule of
> thumb using twice that experimentally determined torque value was a good
> starting point.
>
> But you can also use math if you know the weight of the table you are
> lifting.  I didn't but if I guess at 325 lbs with a 1" lead 4 TPI ACME
> leadscrew with Bronze nut
> https://daycounter.com/Calculators/Lead-Screw-Force-Torque-Calculator.phtml
> I get pretty close to the value I determined experimentally which is 600
> oz-in
>
> So choosing a 600 oz-in motor for roughly 600 oz-in torque determined
> experimentally looked like it would work.  Then looking at the torque curve
> I determined that to get that 600 oz-in torque at the desired speed would
> require a 3:1 reduction belt drive to get the 600 oz-in at target speed.
>  I used a Gecko Drive with a 60VDC power supply and a motor with 6.5A,
> 2.2mH rating.  The calculator here
> https://daycounter.com/Calculators/Stepper-Motor-Calculator.phtml
> shows max 10.5 RPS or 630 RPM and therefore with 3:1 reduction I'd be
> looking at 210 RPM into the lead screw drive.
>
> With a 4 TPI drive that means 52 ipm of table motion.  In reality the best
> I got was 15 ipm before the motor locked up.   Click on the torque curve
> and at 630RPM it's about 1.5NM which is 33% of the 4.5NM rating (640 oz-in).
>
> https://www.automationtechnologiesinc.com/products-page/nema-34/nema-34-640-oz-in-stepper-motor-kl34h280-45-8b-dual-shaft
>
> Long story short is that I updated to a 1200 oz-in motor and had reliable
> motion with a Gecko Drive and 60VDC at 25 ipm from LCNC or MACH3.  This
> moved the torque up but not really the top speed since now this motor has
> 6mH rather than 2.2mH with a 6A drive rating.  I might have been better to
> have changed the reduction belt drive to 2:1 or less.
>
> Back to that pesky power rating.  At 60V the motor calculator suggests
> it's drawing 380W.  Change the voltage to 120VDC and with the same current
> we get twice the RPM and twice the power 750W.  So then theoretically I
> should be able to get 50 ipm reliably if I had a driver that could handle
> 120VDC and if the motor winding insulation was rated for that voltage.
>
> Instead I ball-parked an AC Servo motor, 750W, 3000 RPM, 3.5NM (495 oz-in)
> and max 2000 RPM as the solution.  This drive runs off 220VAC and is rated
> at 3A with a peak torque of 10.5NM.  With 3:1 and that 495 oz-in torque
> value all the way up to 2000 RPM (667 shaft RPM or 167 ipm) I can
> comfortably run this motor at 150 ipm without it ever faulting.  That's 6x
> the stepper motor speed.
>
> What that suggests is what everyone already knows is that stepper torque
> falls off badly at higher RPM even with a higher voltage but they are
> really good at low speeds.  Change things to get more torque and if the
> inductance goes up the torque at low RPM is there but not at the high RPM.
> Unless your driver and motor can handle a correspondingly higher voltage.
>
> But look around and most drivers for steppers are in the 24V to 80V range
> so with steppers,  motor power in watts means nothing if you can't get the
> voltage up.   Now if you go 1:1 then 0.25"/200 results in a resolution of
> 0.00125".  Torque falls off to 50% with half stepping and micro-stepping
> results in a 70.7% reduction in max torque.  And Micro-stepping also can
> require as many as 4 steps before static torque is overcome and the motor
> shaft turning.  That's just due to the nature of the current waveform so
> for accuracy you must design for full step resolution.
>
> John
>
>
> > -Original Message-
> > From: Thaddeus Waldner [mailto:thadw...@gmail.com]
> > Sent: February-05-22 12:40 PM
> > To: Enhanced Machine Controller (EMC)
> > Subject: Re: [Emc-users] What Would You Suggest?
> >
> > I believe the flat part of the curve is defined by the motor thermal
> limits. Is this correct?
> >
> > > On Feb 5, 2022, at 11:44 AM, John Dammeyer 
> wrote:
> > >
> > > ?All fine and good but doesn't help anyone choose a stepper motor.
> For example:
> > > https://www.geckodrive.com/support/step-motor-basics.html
> > > tries to explain corner speed but fakes the curves by showing that the
> torque of a stepper motor is constant up to a certain speed.
> > I've yet to see any curves of real motors that look li

Re: [Emc-users] What Would You Suggest?

[John Dammeyer]

No.  The motors are designed to handle N amperes although they get quite warm 
that should be a 24/7 rating. Because they get so warm many drivers back the 
current off when the motors have been idle for a period of time.  

Let's take a step backwards with a long explanation.   When I first sized the 
motor for the knee mill I tested how much torque was required to overcome 
'static' friction by attaching a bar and adding weight until it started 
turning.  The 'kinetic' friction is always less so as a rule of thumb using 
twice that experimentally determined torque value was a good starting point.  

But you can also use math if you know the weight of the table you are lifting.  
I didn't but if I guess at 325 lbs with a 1" lead 4 TPI ACME leadscrew with 
Bronze nut
https://daycounter.com/Calculators/Lead-Screw-Force-Torque-Calculator.phtml
I get pretty close to the value I determined experimentally which is 600 oz-in

So choosing a 600 oz-in motor for roughly 600 oz-in torque determined 
experimentally looked like it would work.  Then looking at the torque curve I 
determined that to get that 600 oz-in torque at the desired speed would require 
a 3:1 reduction belt drive to get the 600 oz-in at target speed.   I used a 
Gecko Drive with a 60VDC power supply and a motor with 6.5A, 2.2mH rating.  The 
calculator here
https://daycounter.com/Calculators/Stepper-Motor-Calculator.phtml
shows max 10.5 RPS or 630 RPM and therefore with 3:1 reduction I'd be looking 
at 210 RPM into the lead screw drive.  

With a 4 TPI drive that means 52 ipm of table motion.  In reality the best I 
got was 15 ipm before the motor locked up.   Click on the torque curve and at 
630RPM it's about 1.5NM which is 33% of the 4.5NM rating (640 oz-in).
https://www.automationtechnologiesinc.com/products-page/nema-34/nema-34-640-oz-in-stepper-motor-kl34h280-45-8b-dual-shaft

Long story short is that I updated to a 1200 oz-in motor and had reliable 
motion with a Gecko Drive and 60VDC at 25 ipm from LCNC or MACH3.  This moved 
the torque up but not really the top speed since now this motor has 6mH rather 
than 2.2mH with a 6A drive rating.  I might have been better to have changed 
the reduction belt drive to 2:1 or less.

Back to that pesky power rating.  At 60V the motor calculator suggests it's 
drawing 380W.  Change the voltage to 120VDC and with the same current we get 
twice the RPM and twice the power 750W.  So then theoretically I should be able 
to get 50 ipm reliably if I had a driver that could handle 120VDC and if the 
motor winding insulation was rated for that voltage.

Instead I ball-parked an AC Servo motor, 750W, 3000 RPM, 3.5NM (495 oz-in) and 
max 2000 RPM as the solution.  This drive runs off 220VAC and is rated at 3A 
with a peak torque of 10.5NM.  With 3:1 and that 495 oz-in torque value all the 
way up to 2000 RPM (667 shaft RPM or 167 ipm) I can comfortably run this motor 
at 150 ipm without it ever faulting.  That's 6x the stepper motor speed.

What that suggests is what everyone already knows is that stepper torque falls 
off badly at higher RPM even with a higher voltage but they are really good at 
low speeds.  Change things to get more torque and if the inductance goes up the 
torque at low RPM is there but not at the high RPM.  Unless your driver and 
motor can handle a correspondingly higher voltage.

But look around and most drivers for steppers are in the 24V to 80V range so 
with steppers,  motor power in watts means nothing if you can't get the voltage 
up.   Now if you go 1:1 then 0.25"/200 results in a resolution of 0.00125".  
Torque falls off to 50% with half stepping and micro-stepping results in a 
70.7% reduction in max torque.  And Micro-stepping also can require as many as 
4 steps before static torque is overcome and the motor shaft turning.  That's 
just due to the nature of the current waveform so for accuracy you must design 
for full step resolution.

John


> -Original Message-
> From: Thaddeus Waldner [mailto:thadw...@gmail.com]
> Sent: February-05-22 12:40 PM
> To: Enhanced Machine Controller (EMC)
> Subject: Re: [Emc-users] What Would You Suggest?
> 
> I believe the flat part of the curve is defined by the motor thermal limits. 
> Is this correct?
> 
> > On Feb 5, 2022, at 11:44 AM, John Dammeyer  wrote:
> >
> > ?All fine and good but doesn't help anyone choose a stepper motor.  For 
> > example:
> > https://www.geckodrive.com/support/step-motor-basics.html
> > tries to explain corner speed but fakes the curves by showing that the 
> > torque of a stepper motor is constant up to a certain speed.
> I've yet to see any curves of real motors that look like that.
> >
> > Perhaps you can find some and post those that do show flat torque up to a 
> > 'corner speed'?
> >
> > In either case that doesn't really help anyone choose a motor so that Gecko 
> > article and corner speed are effectively techo-babble
> framed in a way to help them sell their drivers which are limited to 80VDC.
> 

Re: [Emc-users] Execute MDI command, from hal pin

[Kenneth Lerman]

I've implemented some safe probing routines here
.
They are examples of subroutines that might be called by external switches
-- or in my case from buttons on a pendant.

Ken
Kenneth Lerman
55 Main Street
Newtown, CT 06470



On Sat, Feb 5, 2022 at 6:54 PM Kenneth Lerman  wrote:

> The other half of the pins that Sebastian mentioned is that the commands
> must be defined in the .ini file.
>
> This
>  
> will
> tell you that you can put things like:
>
>-
>
>*MDI_COMMAND = G53 G0 X0 Y0 Z0* - An MDI command can be executed by
>using halui.mdi-command-00. Increment the number for each command listed in
>the [HALUI] section.
>
> The first such entry will be executed when halui.mdi-command-00 is
> asserted. The second corresponds to -01, etc.
>
> You might then write subroutines to do such things as jog in +X, -X, +Y,
> -Y, etc. Those subroutines might save the direction of the last jog. You
> could also have a subroutine to use your touch probe. It could be "smart
> enough" to probe in the direction of the last jog.
>
> Ken
>
>
> Kenneth Lerman
> 55 Main Street
> Newtown, CT 06470
>
>
>
> On Sat, Feb 5, 2022 at 5:56 PM Sebastian Kuzminsky 
> wrote:
>
>> On 2/5/22 1:13 PM, Nicklas SB Karlsson wrote:
>> > Anybody know about a method to execute a MDI command with g-code then
>> on
>> > a hal pin flank?
>>
>> I'm not entirely sure what you're asking for, but one of my machines has
>> a hardware button coming in to HAL through a gpio, connected to a halui
>> input which triggers an MDI command.  Check out halui.mdi-command-XX:
>>
>> http://linuxcnc.org/docs/devel/html/man/man1/halui.1.html#PINS
>>
>>
>> --
>> Sebastian Kuzminsky
>>
>>
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>
>

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Execute MDI command, from hal pin

[Kenneth Lerman]

The other half of the pins that Sebastian mentioned is that the commands
must be defined in the .ini file.

This

will
tell you that you can put things like:

   -

   *MDI_COMMAND = G53 G0 X0 Y0 Z0* - An MDI command can be executed by
   using halui.mdi-command-00. Increment the number for each command listed in
   the [HALUI] section.

The first such entry will be executed when halui.mdi-command-00 is
asserted. The second corresponds to -01, etc.

You might then write subroutines to do such things as jog in +X, -X, +Y,
-Y, etc. Those subroutines might save the direction of the last jog. You
could also have a subroutine to use your touch probe. It could be "smart
enough" to probe in the direction of the last jog.

Ken


Kenneth Lerman
55 Main Street
Newtown, CT 06470



On Sat, Feb 5, 2022 at 5:56 PM Sebastian Kuzminsky  wrote:

> On 2/5/22 1:13 PM, Nicklas SB Karlsson wrote:
> > Anybody know about a method to execute a MDI command with g-code then on
> > a hal pin flank?
>
> I'm not entirely sure what you're asking for, but one of my machines has
> a hardware button coming in to HAL through a gpio, connected to a halui
> input which triggers an MDI command.  Check out halui.mdi-command-XX:
>
> http://linuxcnc.org/docs/devel/html/man/man1/halui.1.html#PINS
>
>
> --
> Sebastian Kuzminsky
>
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Execute MDI command, from hal pin

[Sebastian Kuzminsky]

On 2/5/22 1:13 PM, Nicklas SB Karlsson wrote:
Anybody know about a method to execute a MDI command with g-code then on 
a hal pin flank?


I'm not entirely sure what you're asking for, but one of my machines has 
a hardware button coming in to HAL through a gpio, connected to a halui 
input which triggers an MDI command.  Check out halui.mdi-command-XX:


http://linuxcnc.org/docs/devel/html/man/man1/halui.1.html#PINS


--
Sebastian Kuzminsky


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Execute MDI command, from hal pin

[andy pugh]

On Sat, 5 Feb 2022 at 20:24, Nicklas SB Karlsson  wrote:
>
> Anybody know about a method to execute a MDI command with g-code then on
> a hal pin flank?

What is the MDI_COMMAND?

I think that if it is pure G-code then it can be an OCALL
and you can use that same format in your G-code.

Alternatively, perhaps configure it as a use M-code or a remapped G-code.

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1912


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] What Would You Suggest?

[Thaddeus Waldner]

I believe the flat part of the curve is defined by the motor thermal limits. Is 
this correct?

> On Feb 5, 2022, at 11:44 AM, John Dammeyer  wrote:
> 
> All fine and good but doesn't help anyone choose a stepper motor.  For 
> example:
> https://www.geckodrive.com/support/step-motor-basics.html
> tries to explain corner speed but fakes the curves by showing that the torque 
> of a stepper motor is constant up to a certain speed.  I've yet to see any 
> curves of real motors that look like that.
> 
> Perhaps you can find some and post those that do show flat torque up to a 
> 'corner speed'?
> 
> In either case that doesn't really help anyone choose a motor so that Gecko 
> article and corner speed are effectively techo-babble framed in a way to help 
> them sell their drivers which are limited to 80VDC.
> 
> Similarly the stepper motor suppliers provide 1/2 step curves leaving out the 
> resonance point so unless you stay below this 'corner frequency' and never 
> reach that point the torque curves are somewhat obscure.
> 
> Perhaps explain how _you_ choose a stepper motor for a given axis?  What 
> process do you go through to do this?  That might help more.  
> 
> John
> 
> 
> 
>> From: Robin Szemeti via Emc-users [mailto:emc-users@lists.sourceforge.net]
>> 
>> "comment about corner frequency with stepper motors _might_ well be valid
>> as long as the maximum current for each step is reached before or at the
>> end of the ste"
>> 
>> Yes, that is exactly what the corne frequency is ... the step frequency at
>> which the current no longer reaches the desired value before the end of the
>> step. It's obviously dependent on inductance and maximum available drive
>> voltage.
>> 
>>> On Sat, 5 Feb 2022 at 11:50, Robin Szemeti  wrote:
>>> 
>>> John,
>>> 
>>> You are fundamentally incorrect when you state " the torque of the motor
>>> to drop off the faster it goes" .. although the back EMF is correct, with a
>>> modern current limited drive, the torque is flat until the corner
>>> frequency, then drops off  ... up to the corner frequency the torque is
>>> constant with a good current-limited drive, above the corner frequency the
>>> torque drops off, power is constant.  You are perhaps confusing the raw
>>> torque/speed curve of a motor fed from a constant voltage source, which is
>>> useful but is not how they are typically used in practice.
>>> 
>>> 
>>> https://res.cloudinary.com/engineering-com/image/upload/w_640,h_640,c_limit,q_auto,f_auto/image002_bezhrr.jpg
>>> 
 On Sat, 5 Feb 2022 at 09:00, John Dammeyer  wrote:
>>> 
 Hi Chris,
 My issue is that a comment about corner frequency with stepper motors
 _might_ well be valid as long as the maximum current for each step is
 reached before or at the end of the step.  But the motor is turning pretty
 slowly there compared to how they are used in real life.
 
 However the comment about corner frequency with respect to steppers
 perhaps is only backed by alternative facts?
 
 I must admit I've not investigated in detail the closed loop steppers.
 The price of an industrial version I worked with was more than the price of
 an AC servo and at higher speeds I could stop the pulley with my fingers.
 Yes. It faulted.  But that isn't really the point.  The DC and AC servos at
 higher speeds just work better.
 
 Stepper motors work great at low speeds usually directly coupled.
 Contrary to popular belief the micro-stepping doesn't improve resolution
 but gets rid of resonance and gives the appearance of better resolution.
 But it doesn't change the fact that the current still has to reverse every
 full step.  I believe that in fact Gecko drives improve high speed torque
 by switching back to full step mode above the resonance velocity.
 
 Now instead of 0.707 x max current in both windings (at the most) we're
 back to 100% in both with an increase in torque.  Absolutely nothing to do
 with corner frequencies whatever they might be or how they are determined.
 
 John
 
 
 
 
 
> From: Chris Albertson [mailto:albertson.ch...@gmail.com]
> 
> John,
> 
> You described it correctly.  But I think what Robin meant by "Corner
> Frequency" might be the peak of the power vs. RPM graph.  Basically, the
> frequency where power output starts to fall with RPM.
> 
> But now it can get worse, or really better but more complex.   We have
> these so-called "closed loop stepper drivers and also a few people are
> running the steppers as if they were many-pole BLDC analog (continuous,
> non-stepping) mortors
> 
> On Fri, Feb 4, 2022 at 2:39 PM John Dammeyer 
 wrote:
> 
>> I disagree.  The physics of the motor, which include inductance along
 with
>> the generated back emf from the motor spinning in the magnetic field,
 is
>> what cause the torque of the motor to drop off the 

[Emc-users] Execute MDI command, from hal pin

[Nicklas SB Karlsson]

Anybody know about a method to execute a MDI command with g-code then on 
a hal pin flank?



Nicklas Karlsson



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] What Would You Suggest?

[Jérémie Tarot]

Le sam. 5 févr. 2022 à 18:44, John Dammeyer  a
écrit :

> All fine and good but doesn't help anyone choose a stepper motor.
> ..
> Perhaps explain how _you_ choose a stepper motor for a given axis?  What
> process do you go through to do this?


Earlier today reading this thread, I thought such a stepper selection how
to would be a priceless addition to our implementor documentation! Same for
other kinds of motors, btw.

Any of yall would be willing to contribute, even in part, or just with good
actionnable references, to this document(s)?

TY
Jérémie

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] What Would You Suggest?

[John Dammeyer]

All fine and good but doesn't help anyone choose a stepper motor.  For example:
https://www.geckodrive.com/support/step-motor-basics.html
tries to explain corner speed but fakes the curves by showing that the torque 
of a stepper motor is constant up to a certain speed.  I've yet to see any 
curves of real motors that look like that.

Perhaps you can find some and post those that do show flat torque up to a 
'corner speed'?

In either case that doesn't really help anyone choose a motor so that Gecko 
article and corner speed are effectively techo-babble framed in a way to help 
them sell their drivers which are limited to 80VDC.

Similarly the stepper motor suppliers provide 1/2 step curves leaving out the 
resonance point so unless you stay below this 'corner frequency' and never 
reach that point the torque curves are somewhat obscure.

Perhaps explain how _you_ choose a stepper motor for a given axis?  What 
process do you go through to do this?  That might help more.  

John



> From: Robin Szemeti via Emc-users [mailto:emc-users@lists.sourceforge.net]
> 
>  "comment about corner frequency with stepper motors _might_ well be valid
> as long as the maximum current for each step is reached before or at the
> end of the ste"
> 
> Yes, that is exactly what the corne frequency is ... the step frequency at
> which the current no longer reaches the desired value before the end of the
> step. It's obviously dependent on inductance and maximum available drive
> voltage.
> 
> On Sat, 5 Feb 2022 at 11:50, Robin Szemeti  wrote:
> 
> > John,
> >
> > You are fundamentally incorrect when you state " the torque of the motor
> > to drop off the faster it goes" .. although the back EMF is correct, with a
> > modern current limited drive, the torque is flat until the corner
> > frequency, then drops off  ... up to the corner frequency the torque is
> > constant with a good current-limited drive, above the corner frequency the
> > torque drops off, power is constant.  You are perhaps confusing the raw
> > torque/speed curve of a motor fed from a constant voltage source, which is
> > useful but is not how they are typically used in practice.
> >
> >
> > https://res.cloudinary.com/engineering-com/image/upload/w_640,h_640,c_limit,q_auto,f_auto/image002_bezhrr.jpg
> >
> > On Sat, 5 Feb 2022 at 09:00, John Dammeyer  wrote:
> >
> >> Hi Chris,
> >> My issue is that a comment about corner frequency with stepper motors
> >> _might_ well be valid as long as the maximum current for each step is
> >> reached before or at the end of the step.  But the motor is turning pretty
> >> slowly there compared to how they are used in real life.
> >>
> >> However the comment about corner frequency with respect to steppers
> >> perhaps is only backed by alternative facts?
> >>
> >> I must admit I've not investigated in detail the closed loop steppers.
> >> The price of an industrial version I worked with was more than the price of
> >> an AC servo and at higher speeds I could stop the pulley with my fingers.
> >> Yes. It faulted.  But that isn't really the point.  The DC and AC servos at
> >> higher speeds just work better.
> >>
> >> Stepper motors work great at low speeds usually directly coupled.
> >> Contrary to popular belief the micro-stepping doesn't improve resolution
> >> but gets rid of resonance and gives the appearance of better resolution.
> >> But it doesn't change the fact that the current still has to reverse every
> >> full step.  I believe that in fact Gecko drives improve high speed torque
> >> by switching back to full step mode above the resonance velocity.
> >>
> >> Now instead of 0.707 x max current in both windings (at the most) we're
> >> back to 100% in both with an increase in torque.  Absolutely nothing to do
> >> with corner frequencies whatever they might be or how they are determined.
> >>
> >> John
> >>
> >>
> >>
> >>
> >>
> >> > From: Chris Albertson [mailto:albertson.ch...@gmail.com]
> >> >
> >> > John,
> >> >
> >> > You described it correctly.  But I think what Robin meant by "Corner
> >> > Frequency" might be the peak of the power vs. RPM graph.  Basically, the
> >> > frequency where power output starts to fall with RPM.
> >> >
> >> > But now it can get worse, or really better but more complex.   We have
> >> > these so-called "closed loop stepper drivers and also a few people are
> >> > running the steppers as if they were many-pole BLDC analog (continuous,
> >> > non-stepping) mortors
> >> >
> >> > On Fri, Feb 4, 2022 at 2:39 PM John Dammeyer 
> >> wrote:
> >> >
> >> > > I disagree.  The physics of the motor, which include inductance along
> >> with
> >> > > the generated back emf from the motor spinning in the magnetic field,
> >> is
> >> > > what cause the torque of the motor to drop off the faster it goes.
> >> > >
> >> > > The problem is to spin a stepper motor you have to not just change to
> >> a
> >> > > new winding like a DC motor does but completely reverse the direction
> >> of
> >> > > the current

Re: [Emc-users] What Would You Suggest?

[andy pugh]

On Sat, 5 Feb 2022 at 12:34, Stuart Stevenson  wrote:
>
> Guys,
> Causing me to 'think' this early hurts my head but I need to express this
> thought. It may be totally crazy but is there a way to 'double wind' a
> motor to cancel/use the back emf?

Yes, absolutely.
Unfortunately this also cancels out all the torque :-)

(In fact this is the way to identify the pole pairs and pole
directions on an 8-wire motor, by working out what combinations make
the rotor hard to turn, and which not, when the wires are shorted in
various combinations.)

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1912


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] What Would You Suggest?

[Stuart Stevenson]

Guys,
Causing me to 'think' this early hurts my head but I need to express this
thought. It may be totally crazy but is there a way to 'double wind' a
motor to cancel/use the back emf?
thanks
Stuart


On Sat, Feb 5, 2022 at 5:55 AM Robin Szemeti via Emc-users <
emc-users@lists.sourceforge.net> wrote:

>  "comment about corner frequency with stepper motors _might_ well be valid
> as long as the maximum current for each step is reached before or at the
> end of the ste"
>
> Yes, that is exactly what the corne frequency is ... the step frequency at
> which the current no longer reaches the desired value before the end of the
> step. It's obviously dependent on inductance and maximum available drive
> voltage.
>
> On Sat, 5 Feb 2022 at 11:50, Robin Szemeti  wrote:
>
> > John,
> >
> > You are fundamentally incorrect when you state " the torque of the motor
> > to drop off the faster it goes" .. although the back EMF is correct,
> with a
> > modern current limited drive, the torque is flat until the corner
> > frequency, then drops off  ... up to the corner frequency the torque is
> > constant with a good current-limited drive, above the corner frequency
> the
> > torque drops off, power is constant.  You are perhaps confusing the raw
> > torque/speed curve of a motor fed from a constant voltage source, which
> is
> > useful but is not how they are typically used in practice.
> >
> >
> >
> https://res.cloudinary.com/engineering-com/image/upload/w_640,h_640,c_limit,q_auto,f_auto/image002_bezhrr.jpg
> >
> > On Sat, 5 Feb 2022 at 09:00, John Dammeyer 
> wrote:
> >
> >> Hi Chris,
> >> My issue is that a comment about corner frequency with stepper motors
> >> _might_ well be valid as long as the maximum current for each step is
> >> reached before or at the end of the step.  But the motor is turning
> pretty
> >> slowly there compared to how they are used in real life.
> >>
> >> However the comment about corner frequency with respect to steppers
> >> perhaps is only backed by alternative facts?
> >>
> >> I must admit I've not investigated in detail the closed loop steppers.
> >> The price of an industrial version I worked with was more than the
> price of
> >> an AC servo and at higher speeds I could stop the pulley with my
> fingers.
> >> Yes. It faulted.  But that isn't really the point.  The DC and AC
> servos at
> >> higher speeds just work better.
> >>
> >> Stepper motors work great at low speeds usually directly coupled.
> >> Contrary to popular belief the micro-stepping doesn't improve resolution
> >> but gets rid of resonance and gives the appearance of better resolution.
> >> But it doesn't change the fact that the current still has to reverse
> every
> >> full step.  I believe that in fact Gecko drives improve high speed
> torque
> >> by switching back to full step mode above the resonance velocity.
> >>
> >> Now instead of 0.707 x max current in both windings (at the most) we're
> >> back to 100% in both with an increase in torque.  Absolutely nothing to
> do
> >> with corner frequencies whatever they might be or how they are
> determined.
> >>
> >> John
> >>
> >>
> >>
> >>
> >>
> >> > From: Chris Albertson [mailto:albertson.ch...@gmail.com]
> >> >
> >> > John,
> >> >
> >> > You described it correctly.  But I think what Robin meant by "Corner
> >> > Frequency" might be the peak of the power vs. RPM graph.  Basically,
> the
> >> > frequency where power output starts to fall with RPM.
> >> >
> >> > But now it can get worse, or really better but more complex.   We have
> >> > these so-called "closed loop stepper drivers and also a few people are
> >> > running the steppers as if they were many-pole BLDC analog
> (continuous,
> >> > non-stepping) mortors
> >> >
> >> > On Fri, Feb 4, 2022 at 2:39 PM John Dammeyer 
> >> wrote:
> >> >
> >> > > I disagree.  The physics of the motor, which include inductance
> along
> >> with
> >> > > the generated back emf from the motor spinning in the magnetic
> field,
> >> is
> >> > > what cause the torque of the motor to drop off the faster it goes.
> >> > >
> >> > > The problem is to spin a stepper motor you have to not just change
> to
> >> a
> >> > > new winding like a DC motor does but completely reverse the
> direction
> >> of
> >> > > the current through the winding.  In order to do that you have to
> >> deal with
> >> > > the collapsing magnetic field and counter the resulting generated
> >> voltage
> >> > > which is based on the inductance of the windings.
> >> > >
> >> > > That's why the winding voltage of a stepper motor might be only 2V
> to
> >> get
> >> > > the rated 3A but you need 48V to make it turn quickly.  And because
> >> of  the
> >> > > inductance and collapsing field,  time is required to change the
> >> direction
> >> > > of the current through the winding.  If that time is longer than the
> >> next
> >> > > direction change then you never reach max current through the
> >> windings and
> >> > > you don't develop full torque.  That's why a stepper motor

Re: [Emc-users] What Would You Suggest?

[Robin Szemeti via Emc-users]

 "comment about corner frequency with stepper motors _might_ well be valid
as long as the maximum current for each step is reached before or at the
end of the ste"

Yes, that is exactly what the corne frequency is ... the step frequency at
which the current no longer reaches the desired value before the end of the
step. It's obviously dependent on inductance and maximum available drive
voltage.

On Sat, 5 Feb 2022 at 11:50, Robin Szemeti  wrote:

> John,
>
> You are fundamentally incorrect when you state " the torque of the motor
> to drop off the faster it goes" .. although the back EMF is correct, with a
> modern current limited drive, the torque is flat until the corner
> frequency, then drops off  ... up to the corner frequency the torque is
> constant with a good current-limited drive, above the corner frequency the
> torque drops off, power is constant.  You are perhaps confusing the raw
> torque/speed curve of a motor fed from a constant voltage source, which is
> useful but is not how they are typically used in practice.
>
>
> https://res.cloudinary.com/engineering-com/image/upload/w_640,h_640,c_limit,q_auto,f_auto/image002_bezhrr.jpg
>
> On Sat, 5 Feb 2022 at 09:00, John Dammeyer  wrote:
>
>> Hi Chris,
>> My issue is that a comment about corner frequency with stepper motors
>> _might_ well be valid as long as the maximum current for each step is
>> reached before or at the end of the step.  But the motor is turning pretty
>> slowly there compared to how they are used in real life.
>>
>> However the comment about corner frequency with respect to steppers
>> perhaps is only backed by alternative facts?
>>
>> I must admit I've not investigated in detail the closed loop steppers.
>> The price of an industrial version I worked with was more than the price of
>> an AC servo and at higher speeds I could stop the pulley with my fingers.
>> Yes. It faulted.  But that isn't really the point.  The DC and AC servos at
>> higher speeds just work better.
>>
>> Stepper motors work great at low speeds usually directly coupled.
>> Contrary to popular belief the micro-stepping doesn't improve resolution
>> but gets rid of resonance and gives the appearance of better resolution.
>> But it doesn't change the fact that the current still has to reverse every
>> full step.  I believe that in fact Gecko drives improve high speed torque
>> by switching back to full step mode above the resonance velocity.
>>
>> Now instead of 0.707 x max current in both windings (at the most) we're
>> back to 100% in both with an increase in torque.  Absolutely nothing to do
>> with corner frequencies whatever they might be or how they are determined.
>>
>> John
>>
>>
>>
>>
>>
>> > From: Chris Albertson [mailto:albertson.ch...@gmail.com]
>> >
>> > John,
>> >
>> > You described it correctly.  But I think what Robin meant by "Corner
>> > Frequency" might be the peak of the power vs. RPM graph.  Basically, the
>> > frequency where power output starts to fall with RPM.
>> >
>> > But now it can get worse, or really better but more complex.   We have
>> > these so-called "closed loop stepper drivers and also a few people are
>> > running the steppers as if they were many-pole BLDC analog (continuous,
>> > non-stepping) mortors
>> >
>> > On Fri, Feb 4, 2022 at 2:39 PM John Dammeyer 
>> wrote:
>> >
>> > > I disagree.  The physics of the motor, which include inductance along
>> with
>> > > the generated back emf from the motor spinning in the magnetic field,
>> is
>> > > what cause the torque of the motor to drop off the faster it goes.
>> > >
>> > > The problem is to spin a stepper motor you have to not just change to
>> a
>> > > new winding like a DC motor does but completely reverse the direction
>> of
>> > > the current through the winding.  In order to do that you have to
>> deal with
>> > > the collapsing magnetic field and counter the resulting generated
>> voltage
>> > > which is based on the inductance of the windings.
>> > >
>> > > That's why the winding voltage of a stepper motor might be only 2V to
>> get
>> > > the rated 3A but you need 48V to make it turn quickly.  And because
>> of  the
>> > > inductance and collapsing field,  time is required to change the
>> direction
>> > > of the current through the winding.  If that time is longer than the
>> next
>> > > direction change then you never reach max current through the
>> windings and
>> > > you don't develop full torque.  That's why a stepper motor with a 24V
>> power
>> > > supply has the same holding torque as one with a 48V power supply.
>> The
>> > > current limiting of the drive holds the winding current at 3A.  But
>> run it
>> > > at 24V or at 48V you get a totally different torque curve.
>> > >
>> > > If you are going to mention something called the corner frequency of a
>> > > stepper motor+drive please show us the graphs and specifications.  I
>> > > haven't been able to find that rating on any stepper motor.
>> > >
>> > > Perhaps you can point it for this one?
>> > > htt

Re: [Emc-users] What Would You Suggest?

[Robin Szemeti via Emc-users]

John,

You are fundamentally incorrect when you state " the torque of the motor to
drop off the faster it goes" .. although the back EMF is correct, with a
modern current limited drive, the torque is flat until the corner
frequency, then drops off  ... up to the corner frequency the torque is
constant with a good current-limited drive, above the corner frequency the
torque drops off, power is constant.  You are perhaps confusing the raw
torque/speed curve of a motor fed from a constant voltage source, which is
useful but is not how they are typically used in practice.

https://res.cloudinary.com/engineering-com/image/upload/w_640,h_640,c_limit,q_auto,f_auto/image002_bezhrr.jpg

On Sat, 5 Feb 2022 at 09:00, John Dammeyer  wrote:

> Hi Chris,
> My issue is that a comment about corner frequency with stepper motors
> _might_ well be valid as long as the maximum current for each step is
> reached before or at the end of the step.  But the motor is turning pretty
> slowly there compared to how they are used in real life.
>
> However the comment about corner frequency with respect to steppers
> perhaps is only backed by alternative facts?
>
> I must admit I've not investigated in detail the closed loop steppers.
> The price of an industrial version I worked with was more than the price of
> an AC servo and at higher speeds I could stop the pulley with my fingers.
> Yes. It faulted.  But that isn't really the point.  The DC and AC servos at
> higher speeds just work better.
>
> Stepper motors work great at low speeds usually directly coupled.
> Contrary to popular belief the micro-stepping doesn't improve resolution
> but gets rid of resonance and gives the appearance of better resolution.
> But it doesn't change the fact that the current still has to reverse every
> full step.  I believe that in fact Gecko drives improve high speed torque
> by switching back to full step mode above the resonance velocity.
>
> Now instead of 0.707 x max current in both windings (at the most) we're
> back to 100% in both with an increase in torque.  Absolutely nothing to do
> with corner frequencies whatever they might be or how they are determined.
>
> John
>
>
>
>
>
> > From: Chris Albertson [mailto:albertson.ch...@gmail.com]
> >
> > John,
> >
> > You described it correctly.  But I think what Robin meant by "Corner
> > Frequency" might be the peak of the power vs. RPM graph.  Basically, the
> > frequency where power output starts to fall with RPM.
> >
> > But now it can get worse, or really better but more complex.   We have
> > these so-called "closed loop stepper drivers and also a few people are
> > running the steppers as if they were many-pole BLDC analog (continuous,
> > non-stepping) mortors
> >
> > On Fri, Feb 4, 2022 at 2:39 PM John Dammeyer 
> wrote:
> >
> > > I disagree.  The physics of the motor, which include inductance along
> with
> > > the generated back emf from the motor spinning in the magnetic field,
> is
> > > what cause the torque of the motor to drop off the faster it goes.
> > >
> > > The problem is to spin a stepper motor you have to not just change to a
> > > new winding like a DC motor does but completely reverse the direction
> of
> > > the current through the winding.  In order to do that you have to deal
> with
> > > the collapsing magnetic field and counter the resulting generated
> voltage
> > > which is based on the inductance of the windings.
> > >
> > > That's why the winding voltage of a stepper motor might be only 2V to
> get
> > > the rated 3A but you need 48V to make it turn quickly.  And because
> of  the
> > > inductance and collapsing field,  time is required to change the
> direction
> > > of the current through the winding.  If that time is longer than the
> next
> > > direction change then you never reach max current through the windings
> and
> > > you don't develop full torque.  That's why a stepper motor with a 24V
> power
> > > supply has the same holding torque as one with a 48V power supply.  The
> > > current limiting of the drive holds the winding current at 3A.  But
> run it
> > > at 24V or at 48V you get a totally different torque curve.
> > >
> > > If you are going to mention something called the corner frequency of a
> > > stepper motor+drive please show us the graphs and specifications.  I
> > > haven't been able to find that rating on any stepper motor.
> > >
> > > Perhaps you can point it for this one?
> > > http://www.automationtechnologiesinc.com/download/9259/
> > >
> > > And explain how you determined that corner frequency?
> > > John
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > > > -Original Message-
> > > > From: Robin Szemeti via Emc-users [mailto:
> > > emc-users@lists.sourceforge.net]
> > > > Sent: February-04-22 2:01 PM
> > > > To: Enhanced Machine Controller (EMC)
> > > > Cc: Robin Szemeti
> > > > Subject: Re: [Emc-users] What Would You Suggest?
> > > >
> > > > What people continually get totally wrong with steppers is failing to
> > > > understand th

Re: [Emc-users] What Would You Suggest?

[John Dammeyer]

Hi Chris,
My issue is that a comment about corner frequency with stepper motors _might_ 
well be valid as long as the maximum current for each step is reached before or 
at the end of the step.  But the motor is turning pretty slowly there compared 
to how they are used in real life.

However the comment about corner frequency with respect to steppers perhaps is 
only backed by alternative facts?

I must admit I've not investigated in detail the closed loop steppers.  The 
price of an industrial version I worked with was more than the price of an AC 
servo and at higher speeds I could stop the pulley with my fingers.  Yes. It 
faulted.  But that isn't really the point.  The DC and AC servos at higher 
speeds just work better.

Stepper motors work great at low speeds usually directly coupled.  Contrary to 
popular belief the micro-stepping doesn't improve resolution but gets rid of 
resonance and gives the appearance of better resolution.  But it doesn't change 
the fact that the current still has to reverse every full step.  I believe that 
in fact Gecko drives improve high speed torque by switching back to full step 
mode above the resonance velocity.  

Now instead of 0.707 x max current in both windings (at the most) we're back to 
100% in both with an increase in torque.  Absolutely nothing to do with corner 
frequencies whatever they might be or how they are determined.

John





> From: Chris Albertson [mailto:albertson.ch...@gmail.com]
> 
> John,
> 
> You described it correctly.  But I think what Robin meant by "Corner
> Frequency" might be the peak of the power vs. RPM graph.  Basically, the
> frequency where power output starts to fall with RPM.
> 
> But now it can get worse, or really better but more complex.   We have
> these so-called "closed loop stepper drivers and also a few people are
> running the steppers as if they were many-pole BLDC analog (continuous,
> non-stepping) mortors
> 
> On Fri, Feb 4, 2022 at 2:39 PM John Dammeyer  wrote:
> 
> > I disagree.  The physics of the motor, which include inductance along with
> > the generated back emf from the motor spinning in the magnetic field, is
> > what cause the torque of the motor to drop off the faster it goes.
> >
> > The problem is to spin a stepper motor you have to not just change to a
> > new winding like a DC motor does but completely reverse the direction of
> > the current through the winding.  In order to do that you have to deal with
> > the collapsing magnetic field and counter the resulting generated voltage
> > which is based on the inductance of the windings.
> >
> > That's why the winding voltage of a stepper motor might be only 2V to get
> > the rated 3A but you need 48V to make it turn quickly.  And because of  the
> > inductance and collapsing field,  time is required to change the direction
> > of the current through the winding.  If that time is longer than the next
> > direction change then you never reach max current through the windings and
> > you don't develop full torque.  That's why a stepper motor with a 24V power
> > supply has the same holding torque as one with a 48V power supply.  The
> > current limiting of the drive holds the winding current at 3A.  But run it
> > at 24V or at 48V you get a totally different torque curve.
> >
> > If you are going to mention something called the corner frequency of a
> > stepper motor+drive please show us the graphs and specifications.  I
> > haven't been able to find that rating on any stepper motor.
> >
> > Perhaps you can point it for this one?
> > http://www.automationtechnologiesinc.com/download/9259/
> >
> > And explain how you determined that corner frequency?
> > John
> >
> >
> >
> >
> >
> >
> >
> > > -Original Message-
> > > From: Robin Szemeti via Emc-users [mailto:
> > emc-users@lists.sourceforge.net]
> > > Sent: February-04-22 2:01 PM
> > > To: Enhanced Machine Controller (EMC)
> > > Cc: Robin Szemeti
> > > Subject: Re: [Emc-users] What Would You Suggest?
> > >
> > > What people continually get totally wrong with steppers is failing to
> > > understand that the maximum power is delivered at the corner frequency,
> > and
> > > power output is constant above that.
> > >
> > > If you have an application that needs to move at say 2m a minute and your
> > > stepper stalls, there seems to be some crazy logic that says to people
> > "Oh,
> > > the stepper stalled because it was going too fast, I need to change the
> > > gearing so the motor spins more slowly" .. which is of course ass
> > backwards.
> > >
> > > The stepper stalled because the power output of the motor was less than
> > the
> > > power requirement of the machine ... to increase the power output of the
> > > motor, you need to spin it faster, not slower.  Steppers motors are
> > capable
> > > of excellent performance but they do need to be used correctly ... sadly,
> > > in most amateur applications they are not.
> > >
> > > If the corner frequency with your drive and voltage is at around 2000

Re: [Emc-users] What Would You Suggest?

[Gregg Eshelman via Emc-users]

What motor driven surface grinder with power feed *doesn't have* automatic 
reversing in X and adjustable auto feed step in Y? Even ones 100 or so years 
old had features like adjustable X travel stops. I'd bet there were grinders 
with Y travel stops so an operator could set it and forget it, then come back 
later to crank the table back on the Y axis, turn the spindle down a bit then 
set it off on making the next pass.


   On Friday, February 4, 2022, 02:07:48 PM MST, John Figie 
 wrote:  
 
 On Fri, Feb 4, 2022, 2:55 PM dave engvall  wrote:

> Hi,
> I seem to remember a crank as in crankshaft lashup to drive the table.
> Personally I think the hydraulic setup is better but harder to achieve.
> The free lunch is hard to find.
> Dave
>
>
> Hmm

I have a Covel hydraulic automatic surface grinder. I would rather have
motors - maybe a retrofit someday. The issues with hydraulic are power
consumption but the pump and leaking oil because it's old.  
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] What Would You Suggest?

[Chris Albertson]

John,

You described it correctly.  But I think what Robin meant by "Corner
Frequency" might be the peak of the power vs. RPM graph.  Basically, the
frequency where power output starts to fall with RPM.

But now it can get worse, or really better but more complex.   We have
these so-called "closed loop stepper drivers and also a few people are
running the steppers as if they were many-pole BLDC analog (continuous,
non-stepping) mortors

On Fri, Feb 4, 2022 at 2:39 PM John Dammeyer  wrote:

> I disagree.  The physics of the motor, which include inductance along with
> the generated back emf from the motor spinning in the magnetic field, is
> what cause the torque of the motor to drop off the faster it goes.
>
> The problem is to spin a stepper motor you have to not just change to a
> new winding like a DC motor does but completely reverse the direction of
> the current through the winding.  In order to do that you have to deal with
> the collapsing magnetic field and counter the resulting generated voltage
> which is based on the inductance of the windings.
>
> That's why the winding voltage of a stepper motor might be only 2V to get
> the rated 3A but you need 48V to make it turn quickly.  And because of  the
> inductance and collapsing field,  time is required to change the direction
> of the current through the winding.  If that time is longer than the next
> direction change then you never reach max current through the windings and
> you don't develop full torque.  That's why a stepper motor with a 24V power
> supply has the same holding torque as one with a 48V power supply.  The
> current limiting of the drive holds the winding current at 3A.  But run it
> at 24V or at 48V you get a totally different torque curve.
>
> If you are going to mention something called the corner frequency of a
> stepper motor+drive please show us the graphs and specifications.  I
> haven't been able to find that rating on any stepper motor.
>
> Perhaps you can point it for this one?
> http://www.automationtechnologiesinc.com/download/9259/
>
> And explain how you determined that corner frequency?
> John
>
>
>
>
>
>
>
> > -Original Message-
> > From: Robin Szemeti via Emc-users [mailto:
> emc-users@lists.sourceforge.net]
> > Sent: February-04-22 2:01 PM
> > To: Enhanced Machine Controller (EMC)
> > Cc: Robin Szemeti
> > Subject: Re: [Emc-users] What Would You Suggest?
> >
> > What people continually get totally wrong with steppers is failing to
> > understand that the maximum power is delivered at the corner frequency,
> and
> > power output is constant above that.
> >
> > If you have an application that needs to move at say 2m a minute and your
> > stepper stalls, there seems to be some crazy logic that says to people
> "Oh,
> > the stepper stalled because it was going too fast, I need to change the
> > gearing so the motor spins more slowly" .. which is of course ass
> backwards.
> >
> > The stepper stalled because the power output of the motor was less than
> the
> > power requirement of the machine ... to increase the power output of the
> > motor, you need to spin it faster, not slower.  Steppers motors are
> capable
> > of excellent performance but they do need to be used correctly ... sadly,
> > in most amateur applications they are not.
> >
> > If the corner frequency with your drive and voltage is at around 2000
> steps
> > per second and you are only ever delivering 1000 steps per second, you
> can
> > never got more than half the mechanical power out that the motor is
> capable
> > of.
> >
> > On Fri, 4 Feb 2022 at 17:13, John Dammeyer 
> wrote:
> >
> > >
> > >
> > > > From: Kenneth Lerman [mailto:ler...@se-ltd.com]
> > > > The longitudinal travel is just over a foot, and it takes about 3-1/2
> > > turns
> > > > of the crank to go that distance. I'm thinking around  a second per
> turn
> > > > would be about the maximum. So, that's 60 RPM. I'm thinking of a 1:6
> > > ratio
> > > > on the timing belt pulleys, so that's 360 RPM at the stepper which is
> > > > pretty slow. A full stepping rate would be 200 * 360/60 => 200 * 6
> which
> > > is
> > > > only 1200 steps per second.
> > >
> > > You won't want to run full step.  A minimum should be 8
> micro-steps/step
> > > to avoid resonance and loss of position or lockup.   I'd measure the
> torque
> > > required to move the table by attaching a lever to the hand wheel that
> is
> > > say 1' long.  Set it horizontal and start hanging weight onto the end
> to
> > > get ft-lbs or ft-in until it turns. That's the torque required to
> overcome
> > > static friction.  Double that to choose your motor.
> > >
> > > Say that is 1 ft-lb or 192 oz-in.If you choose 3:1 for your
> reduction
> > > ratio you get 600 oz-in.  Look at the motor torque curve (they are all
> > > different and if the supplier can't give you that buy one somewhere
> else)
> > > and see where the torque drops below 400 oz-in.  Say that's 180 RPM.
> > > That's 3 RPS which multiplied by 2000 steps