RE: [CAD_CAM_EDM_DRO] Re: PWM vs Constant current

I may have some insight here. At one point on my drive building learning
curve I was convinced of the merits of constant current drives. Actually
built a 4A one and ran it for about 6 months too.
http://www.caswa.com/dropbox/files/wired_with_thermal_cutouts_2.jpg
The board on the lower right is a simple half step translator and the two
fans above cool a pair of 2A linear, current regulators.

The reason I felt this was the way to go was I didn't understand enough
about using PWM to drive motors. Particularly low inductance motors. If you
are scratch building drives using discrete components, (ie no driver blocks)
then CC is pretty straightforward.

For drives up to 4A 40V however, there's just no point as the L297/298
combination, the 6203 , or (for smaller motors) the Allegro 3977
microstepper are fantastic and cheap.

Constant current drives will work. And work reasonably well but they will
dissipate VERY LARGE amounts of heat when at rest.

If you have a 40V supply and a 2A 3V motor, then that constant current
supply will be getting rid of (40-3)*2=74W when the motor is not turning. It
gets better as the motor goes faster....

You can compensate for this by having your driver drop the current or
voltage when not turning. Mach3 has an output for this, but for motors 4A or
less, monolithic PWM chips exist and are better.

PK

-----Original Message-----
From: CAD_CAM_EDM_DRO@yahoogroups.com
[mailto:CAD_CAM_EDM_DRO@yahoogroups.com] On Behalf Of Alan Rothenbush
Sent: Tuesday, 25 October 2005 5:57 AM
To: CAD_CAM_EDM_DRO@yahoogroups.com
Subject: Re: [CAD_CAM_EDM_DRO] Re: PWM vs Constant current

On Monday 24 October 2005 11:55, Jarrett & Heidi Johnson wrote:
> Alan, I truely don't know [ hence why I asked]. Below is a paste from him
> which is off a website where he purchased his equipment. I'm sure they are
> trying to promote their product however I just want the final answer that
> isn't bias in any way :-)

I'd wait for more in the group to comment, but what's written below strikes
me
as .. not completely accurate .. to be as polite as I can be.

Others might suggest "total BS" as an accurate description, but I'll remain
polite.

Driving stepper motors is a complex process, but it's also reasonably well
understood. I'd advise you to go to the files section of this group and
study all that's there, paying particular attention to the "white papers"
produced by Mariss Fremanis, but studying all that's there.

To get you started though, I'll give you a GREATLY simplified view of a
couple
of points, that while not being completely technically accurate, will be
close enough to help you being.

Steppers are current operated devices. That is, you must shove electrons
(current) into them to make them work.

Shove more current into them, they make more power.

But you can't put too much current into them or they overheat and fail.
(Typically, the insulation on the wires melts). But even before they fail,
extra current may not help as the current works in conjunction with the
magnetic materials in the stepper and if you've "exhausted" the magnetic
materials, extra current just goes to heat.

So, there is a maximum current for any particular motor, it's based upon
wire
size and magnetic materials, and manufacturers are pretty good at optimizing

the two.


Now, shove current into a stepper motor and it will move to a particular
angle
and stay there, resisting attempts to move it. Larger motors (with larger
amounts of current shoved into them) resist harder. This resistance is the
"holding torque" and is a measure of motor's ability to retain its
rotational
position. That is, it's a measure of the motor's performance when stopped.

For any given motor, this torque is proportional to the current going into
it,
the "resistance to change" increasing as the current increase, right up to
the maximum current.

(The above statement is a generalization and one of those "not completely
technically accurate" referred to above, but it's close enough for this
level
of discussion)

So, you want more power, feed it more current, and if you can't feed it more

current because it will melt down, buy a bigger motor.


Now, all the above implies a means of supplying an optimal amount of current

without supplying too much current, and there are a number of ways of doing
so. The first is to read the manufacturer's spec sheet, where a motor is
typically listed with a voltage and a current spec, that basically says
"this
is the maximum current this motor can take, and if you give it this much
voltage, that's the current it will draw".

One such spec might be 2 volts @ 2 amps, so put 2 volts on a pair of leads
and
2 amps will be drawn.

The question then becomes, if a 2 volt power supply will cause this motor to

draw the full 2 amps it can take, why not just build a 2 volt power supply
and be done with it ?

And the answer is, if all you want to motor to do is sit there, building a 2

volt supply is the right thing to do. But if you want it to rotate ...


At this point, we need to talk a bit about how a stepper motor rotates.
(Again a simplification) Steppers have two windings, and if we shove
current
into one of the windings, the rotor moves to one position and if we shove
current into the other winding, it moves to another. The other key point is

that the direction we "shove" the current in is important .. shove the
current into winding one "east to west" it moves to one position, shove it
into winding 1 (same winding) "west to east" and it moves to another
position.

So, take the simplest possible stepper motor and here's the position/current

table

Winding 1 west to east 12:00
Winding 2 north to south 3:00
Winding 1 east to west 6:00
Winding 2 south to north 9:00

Two coils * two current directions = 4 positions.


OK, you say, how about just building a pair of 2 volt supplies,
electronically
connecting them up as required, and in fact, that would work.

However, here's important technical concept number 1 .. it takes time to
shove
that current into the winding. (I'm going to assume from here on that we
want
the motor to rotate.) If we want full power out of motor, we have to wait
until we've filled a winding up before going onto the next winding.

The faster we can get current into a winding (and out again .. more on this
later), the faster a motor will turn. Now we could just stuff less current
into a motor and it would turn faster, but at lower power, and we want all
the power we've paid for.

The ONLY way to get that current into a winding is to push harder, which
electrically means using a higher voltage. But if we use a higher voltage,
we'll get more current (current is directly proportional to the voltage
applied) and we'll melt the motor down.

So there's our quandry .. how we both use a higher voltage while not
exceeding
the current ? There are three solutions, a resistor, a chopper source and a

constant current source.

The resistor is the cheapest solution, but one providing only a slight
advantage .. it does have its place, but not in the context of machine
tools.

Both chopper and constant current are more complex than a single resistor,
but
both work very well at using a high voltage to shove just the right amount
of
current in, but no more, as quickly as possible.

A basic Constant Current (CC) method is very simple to design and very
reliable, but has the disadvantage of being extremely inefficient, requiring

a much bigger power supply and turning all of that extra power into heat.

It's no surprise that a CC design would appear as a kit .. they don't have
to
include the power supply and the cost of that supply does not appear in the
calculations.

Chopper supplies are somewhat more complex than a CC supply.. but only a
bit .. and are WAY more efficient, easily saving in power supply costs what
they eat up in extra parts.

Really, the market has spoken .. I would guess that 99.9% of all commercial
stepper controllers for the machine industry built in the last ten years are

chopper designs.


HOWEVER, getting the current into the coils is only half the problem .. you
have to get the current OUT of the coil as well before shoving current into
the next one.

And it is here that the K142 design appears to fall down. I say "appears"
because I have not seen a schematic for it, only a pic of the board and so
I've got to guess as to what's going on.

Suffice to say (that's shorthand for "I'm tired of typing, so I'm going with

short explanations") getting the current out .. aka, "current decay" is
just
as important in the whole scheme of performance as getting the current in,
and it's not really related to CC or chopper designs. Either can have
sophisticated or primitive approaches to the question of getting the current

out.

Do some reading on this topic.


Finally, the big issue, microstepping, which I'll cover VERY briefly. The
position of a rotor is related to the direction and magnitude of the current

in the coils .. it's a vector quantity.

Take our example above

Winding 1 west to east 12:00
Winding 2 north to south 3:00

What would happen if we did both ? That is, supplied current to both
windings
at the same time, in the manner shown above ?

The rotor would point to 1:30 !

So if, using our primitive stepper above, we supplied current to one
winding,
then both, then one, then both, and so on, in the right sequence and the
right direction, we would get 8 positions, not just 4.

In fact, the position of the rotor is (roughly) determined by the formula

angle = sin(current in winding 1) * cos(current in winding 2)

Say, with our 2 amp motor, we shoved 2 amps into winding 1 W->E but only 1
amp
into winding 2 N->S, where would the rotor be ? Roughly 1:00

Why is this important ? NOT because of what you're thinking, greater
positional accuracy, because you don't get that (I said "roughly" earlier)
but because of an annoying flaw in stepper motors known as "resonance".

At some rotational speed, a stepper makes no power .. it is unable to move
from position to position. Won't go into why, it just does.

The RPM at which this happens is roughly related to the size of the steps.
In
our simple example, steps are 90 degrees wide, but in example two, they are
only 45 degrees wide and so the RPM at which the motor becomes a problem is
doubled.

"Microstepping" .. the 1:00 scenario .. increases the number of intermediate

steps, often to the point where resonance is no longer an issue.

Microstepping is an absolute requirement in all modern stepper systems, and
I
see no evidence (from staring at the PCB) that the K142 supports this in any

way.

Read about microstepping.



Hope this helps .. gotta go.


Alan
--
Alan Rothenbush
Academic Computing Services
Simon Fraser University
Burnaby, B.C., Canada


Before me things create were none, save things
Eternal, and eternal I endure.
All hope abandon ye who enter here.



Addresses:
FAQ: http://www.ktmarketing.com/faq.html
FILES: http://groups.yahoo.com/group/CAD_CAM_EDM_DRO/files/
Post Messages: CAD_CAM_EDM_DRO@yahoogroups.com

Subscribe: CAD_CAM_EDM_DRO-subscribe@yahoogroups.com
Unsubscribe: CAD_CAM_EDM_DRO-unsubscribe@yahoogroups.com
List owner: CAD_CAM_EDM_DRO-owner@yahoogroups.com, wanliker@...,
timg@...
Moderator: pentam@... indigo_red@... davemucha@...
[Moderators]
URL to this group: http://groups.yahoo.com/group/CAD_CAM_EDM_DRO

OFF Topic POSTS: General Machining
If you wish to post on unlimited OT subjects goto:
aol://5863:126/rec.crafts.metalworking or go thru Google.com to reach it if
you have trouble.
http://www.metalworking.com/news_servers.html

http://groups.yahoo.com/group/jobshophomeshop I consider this to be a
sister site to the CCED group, as many of the same members are there, for OT
subjects, that are not allowed on the CCED list.

NOTICE: ALL POSTINGS TO THIS GROUP BECOME PUBLIC DOMAIN BY POSTING THEM.
DON'T POST IF YOU CAN NOT ACCEPT THIS.....NO EXCEPTIONS........
bill
List Mom
List Owner


Yahoo! Groups Links