开云体育

I spent the weekend and early part of this week polishing up the software on
my chopper drive and took the plunge last night and hooked it up to my mill.
The first move was real slow and it sounded like someone running a jack
hammer. Most of the racket was coming from the handles that were still
attached to the other end of the lead screws. I removed the handles and was
much better. I ran it through it's complete speed range with a circular
interpolation - only one axis moving though.
There were a couple speeds that made a pretty loud 'buzz' - found out that
was coming from the tooth belt. After a few cycles I went back to 0,0 and it
had not lost any steps.
I had been reluctant to use half-steps because every other step only one
field energizes and I thought it wouldn't have as much torque - after
thinking about it a bit, that's wrong. It still gets full power every other
step and even if it couldn't move at all on the half step it would surely
move on the next full step. So I tried half step and the machine moved very
smoothly at all speeds. Ran several tests and even took some cuts on a
chunk of aluminum. Never lost a step. I'm real happy with this driver.
I'm running the motor at 4amps/phase and at motor lock it's drawing about
1 amp from my 40 volt supply. At 20-25 ipm feed it draws 2 amps from the
supply. (8000 steps per inch). 20 ipm is the fastest speed that the field
coils can reach 4 amps but the machine rapids at 25 ipm just fine.
I think I'll put amp meters on my control, seems like a good indication
of machine friction/cutting loads.

That was all...

Joel

On 3 Mar, CAD_CAM_EDM_DRO@... wrote:

Message: 1
Date: Thu, 2 Mar 2000 11:39:34 -0500
From: "Joel Jacobs" <jj@...>
Subject: Stepper resonance - not a problem

I had been reluctant to use half-steps because every other step only one
field energizes and I thought it wouldn't have as much torque - after
thinking about it a bit, that's wrong. It still gets full power every other
step and even if it couldn't move at all on the half step it would surely
move on the next full step.

I think you're a bit mistaken on a couple of things here.

If the motor stopped on a particular half-position step as a result of
insufficient torque, it may not move on the next "full" step either.
That's the whole point programmers are trying to make with SMOOTH
accelerations tied to hardware timers. Resonance is a problem that can
cause missed steps, but "unsmooth" accelerations will also cause
missed steps.

The constant torque "secret" is to increase the current to the winding,
when only one winding is on. Appropriately scaled, this will result in
constant motor torque for each position.

The theory is clear and well understood. For any given angle, the
current relationship should be sin(a) cos(b). For simple half-step, this
translates to increasing the current by 40% when only a single winding
is on.

Note that at no time does the current through the windings exceed the
current in the both windings on position, so no motor overheating
should result. ( Your 4 amp steppers, for example, will have 8 amps
total in "both windings on" position and 5.6 amps though one winding
only in the "half" step position ) If the motor is stopped in the
one-winding on position, though, the motor current must be reduced, or
the winding will overheat.


I've been working with DanM on this for a bit, and Dan has found that
going to quarter step solves almost all resonance problems, and that
going finer than that, to so-called microstepping is really not
necessary.

Be aware that absolute positioning is not as accurate in any of the
intermediate positions, and this position error is increased as the
steps get finer and finer. ( Yet another reason to avoid uStepping )
Moral: Don't change your pulley ratios to keep the speed up.

Alan


--

Alan Rothenbush | The Spartans do not ask the number of the
Academic Computing Services | enemy, only where they are.
Simon Fraser University |
Burnaby, B.C., Canada | Agix of Sparta

If thats the case, it probly wouldn't work in full step mode either.

You might be right .. theory and the real world are rarely in sync.

I had thought about doing this, my controller has the capability to adjust
the current sense reference voltage from 50 - 100%, with a resistor change
it could do 60/100. But are the motors ratings limited by the temprature
rise - or core saturation?

Good question, although neither may be important. Torque is roughly
proportional to core characteristics, so increasing the current to maintain
a constant torque implies that if the core hasn't saturated with both
windings engaged ( at some particular torque level ), it shouldn't
saturate at the same torque level with only one winding engaged.

Just to be on the safe side, since my motors are quire big enough for my
machine, I mentally derated them just a bit. Instead of running them
at 5 amps per winding in full step mode, I run them in half-step at 4 amps
with both windings on, 5.6 amps with one winding on.

Again, though, this should not be necessary for either temperature reasons
( assuming a reduced current "stop" feature ) or for core saturation reasons.

My controller does a power saver function where the current is reduced
automaticaly if it has nor recieved a step pulse in 1 second - it resumes
full power imediatly when a step pulse arrives.

As does mine, for the reason stated just above.

This is the second time If seen 1/4 stepping mentioned - could you explain
how this works? Do you have to be able to set different current levels
simultaniously when two coils are energized?

Exactly, in the proportion mentioned - sin(a) cos(b). The relationship
is actually the same for all modes, full, half, quarter, eighth, etc.
Increasing the number of intermediate steps does mean increasing the number
of discrete current levels. This usually ( but not necessarily ) means
a microprocessor to assist things. It IS possible with discrete logic,
but ... <G>

Alan

--

Alan Rothenbush | The Spartans do not ask the number of the
Academic Computing Services | enemy, only where they are.
Simon Fraser University |
Burnaby, B.C., Canada | Agix of Sparta