motor control

[cabber]

Hi,

I have a dc motor that normally rotates in one direction for steel coil winding. I have noticed that after threading the coil onto the mandrel (beginning to wind the coil), the shaft first rotates in the opposite direction (slightly) and then begins to rotate correctly. I use an Avtron M785 tach (1200 PPR). Is there anything in particular that I should look into to resolve the 'reversing' issue? I am thinking that it is a tach issue.

Thanks

 

[60hzhmmm]

cabber: Does it always run backwards first(ALWAYS) or occasionally? Since this is a DC motor could it be the commutating/brushes in a position where the initial torque is briefly in reverse? I would expect this to be an occasional thing whereas a tach thing might be consistent.. Just a guess.

 

[ScottyUK]

Have you got a PID controller on this winder? The 'I' term could be responsible for this behaviour if a small but persistent error exists, causing PID windup.

 

[jbartos]

Suggestion: It might be a mechanical problem, if applicable. If the motor uses a brake in steady state OFF. When the brake is released, there might be a force or torque in opposite direction, which forces the motor rotate in the opposite direction until it develops enough torque in the designed direction.

 

[cbarn24050]

Hi, you cant drive in reverse unless you have a 4 quadrant drive. Check to see what kind of drive you have.

 

[cabber]

yes, it always runs backwards before speeding up. I'm thinking that the tach might not be zeroed (from another post) and the positioning is wrong which then would contribute some error in the control system. I'll have to check with everyone else's inquiries and get back. Thanks for the continued support.

 

[amptramp]

Cabber:

Food for thought:

What does slightly mean - half a rotation - 3 or 4 rotations?
I would think the amount of reverse rotation could help you zero in on the problem.

Is "threading the mandrel" an auto or manual process?

Is there any kind of tensioning system (mechanical or otherwise) that might be "fighting" you? (This is my suspicion.)

Why do suspect the tach - is your application really positioning the winder? Is the tach being used as a data source for a counter? Is the tach for velocity feedback (speed control)?

Finally, no offense intended, could the reversal be normal? Is the reversal detrimental to the operation. If not, the problem could be "not a problem". I have been burned seeking solutions to a problem that wasn't. You know the old saying - If it ain't broke, don't fix it!

 

[cabber]

I totally agree with the above statement. But the situation is affecting product quality by producing creases at the eye of the coil. The mandrel reverses approximately 2 inches (not that much of a move, but enough to affect product).

I don't believe it to be the tensioning system bc when the mandrel reverses, it creates slack, which means that the bridal system beforehand is not pulling back. The tach is used as a speed fdbk that goes into a speed amp.

The drive I'm using is the GF2000.

 

[imok2]

The mandrel reverses approximately 2 inches (not that much of a move, but enough to affect product).
Just a thought ,but is the mandrel seeking a be balance position on start?

 

[cabber]

I'm thinking that it is. I'm a new person working on this system that was put into place approx 5 years ago. So I'm still getting familiar with the control issues.

 

[edison123]

cabber,

If your dc motor has a differentially compunded series field winding and if the starting inrush current is high enough, then the motor could start in the wrong direction. Check if your motor is differentially compounded and change to cumulative compounding.

Differential compound - series opposes shunt (main) field.

Cumulative compund - series field aids shunt (main) field.

 

[jbartos]

Suggestion to the original posting: Please, provide the motor nameplate data and associated control scheme data.

 

[cabber]

Motor:
HP- 400
Arm. V- 250
Arm. I- 1300
Base Speed- 300 RPM
Top Speed- 1340 RPM
Field Excitation- 29.5A
Field R- 2.85Ohms

Generator:
Power- 320 KW
Volt- 250
Load I- 1280A
Field Excitation- 18.8A
Field R- 16.5Ohms

Gear Ratio:
1.55

Inertia:
25.717 lb-ft^2 (full load)
3.117 lb-ft^2 (no load)

Speed Ref and Speed Fdbk (from tach) goes into a Speed Amplifier. An error is calculated and goes into a Current amplifier along with a tension ref and a current fdbk. another error is calculated and goes into a phase controller which then supplies the motor with voltage. (from how i understand it to be)

 

[ScottyUK]

Is the motor energised only when it is required to move, or is it a true servo with holding capability at zero speed? I'm guessing that you energise it when you need it - the heat losses would be outrageous otherwise.

So if my assumption is true, here's a scenario:

Consider what happens if the measured speed has a calibration zero error of, say, +0.1%. If the reference is true zero, then your controller sees a speed error of 0.1%.

How does the controller behave with such an error? If you have an integral controller, your error will cause the demand to your motor to build up in the negative direction - i.e. it will try to reduce the error. But the motor is turned off, the motor can't slow down and the error persists. When the motor is first turned on, it runs backward as the controller tries to drive the error to zero (i.e. drive the actual speed to -0.1%). As the speed reference ramps up, the error becomes negative, and the controller accelerates the motor up to running speed in the forward direction.

 

[android125]

Hi Cabber
Question :has the machine always been prone to do this or could it have slowly developed the tendancy as it grew older?
Wear creates mechanical play or backlash - could this have something to do with it ?

Some kind of simple ratcheting device could prevent unwanted backwards movement.
Difficult to say, having not seen the machine.

 

[cabber]

I am taking in everyone's consideration and I would like to say thanks for giving me some ideas. I'll post back up when I track down this issue. Thanks again everyone.

 

[nbucska]

look up the maker of the motor in the

http://www.thomasregister.com

and consult them.

<nbucska@pcperipherals.com>

 

[jbartos]

Suggestion: It appears that so many calculations being performed raise a question about initial conditions and their proper setting or resetting.

 

[cabber]

okay the problem resides when a reel in the beginning of the process switches over from being speed regulated to being voltage regulated. the switch causes tension to drop 65%, which means less control between the entry reel and the exit reel. the exit reel, during this state, backs up due to the movement of the entry reel. we don't use brakes, during the transition, and the solution will be software based.

that's the problem, and i'm looking for a solution so i'll update this when i figure it out.

thanks again for everyone's help.

 

[jbartos]

Comment on the previous posting: That problem may need some artificial intelligence.