MAM instruction help

Darkzadow

Member
Join Date
Nov 2011
Location
Phily
Posts
94
Hi I am having trouble with a Motion action Move CMD. The drive is not exactly tuned so that is part of the issue but I have a question about the Command position reaching set point. During the move the the actual position follows the command position at a close rate of about .05 inches. When command position reaches he set point my velocity changes from the programmed speed of .005 inches a second to .1 or .15 and oscillates to bring the position in line with the command position.

Is there a way to not have the speed ramp way beyond my programmed move. if not is there a way to limit the magnitude of the velocity after command position reaches set point?

Thanks any help is apreciated I am working on tuning but it has to be by hand because underload the drive must follow closely with 3 other drives
 
tuning matters lots
Under load - if the axis cannot follow the command reference then the tuning values must be fixed before starting the application testing

This is a RS Logix 5000 system ? Version?
Kinetics 6000?
MO2AE?
 
Last edited:
This is rslogix 5000 v16 M02AE card to an Allen Bradley 1391 ac servo drive to an 1326 ac servo. The system has 4 identical servos that raise and lower a quench press. Using the auto tune is not possible under load. The servo seems to follow the command position closely just that when the command position reaches set point even with .05 inches remaining the servo hunts for position very aggressively. hunting is ok for this application but it goes from the .05 inches per second travel speed to .1 inches per second while hunting. I want to stop or limit the aggressiveness of the hunting or eliminate if possible. I do not know what parameters control the hunting as it does not follow the parameters of the MAM block and exceeds parameters loaded in the axis. IE: i set the max speed and acceleration to just above MAM parameters.
 
do all four servos control one corner each of the same press?
So if one servo moves too far compared to the others the press twists and locks up
(a basic diagram may assist)

I have some ideas but need the mechanical layout
 
Yeah each one controls a corner. The bottom of the press is fixed while the top is servo driven/ free floating. This press has line rolls on the top and bottom as this press is used to prevent plates of steel from warping during a quench process to drives only need to maintain a position and not apply torque and flatten an object. The press has 12 inches of usable movement, and will run off the screws at 14 inches. The drives are 3000 rpm capable drives but I am limiting them to 2000 rpm as my max usable speed(.05 inches per second goal). The drive is going to a 10:1 gearbox to a 48:1 jack screw so I have 480 revolutions per inch on my drive. I have 1 inch of error tolerance before frame locks severely. I am not sure what axis values are appropriate so I have 100 as P gain 2.5 as I gain(instruction set says it is a usual set of values). All in all it is a simple system but I am upgrading from 4 PLC5 IMC cards and the manufacturer control system.

For my next try at this Saturday Im going to have Velocity Feedforward and accel feed forward at 100% and have integrator hold enabled.

For axis I am entering .066 inch/sec as max speed, .1 as max acceleration and have Jerk set to 60% of time. I was using trapezodial movement and will try S-curve. I suspect that I will still have the large oscillation but I am hoping that Integrator Hold and the Feed Forward settings will reduce the oscillation and keep the drive closer to the command position during MAM.

I dont see how I can tune the drives under load without wracking the frame. I can uncouple one of the drives easily and tune it with no load but press has significant inertia. the screw jacks will hold the press in place when all drives are uncoupled. I could live with a drive that took longer to reach a point over having oscillation(current oscillation is violent using max speed of the axis which i think is because Integrator hold was not enabled giving me a large integral value during a MAM before command position reaches setpoint and as the drive was .05~.07 inches behind caused large corrective movement at limits drive, All wild speculation on my part).
 
Last edited:
Also I do not know what is appropriate accel and jerk values. I will be trying to trend normal movement from the inplace control system to see if I can pick it up from there.
 
sounds like fun

Get the Trending to work before going further - need Position FB , position error, velocity, accel and torque (if in torque mode) on every axis

think about using 1:1 MAG on three axis to the third axis which you use a auto tune on (need same scaling etc) Note only possible after basic tuning is operational
As always try small steps first eg start 1/2 inch below level and tune up to 1/2 inch above level (gives you 1 inch of travel) At 30% of top speed using 80 percent torque with a 1" travel limit

as a lift only need to tune in the up direction

is the MO2AE in Torque or velocity mode?
are the 1391 new or existing (if so are you using the same mode and scaling)
 
I am in velocity mode and the 1391 are existing. All servo motors and drives are identical. I have trending when I use the rslogix 5000 M02ae I will be trying to trend the existing controls after startup. I did not think you could use MAG on an axis that you are tuning. When tuning I thought the axis had to be off. Also, if you know , does gearing compensate if one drive has unexpected drag or lag.
 
How does one first tune the other three axes that are geared to the one being tuned? We recommend sending the same open loop command to all axes at the same time but have a skew detection.

The maximum speed should be just that or the feed forward gains will be wrong. The maximum speed should be the speed obtained when the control output is 100% or 10 volts. If you do do this then the feed forward gains will be wrong.

The way the M02AE works internally is that there is a outer position loop and inner velocity loop. The outer position loop outputs a velocity reference to the inner velocity PI controller. Therefore the feed forward should be passed through with a value of 1 or 100% in most cases but that will only be right if the maximum speed is set correctly.
 
So on the axis properties I should have the maximum achievable speed not just Max speed I would like the drive to be limited to. And for the tuning, I want all settings to be identical, have the 3 geared to the one to be tuned, and have a compensation routine to keep the drives within allowable tolerance of each other. or at least one to halt the press if they differ to much.
 
So on the axis properties I should have the maximum achievable speed not just Max speed I would like the drive to be limited to. And for the tuning, I want all settings to be identical, have the 3 geared to the one to be tuned, and have a compensation routine to keep the drives within allowable tolerance of each other. or at least one to halt the press if they differ to much.

MO2AE settings to a velocity VSD Using Version 19.01
Do not get confused about max speed - Look at the Output tab the % per (inch/s) is the value that Peter is talking about so that the Feedforward works correctly - The value on the Dynamics Tab is where the you can limit the Max Speed for your application
M02AE Output.png
M02AE Dynamics.png

On the Tune Tab you can restrict the travel to 1 inch and use a % of the application speed - Direction is Forward Unidirectional (assuming that up is forward)
Damping factor is by default 0.8 which gives you a little bit of overshoot (hot tune setup) - I usually use a damping factor of 1 which gives me a critically damped system
Position error Integrator - This (Rockwell recommendation) is for high precision systems only - I usually do not use any integral until you have the system operating reasonably (Gains tab)
Velocity Feedforward - I usually use this every time ( Gains Tab feedforward = 100)
Output filter - Advanced discussion needed to set correctly manually (turns on low pass filter and sets the bandwidth on the output tab)- rough setup is try not to use it on normal systems, on high inertia systems it may be required
M02AE Tune.png

M02AE Gains.png

I want the servo output level on my trends so set up the servo tab as shown

M02AE Servo.png
 
Hey I just want to say thanks for the tuning while geared and the help on the movement trending. I got the axis's to move without hunting. Its scary precise now, with no visible hunting. Now I just have to have it home correctly according to load cells which is fickle at best as whole frame is sagging from age and rolls in need of replacing. Thanks for the help I couldn't have done it without you. :)
 
Glad to be of assistance. 🍻

Thank you for informing us on how it all went, its surprising how many people do not report back about how our information helped.
P.S. we also like to know to improve our knowledge.
 

Similar Topics

Hi guys, beginner here. I need to be able to stop an incremental MAM instruction in Studio 5000 and then start it back. But the catch is I...
Replies
8
Views
354
Hi All, I currently have an intermittent issue on a CompactLogix 1769-L30ERM with firmware 28.11. When I get to the machine, all conditions for...
Replies
5
Views
3,324
Hi I am new to the motion control. We have two kinetix 7000 which are controlled by controllogix 5000. Both servo drives mostly work fine...
Replies
11
Views
5,286
We have a servo motor for a tubing coiling (rotating) arm. I'm using RSLGX5000 (V19.01.00), CTLGX 5000 (1756-L61), and Kinetix 6500. The coiling...
Replies
13
Views
9,949
Howdy guys - Looking for some insight on the MAM instruction. Had an interesting one today: I performed a controller upgrade on an older system...
Replies
1
Views
456
Back
Top Bottom