PID - The most difficult system so far.

Re: There is more to this than geting the lowest ISE

Peter Nachtwey said:
The maximum speed of the system is only 50 inches per second. WHY?
Does anybody know?

The system gain G is 5 inches per second per volt. The controller can output 10 volts. 5 inches per second per volt * 10 volts = 50 inches per second. This is the principle behind feed forwards. If we want to move the actuator at 5 inches per second we output 1 volt.

Code:
Desired Speed     Control Output
  -50                 -10
  -40                  -8
  -30                  -6
  -20                  -4
  -10                  -2
    0                   0
    5                   1
   10                   2
   15                   3
   20                   4
   25                   5
   30                   6
   35                   7
   40                   8
   45                   9
   50                  10

This is a key point.

control output * Gain = Velocity

If the desired actuator speed is 4.7 inches per second then we need to output:

Code:
velocity
-------- = control output
  gain  

or

4.7 inches per second per volt 
------------------------------  = .94 volts.
     5 volts

This is a key point!!! Think about this. One can predict how much output the system requires to move using feed forwards. This was demonstrated by the previous spread sheet that uses only feed forwards for control.

I hope you guys are reading the comments in the Excel spreadsheet. The cells with commensts have a little red triangle in the upper left hand corner.
 
WHERE TO STICK IT

code:--------------------------------------------------------------------------------

velocity
-------- = control output
gain

or

4.7 inches per second per volt
------------------------------ = .94 volts.
5 volts

--------------------------------------------------------------------------------

Peter for give me,BUT This is a key point!!! I`m missing it!!!
On the FF sheets we could see if the system wasn`t responding
increase acc ff if it response good you don`t need to accel as fast
I`ve read and reread your last post I may be brain dead but
I don`t know what to do with the .94 volts?? đź“š banghead
 
Re: WHERE TO STICK IT

Originally posted by Thomas Sullens
Peter for give me,BUT This is a key point!!! I`m missing it!!!
On the FF sheets we could see if the system wasn`t responding
increase acc ff if it response good you don`t need to accel as fast
I`ve read and reread your last post I may be brain dead but
I don`t know what to do with the .94 volts??

You don't do anything with the .94 volts and the PID doesn't either. That is the point. If this system is to move at 4.7 inches per second it will take .94 volts. This output can come from the feed forwards or the PID or a combination of the two. If the feed forward is set correctly at 1/G or .2 volts per inches per second then there will be no or little error for the PID to correct.

I will make another FF spread sheet wit a real move instead of a sinewave.
 
accel FF

OK
Kp=7.1
Ki=32.5
Kd=.002
Kv=.2
Ka=.0056
Kj=.0001
66.194
System GAIN = 5"per sec per volt of output and what your saying is
IF you know your systems G then 1/G = Kv which is
the same as Velocity FF is a give me. So you stick that in the Kv
and leave it alone and tune the rest. Just in case who ever designed
the system forgets to put a sticker on the machine telling you what the system gain is looks to me like now your going to have to tune
the system to find out what the gain is. So now we will use Pro. G
get some control then use Ext. FF then find the difference between
hi and low volts on the drive and use this for accel FF. Then go back
to Pro. G and tune the rest. My question is how can you find the system G without tuning the system?
 
WHAT'S THIS? THE DESIGNERS FORGOT THE GAIN LABEL?

Thomas Sullens said:
My question is how can you find the system G without tuning the system?

It should be a law. The mechanical designers should put a label on the machine that indicates the system gain or better yet the transfer function. That would make life easier for the control guys.

Seriously, the simple thing to do is to output 1 volt using a open loop command and check how fast the axis moves. In the case of a motion system you can output 1 volt and measure the speed. If the speed is 5 inches per second then the system gain is 5 inches per second per volt.

Check my past post in reply to what I do when I tune up a system.
The first thing I usually do is make an open loop move.

Last Thursday when I was tuning the electric servo on the trainer I noticed that it took 6 volts to go 20 inches per second by looking at the graph.

Code:
20 inches per second
--------------------- = 3.333 inches per second per volt.
      6 volts

Since Kv is 1/G, Kv is 1/3.3333 or .3 volts per inch per second.

Tuning the system with the a dead band is more problematic like your Nacchi valve.

With out the gain label you had better have tuning tools otherwise there is a lot of guessing, trial and error.
 

Similar Topics

Hi, I would like to assemble a simulator/practice booster pump system that uses PID to maintain steady water pressure under various outlet demands...
Replies
0
Views
82
Hello, I have a motor that we are sending a RPM Speed Output from 0-100% to the VFD. However, the Motor HP needs to be limited to 6000 HP and the...
Replies
3
Views
91
I have S7 1512C controler for controlling 48 PID temperature loop, the output is PWM. Please I need the best, most efficient way to write the...
Replies
13
Views
604
Hi all, I'm having trouble solving a problem I've been working on for several months, and thought you might like a stab at it. The machine runs...
Replies
22
Views
954
How can I connect PID Output to a valve. In ladder logic program is there any logic do I want to add between valve and PID? PV=SP What will be the...
Replies
7
Views
418
Back
Top Bottom