Announcement

Collapse
No announcement yet.

I'd like your opinion of PID Flow Control please!

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts


  • I'd like your opinion of PID Flow Control please!

    Hi All,

    BACKGROUND:
    This unit is used to dispense Urethane (Glue) by pumping a liquid material out to a dispense head which applies the Urethane to a substrate on a moving conveyor line. The idea is to use the flowmeters in a closed loop system and let the operator set the desired flow rate and have the PAC control the flow based on how fast the conveyor is moving (i.e. the faster the conveyor the more glue you need to pump).

    I am using a P3K with a High Speed Input Module to read the pulses from a VSE (gear type) flowmeter (pumping urethane resin). Resin is pumped with a PD internal gear style Viking Pump coupled to a 3HP Inv. Duty motor.

    The material is discharged from the pump, enters a heat exchanger (temp control), then enters the flowmeter, then exits and is delivered to the dispense head.

    Motor is controlled by GS2 VFD using the PID Loop instruction in the P3K (I do have an EDRV-100 module for this drive but have decided to use a Analog Output of the P3K to the AI of the GS2 as I feel the resolution is better.

    I have used this same setup in the past and it works well HOWEVER, the distance the material travels from the pump to the dispense head is relatively
    short approx 10'/3M. I am working on a new machine and the distance is FAR greater to thedispense head 60'/18M. There is a noticeable loss in accuracy and it seems that the PID/Motors seem to be not as stable (more/constant flow changes).

    I assume that this is happening as result of the greater distance to the dispense head and the fluid/hydraulic mechanics occuring in the piping. Honestly, these fluid variables are "above my pay grade" and I am left to wonder what I can do to improve accuracy.

    QUESTION
    Do you think that relocating the flowmeter CLOSER to the dispense head would improve my accuracy or create more instability in the system because theoretically there would be more of a time delay between the motor/pump speed changes and the reaction of the flowmeter (and of course the processing time in the P3K).

    I have seen this setup work DEAD NUTS perfectly in the past and cannot help but feel I am missing something here to get it to work as well in this application.

    I'd LOVE to hear from any/all of you on your thoughts and suggestions!

    Reagrds,
    Sid


  • #2
    Initial thought after reading your post:

    Is there air in the line between the pump and the dispensing head?
    If there is a high spot in the line, where an air bubble could sit, it could be compressing and decompressing which could really mess with the flow meter vs. the actual flow rate at the dispensing head.
    Circumstances don't determine who we are, they only reveal it.
    Jason Wolthuis

    Comment



    • #3
      Originally posted by sshneider View Post
      .

      QUESTION
      Do you think that relocating the flowmeter CLOSER to the dispense head would improve my accuracy or create more instability in the system because theoretically there would be more of a time delay between the motor/pump speed changes and the reaction of the flowmeter (and of course the processing time in the P3K).

      I have seen this setup work DEAD NUTS perfectly in the past and cannot help but feel I am missing something here to get it to work as well in this application.

      I'd LOVE to hear from any/all of you on your thoughts and suggestions!

      Reagrds,
      Sid
      From your description it sounds like you have done this before and it worked for you. I'm assuming you have checked all the usual culprits like noise affecting your highspeed counter. If air pockets or entrained air are the problem moving the flowmeter closer to the dispenser may or may not help depending on how the flowmeter handles entrained air.
      PERCUSSIVE MAINTENANCE: The fine art of whacking the devil out of an electronic device to get it to work again.

      Vaughn

      Comment



      • #4
        How easy is it to increase your line size going to the process? You are probably fighting a tremendous amount of backpressure at that distance (depending on the fluid viscosity). I've had to rework plumbing on similar setups when I realized how much pressure there was (you may want to sacrifice a gauge just to see).

        How are you identifying the issue? Is it visible flow issues at the process, or irratic readings from the flowmeter? If you have irratic flowmeter readings (and it sounds like the pump is 'hunting' for the right speed), I'd move the meter closer to the pump (before the heat exchanger) to take out any possible 'lag' between the pump and the flowmeter. In a perfect world, as long as the fluid is coming out of the PD pump at the right rate, you will get it at the right rate at the other end of the line.

        If the flowmeter signal looks okay and the process flow is irratic, I'd try taking the heat exchanger out of the loop and see how it looks. You could possible have an air entrapment problem that you never had before because the additional backpressure is not letting the air escape. This seems like kind of a long-shot though.


        We have a very low viscoisity epoxy hardener we pump with the exact pump you are using (even the same HP if I remember correctly). Even though it is a positive displacement pump, the fluid can 'slip' in the pump (I shut the pump outlet valve completely off during startup trying to set the overpressure switch and never got over 60psi). I would only worrry about this if the Urethane is thinner that water (the hardener I'm talking about is). Again, this is a long-shot.
        Last edited by pressgrove; 11-18-2012, 03:00 PM.

        Comment



        • #5
          Hey Fellas, Thanks for all your suggestions thus far! I think the size of the piping is OK as there does not seem to be that much pressure either at the head or the pump (I have gauges in both places).

          Viscosity of the material is approx 1500 cps (looks like BBQ Sauce) Temp = 85F. I don't think there is slippage in the pump and I have disabled the PRV on the pump as I have Pressure Transmitters set up to alarm/shut down on high/over pressure.

          Now the air entrapment is a good catch- i don't know why I didn't think of checking that before? This merits more investigation-Thanks!

          Yes-Motor/Pump is erratic- constantly searching for PID SP target. It does arrive at the target but never stays steady for longer than a couple of seconds. Sounds like too much PID gain right? But K= .85, I = 2.5 & D = 0. Seems to me that this loopis working more off of Integral than Gain?

          Perhaps this is an application where the standard PID instruction just wont work? I guess I could write some kind of rung that does something similar (compares values and adjusts pump speed accordingly) but do it in a very "Black/White" way so the pump is more stable?? I dunno, seems like re-inventing the wheel.

          I wish the P3K PID instruction had a bit more "tweakability" to fine tune it a bit more but, people in Hell want Ice water.

          Comment

          Working...
          X