Mixer Discharge Control?

Here's one I'm thinking about so I thought I'd ask the experts:
I have a sand mixer in a foundry that mixes recycled sand, new sand, clay
and water continuously. The average time from intake of the mixture to discharge is 1 minute 35 seconds. The mixer is capable of producing between 40 and 125 tons per hour of mixed material.
The current system uses an in-place PLC-5 and hardware to constantly sample the sand for the correct 'compactibility' and 'strength'. The controller meters the water and clay to correct and reach the target.
The discharge "door" on the mixer is also controlled by the -5 in three positions.... Open, Closed, and Float. Float is the optimal mixing scenario because the door allows only what material is pushed against it to exit onto a discharge conveyor. This makes for more efficient mixing.
The door's open, closed and float settings are controlled by the mixer motor amperage. A typical setting, at 50 tons per hour, would be 125 amps (close), 135 amps (open), >125 or <135... float. This works well if... if...
The door also has flow controls on the pneumatic cylinder that move it. If the flow controls (particularly the exhaust on the cylinder extend) are set correctly all works well.
So here's the problem... every time the throughput is increased (tons per hour) the manual flow control must be adjusted to hold the discharge door in a correct position.
My thought is to use a proportional regulator (I like Numatics) on the single exhaust line and hold X pressure on the door cylinder... keeping it closed during the 'float' at a certain pressure every time. P/R's like the Numatics line have command AND feedback lines... I'm thinking I could tune this thing pretty well.
I know that a servo on the door is the best solution, but this is a dirty, moist environment and I'm a little leery of that.
My question is: Do you think proportional pneumatic control in this situation will help?
Thanks,
Jake
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Jake,
Have I got this right? The intermediate 'float' state isn't simply floating on the pressure of the mix. It is a fixed, partially open limit stop. The optimal degree of opening depends on the throughput. You would like have an automatically controllable limit stop to provide this controlled degree of partial opening.
Here is a suggestion: Get a valve actuator from someone like Fisher (Emerson). Send it a 4-20 mA signal from the PLC. This signal would represent 0-100% of valve travel. The actuator should be firmly braced with a rod in place of the valve stem directed towards the door. The door can only open until it is stopped by the actuator. Typical pneumatic actuators operate with an internal pressure of 0-30 psig against a spring. If the door produces enough force to drive the actuator out of position, a positioner can be added to provide internal feedback correction. Get the old, cheap models. Fancy Fieldvue with diagnostics, etc. is unlikely to add value in this application.
Do you have a rate measurement for the PLC to use to set the optimal opening? If not, enter the value manually and let the PLC calculate the optimal position from a table.
Walter.

between
sample
scenario
onto
motor
(close),
set
in
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

floating
Sorry, I must have explained it a bit vague. There is no 'hard' stop. When in 'float', both sides of the door valve are at neutral and exhausted to atmosphere. The only thing keeping the door at a open float position are the flow controls installed on both sides of the door actuator piston.

Walter, I think a proportional regulator would do the same thing on the existing actuator piston. Am I wrong here? It'll take 4-20 and let me drop pressure off the exhaust until I get to X. It will also give me feedback on what pressure still exists in the exhaust line for seeking.

I was hoping to use the tons-per-hour from the -5 as a primary point, and then fine tune on the amp readings coming from the mixer. All are available to the PLC-5 either through direct backplane or DH+ links.
Jake

clay
If
per
door
it
the
tune
dirty,
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
What do you mean by, "... the flow controls installed on both sides of the door actuator piston"?

controller
three
exit
it.
are
keeping
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Here's the scenario: Mixer has amp setpoints of 125 amps to close the door and 135 amps to open the door. Mixer begins charging to 125 amps.
The close valve is de-energized at 125 amps so that the pneumatic cylinder has no <positive> pressure on either side. Each side of the cylinder has manual flow controls that only allow so much CFM to escape from the cylinder in any given time... so far, so good.
What happens in practice is that the load against the cylinder "loads up" every time the mixer tries to push material out.... pushing the door farther and father open with each pass. The back side of the cylinder (extend) loses holding force with added load.
I quoted a Numatics electronic regulator today for $800... the customer said no-go. Back to the drawing board...
Jake
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.