Advice

[Bob O]

Thanks Bernie.

Sorry for the other post must have hit the wrong button. ooopps

 

[Terry Woods]

Actual Weight
---------------- x Current Speed Value = New Speed Value
Target Weight


So... if the Target is 100-lbs and the Actual is 80-lbs then the Speed Value is reduced to 80% of the Current Speed Value. The box was not filled enough so you slow down the speed of.... what?

You said that you...
"...send the new value to the drive to speed up or slow down the machine. Changing the machine speed increases or decreases the amount of product in the container (time fill machine)."

So... by "changing the machine speed" are you changing...

The filler feed-speed?
This would be the speed of the product material as it is fed into the box.

Or, the speed of the box as it moves by the filler?
In this case, the filler is feeding material constantly while the box moves by under the feed discharge. Slower equals more while faster equals less.

Are you trying to control the speed of the walking beam?

If the workers can turn the box around...

Aaahhhh... Could it be...

(so many conflicting pieces of information...)

You say it's a "time-fill machine".
Is it the case that the walking beam carries a box into the filler area? Once the box is in the fill area the feeder begins feeding for x-seconds. (You must have an Eye or something to say "Feed me, Seymour!")

Meanwhile the walking-beam keeps right on walking. It might take two or three steps to move the box through the filler area. The timer on the filler stops the filler before the box leaves the area. This would help reduce spillage, I should think.

Then, at some point the box is transfered onto a conveyor which brings the box to the load cell where the box is weighed.

And so... are you really trying to control the "dwell-time", that is, the time that the box stays in the feed area.

No... that can't be right.

If the feed is timed then no matter how slow the beam walks, only so much material is gonna load.

If the beam walks faster than you end up with spillage as the box leaves the area before the fill is done.

You gotta be trying to synchronize the fill timer with the walk-beam speed...

(Reverse Engineering! What a Hoot! ...sometimes.)

 

[Thomas]

Another approach

I think Bernie has a good approach for your situation since you cannot weigh as you are feeding. I would guess that precise accuracy isn't a priority.

If accuracy is a priority, what about an intermediate hopper so you could weigh the material before boxing it?

 

[Bob O]

I agree with you Thomas. I believe Bernies approach is the correct road to take. I was headed down a similar path but I never thought about not adjusting if within "X" amount of the target. I will adjust in small steps as Bernie stated.
It sounds like all or most that have provided feedback on this issue agree not to use a PID for this application. I have never used PID and was wondering if this was an application for it.

Again Thanks for all the advice and taking it easy on me.

Bob

 

[TheCanadianBear]

Feeder Problem

I currently am running a blending system that fills a hopper with seven different materials of different densities to make a blend of one material having a batch weight total of 20lbs. Then fills a box (lined with plastic) and goes of for taping and palletized.

I have used this process now for five years and is .5 PerCent accurate when I had it qualified.

 

[Bob O]

TheCanadianBear What method are you using to do this? PID or one of the above mention ideas.

Thanks again
Bob

 

[rsdoran]

I didnt understand what you were doing at first, I think I do now. You are using the first box to set the speed for the next box hoping to maintain a constant speed ...ie a constant weight.

I would consider feeding the material into a scaled feeder and it dispense the specified weighed amount. Something like these Accurate models:http://www.accuratefeeders.com/weighbelt_beltscales.html

This would insure that when the operator put in 5kg the weight would be more accurate. I also have worked with a blending system that have to be fairly precise.

I agree with bernie's method if the density etc doesnt change then you should be able to determine an approximate "time" for the first box then use the final weight to adjust the speed.

The batching system I worked with so looonnnnggg ago had one material (ground styrofoam) as a filler, the styrofoam originally was a made board product that had a specified density. They converted to using "used" styrofoam, the kind used in packaging. The density became a problem for awhile because it would deviate from .5 to over 1 lb when the desired was .7 ( I believe this was per cubic foot but its been a long time).

The system also used things like ground carbon black which was fed in very small amount, .3 lb.

It was things like this that taught me how to use compare and create averaging routines. The scales could deviate a little because of vibration. It was a great learning experience.