Loss of weight simulation

Hi,

I need to simulate a loss of weight program.

I have managed to alter the weight of the bin by having a value "A" and taking a fixed amount off it every time a timer gets a DN bit (to empty), and obviously adding an amount when I need to simulate a fill. I have used the timer to slow the rate that the bin fills and empties. This works ok but it is not accurate, see below.

The tricky bit seems to be the valve at the bottom of the weigh bin. It is always open (50%) and varying its position. How can I vary the weight "A" by the position of the valve? If say the valve can be positioned between 0-100% open.

Also, as the bin empties the rate of emptying reduces as there is less material pushing the existing material out of the bin. Assuming a fixed position of the valve. The inverse happens during the fill.

I know if I have got my logic right when the bin is filling the valve will start to close (40%) and when the bin is emptying the valve will start to open (60%). Can anyone help me simulate the weight?

Ben

Sounds like its time to get out the old physics and hydraulic books.

You need to calculate pressure head created in the tank which is a direct correlation of the height of material. To simulate this you are going to need to use alot of math instructions to very the preset of your "Emptying Timer". The higher the fluid column the lower the preset will be therefore making your timer cycle faster and empty quicker.

As far as weight goes you need the dimensions of the tank to calculate volume. If you know the weight of a given volume of your fluid you can then calculate total weight using just the fluid height.

I know I didn't give specific ansewers but I don't have all the formulas with me and I would hate to take all of the fun away from you.

Hope this little bit helps

Tim

You will also help your accuracy a little bit if you use a timed interrupt (if your plc has one). This will keep the time between calculations more consistent and help maintain the accuracy of your sum. If you don't have a timed interrupt using the timer done is fine. Just keep track of the preset as you will use it later

Also try to keep the values in user units as much as possible and scale them where you add to and subtract from your total. Calculate your mass flow rate in pounds/minute or some other appropriate unit so you can use normal equations that make sense. Then convert this number to pounds (or some other unit) per scan and add or subract this from your total as required.
I have always used this approach (change what you add to the total as opposed to how often you add it) because it leads to more straight-forward evaluation.

Keith

Sounds like integration to me. Calculus makes me choke.

Fortunately, a PLC (I assume you're using a PLC or similar device) can handle this type of problem in a number of ways:

1. Create a formula to simulate your system. If your valve position is directly related to the weight in the bin and both variables are proportional to time, then you should be able to relate the two and write an equation. *GAG* You could then have the PLC solve the equation based off the value in a timer.

2. Take samples of your real world data and create an accumulating value. You're ALREADY doing this. The "problem" here is you're sampling, so you'll never have 100.000000% accuracy. But you should be able to get close enough, in most cases, to satisfy your design requirements.

3. If your system is predictable, create a table (also not 100% accurate). Select a value from the table based on your time period. You can use data, rather than calculation, to fill in the table if you have to. This one is a memory hog, and is tricky to program, but it can save time.
(I learned this one in an economics class. You spend weeks learning those silly formulas with all the exponents then..."OH LOOKIE! Here's the table of standard multipliers, sorry about all that math.")

Here are my two cents on your problem:

ben said:

Hi,
The tricky bit seems to be the valve at the bottom of the weigh bin. It is always open (50%) and varying its position. How can I vary the weight "A" by the position of the valve? If say the valve can be positioned between 0-100% open.

Click to expand...

Always open (50%) AND varying position? Does the valve vary between 50(min)-100%; or have you tested your program with the valve open 50%, found out that it works, and then tried to adjust the valve?

Also, as the bin empties the rate of emptying reduces as there is less material pushing the existing material out of the bin. Assuming a fixed position of the valve. The inverse happens during the fill.

I know if I have got my logic right when the bin is filling the valve will start to close (40%) and when the bin is emptying the valve will start to open (60%). Can anyone help me simulate the weight?

Click to expand...

You need to clarify something: are you using the valve to create a constant flow rate?
If so, you can compensate for the problem caused by the weight of the material in the bin. Your real problem, then, is figuring out what position the valve should be in based on the amount of material.

Otherwise:
What you need to do is create a multiplier, based on the amount of material already in the bin and/or the position of the valve. Multiply the amount you add/subtract with each time period by the multiplier.

If you have one variable (weight OR valve position), the multiplier will be easier to calculate.

AK