Square Root Extraction of 4-20mA Instruments

Hi,

I'm in the process of moving about 100 signals across from an old PLC system to a new CLX one.

Some of the readings( flowmeters to begin with) are not as expected and I've being told thatin the old PLC (which I have no access to logic) there is Square root extraction logic performed on the signals.

Has anyone seen this before? What is required? Is it basically just a math operation on the scaled or raw value?

Help!!!

Square Root is available as an instruction in the Control Logix family.

They must be using a differential pressure device to calculate flow.

The square root of the differential pressure, for example across an orifice plate, is proportional to the flow.

You really need that old program or you'll have alot of work to do, but you probably know that already.

On the flow measuring device there should be a tag that indicates the meter's rated differential pressure (usually inches of water in the US) and corresponding flow rate. The transmitter should have a tag indicating transmitter max range - that is, the inches of water at 20 mA.

In your program convert the 4-20 mA to engineering units of inches of water. The flow will be:

Meter Full Scale x (Sqrt(Actual inches H2O / Meter Design Diff Pressure))

Yes I had a feeling I had my work cut out. Trying to get access to the the program but not looking too good yet.

Unforuntately the instrument label dosent tell me anything. They are approx 30 years old I'm told.

The problem came to light when we moved the first number of signals across. The span of signal 1 for sake of argument (according to spec was 0-1 cum/s). I measured the output from transmitter to be 12mA. Using my standard scaling block my result was 0.5 cum/s as expected.

I googled this SQRT extraction today and write a small routine as follows::

1. VA=(I-4)/16 /Normalise Input
2. R=SQRT(VA) /Sqaure Root
3. VA_Actual= (R * 16)+4

4. I then pass VA_Actual into my scaling block which is scaled for 0-1 Eng Units.

Again the results are not as expected. I was expecting a value of 0.26, my result was 0.64

Does this make any sense?

Thanks

Can the flow meter do the math for you?

I never succeeded with this kind of flow meter mesuring air flow....All the time it was out of range, dirty etc....

And it is worst than what i have tried because yours are 30 years old...!!! I may imagine how dirty are the holes

Are they controling critical things ? Could you event think about replacing them with a differentiel temperature mesurement ones ? You can have some with a linear 4-20mA range etc

Jeff23spl said:

I never succeeded with this kind of flow meter mesuring air flow....All the time it was out of range, dirty etc....

And it is worst than what i have tried because yours are 30 years old...!!! I may imagine how dirty are the holes

Are they controling critical things ? Could you event think about replacing them with a differentiel temperature mesurement ones ? You can have some with a linear 4-20mA range etc

Click to expand...

Yes thats the problem I'm expericing. No consistancy...some appear right and others do not for the same calculation. Why is this?Doubt the client is going to agree to 100+ new instruments.

So will it work for other flows other than air?

Hi if you suspect a bad flowmeter (s) why not hire or buy a good flowmeter and put in in series to allow a benchmark calibration /validation of the old ones

i can't help on mesuring other kind of flow than air or gaz flow...

But with diff pressure meter, a bad meter need to be cleaned out and usually go back to work for....a while.... that is sometimes too short...So most of our air project are know made with diff temerature meter instead but i have not much experience for other media than air

MorphuisOGrady said:

Again the results are not as expected. I was expecting a value of 0.26, my result was 0.64

Does this make any sense?

Click to expand...

what the?
the SQRT of 0.5 is 0.707 so getting 0.64 is close
also 0.5*0.5 = 0.25 which is a square

PS i am learning something new about flow rates and differential pressure

More Detail

Been doing more research - Tom has the formula correct - see below for more detail in a orifice plate
http://www.efunda.com/formulae/fluids/calc_orifice_flowmeter.cfm

You need to isolate your problem. It doesn't appear that you have identified the source as the primary element (orifice plate, pitot tube, venturi, whatever), the transmitter, the PLC input, or the program logic.

Start at one end of the chain and work through. Get an inexpensive DP guage, and make sure the impulse lines (tubing to the high and low taps on the element) and orifices in the element are clear and you are getting an accurate dp (differential pressure) reading. Check at a couple of different flow readings.

Then hook up the transmitter, and with a multimeter make sure the output signal matches the dp at both flow rates.

Then check the raw data to the PLC input and make sure the transmitter output is being read by the PLC correctly.

Then you should check the results of the calculation in the PLC and compare it to both a manual calculation ant the actual flow rate at the meter.

Note: these kinds of flow elements are essentially very reliable unless dirty (as someone above pointed out). They read volumetric flow, not mass flow like a thermal mass flow meter. If you want mass flow you need to correct with actual pressur and temperature of the fluid.

First of all I assume that your analogue input to the PLC is configured for 4 to 20mA input.
That being the case then given your span of 0-1 cum/s (which I assume is 0-1 cubic metres/ sec) and your 12mA signal which is 50% of differential pressure span then the digitised input to the PLC will be equivalent to 50% or 0.5 per unit value.

SQRT(input) = SQRT(0.5) = 0.7071
0.7071 x Max Flow Value = 0.7071 x 1 = 0.7071 cum/s

This corresponds to 70.71% of maximum flow rate which what it should be for 50% of Differential Pressure. Hope this helps you.

I learned stuff too thanks to the Micheal link, it will help me out to get the equation to find mass flow with volumetric flow etc...

By the way their is other mecanical things to look for when using any kind of flow meter....Having 90deg or piping not being straight near the reading will greatly affect it etc...