double check on current loop

Hi All,
I don't do much analog, and I do even less with 4-20ma current loops. Here's what I'm thinking of doing; I'd like a double check so I don't do something silly.

I'm using these transducers to sense the height of a 48" water column in two seperate water tanks. I'll place it at the output of each water tank I'm sensing, which is at the bottom.

The important specs are:
0-5 PSIG
4-20ma output

http://www.dwyer-inst.com/Products/Product.cfm?Group_ID=98&sPageName=ModelChart#specs

I plan to use this panel meter:

http://www.lascarelectronics.com/temperaturedatalogger.php?datalogger=390

It can accept and display two separate voltage inputs with ranges to 0-40vdc (zero and span are adjustable in software).

I plan to convert my 4-20ma signals to voltage inputs at the meter with precision 500 ohm resistors across the input, to ground.

http://www.mouser.com/ProductDetail/Vishay/CPF1500R00FKE14/?qs=TaOZSEYtRiU3zlX9L2fWjg%3d%3d

Sensors will be tied to the meter with Belden 8760 cable; it's 18-2 twisted pair with a drain wire. One-way distance for the cable runs will be approximately 150'; 300' loop.

I plan to feed the whole thing with a 24vdc power supply. According to the data sheet, the meter seems to be OK with it's power supply sharing a ground with the analog inputs. I realize that I will have a live zero and that's fine, as I can account for that in the meter's setup.

Since the meter's zero is adjustable, I will empirically determine what my setting for "empty tank" will be by measuring the voltage across the resistor with the tanks empty.

Likewise, I'll determine what my setting for "full tank" will be by simply filling the tanks with water and measuring the voltage across the resistors. This should give me my span so the meter displays "full" when the tanks are full.

The relationship between water column height and tank fill should be directly proportional, as the tanks I'm sensing are simple cylinders.

Anyone see any pitfalls with this setup? Thanks for the help!


-rpoet

It seems to me you are on the right track. One suggestion I would make is to rescale the instrument to 1.7328 psi this would mean at 48 in of water (100%) = 20ma thus 0 psi (0%) = 4ma. The reason I suggest this is because you will get the full scale of the 4-20ma circuit with more resolution.

Hope this helps

Joe

I just reviewed the transmitter service manual page 3. If you terminate from terminal 3 of the transmitter to the input of the receiver (display) this is the voltage output no Ma output. You should not need the resistor for your application just change the range values in the receive to accommodate the output of the transmitter. It should be close to 0-10 volts

Jwillis said:

I just reviewed the transmitter service manual page 3. If you terminate from terminal 3 of the transmitter to the input of the receiver (display) this is the voltage output no Ma output. You should not need the resistor for your application just change the range values in the receive to accommodate the output of the transmitter. It should be close to 0-10 volts

Click to expand...

I can indeed order the sensor with a 0-10v output. I was planning on using the current output option as the run from the tanks to the display is a fair distance (300' round trip), and a current loop will effectively deal with the distance without having to deal with the voltage drop in the cable; the signal will be converted to a voltage at the meter. It should be more immune to noise as well; the sensor cables will be running near some 120vac wiring. Different underground PVC conduits, but relatively near nonetheless.

I have to order the sensor with a 0-5psig range, as the lead time is too long for me to wait for a custom part. 0-5psig is the closest range I can get; 48" WC is just less than 2psi. I think that's a reasonable amount of the range to get the accuracy I need. I can't rescale the sensor, but I can rescale the display's zero and range so I should be good. Accuracy of +/- a few percent is plenty accurate for this application. Cost is definitely a factor with this project as well and standard parts will help the bottom line.

Thanks for the double check and the suggestions.


-rpoet

rpoet said:

The important specs are:
0-5 PSIG
4-20ma output

Click to expand...

what model sensor ? 0-5 PSIG or 0-15 psig ?

osmanmom said:

what model sensor ? 0-5 PSIG or 0-15 psig ?

Click to expand...

0-5 PSIG

According to the Dwyer configurator: 628-06-GH-P2-E7-S1


-rpoet

Good luck on the project sounds like you have it under control.

are the tank open or close,

if empty what the pressure inside the tank.

check with instrument gauge

osmanmom said:

are the tank open or close,

if empty what the pressure inside the tank.

check with instrument gauge

Click to expand...

Tanks are vented. They're rain water storage tanks. The drain fitting is at the bottom of the tank, so there's zero pressure when the tanks are empty. It should be just less than 2 PSIG when the tanks are full.


-rpoet

If the sensor is going to be at the output, what happens with level while you are draining it? It may give you a false reading.
George

Buy another meter and use 4-20ma - hate voltage analogues - so prone to noise and induced voltage it is not funny.

+1 on getting a 4-20 mA meter and just putting it in the loop ...

George Graziano said:

If the sensor is going to be at the output, what happens with level while you are draining it? It may give you a false reading.
George

Click to expand...

My first thought as well. The pressure sensing point should always be in a spot where the water isn't moving. Otherwise velocity head will cause your reading to be lower than the actual height of the water. This may or may not be significant depending on fluid velocity in the pipe and the accuracy demanded by the application.

Tom Jenkins said:

My first thought as well. The pressure sensing point should always be in a spot where the water isn't moving. Otherwise velocity head will cause your reading to be lower than the actual height of the water. This may or may not be significant depending on fluid velocity in the pipe and the accuracy demanded by the application.

Click to expand...

Not a big deal at all in this particular application. The tanks feed drip irrigation for vegetable gardens. Nowhere near enough flow to cause a noticeable error.

danatomega said:

+1 on getting a 4-20 mA meter and just putting it in the loop ...

Click to expand...

The two counter arguments to that are:

- If you want to just put it "in the loop", you still need a load resistance for the loop, usually of minimum 250 ohms (you would have to read the manual for the transmitter to confirm for the particular device). Usually this would be done by ultimately wiring it to an analog input on a PLC, but it doesn't sound like there's a PLC in the equation here, so then you have to do a series resistor anyway

- 4-20mA is much more noise resistant and all, but what does your PLC do with this 4-20 at it's input terminals? Puts a 500ohm resistor across it and measures the voltage. Which is exactly what the OP is proposing here, only his resistor will be outside the box, not inside it. The arguments about noise and interference are irrelevant as the whole circuit right up to the terminals of the meter IS a 4-20mA loop