0
\$\begingroup\$

Hoping I can gain some clarity on an issue that I am having -

We recently installed a new system to running a VFD controlled pump, controlled by a pressure transducer. The pressure transducer we selected is a TD1000 from transducers Direct.

Pump we are using.

Motor on pump: https://product-selection.grundfos.com/us/products/mge-mle/mge132se-98988681?pumpsystemid=2245336963&tab=documentation

https://api.grundfos.com/literature/Grundfosliterature-4912197.pdf

I/O wiring diagram from manufacturer. Transducer is wired into pin #8 (+24VDC), and pin #14 (AI3). AI3 is configured in the parameters as a 4-20mA signal.

enter image description here

When connected to the VFD on the specified analog input channel, the reading from the pressure transducer would not display a pressure normally continually, but intermittently it would work. Trying another 4-20mA sensor proved that everything else was working as it should. Logically we wondered if the sensor was bad, but after trying the TD1000 sensor in another system (completely different application) the sensor functioned correctly with an accurate reading. When tried it again our new application, it does not work.

The transducer manufacturer is speculating that the issue could be with the input card resistance, and the controller therefore is sensing that the transducer is being detected as a short and cycling power.

Any insight on why this would not work with this specific application, but works fine with others, and other pressure transducers work with this project? After some research online, I am curious if it has to do with the < 100ohm Input Impedance specified in the sensor datasheet. This is just a hunch that I could be completely off base on - thus the question.

\$\endgroup\$
5
  • \$\begingroup\$ Are pin 3 and pin 9 the same ground? \$\endgroup\$
    – Jeroen3
    Commented Jan 19 at 14:23
  • \$\begingroup\$ Can you provide a link to the VFD user manual rather than the web catalog page. Thanks. \$\endgroup\$
    – Transistor
    Commented Jan 19 at 14:36
  • \$\begingroup\$ Thanks for the manual link. It's not clear to me how you're supposed to configure AI3 as 4 - 20 mA rather than 0 - 10 V. \$\endgroup\$
    – Transistor
    Commented Jan 19 at 16:37
  • \$\begingroup\$ @Transistor see page 36-37 of the api.grundfos link. This setting is all done through the display on the VFD. Settings ->Analog Inputs ->Analog Input 3, setup ->Analog input 3, elect. signal, then a selection between 0-10v,4-20mA, and 0-20mA. \$\endgroup\$ Commented Jan 19 at 16:56
  • \$\begingroup\$ @Jeroen3 Resistance values between AI3 and ground pins were the same. \$\endgroup\$ Commented Jan 19 at 17:03

1 Answer 1

1
\$\begingroup\$

schematic

simulate this circuit – Schematic created using CircuitLab

Figure 1. Simplified circuit.

A 2-wire, 4 - 20 mA transducer "steals" some power from the loop. To do this it must have a minimum voltage drop across it.

enter image description here

Figure 2. Extract from the datasheet.

In Figure 2 we can see that the transducer needs a supply voltage of at least 7.5 VDC above the maximum output voltage (which will occur at 20 mA). (The 0.3 V specification is for 0 - 10 V versions.)

Debug steps:

  • Power down the VFD and measure the resistance between the analog input and ground. I would expect you to find that it's 500 Ω as this gives 10 V for a 20 mA input and that would support the 0 - 10 V control voltages common on industrial drives (by simply removing the jumper to the current shunt resistor).
  • Now calculate the maximum input voltage to the VFD using V = IR = 0.02 × R measured above. We're expecting 500 Ω so at 20 mA that's 10 V.
  • Your minimum supply voltage for the transducer must be the maximum input voltage calculated above + 7.5 V. That would be 17.5 V in our example.

The 100 Ω output impedance is only relevant for the voltage output version. The 4 - 20 mA version is a constant current drive and they are have the characteristics of infinite output impedance (when operated within their compliance range).

Report back!



From the comments:

Measuring the resistance between the analog input (AI3, #14) and the ground pin (#9) with no power to the VFD gives a value of 122 kilohms. This seems abnormally high.

That's got to be wrong. There must be a jumper missing. See my Figure 1.

Checking to the same pins with power, gives a value of 292.2 ohms.

You can't make a multimeter resistance measurement on a powered circuit. The multimeter measures the voltage drop caused by a test current from the meter. If there is another source of current in the powered circuit then your reading is nonsense.

Checking voltage differential to AI3, and a 24V source, pin #8, shows 23.83V.

There's not a lot of relevance to that reading other than it looks like the 24 V supply is good and there is a path to ground through A13.

With the transducer wire connected to AI3, voltage differential shows 22.58V (at 0 psi).

A reading from AI3 to GND would be more useful but from the previous reading we can calculate that VAI3 = 23.83 - 22.58 = 1.25 V. If your transducer is giving 4 mA at 0 psi then the AI3 shunt resistor value would appear to be RAI3 = V / R = 1.25 / 0.004 = 312 Ω which isn't too far from your 292 Ω in the manual (which I'd like to confirm).

If the 292 Ω value is correct then at 20 mA we'd have VAI3 = IR = 0.02 × 292 = 5.84 V.

\$\endgroup\$
8
  • \$\begingroup\$ Measuring the resistance between the analog input (AI3, #14) and the ground pin (#9) with no power to the VFD gives a value of 122kohms. This seems abnormally high... I specifically double checked this value and confirmed the "K". Checking to the same pins with power, gives a value of 292.2 ohms. Checking voltage differential to AI3, and a 24V source, pin #8, shows 23.83V. With the transducer wire connected to AI3, voltage differential shows 22.58V (at 0 psi). These measurements are with a working transducer, currently the non-working transducer is being inspected by the manufacturer. \$\endgroup\$ Commented Jan 19 at 13:52
  • \$\begingroup\$ The pump controller manual says Ri is equal 292 Ω. This sensor requiring 7.5Vdc is a problem, because there is no way to make this from 24V while maintaining full range. (eg, you'd need 1.8k shunt, but then 20mA takes 37V) I'd assume the psu needs to be at least 7.5, so 5V won't work but 12V will. \$\endgroup\$
    – Jeroen3
    Commented Jan 19 at 14:17
  • \$\begingroup\$ @StevenAnderson, see the update. Can you provide the manual direct link as requested below your question. Thanks. \$\endgroup\$
    – Transistor
    Commented Jan 19 at 14:45
  • \$\begingroup\$ @Transistor, I have updated the initial post with motor information. \$\endgroup\$ Commented Jan 19 at 15:26
  • \$\begingroup\$ @Jeroen3 Not quite wrapping my head around what you are meaning. Can you explain the last sentence of your post? TIA. \$\endgroup\$ Commented Jan 19 at 15:28

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.