What is your nomenclature for 4-20mA speed signals?

dogleg43 said:

Isn’t this kind of a point of reference thing? Kinda like using the terms upload and download?

Click to expand...

Some of the posts are about that, but the original question was really about what people use for setpoint (reference vs command) and what people use for process variable (feedback vs reference).

jhenson29 said:

...if I have a VFD setup with a positive feedback loop because the process feeds back into the reference to keep the process at a limit set by another device, then the reference comes from the process.

Click to expand...

Even you are still calling the value that goes into the VFD, as the setpoint, the "reference." That is the what I have been trying to say all along, and it is the opposite of the OP's convention.

I already said that using the phrase "from the process" as the source of the reference was a mistake on my part.

The OP said they use

"reference" for the signal that indicates current VFD speed.

Click to expand...

i.e. not the setpoint, but the actual motor speed that comes back from the VFD. It does not make sense, to me, to call the measurement of the thing being controlled the reference.


But that's just my nose.

drbitboy said:

It does not make sense, to me, to call the measurement of the thing being controlled the reference.

Click to expand...

I understand that we agree to call the setpoint the reference and to call the process variable the feedback.

We just don’t seem to agree that they can the same thing in a positive feedback loop, which I only mentioned because I don’t agree it’s strictly excluded as you stated.

But I don’t know how else to explain it, so I’ll just leave it there.

Nose is looking good drbitboy

jhenson29 said:

We just don’t seem to agree that they can the same thing in a positive feedback loop, which I only mentioned because I don’t agree it’s strictly excluded as you stated.

Click to expand...

Nope sorry, a "reference" is something you can "look-up". Most often fixed but can be changed.

You cannot call the feedback of a process variable a reference, unless (cough) it is used as a reference to another control action.

And so we disappear up our own tail-pipes...


'nuff said methinks

daba said:

Nope sorry, a "reference" is something you can "look-up". Most often fixed but can be changed.

Click to expand...

A reference is a source of information.

Which is why the OP thought it reasonable to use as the name for process variable and why i said (more then once) it doesn’t matter too much what you call them so long as it’s clear and consistent.

And note that’s different from what I was saying to drbitboy where I literally meant taking the process variable to the setpoint for the same control loop, either with scaling to force an error term for without scaling to force a zero error term (and in either case...the reference is coming from the process, whether you like it or not...).

Get out of your “tail-pipe.

jhenson29 said:

I understand that we agree to call the setpoint the reference and to call the process variable the feedback.

We just don’t seem to agree that they can the same thing in a positive feedback loop, which I only mentioned because I don’t agree it’s strictly excluded as you stated.

But I don’t know how else to explain it, so I’ll just leave it there.

Click to expand...

Sigh, I'll try again (without getting my nose out of joint).



#1 I already said I understood cascade, and I misspoke when I wrote "the process cannot provide a reference;" if that is the statement that generated this objection, then you are half a dozen posts out of sync.



#2 I think I understand what you are saying: e.g. two motors, MotorSpeedRef0 is supplied by an external, operator supplied setpoint, and the VFD returns a signal MotorSpeedPV0 that is the measured speed of the motor. MotorSpeedPV0 could be cascaded* as the reference MotorSpeedRef1 for the second motor, perhaps scaled by +/- a few percent with the intent that the second motor speed follows the first and torque-limited to provide tension control on the process between the motors. Yes, in that case the "feedback" signal becomes a "reference," but only downstream of where it is considered to enter the second motor's VFD, and not before.


#2.1 But that is not at all what OP was asking about.



* Yes, I am aware this is not a canonical use of "cascade,"

If I have an analog output that tells a VFD how fast it should be running, I refer to that analog output as a speed reference.

If the speed reference value is derived from the output of a PID controller, I'd refer to it as the CV, or control variable.

Properly naming I/O points is a very important subject that is often overlooked, especially by newbies.

This discussion about analog signal naming is similar to the age-old topic on discrete signals where the name reflects the condition in the energized position.

Relative to this analog discussion, I prefer “speed command to VFD” and “speed feedback from VFD”. Most descriptor fields permit many characters so don’t be too lazy to use them.

Remember, many systems have design & requirement documents, sequence of operations, etc. that will be created. Others referring to those documents in the future need to have clear descriptions for clarity.