Zero/span "hypothetical" question

[shiftman]

A level transmitter (0-100 in/h20 = 4-20mAdc) is bench cal'ed and installed on a liquid tank. Later, the transmitter is essentially re-ranged whereby 0-100 in/h20 is now equal to 20-4 mAdc output. Given the inversion, would you now also invert your calibration use of the zero and span adjustments as well?, or would you calibrate this new configuration using the old zero/span conventions? To be honest, I can actually see doing it both ways.

From my limited research, it appears that you can find multiple definitions for what zero and span actually mean. It appears that the same holds true for definitions of "suppression", "elevation" and "wet-leg" as well.

At any rate, I enjoy "hypotheticals" - as they get you into thinking mode, and I appreciate any and all responses.

 

[xnuke]

There is an ISA standard for process instrumentation terminology - ISA 51.1. It should have standard definitions for each of the terms you mentioned.

xnuke
"Live and act within the limit of your knowledge and keep expanding it to the limit of your life." Ayn Rand, Atlas Shrugged.
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[shiftman]

Yes, I have read that one - along with several others. How would you go about using the zero and span adjustments in the scenario listed above?

 

[waross]

Leave the calibration as it is and switch the controller to reverse acting.
Or-
Interchange the impulse lines.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[shiftman]

This is a dp level transmitter for indication only. No controlling action in this scenario. Additionally, the swapping of the sense lines wouldn't be permitted under configuration control.

Therefore, at some point, a future calibration would be required and adjustments to the zero and span might need to be made. In such a case, would you then use the zero adjust to correspond to minimum output or minimum Delta P?

 

[IRstuff]

You've not stated how "the transmitter is essentially re-ranged whereby 0-100 in/h20 is now equal to 20-4 mAdc output" Where is that done, and how? If it's purely a digital process, then the analog calibration should remain valid. If this is performed via some analog process, then the unit needs to be recalibrated.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[shiftman]

For the sake of discussion, let's assume that it is an older analogue (electronic and not pneumatic) transmitter that is now due for a routine calibration check.

After a nine point check, the instrument needs adjustment. With both hi-side and lo-side legs equalized (zero delta p), the transmitter output is slightly above the acceptance criteria for 20mAdc. To bring the reading within tolerance, would you adjust the zero pot or the span pot?

 

[waross]

This is instrument specific in a couple of ways. It could be either or both.
1> Do you adjust range or zero? That depends on the interaction between the range and zero adjustments.
It may also depend on the zero setting. A one-point-error may be either range or zero, even with no interaction.
The instrument may not be capable of measuring a negative differential pressure. You must have an instrument that will accept a range setting of - 100 in/hg to + 100 in/hg.
You have stated the range as 0-100 in/hg. I doubt that you can accomplish what you want with that instrument.
If you indicating instrument is a newer electronic device you may be able to program the results you want into the indicating instrument.
Hypothetically of course.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[IRstuff]

I agree it could be either or both:

http://iamechatronics.com/notes/gen...zero-and-span-adjustments-analog-transmitters

The zero changes the offset, and the span changes the gain. Depending on your specific error curve, you may need to change both. The objective would be to add error to the zero-error points so that the points that are too far out of bed are brought back in. Assuming a simple, say, concave down nonlinear transfer curve, shifting the zero would bring the hump closer to the ideal, but the endpoints would no longer be ideal. Your calibration person should be able to handle this, since it's their job to bring all points within tolerance.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[shiftman]

With a range in inches of water (in/h20) it should be possible but I cannot say for sure.

I can see a possible issue where a technician accustomed to a 0-100 (in/h20) /4-20madc output transmitter would be unsure if a transmitter calibrated for 0-100 (in/h20)/ 20-4 mAdc output would require a zero or span adjustment if the 20mAdc reading was out of tolerance.

After all, in one scenario, the 20mAdc signal represents the max delta P reading and in the other, it represents the minimum delta P reading.

 

[IRstuff]

I think you are over-analyzing. Unless your calibration person is simply inept, or too green, there should be no reason to not expect a complete calibration. Given that are only degrees of freedom, and they are well-defined, even if we don't know what the definitions are, and they amount to a gain and offset of some sort, so there's only so much that can be done to calibrate, and there cannot be so large a change that would muck things up, since you still have to meet calibration requirements.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[shiftman]

Not all calibration procedures tell you exactly what to do step by step. In fact, many utilize "skill of craft" in order to perform calibrations of instruments on non-safety related systems. As such, it isn't necessarily a matter of being "inept" but possibly "unfamiliar" in this instance.

If I calibrates a dry-leg delta P level transmitter with a 0-100 in/h2o range and a 4-20 mAdc output and both legs vented to atmosphere (zero differential), I adjust the zero to 4 mADC and the span to 20 mAdc. Pretty basic stuff.

If I encounter an obscure wet-leg delta P level transmitter, with a 0-100/h20 range and a 20-4 mAdc output, and I have zero differential and 21.3 mAdc output, would I then use the span or zero to set the 20mAdc to its acceptable output value? I know that I will go back and forth with both adjustments to bring both the 4 and 20 mAdc values into spec, but which adjustment do I use for which output?

That is really all that there is to this question............

 

[waross]

THIS SETTING IS INSTRUMENT SPECIFIC AND NOT ALL INSTRUMENTS ARE CAPABLE OF THIS SETTING. Some may be capable only IF THE DP CELL IS PHYSICALLY REVERSED.
Or a standard instrument may be installed with the impulse lines reversed and the zero depressed by the amount of the range.
What you want to accomplish is fairly straight forward with any standard instrument. The impulse lines are simply reversed and then calibration is straight forward.
If you want to accomplish your wishes with calibration settings only, then you must use an instrument that is capable of measuring a NEGATIVE RANGE.
If the instrument is capable of a negative range setting then you are probably turning the range adjustment the wrong way. When you get close to zero keep turning the adjustment past zero. If you can't do this on your specific instrument then you can't.

If I calibrates a dry-leg delta P level transmitter with a 0-100 in/h2o range and a 4-20 mAdc output and both legs vented to atmosphere (zero differential), I adjust the zero to 4 mADC and the span to 20 mAdc. Pretty basic stuff.

Yes trying to calibrate with both legs with vented to atmosphere sounds like the basic definition of inept.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[shiftman]

A delta P transmitter is a delta P transmitter whether both legs are filled with air, vapor or some type of liquid (A.K.A. dry leg vs. wet leg). After all, it is only measuring differential.

That said, calibrating a DP with both legs at atmosphere (or the lo-side at atmosphere and the hi-side at 0.00 psig (with the pressure source vented to atmosphere)for a 4 mAdc output is quite common in industry. In fact, it is standard in many industries.

Additionally, the instrument does not have to be capable of reading or outputting in the negative. In fact, it used to be quite common in the chemical industry to set up (calibrate) old analog wet leg level transmitters (that were not capable of reading a negative value) on a work-bench "dry", install the instrument in the system, valve the low side to the reference leg, shift the zero to read 4.0 mAdc (elevate or suppress-depending on which definition convention you subscribe to) and return to service.

That said, smart transmitters can be done differentially, but old wet-leg analog transmitters that are installed in systems where the liquids are either contaminated or radio-active are rarely removed and bench-cal'ed. These are usually done in-situ with a "water box" or some other variation of a head chamber device.

And yes, trying to calibrate beyond the zero dp output with BOTH legs vented to atmosphere would not work. But since that was never the premise of the discussion, it is not relevant.

The whole premise of the question is quite simple. If you invert the output of the transmitter, do you invert the function of the zero and span adjustments as well? I will certainly try sometime if the opportunity presents itself................

 

[roydm]

Looking at the transmitter as a simple op amp the zero pot sets the offset and the span pot sets the gain

Therefore if you invert the range from 4 - 40 to 20 - 4 the zero pot still sets the offset and the span pot still sets the gain

 

[roydm]

Actually this scenario is not that obscure, it happens often on a hot tank where vapour will condense to fill the upper connection (wet leg)
On your hypothetical 100 inch tank the DP with the tank empty is 100 inches, with the tank full it is 0 inches.
If I remember correctly you use the zero screw to adjust the full tank, Span for empty, forget about level, the transmitter is measuring DP

Actually it gets even trickier because the density of the cold liquid in the wet leg may be different than the density of the liquid in the tank.

Many of the old analog transmitters were not capable of measuring a negative DP, in the case of a Foxboro singing wire type the wet leg would be
connected to the HP connection.

Roy

 

[Parchie]

Perhaps a simple amplifier is all you need. A 20ma representative signal (high) will be inverted to a "low" one!