Descrepancy Between Measuring Ranges of Process and Control System

Whatever the reading other than 0.1m, the reading will be out by about 2¹/₂%.

Given the restrictions on the programmed system, the simplest things to do are either:

• to re-range the instrument so that it transmits over a range 0.1-4m, or
• to adjust the zero back to where it was and live with the 0.1m loss-of-measuring-range at the bottom end.

Remember to make the as-built documentation match the installed equipment, whichever option is chosen.

There are two things you need to be concerned about - Safety and Cost - and your concern should follow in that order.

Does this error does lead to an increase in harm to people or the environment?

If so, you should make the client aware of the increase in harm (in writing) which should compel them to stop and make the necessary changes. If the harm is of a serious nature (e.g. death or severe injury is likely if changes are not made) and the client refuses to make changes, you may be required by law to report this issue to the appropriate governing body to push for resolution against the wishes of the client.

Does this error lead to an increase in cost (e.g. operating expense or life cycle cost due to a decrease in mean time between failure of some or all of the system components)?

If it does, you should prepare a report or provide an analysis, to the best of your ability, detailing the expected increase in cost and present the report or analysis to the client. The client can then evaluate the cost and make the appropriate decision about what to do. If they elect to do nothing, it is their perogative to do so and you can rest easy knowing that you have made them aware of the issue and that it is their decision to incur the increase cost.

If the discrepancy just makes you feel bad but doesn't cause harm or lead to an increase in cost, I would say relax and make sure you avoid this mistake in the future.

Thanks a lot for your replies. So basically the modification is to be carried out on a pumping station of a power plant. The sensors are non-classified, safety-wise. The only risk is flooding of the pumping station when the water overflows the sump pit (That would take some time though). The control system is designed to trip off the sump pumps when the 3.5m level is reached to avoid overflow anyway. The only thing is that from my calculations, instead of the control system seeing 3.5m for example, it reads 3.415m.

There appears to be enough margin between 3.5m (max level threshold) and 4m so I guess even it would take 3.6m physical level to trip the pumps at control system level (since the control system would see 3.5125m).

I came up with the following using excel spreadsheet:

>I have got a 4-20mA level sensor with a measuring range of 0-4m.

Exactly what brand, model and technology level instrument does not allow you to set the LRV (lower range value:zero) and URV (upper range value: span) to whatever you want it to be, so that 4.0mA = 0.1m, 20.0mA = 4.0m?

Hello Iris, Thanks for your reply. I actually thought of the same thing. Since the sensor is calibrated to measure 0-4m, I thought that with the offset, we had to respect the 4m range by calibrating as 0.1 -4.1.

The instrument is a VEGAWELL 52. I would look up the specs again.

What's the problem with the way things are now.

The sump won't ever over flow and do you really care if there's some residual water in the bottom?

This looks like a non-problem to me, unless I'm missing something, again.

Thanks lyn,

Like you rightly pointed out, the sump would not overflow. However, there is going to be a descrepancy between the control system level reading and the physical process level reading, meaning that the threshold levels that actually trigger the sump pumps would be different from specification. So what I basically did was find out the corresponding mA reading for a sump level and then used the same mA reading to calculate what the control system would read. That is what I have got in the excel screen above. Is my logic right?

Thanks a lot.

Wow, thanks a lot for your analysis Iris - This forum is wonderful. Speechless.

The VEGAWELL sensor is indeed submersible and non-adjustable. Basically from all inputs I have received, one thing is sure. The sump would not overflow because we have sufficient margin. The system is also not a safety-classified system.

Thanks a million.

I'm negligent in my social graces, I haven't welcomed you, divine, to the forum. Welcome.

As a first time poster, you should be aware that you are an unusal participant because you actively answer questions and dialogue with responders.

That is, frankly, atypical. Most OPs (original posters) never answer questions or even reply after their initial post.

Hope to hear from you again.

Iris

Iris, my head is spinning, I think. And I am not unfamiliar with control systems at all.

Here is a supposedly linear measuring gizmo. It is mounted at whatever height.

The "bottom" reading is the Bottom Set Point at X mA (at whatever liquid level).

The "top" reading is the Top Set Point at Y mA (to prevent overflow, or whatever).

So, X mA turns on the pump. Y mA turns it off.

The actual mounting of the sensor ought to be irrelevant, as in the controls we react to X and Y only.

Am I missing something here? Thanks.

You must have forgotten 'elevation' or 'suppression' of the zero point of the span?

Where there is a continuous output from the measuring device, you don't dismiss the actual measuring range in favour of a key set-point marker/ value. In such case, a simple float switch could have been deployed to start/ stop the pump.

Graduated measurement outcomes from elevating the zero, is what Iris has so nicely set out for the OP.

Ooops!...you addressed to Iris personally ? Apologies...

Hello Everyone,

Thanks for the inputs. They have given me more arsenal to resolve this issue. Just to summarize, when my 4-20mA (0.1 -4.1 range) level sensor sees 3.5m, it corresponds to 17.6mA. For 17.6mA however, the control system (0.1m -4m range)would see 3.415m.

I would do a report on the impact - taking cost, safety etc into consideration.

Highly encouraged by the responses.

Thanks a million.

Divine.

That looks like a lot of work.

I's a lot less work to range the instrument appropriately.

_{Welcome to CR4, BTW, and the responses to other readers inputs are valuable in helping the forum answer questions satisfactorily. Keep it up!}

I fully agree with you PWSlack

However the intrument ranges comes as '0-4m' from VEGA and there are no funds to update the control system.

Regards.

Starting with Divine's note #15 I start to get it. But, I admit I have to suspend
disbelief to get there.

1,. The 4m+ long analogue, current sending sensor comes factory calibrated. That is fine.
One does not monkey with sensors. Maintaining factory calibration is essential.

2,. The (replacement) sensor cannot be mounted exactly to correspond to factory caibration.
Common occurrence, ought not be a problem.

3,. It drives an analog meter (I assume, and ok), whose scale is not moveable to take the
mounting offset into account. Annoying, but there are services to print new scale in a
few days, inexpensively.

4,. It is time to set up the controls, to test the setup, and make it to work to the
sponsor's satisfaction. Document, pick up pay, and drive home happily.

5,. But wait! The sponsor refuses 4,. What?!? Discuss, document, pick up pay and go home.

6,. Never mind now, why it happened. Controller is an antique, nobody has the manuals,
sponsor fears to touch the controls. Now we are really down in the weeds. The chart did
show, that bleeding off 0,1mA at 20mA in the loop brings everything into reasonable
aignment. You may offer that as a solution. But, let the sponsor contract that out.
You do not need the liability coming with it.


It does not matter, where the sponsor stopped you in completion of your task.
HE STOPPED YOU. You, being a reasonable person, may plead with him. But, when refusing,
it is his responsibility from then on. Document well, try to help him, but it is his
responsibility from then on. You were hired, you cannot force a sponsor to be reasonable.
He may be left with a paper chart to relate readings to reality. Tough!

We were all looking to the technical side, when the trouble was on the human side.
Which frequently is less tractable.

Hi Iris, Yeah. You got the gist fully now. Excuses if I was not clear enough from the start. As I cannot modify the control system's upper limit to 4.1, I would have to look at tweaking the setpoins. Thanks.

I do not understand how a probe/sensor made for a 0.0m to 4m depth can be too short for a 4m tank??

If the probe measures up to 4m but the probe does not reach to the bottom of the 4m tank does that mean the sensor is not suitable for measuring 0.0m to 4m but actually made for measuring 0.1m up to 4m ????

The transmitter's range is not 'too short', rather its output does not reflect actual level because the sensor does not go all the way to the bottom. There is a site restriction that the control system readings can not be re-configured to match the sensor conditions with the actual level. The transmitter has virtually no adjustment of its zero or span.

So a
4.0mA output representing 0.0m level occurs when the level is at 0.1m
8.0mA output representing 1.0m level occurs when the level is at 1.1m
12.0mA output representing 2.0m level occurs when the level is at 2.1m
20.0mA output will never occur because at maximum tank level, the sensor only 'sees' 3.9m level.

A head pressure sensor can not 'see' any level below its elevation, so it cannot 'see' the level between the bottom and its elevation at 0.1m.

Yes, the sensor is measuring the level from 0.1m to 4.0m, but it is representing that level as 0.0m to 3.9m

Thanks for helping out. This has taken 2 years but thanks a lot. Issue resolved and the measuring ranges were updated on the conrol system and instrument calibrated accordingly.