Pressure Transducer reporting a negative value 1

[CuriousElectron]

Hello Gents,
I'm looking at a trend in PI for a water level value and the value is fluctuating below 0 feet and above 0 feet. To me that is strange, I'd expect the trend to be above 0 feet. What can I infer from this trend? When the sensor is reading & reporting a negative value, is that indicative of the sensor issue or perhaps a comms issue?
Any insight on this would be helpful.
Thank you,
EE

 

[IRstuff]

When the sensor is reading & reporting a negative value, is that indicative of the sensor issue or perhaps a comms issue?

Only if the comms were so poorly designed that missing or corrupt data didn't get flagged as bad data. Also, given that comms problems are ostensibly random, it would be highly coincidental that your readings are only slightly negative, as opposed to hugely positive or negative.

What does the actual data look like?

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[GregLocock]

PI, presumably not being 3, is what?

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[danw2]

A couple common causes of dither: a calibration issue or electrical noise.

Calibration
Whatever measures the water level is probably some instrument with an analog 4-20mA output. If the transmitter's output is out of calibration slightly, the output signal might be 3.99mA or 3.98mA at zero level. The receiver device correctly interprets anything below 4.00mA as a negative value. Or the receiver might be mis-calibrated so that it interprets 4.00mA as 3.98mA and then reports a negative value.

Electrical noise
The signal wiring can pick up electrical noise which creates a signal that dithers around the nominal value.

Communication issue
The field device that's doing the measurement can communicate to the 'system' by either an analog signal (4-20mA, 0-10V) or over a digital bus, like Profibus, Foundation Fieldbus, Modbus, or a wireless protocol. A digital number communicated over a digital bus can be relied upon to not be corrupted by noise, because digital communication protocols as a rule discard any data with corrupted bit(s).

But the reported digital value is only as good as the analog sensing behind it, and all process measurements start at the analog level. There are a number wireless sensors that cannot be field calibrated, whereas their analog counterparts can be. The 'wireless' bandwagon let all sorts very low end, marginal sensors into the field because of the rush to go wireless and the IIoT (industrial Internet of Things) bandwagon is feeding the cheap-but-wireless-sensor approach.

But in the end, does dither around zero level really matter?