AT plant frequent tripping of CW water motors are observed (NOV to MARCH) from several years due to HIGH WINDING TEMP (Normal 85deg, trip 120degC). Three wire RTDs are installed at each winding. Local TI gauge always shows correct reading but same wire after passing through JB travels to control room (Barrier + PLC ). In case of trip PLC shows/logs a millisecond disconnection.

But some times temperature even decreases to 55degc. Atmosphere is quite humid in area as very close to cooling tower. How should i go step by step to solve this problem. How temperature decreases to 55degc?

Thanks

 

How many RTD's per motor?
How many motors show this winter time related high winding temperature trip?
How many motors don't?
Are these motors located outdoors or under roof? Do the motors see outdoor ambient temperatures?

What is your local climate? (northern/sourthern high latitude winter, equatorial? rain forest, dessert?)
Does this motor winding RTD connect to a temperature transmitter?

What is the local indication (local TI gauge)?
- A digital indicator that uses the same or a different temperature sensor than the RTD?

- A bimetallic thermometer?

- A digital display on an RTD transmitter?

- A loop powered digital indicator?

- A panel meter that retransmits the RTD input as 4-20mA?

Is 55°C a scale lower limit value?
What is the range of the temperature transmitter?

Does the 'millisecond disconnection' last only for one millisecond? longer?
Is this disconnection the reason for the trip action?

 

Adeel Imtiaz... several additional questions

1) Are motors mounted vertically?

2) What type of enclosure is provided?

3) Have IR (Insulation-Resistance) tests been done?

4) Is monitoring instrument equipped with open-circuit detector?

Regards, Phil Corso

 

>How many RTD's per motor?

3 RTD for each winding and 2 RTD for bearing Temperature (triping on winding)

>How many motors show this winter time

Total 4 motors are available two motors A&B have high frequency of RTD loop failure problems all motors have local outdoor JBs installed with separate multicore traveling to Control room

>related high winding temperature trip?

Well whenever RTD is disconnected the temperature goes to 150degC (Higher limit) and motor trips at 120degC

>How many motors don't?

All motors show this behavior but some are more frequent then others

>Are these motors located outdoors or under roof?

Motors are located outdoor adjacent to cooling tower with a small shed on it


> Do the motors see outdoor ambient temperatures?

in winter its less then 25degC and in summer its 50-52degC

> What is your local climate?

Usually climate is hot and dry but this problem starts mostly in winter season

>Does this motor winding RTD connect to a temperature transmitter?

RTD are directly terminated on barriers and barriers transmit 4-20MA further to PRASFE RS PLC input card

>What is the local indication (local TI gauge)?

> - A digital indicator that uses the same or a different temperature sensor than the RTD?

A simple Gauge mounted on bearing temperatures separately

>Is 55°C a scale lower limit value?

No tripping is actuated at lower temperature but it is unique observation that some time RTD value starts decreasing. So when we open local JB and reconnect cables in TB all problems are solved but it appears after one month again. Therefore we have started bypassing Local JBs TBs by using a sort of butt connectors and now we are monitoring the result.

> What is the range of the temperature transmitter?

non

> Does the 'millisecond disconnection' last only for one millisecond? longer?

Whenever PLC detects a millisecond break it trips the motor but usually it stays in milliseconds. Before entering in barrier there are two JBs and one interconnection cabinet so we have to check connections in all of these places.

>Is this disconnection the reason for the trip action?

Yes any disconnection of ms will cause tripping in Pro safe RS PLC

 

Adeel, some questions:

1. you said that when you get a trip you 'open local JB and reconnect cables'. Does this mean that the cables in the JB are found broken due to corrosion on the terminal blocks? Do you find water or moisture in the JBs?

2. are the junction boxes you are using suitable for the environment they are working in? My main concern is the high ambient temperature you indicated.

2. Are the junction boxes exposed to direct sunlight? Your high ambient temperature could also be associated with a very high UV index (Sun's ultraviolet index).

3. Are the JBs made of plastic or metal? If plastic then be careful as most plastics do not resist the high UV index at your site. At the end of summer season, your JBs would have been 'burnt' by the sunlight, rendering them no longer water tight.

3. Most probably a silly question, but are you using JBs having environmental protection of at least IP55, and do you have proper IP55 rated cable glands on the cables to the JBs? Do the cables enter the JBs from the top, bottom or side?

4. Did you ever try to spray the JBs terminal blocks with some non-conductive oil spray, to create a corrosion protective coating? I must admit not a neat solution, but sometimes such approaches solve problems. If you try this, make sure the oil you use is compatible with the JB internal components.

One final comment. Your decision to bypass JBs and connect the cables direct will solve most of the problems, especially if the joints are properly sealed by a heat shrinking oversheath, but will this affect future maintenance requirements? Normally JBs help in isolating regions during fault finding, and so eliminating them will render future fault finding more difficult.

 

To summarize, there appears to 2 problems, maybe related to one another:

- one is a lower than expected temperature reading in the winter time.

- the second is the intermittent trips caused by spurious high indicated temperatures.

To clarify, a device that converts RTD resistance to a 4-20mA signal is a transmitter. There might be an intrinsically safe barrier, too, but for that kind of service ? ? I doubt it.

1) Lower than expected temperature

The ways I can think of getting an indicated value to show abnormally low is

- actual cooling of the RTD element, perhaps due to water intrusion. Could condensed moisture be cooling the motor windings?

- water/moisture intrusion affecting the inner sheath RTD insulation. If the 'transition' of an RTD sheath is broken, water could intrude into the sheath.

H. M. Hashemian's book, "Sensor Performance and Reliability", pg 15, cites moisture in the inner sheath insulation "reduc[ing] the effective resistance of the RTD and caus[ing] the sensor to indicate lower than true temperature." His background in the nuke industry supports that statement.

- reduced resistance before the wire termination at the transmitter. Generally oxidation or poor wiring at junction boxes or terminal strips leads to increased resistance, not decreased resistance. Not likely.

- the sensing element is not an RTD, but is a thermistor, which would account for the abnormal low temperature increasing to normal when dirty, oxidized junction box connections are cleaned and retightened (resistance decreases indicating a higher temperature). Is the sensor really an RTD?

- the sensing element is really a thermocouple, not an RTD, and the 30°C offset from abnormal (55°C) to normal 85°C is due to malfunctioning cold junction compensation at transmitter. Multiple failures of the same nature on different devices? Low probability.

- the transmitter's analog input's constant current source does not hold steady, drifts down. How could this happen on multiple commercial devices? Overheating in the summer? Less likely in the winter. Not very probable.

2) intermittent trips
- momentary power drop out. I'd put the RTD transmitters' DC power supply on a battery backed UPS and see if there's still a trip. Power supplies account are about half of all field problems.

- water intrusion into the inner sheath insulation. Hashemian (pg 15) says, "moisture can cause noise at the RTD's output and/or result in erratic RTD output."

Intermittent opens and lower-than-expected indicated values suggests replacing a couple of the most offending RTD's and seeing if the problem doesn't disappear.

David

 

Adeel... While your investigation of the problem continues, I suggest the following plan to eliminate unwarranted shutdowns:

WINDING TEMP RTDs
Reconfigure the tripping logic from 1 of 3 (Non-Voting) to 2 of 3 (Majority Voting!) However, instead of setting all three (3), just set two (2) to trip at 120 Deg C! Set the 3rd one lower, say 110-115 Deg C. In this way, the 3rd one serves as a pre-trip Alarm!

BEARING TEMP RTDs
Are they configured to trip via the winding trip circuit?

Regards, Phil Corso

 

> To summarize, there appears to 2 problems, maybe related to one another:

> - one is a lower than expected temperature reading in the winter time.

> - the second is the intermittent trips caused by spurious high indicated temperatures.

<b>Moderator's note: rest of quoted text snipped by moderator. For complete text please see http://www.control.com/thread/1296883657#1297552259.</b>

We are also experiencing the same problem. Is this issue been resolved? How?

I am doubting our TBs and JBs which are outdoors. I have checked the IP rating and it is IP55. Is this enough for outdoor application? Usually in the summer we get around ambient 40 deg C temp and as high as 30% Humidity. During the winter season, we get around 18 deg C and humidity up to 80%

 

> Local TI gauge always shows correct reading but same wire after passing
> through JB travels to control room (Barrier + PLC ). In case of trip PLC
> shows/logs a millisecond disconnection.

This statement is confusing to me, to have a local TI.

Do you have an indicating transmitter at the motor for each RTD?
Is the signal going to the control room 4-20 mA or 3 wire RTD?
Are your cables twisted shielded?
Where are the shields grounded?
What purpose does the barrier serve, with the RTD in the motor winding it's certainly not intrinsic safe?

I have seen wet terminals make an RTD indicate low temperature, the leakage across is a resistance in parallel.

I would remove all the wires at the junction boxes and inspect for signs of corrosion, dip in petroleum jelly and replace.

If the cables are not shielded correctly perhaps they are picking up some inductance.

Why not just add a timer into the trip circuit, you say it only peaks for a millisecond, a genuine overheat trip wouldn't mind waiting a few seconds.

Roy