Hello folks and merry Christmas.

I don't know where to put this, but it sure involves a few MCU's.

I am having a strange problem on this ship I am on now. It's the level sounding of a tank, it jumps up and down from 0% to 45% at a few seconds interval on my integrated alarm system (IAS)

There are three sensors, two at the bottom to measure the static pressure of the fluid in the tank, and one in the top to measure the atmospheric pressure. I believe the upper one is not in the calculations for fluid level.

I get a good level indication from both bottom sensors, steady at 1,2 meters. But the tank animation jumps from 0% to 45'ish%. As far as I can tell, it must be a software issue, because I've tried with both steady sensors, but the level keeps jumping up and down.

It is difficult to give information on this, because it is all proprietary. It is built up with a pressure tank sensor, going into an Phoenix Contact I/0 block, and passed on to a Phoenix Contact industraial switch, and then up to the main IAS computer. All hardware seems fine.

The strange thing is, that this is the only one who does it, in the entire I/O cabinet.

Maybe a difficult question, since I can't provide more info than this. But maybe someone has some fault finding tips. I'm running low.

 

Divide and conquer.

Assuming you have other tanks and sensors- swap the sensor inputs and see if the problem persists.
Then you know where to start looking.

 

Good idea. I tried to change between different Software ID, but I only got a different up and down value. I will try a physical switch.

 

For some time I worked in preventive maintenance of wave piercing fast ferries built in a shipyard located in Tasmania.

They were fitted with 2 high RPM Caterpillar engines. At full speed, they neared kind of 90 Km/h IIRC. Control system was RACAL ISIS with monitoring terminals on the bridge plus two local more in the ER.

Since vibration is terrible, the norm was to retighten all terminals ( sensors and channels) every two days.

The very first check using some chanels kept free on purpose in each panel, was swapping the problematic with one free.

Swapping sensors was the next action.

 

Yes, usually we swap sensors, but these sensors are in a coffer dam filled with water.

But as far as I can see, the sensor is giving a steady reading of 1,2 meters/1,3 bars, pendeling only a little bit. So it seems ok.