OT - 0 or 1 Status if a shaft is rotating

[MikeW]

I have an application to determine if a pump shaft is turning at least 30 Hz which will be used as feedback that the pump is actually turning. There is a flow switch that confirms flow but, due to the existing valving, it may not register if part of the flow is diverted.

My first thought would be to find a proximeter/ photo detector that could be triggered once or more each shaft rotation and then put this into a stand alone counter that provides a relay output if it's count is above the desired setpoint.

Any other ideas/ suggested hardware that you have used in the past?

 

[Rube]

If the motor is direct-coupled to the pump you can use a current transducer to indicate whether the motor is rotating (or pump, depending on pump type) or move the flow switch upstream of the diversion to indicate output from the pump, no matter where the flow is going.

 

[bernie_carlton]

Crouzet has speed sensing relays. Just hook up an appropriate sensor.

 

[TConnolly]

If the motor is direct-coupled to the pump you can use a current transducer to indicate whether the motor is rotating (or pump, depending on pump type) or move the flow switch upstream of the diversion to indicate output from the pump, no matter where the flow is going.

http://www.loadcontrols.com

One of these unit will tell you if the motor is actually running and precisely how much load is on it. It has an adjustable switch point and a relay output, or an analog output. It can detect if the coupling is broken - or if the pump in running dry, or even measure the viscosity of the fluid being pumped - all from measuring the motor power instead of current.

It may even be able to differentiate if the diverter valve is open or closed due to differences in the load.

 

[Hydro Power Guy]

Depends on the accuracy that is needed, but I have used tach tape with speed sensing probes and controller such as Airpax. Depending on the speed and your plc equipment, you might not need an independant module such as the tachpak 3, you could bring the signal into a DI or a high speed counter card. If you need a high degree of accuracy, then you will need a machined target.

 

[MikeW]

Thanks for the prompt replies

Rube, the current transducer has limited use as it would tell the motor's turning but may not sense the pump's response accurately.

bernie, the Crouzet on the url only goes to 230V, I have a 460V inverter, my bad for not giving enough info on the front end. Looks like a good unit, I'm filing it away for future reference, din't check to see if they make a 460V unit.

Alaric, I have seen these before but have never used them, they look promising for this and especially another application that is on the horizon.

 

[Tom Jenkins]

I've used prox or photo switches on the pump shaft and pulse to analog convertors from Red Lion for similar aplications in the past. If you have a PLC with an anlog input in the system you can then measure the actual speed.

If you want a stand alone system Red Lion has zero speed switches that would work. You could also use a pressue switch if you have a check valve after the pump that doesn't open until flow occurs. A suction switch on the inlet is another option.

Remember that a pump working against a static head can rotate backwards if you don't have a backstop or good check valve. Something to factor into your thinking.

 

[rsdoran]

The simplest thing I can think of is add another flow monitor or sensor on the pump output, that will tell you its pumping (and possibly what rate if necessary) regardless of any diverters down the line.

If you want to monitor the motor shaft a mag pickup is a common method, this can be used to determine speed of the shaft etc.

If the motor has a VFD it should be able to supply frequency, current information and more. How this information is used can vary depending on circumstances.

 

[leitmotif]

I think the first and maybe key issue is what does the pump do?
Does it pump water or ?? to keep things cool or warm, or pump water to a tank for use in a process, sewage, hydraulic pump, etc, etc??

Seems to me to install instrumentation to ensure the pump task is being fulfilled. The flow monitor can be useed to do 3 things
1. Confirm pump is running
2. Act as feedback to VFD for speed control
3. record water or whatever consumption

IF you have a VFD why do you have a bypass valve? A concept of VFD on pumps is to reduce power consumption by eliminating bypass (and wasted power) and pump only what is needed to do the job as it is needed. Half the speed and reduce power by a factor of 8. I admit this cannot be done in all cases.

Dan Bentler

 

[Bruce99]

Flow

Tom is right for the best and cost effective method. I use a prox on the shaft to detect either a key or a welded nut or bolt to the shaft. This tells us it turns, not if it is pumping. I use current transducers for this. Cavitation can be detected by a low current and bearing trouble can be tracked also. A good month of operation can produce a trend that can be used for PM maitenance. If the motor is changed run the motor for a period of time to get a good sample of no load current. Once coupled, trend its current and build a history. In large pumps I use vibration sensors on the motor and the pump. Once a trend is made, problems can be caught before they get out of hand.

 

[MikeW]

Life is not simple in the test lab ...

Since the original posting, some more detective work has occurred. I'm helping a new engineer "learn the ropes" of applying controls to various systems so some of my info was incomplete as he was doing the footwork.

The ~15 year old VFD drive does indeed have an optional output board that was installed sometime after the initial installation and the drawings were never updated. (Change control was poor around here until recently). Thus, I can determine that the motor has been told to start and whether it thinks it's running. Additionally, there is a plugged 1/2" NPT tap that is on a straight section of the discharge line that was very hard to see until you got up into the 8" pipes. With this available, I'm leaning to get a flow switch (an ifm efector type) installed as the simplest "flow/ no flow" indicator. Not sure why there was not one already installed at that location. There will be no "nuclear meltdown" if the pump fails but a large block of ice can be created that will ruin a large process chiller if evasive action is not taken.

The application is basically a process cooling and heating application with water or a water/ glycol mixture as the transport fluid depending on the test de jour, no two days are the same around here. Since it's a "new day everyday", control tuning is always a joy since one day you will be running at 5% capacity at 35F then 95% at 85F two hours later and then "can you dynamically change the conditions to go for 10% to 85% in 30 minutes and repeat this as a step/ ramp/ or sine wave response?" It's a great challenge that should keep me employed for as long as I want. Bypass valves are used to allow the operator to change the loop configuration depending on what method is used to heat/ cool the fluid. There are way too many ways this is done to go into here unless I'm going to do a lot of typing and furnish some piping schematics.

We check vibration at least once a year to check for overall mechanical system health.

Thanks for all the responses and chalk up another problem solved by the PLCS crew.