Large Pump/Electric Motor Control: Best Practices

[ryangriggs]

I'm starting on a project that will soon involve controlling two large water pumps (4160 VAC, not sure the HP).

I'm curious: what generic "best practices" have you learned in regard to controlling large motors/pumps using PLCs? Any advice is appreciated - doesn't need to be application-specific.

For example, I know it's a good idea to implement timers to prevent fast state changes (i.e. wait X seconds before stopping or starting after the Call To Run changes state).

Also, alternate between pumps to distribute wear.

What tips, tricks, do's and don'ts have you learned along your way? Anything is appreciated which would help create a better/safer/more robust control algorithm.

 

[jraef]

Being that they are pumps, are you going to be using VFDs? If so, implementing PID control loop is the optimum method.

 

[ryangriggs]

No, these are not connected to VFDs unfortunately. They are part of an older installation (c. 1985) and we're working on upgrading the controls to modern PLCs.

I'm looking for generic advice about controlling motors/pumps in general. I.e. What to do, what not to do, to keep users safe and make the pumps run happily. Your own experiences/stories/tips/tricks etc would be helpful.


Thanks!

Being that they are pumps, are you going to be using VFDs? If so, implementing PID control loop is the optimum method.

 

[Tom Jenkins]

The best practices depend on type of pump to some extent. For example, vertical turbine vs. end suction. In general consider bearing temperature monitoring for pump and motor, vibration monitoring for case and possibly bearings, suction and discharge pressure, motor amps or kW.

Talk to a couple of pump suppliers. They will undoubtedly have recommendations and will have to mount much of the instrumentation.

 

[ryangriggs]

Thanks Tom.

The best practices depend on type of pump to some extent. For example, vertical turbine vs. end suction. In general consider bearing temperature monitoring for pump and motor, vibration monitoring for case and possibly bearings, suction and discharge pressure, motor amps or kW.

Talk to a couple of pump suppliers. They will undoubtedly have recommendations and will have to mount much of the instrumentation.

 

[Saffa]

Alternating equipment to prevent excessive wear = good.

Alternating exactly equally between two pumps = not so good according to what I was taught in Reliability Centred Maintenance.

Equal wear can give reduced system reliability if failures are linked to operating hours, e.g. bearings and seals. You have a higher likelihood of both pumps having problems within a small time period. Using a 60/40 ratio gives you a standby unit which is less worn.

This also helps spread your maintenance / asset renewal budget.

I personally like Multiple Start Prevention Timers for big motors. If it's run within the last X minutes, block a further start until that time has expired. The motor supplier should be able to provide maximum starts per hour. Make sure to display on the HMI or SCADA that the pump is being inhibited by this protection to prevent the operator thinking "That %€£&*ing PLC is on the blink again".

 

[ryangriggs]

Thanks Saffa. This is exactly the type of info I'm looking for.

I definitely see your point about 60/40 runtime preventing both pumps wearing out at the same time. I had not thought of that.

I think our particular motors are allowed 2 starts per hour. But not sure if the motor controller enforces that limit or if the PLC will need to do it. I'll have to find out.

Why is it bad to start the motors too frequently? I understand that you wouldn't want to cycle power rapidly, but why only 2 starts per hour? Is the starting process extremely hard on the motor?

I also like your idea of showing that the start is inhibited in the HMI. Great idea: not needing to wonder why the motor's not starting - sounds like you've been there before haha.

Again thanks!

 

[Saffa]

Why is it bad to start the motors too frequently? I understand that you wouldn't want to cycle power rapidly, but why only 2 starts per hour? Is the starting process extremely hard on the motor?

I also like your idea of showing that the start is inhibited in the HMI. Great idea: not needing to wonder why the motor's not starting - sounds like you've been there before haha.

Again thanks!

DOL starting of motors causes currents of 6-8 times normal running current. That results in rapid heating and mechanical stresses of the motor windings. Doing that too often and the windings can be damaged.

And yes... at one place I worked they had the MSPT feature but no indication. At least one call a week for "I just ran this pump and now it won't go again!"

 

[ryangriggs]

Got it.

DOL starting of motors causes currents of 6-8 times normal running current. That results in rapid heating and mechanical stresses of the motor windings. Doing that too often and the windings can be damaged.

LOL I know the feeling.

And yes... at one place I worked they had the MSPT feature but no indication. At least one call a week for "I just ran this pump and now it won't go again!"

 

[jraef]

As part of the starts per hour issue is a similar control strategy called a "back spin timer" that disallows even the 2nd start in an hour if the motor is still spinning when you re-close the run command.

As Tom Jenkins said, this may depend on the pump and system involved and it may or may not be part of any pump motor protection relay already assigned to it, so implementing these in the PLC may end up with redundancies that cause frustration. You need to vet this with the pump supplier. In many cases, all they will want from the PLC is a Run command in the Auto position of an HOA switch. But one thing I always like to implement in the PLC is to have an input from BOTH the Hand and Auto position so that operators know in the HMI if someone is running it in hand. That can be critical information. I've seen people insist on knowing if it is in the Off position, but you can do that logically in the PLC; if not A or B must be C...

 

[ryangriggs]

Good Point jraef. I will try to get specs on the pumps and controls.

The valve/pump control panel has a HOA switch, but I spoke with the mfr who says the HOA switch doesn't have dedicated feedback connections for a PLC to monitor.

I hoped to at least monitor this in the PLC as you suggested, so I can show the position of the switch to operators. Hopefully will be able to tap existing pilot circuits to the switch for some sort of feedback.

As part of the starts per hour issue is a similar control strategy called a "back spin timer" that disallows even the 2nd start in an hour if the motor is still spinning when you re-close the run command.

As Tom Jenkins said, this may depend on the pump and system involved and it may or may not be part of any pump motor protection relay already assigned to it, so implementing these in the PLC may end up with redundancies that cause frustration. You need to vet this with the pump supplier. In many cases, all they will want from the PLC is a Run command in the Auto position of an HOA switch. But one thing I always like to implement in the PLC is to have an input from BOTH the Hand and Auto position so that operators know in the HMI if someone is running it in hand. That can be critical information. I've seen people insist on knowing if it is in the Off position, but you can do that logically in the PLC; if not A or B must be C...

 

[James Mcquade]

GET familiar and know NFPA70E - arc flash by heart !!!!
GET familiar and know NFPA70E - arc flash by heart !!!!
GET familiar and know NFPA70E - arc flash by heart !!!!
Know the flash boundaries and barrier limits and have them in place !
Have a watch in place when working on the equipment.

YOU MUST have the correct arc flash gear, meters and tooling if you work on the system while energized.

DO NOT, I Repeat, DO not have or allow any standard tools or electrical meters
of any kind in the room or area when the equipment is energized !!

I cannot stress this enough !!

I saw the result of trying to use the wrong volt meter on a 4160 volt circuit!
this is not a joke and I will not comment on it any further!
Please heed my warning and be safe.
I don't care how hot the Cat 4 or higher suit and equipment is, keep it on.

regards,
james

 

[James Mcquade]

Everyone,

I do apologize if my previous offends you.
After thinking about it, what I wrote can be considered extreme,
but after going to the plant with the 4160 volt arc flash and seeing the damage, you'll understand. I would not wish that on anyone.

james

 

[ryangriggs]

Hi James, thank you very much for the words of warning. No offense taken. I will definitely study the NFPA standards.

It's not "extreme" if you're speaking from experience.

As I told a friend, I have no intention of getting anywhere near the high voltage systems without ensuring I have the proper training and PPE. 120VAC single phase is enough for me!
Again thanks for your advice.

Everyone,

I do apologize if my previous offends you.
After thinking about it, what I wrote can be considered extreme,
but after going to the plant with the 4160 volt arc flash and seeing the damage, you'll understand. I would not wish that on anyone.

james

 

[James Mcquade]

ryangriggs ,

not trying to start an argument, but 120 volts is more dangerous than 220 or 440.

a 120 volt circuit will cause your muscle to contract and you cannot release the circuit.

I don't know your background, but get an electrical engineer involved with your design !!

you might also google arc flash videos and that will give you an idea of what I saw.

james.