how to set PID control in VFD that operate a centrifuge pump transporting juice being concentrated 1

[systemsaver]

Dear engineers
I have a customer that is using a Siemens VFD manually to operate a centrifugal pump that transports juice with different Brix grade and the only reason he is using the VFD is for the start of the pump and is not getting a benefit of the affinity law to save energy running the pump.
Would someone help to set up a PID control for the VFD?
I would appreciate very much any help you can provide us
Thank you.

 

[3DDave]

Each different grade potentially needs a different speed for the pump; why can't the operator set that on the VFD controller manually? If not manually, how can the different grade materials be detected?

One needs to have some feedback for the PID to P, I, and D on.

 

[systemsaver]

Thank you 3DDave for your promtly reply.
When the juice has a higher-grade Brix the pump does more effort to transport it. The centrifuge pump is part of a concentration System. The idea I think, if I am not wrong, it is to work with a torque sensor in such a way that when the juice liquid is not heavy the pump speed can be reduced and therefore the electrical power required to transport it. How to set up the values in the PID is the question. Thank you for any help in this matter.

 

[jraef]

A PID loop needs to "follow" something called a "Process Variable" (or "PV"), maintain a "Setpoint" (or "SP"), and actuate a "Control Variable" ("CV"), while avoiding deviations, a.k.a. "errors", as much as possible. So what's you goal here? Flow rate that varies with viscosity? Pressure in a vessel? Level in a tank? You have to start with what your PV is as it relates to the SP. You cannot have "energy" as a PV, because the natural result of that would be for the PID controller to shut down since that would save the most energy! But energy savings is something that will the RESULT of using a PID controlled VFD on a centrifugal pump.

So let's say for now that your PV is to be a flow rate. You need a transducer to measure that and give you an analog signal value, that goes into your PID controller as the PV. Then you need to decide what value of flow rate that you want to maintain, that is your SP. The Speed of the VFD then becomes the CV, and will vary as necessary to maintain the SP, using the acceptable errors of the response rates, being PROPORTIONAL ("P") to the change in the PV, INTEGRAL ("I") to the rate of change in the PV, and DERIVATIVE ("D") of the number of changes in a given period of time.

A classic example I like to point out to people who think they have never used a PID loop, is that the "Cruise Control" in your car is in fact a PID loop controller! The speedometer is the PV, the speed setting button on your control button is establishing the SP, and the throttle servo that changes your engine power is the CV. In your car, you don't have access to change the P, I and D functions, that was done for you at the car factory, but they are in there, somewhere.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[edison123]

jeff, as usual, clear explanation.

Muthu

http://www.edison.co.in

 

[waross]

Are you lucky enough that the juice is electrially conductive?
If so a magnetic flow meter may be used to deliver at the same transport velocity regardless of changes in SG.


--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[LionelHutz]

Don't fall into the fallacy of believing a reduction of the power input into the motor means you're saving energy. Energy is work done and the work done should be considered as the product due to operating the motor. In this case, it could be considered the amount of juice concentrated in a period of time. If the volume of concentrated juice produced per hour goes down faster than the power drawn by the motor then the speed reduction actually costs energy vs saving energy.

If you're filling a tank, you can often show that running the motor and pump at say half speed will actually take longer than double the time since the pump efficiency drops at half speed. This means the efficiency of the system goes down and it requires more energy to fill the tank even though the power into the motor is reduced.

 

[systemsaver]

Jraef thank you for that straight clear explanation how to use a PID control for the VFD running a centrifugal pump,
"Energy savings is something that will the RESULT of using a PID controlled VFD on a centrifugal pump" That is clear because
a centrifugal pump follows the affinity law, the relationship between, flow, power and pressure.
This pump works in a closed loop with a juice concentrator unit which evaporates water from the juice till it has a 50 grade Brix where the viscosity is higher than the beginning. This means less power is required at start to pump the juice.
What type of transducer should be used in this application?
Something else.
A second pump discharge the concentrated 50 grade Brix juice but in such a way that the new not concentrated juice does not lower too much the Brix that is in the closed loop with the concentrator.
Sorry if I am not making myself clear. Next week I am going to the customer to see this particular case. If you have experience with this type of application, I would be happy to get some tips about how to save energy automatically.
Waroos, thank you for your comments.

LioneHutz,
Good point. Thank you.

 

[LionelHutz]

LOL, assuming an automatic energy savings because the affinity laws told you so is a path to failure. It'd be nice if it was that easy.

I hate to tell you, but VFDs use PID loops for process control, not for energy savings. You do it because you need a certain flow rate, or a certain volume in a tank, or for some other reason. The required flow rate or tank volume dictates the VFD speed, not the amount of energy being saved. Sometimes, the slower pump speed saves energy, but other times it does not.

 

[jraef]

A better way to look at it is that IF you were going to use PID control on a throttling valve as a way of attaining flow control, the VFD will save energy COMPARED TO the use of the throttling valve. But the use of PID control is not in and of itself done for the purpose of saving energy directly.

That said, if there is currently a totally manual process of inspection, determination of needs and some manual process of adjustment to attain the desired results, automating that process through the addition of effective process control can absolutely save energy. But to be honest, that is usually a lot more involved than just adding a PID loop controller.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[LittleInch]

The VFD might save energy compared to a control valve.

Don't forget the VFD itself uses about 8 to 10% of full load of the motor and efficiency of the motor and pump can vary depending on speed.

Don't forget also that the power absorbed by the motor reduces as flow reduces so it isn't always s benefit if your trying to save money. All depends on the specifics.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[FacEngrPE]

only reason he is using the VFD is for the start of the pump

I would not want to try to sell energy savings by adding flow control to the VFD's duties.
It could be that there is a energy savings that could be measured, but it is just as likely as noted above that the energy consumption will go up.
In this case as the VFD losses are part of it's use as a motor starter, you can ignore them when calculating energy savings. If this application would benefit from constant flow, you need a food service flow meter. If all that the application cares about is filling the tank, and stopping when full, on / off level control is enough.

 

[jraef]

Don't forget the VFD itself uses about 8 to 10% of full load of the motor and efficiency of the motor and pump can vary depending on speed.

That has not been true for decades... Almost all modern VFDs are around 97% efficient at full load, dropping to around 93% (depending on mfr) at 40%

I'll just point out that energy savings on variable flow applications with CENTRIFUGAL machines (pumps and fans) using VFDs is not a theory or a hoax, it is a proven fact borne out by innumerable site studies involving utility rebate program for the past 3 decades. The savings come from the DIFFERNCE in losses though OTHER methods of flow restriction compared to using the changing motor speed.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[LittleInch]

Ok, maybe a bit high, but I'm perhaps more used to larger wattage motors where the heat that is produced by these VFD units then requires more money spent on A/C units both CAPEX and OPEX to cool the electrical room down. But even 5% extra load is significant over the long term and certainly in terms of heat. For a 100kW load, you're then having to get rid of 3 to 5kW of heat inside your electrical room.

They can save energy for sure, but there is no guarantee that this saving is either present or of a value that is significant. Nothing in this world is a guaranteed money saver.

If chosen well then yes, VFDs are good, especially if they operate at relatively low power for a lot of the time and then full power in shorter bursts, but many times it is just assumed it will reduce energy / costs which in practice may not appear or be a smaller saving than might be first thought once you add in all the other losses and additional cooling / ventilation etc



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[LionelHutz]

I'll just point out that energy savings on variable flow applications with CENTRIFUGAL machines (pumps and fans) using VFDs is not a theory or a hoax, it is a proven fact borne out by innumerable site studies involving utility rebate program for the past 3 decades. The savings come from the DIFFERNCE in losses though OTHER methods of flow restriction compared to using the changing motor speed.

Right, it's possible. Applications requiring a varying flow can save money using a VFD. There's no guarantee that an application with varying flow demands will save money.

However, "I'm going to use a VFD because the affinity laws tell me it'll save money" isn't a great way to engineer it. I've seen VFD rebate programs that paid out if you had a motor driving a pump or fan with no consideration of the use or need for varying flow.

Varying speed because "different viscosities require different power." isn't a much better approach.

Most people never even use a power meter to confirm anything and many of those don't understand the power demands of various applications. Still others don't understand that you have to look at the energy efficiency of the whole system, not just the power input into a motor.

There was an idiot on here a bunch of years ago going on about how he tested a VFD on a crappy vent fan and his testing showed that the power draw varied linearly with the speed. That thus proved that ANY application slowed to 1/2 speed would only use 1/2 the power and therefore 1/2 the energy. He was peddling that to everyone like it was gospel with 100% conviction.

 

[edison123]

In India, I have seen a booming market for VFD's. Many of my clients in traditionally conservative process industries like cement, chemical. steel, sugar, glass making, tyre manufacturers etc are replacing their time tested DC and slipring motors with VFD driven AC motors in the last two decades. Some of the VFD's are even in MW range. I have in my shop right now two 3.3 KV, 2.5 MW Banbury motors driven by VFD for servicing.

Almost all of them have claimed they are seeing their utility bills reduced significantly post VFD and they like the less maintenance of the cage motors.

As a motor man, I prefer VFD's since they prevent thermal and mechanical shocks during starting.

VFD is power saving is not a myth given the actual reality on the ground.

Muthu

http://www.edison.co.in

 

[LittleInch]

Not exactly comparing like with like though are they?

Just replacing some ancient 50-60% efficient DC motor with 95%+ new A C motor would have done some of that....

No one is saying it's a myth, just that it's not a simple as saying it will ALLWAYS save you money overall.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[edison123]

Agree it's apples and oranges, but these oranges still had 90+ efficiency, not 50 to 60%. May be using higher efficiency cage motors off-sets whatever additional VFD loss but my clients are switching over to VFD due to utility cost saving, less maintenance headaches and option of variable speeds.

Muthu

http://www.edison.co.in

 

[LittleInch]

A lot depends on how the users use the motor. Running 75% of the time at 50% load will get you some savings for sure and also the soft start helps. but run these at 90% for 905 of the time and they are much less likely to, unless their existing motors were very inefficient. And if by "time served" you mean something built in the time of the Raj then this is possible.... ;-)

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[systemsaver]

Thank you engineers for sharing your experience using VFD with centrifugal pumps with a liquid that does not have a fixed viscosity. On Friday we are going to install a network analyzer for the input power before the VFD to record every minute of active power, power factor, current, voltage, and active energy for a week and only then we will review the data to see if it is ´possible to install a flow or pressure transducer together with a PID control to get energy savings.