How to reduce energy consumption (motors) 3

[rayhud]

Hi,

I'm a student, and i have been given a task during my internship, which is a study of the energy consumption of all the motors inside the enterprise, in order to try and come up with a plan to reduce their consumption.

I've done some research, and was able to determine that in order to do so, a speed control of motors so as to adapt it to the load is necessary, as well as the optimisation of their efficiency and regular maintenance.

The problem is that i can't figure out a way to start working on this project, so what i need to know is how to be able to give a solution through analyzing the motors' energy consumption ? and what are the points i need to focus on ? or the motors i need to target ? and which strategy is to adopt in order to be able to succeed in this task ?

And if there are more ideas on the solutions to reduce motors' energy consumption, please feel free to enlighten me.

Thank you in advance.

 

[racookpe1978]

Start by figuring out the problem.

How many motors?
What are those motors: size, current, voltage, and load (pump, machine tool, HVAC, fan, air compressor, roller, or printing press or crane or whatever)?

What are their running speed and the duration of each run and the existing process control type (continuous, intermittent, speed-controlled, throttle valve controlled, speed independent, backup, low-speed (roller or printing press) or high-speed (food canning machine)?

Then look at the list again and pick the first few that "might" benefit the MOST by a VFD.

Size of VFD, price, installation price, problems, cost of current power, potential for changing current operations WITHOUT using a VFD (changing to run that pump at a different "cost point" of the day, or change its running time from continuous to standby, etc.

But first, get the list.

 

[rayhud]

Thank you so much racookpe, that really helps point me in the right direction, i will keep you posted while the work progresses and in case there was a problem or something.
Again, thank you so much.

 

[sandeepelect1]

racookpe, is absolutely right.

you can use a power analyser to collect the actual data of the motors (i.e. Voltage, Current, Harmonics,Power Factor) and then you can compare it with designed data.



Regards,

 

[mikekilroy]

When all seems too overwhelming to even start, stepping back and beginning a list as Racook suggests often gets the process rolling. Great advice worth purple star!

http://www.KilroyWasHere<dot>com

 

[waross]

Motors are pretty efficient. I would look at each major motor and associated load. What is the mass of fluid pumped against what head? How much power is the motor using to do the job?
Look for hot spots. That will pinpoint areas of losses. Use an infrared spot meter. If possible measure the temperature of the fluid before the suction and after the pump discharge and at the final discharge. Look for delta T across control valves. Mass pumped per time unit times delta temperature will give some indication of pumping losses.
Look for underloaded and overloaded motors and run the numbers on appropriate sized motors.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[LionelHutz]

The big one is to look for application with waste. By this, I mean a fan that is blowing more air than necessary or a pump that is recirculating the liquid. These are easy places to gain efficiency.

Next, look for applications that leave the motor running without any load. Generally speaking, motors have a fairly flat efficiency curve from around 40% load to 100% load. So, if the load remains within that range the efficiency will be good. But, the application which unloads the motor has zero efficiency during those times.

Lastly, take a look for old motors. Often, older motors aren't that efficient when compared to a new motor.

With the exception of the above mentioned waste applications, a VFD should be used for process control without expecting any energy savings. You really can't just change the speed of a motor and save energy without changing the process. A process that requires a certain amount of work gets done will require a certain amount of HP to do this work. Just changing the speed of the motor won't change the HP the load requires. All you're doing is trading rpm for torque. I'll give you an example. Your process pumps water into a tank and requires 50,000 gallons of water per day. The pump is turned on and off via level controls. Just switching to a VFD and running the pump slower all day won't save energy. In fact, it's less efficient. The process still requires the same volume of water each day and adding the VFD had added losses which makes your process less efficient.

 

[jraef]

I am a denizen (junkie?) of several Internet forums for electrical and control related fields, I have to admire your tenacity in finding them all and posting this same question in every one (that I am aware of anyway) in an attempt to get help with your homework. That said, a couple of nuggets of advice relating to your future.

1) Getting othe people do do your work for you is a Management function, not an Engineering function. Collaboration is great, but collaboration in this case would mean do your own work first and if you get stuck, present a specific problem you are having and ask for help, don't just blast out a plea for where to start.

2) Asking 200 electrical professionals for an opinion on something this open ended is likely to net you 600 different opinions. You are more likely than not to end up even more stuck than before you started.

Jump in on your own, even if you have to intuit the beginning, then ask for help with specifics. In the end you will be an Engineer, not a middle manager somewhere pushing paper across a desk waiting to be downsized.


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[waross]

Well said Jeff.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jraef]

Notice that I resisted the temptation to be fatalistic and say "you will be an Engineer waiting to be downsized..." Things seem to have improved for our lot lately, at least around here. Middle managers on the other hand, not so much. We just had a shakeup and they "consolidated" a lot of middle management positions. I always thought there were too many layers anyway, it's just sad to see people who started their careers as Engineers, then switched to Management for more money, get it in the neck. Then it's even more ironic that it's happening right when Engineers are riding higher, but they have lost their skills and likely can't go back.


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[rhatcher]

I agree that we shouldn't be helping the OP with his homework. However, for the sake of any others following this post, I would like to correct something from a previous post.

Running a pump at reduced speed will result in energy savings. The pump affinity laws state that the flow is proportional to speed, the torque is proportion to the speed squared, and the power is proportional to the speed cubed.

In the example given, a pump is cycling on and off, controlled by a level switch, to fill a tank. For the sake of discussion, let's assume that the duty cycle for the pump is 50% on, 50% off.

Let's install a VFD and run the pump at half speed. Flow will be reduced to half so we must run the pump continuously to keep the tank full. However, power consumption will be reduced to one half cubed (0.5^3 = 0.125). So, while we must run the pump twice as long, at 12.5% power this results in a total energy consumption of 25% of the original case.

Yes, the VFD will not be 100% efficient. Yes, additional losses will be introduced in the motor due to harmonic heating. However, the motor's losses for friction and windage as well as stator and rotor I^2R will be reduced. Off the top of my head I would say that it would be reasonable to expect that the net increase in losses from operation on the VFD would only be a few percent at most, leaving us operating at 27-28% of the original energy consumption.

 

[BrianPetersen]

^ unless you are running a centrifugal pump against a fixed head pressure, and the proposed new operating speed is no longer sufficient for the pump to develop enough head pressure to pump anything, in which case it now becomes 0% efficient! You have to find the right balance ...

No one said it was easy ...

 

[LionelHutz]

LOL, despite what some perfect law states, you can't do the same work with less energy unless you improve the efficiency of the system.

Apparently, running the pump at 1/2 speed has improved the efficiency of the motor and pump system 4 times? Something doesn't compute with that claim....

 

[1gibson]

Rhatcher did say "for the sake of discussion" so let's not get too upset about it.

1/2 speed would mean 1/4 of the head, so the pump would have to be grossly oversized to begin with. Most pumps are just somewhat oversized.

The real savings come with slightly reduced speed, to target the pump operating at it's best efficiency point. You will find a speed that matches BEP with system resistance. It would be hard to come up with real numbers where a lower speed shows improved efficiency from there.

 

[jraef]

How a VFD can save energy is by reducing losses in a variable flow system from OTHER means of FLOW CONTROL. It is simply a more efficient way of accomplishing it. If for some reason a tank fill operation MUST employ variable flow input, for example to retain a specific tank level at all times within a narrow band and a variable outflow, then using a proportional valve and running the pump at full speed will consume more energy than running the pump at reduced speed to match the outflow. But using a VFD to replace a simple fill / drain on-off system is not a good energy savings plan all by itself.

By the way,

Let's install a VFD and run the pump at half speed. Flow will be reduced to half ...

Therein lies the mistake most people make in this scenario. 1/2 speed does NOT equal 1/2 flow. Static head must be overcome FIRST. There is now way to directly equate speed and flow without knowing ALL of the dynamics of the system. I have seen many many marketing pieces, some from my own company, that make this equivocation error, but if you look at the data files behind the slick charts and graphs, you see that the Engineers who actually created them did NOT make that mistake, the energy savings charts clearly say FLOW, not speed, but the Affinity law used for the argument says SPEED, not flow. It's typically the marketing people who make the leap.

For example, here is an image of a pumping system performance curve showing a VFD vs a throttling valve.

In reality the VFD energy curve would likely have a straight line drop off of energy savings at around that 40% flow spot, maybe 30%, but usually no more than that, because for most pump designs that is the point in the speed curve where you no longer overcome the static head. So running the pumps with a VFD slower than that cannot offer any further savings, because if you don't overcome the static head, your fluid is no longer flowing and you are completely WASTING the energy. But the "belly of the beast" in most pumping systems is in the 40-90% flow range, right were the system efficiency is best using speed control. But you will see that if you look at 50% flow, it is roughly 25% energy, not 12.5% as per the affinity law for speed. This is more correct.

But compare that to this fan chart, where someone put in a parenthetical comment of (RPM) as if it were the same as CFM (flow), then used the affinity law RPM values on the X axis, not the Flow values. So if you extrapolate the 50% flow rate you see the 12.5% power consumption, but this is NOT correct. This happens a LOT and leads to a lot of confusion.

"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[waross]

If you are pumping against a dynamic head that is appreciably higher than the static head, then pumping at a lower rate for a longer time should reduce the dynamic head.
The static head is related to the work done and the difference between the static head and the dynamic head represents system losses.
Now run the numbers of the pump efficiency at the lower pumping rate, the motor efficiency and VFD losses and see if there is a real saving. Could go either way.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[LionelHutz]

I've seen a pump lifting from a well into a tank not move any water at 48 hz or 80% speed and another pump from a tank into a pressurized system have no flow at 57hz or 95% speed. After seeing how pumps work in the real world, it becomes obvious just how much BS is in most affinity law marketing claims.

The simple fact is that the "correction" by rhatcher is completely wrong. You might save some energy

As for using a throttling valve with a on-off pumping application. I never posted such a thing and that'd be rather silly to employ 2 different methods of pump control at the same time. Still, if you had properly designed the whole pumping system to operate with the throttling valve instead of on-off control, I bet you'd find there isn't much for energy savings by switching to a VFD.

 

[LionelHutz]

I was going to post that you might save some energy under the right conditions but if you expect to install a VFD and reduce the energy multiple times you will be sadly disappointed.

 

[ScottyUK]

I think "completely wrong" is a bit harsh, but VFDs don't universally save money as some would like to believe.

In high head applications like boiler feedwater pumps they are borderline useless because the controlling range from zero flow to full flow occurs in a very small frequency band near base speed as you've already noted, unless the boiler itself operates at a variable pressure. A pump operating at a more modest pressure which regulates discharge pressure by returning part of its output to the suction side will almost certainly show a reduction in energy input if the control scheme is modified to suit the VFD control. Whether that equates to a net saving depends on the costs of the changes, the cost of the energy, the value of additional maintenance and the value of any maintenance eliminated.

 

[LionelHutz]

No, it wasn't harsh. The post completely misapplied the affinity laws making it completely wrong.