Harmonics and VFDs 22

[Marke]

One of the areas that I operate in, has a high density of VFDs on pumps on relatively weak supplies. The result is that the high levels of harmonics on the VFD inputs has accumulate in the supply and is causing a high THD of the supply voltage. While we do have supply regulations covering harmonics, in this instance, the harmonics are higher than they should be.

There is an option of using zig zag transformers and six phase rectifiers as a means of reducing the harmonics drawn by drives however in this case, the drives are already installed.

There is a transformer for each drive and sizes range up to 200KW.

One thought that I had, was that for future installations, and there are new installations going in all the time, that the new supply transformers be designed with a zig or a zag winding to give a phase shift, and install equal loading on the leading and lagging phase shifts. This should act like a twelve pulse input on one drive, only it will be across two drives.

Any thoughts on this??

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[gepman]

It is nice to know that some things transcend language. I could see it even though my French is non-existant.

I wonder what the prices of these units are versus an active power correction unit for a given amount of distortion.

 

[Valvecrazy]

"I don't expect to convince you since I can only offer engineering principles, calculations, manufacturers literature, articles, research publications, and similar material while you offer your unsupported statements that valves save more energy than VFDs."

I was doing the math and my own research 20 years ago, when I was hyped up about VFD's like you are now. As you have already proven, there are many ways to doctor the math to make a VFD look really good. The math has not proven itself accurate with the many thousands of installations I have seen in real world applications.

I know that a VFD could be more economical if the head lowers to 200' or 115' instead of 231'. However, the well has a 231' pumping level, or the golf course sprinklers need 100 PSI form the booster pump no matter if running 1 sprinkler or all of them at once. The well is not going to make itself more shallow just to make a VFD look better. The sprinklers at the golf course are not going to give the coverage they should at 87 PSI to make a VFD more efficient, they need 100 PSI all the time.

If you want to add in another 14' of head from friction loss, then the pump now needs to deliver 1200 GPM at 245' of head. If you slow the pump down where it can only produce 200' of head, then the water is not even getting to the top of the well as the lift is 231'. The 14' of friction loss is going to be reduced as the flow rate is reduced. So the head can vary a little form 1200 GPM at 245' to 100 GPM at 231'. Other than that, you cannot reduce the head requirement just to make the power requirements of a VFD look better. If you are going to run a 1200 GPM pump at 240 GPM all the time, as per your example, then yes a VFD will save considerable energy. However, if that is all you are going to use, then replacing the 100 HP pump with a correctly sized 25 HP will cut the electric bill in half compared to using the VFD.

What I find in the field, is that most pumps are run at maximum capacity about 80 to 90% of the time, and the variable flow is only used when filling a spray tank or something else intermittent. This makes the electric bill increase for installing a VFD, after the math had already promised energy savings.

On the systems that I work with, if the system curve makes much difference, then somebody messed up and put in way too small of a pipe line. As per the ABB calculator, a reduction in head is really what makes a difference in power consumption, not the VFD. In the section labeled "head above an open valve", it should read, how much head can be reduced. Using pipe with less friction loss or going to a lower pressure irrigation system should be what the cash incentive is paid for. This will decrease the cost per gallon pumped, while decreasing the pump speed with a VFD increases the cost per gallon pumped.

"Efficiency will be reduced when a motor is operated on a bus with harmonic content. The harmonics present will increase the electrical losses which, in turn, decrease efficiency. This increase in losses will also result in an increase in motor temperature, which further reduces efficiency."

Then when transformers or line filters are added to reduce the harmonics, the input voltage is decreased and the amperage and heat increase even more.

I have been playing with the ABB calculator, and I find it to be as bad as the calculator for PG&E, that you already agreed was flawed. The ABB calculator also makes no provision for looking at the pump curve. How can you make a comparison if you don't know what the pump will do naturally? Considering the source, this calculator is designed to make a VFD look good no matter what.

I don't believe a Drive can be 98% efficient. I always see at least a 5% loss in pump output when I use a Drive. This usually causes more of a loss than the 7 PSI friction loss across a valve. This means at full flow there would be substantial savings for just pulling out the VFD.

I did calculate an actual savings of $800.00 per year when flow rates below the maximum are used for a majority of the time. This is no where near the $4,600.00 savings calculated by the ABB software. At $800.00 actual savings per year, the VFD would have to last many years to pay for itself. You might get a year or two of savings if the VFD last longer than it takes to payout, which is not likely.

If the flow required varies widely, even when the head remains constant, there are some energy savings by using a VFD. It is just never as much as exaggerated by the math or any Energy Savings Calculator concocted by VFD manufacturers. Even when a VFD can save $1,000.00 per year in energy, if you have to replace a $5,000.00 Drive every 5 years, you are not really saving anything. I know there are a few old drives out there but, very few last long enough to pay out. See the attachment below of an ABB drive that only lasted a couple of months. This is what I am used to seeing in the field, and why I replace VFD's with valves every chance I get.

Then lets not forget what started this thread, which was harmonics. Harmonics is "dirty power", which causes everything else on the same electric grid to become less efficient. I have even seen several installations where the air-conditioning for the VFD uses more power than the pump itself. How can your math or these calculators figure the losses elsewhere in the grid and add them back into the calculation? Harmonics are just one of many negative side effects of VFD control. Resonance frequencies, bearing currents, voltage spikes, RFI, nuisance trips, required technical assistance, and a several other problems go away when you eliminate a VFD.

Here is what you should learn from this;
1) Just adding the VFD without decreasing the head costs you more energy per gallon.
2) You should like Valves. Valves let you use the full range of the pumps capacity without any of the negative side effects of VFD.
3) You can save money using a Valve if you select a pump curve with a steep Brake Horse Power slope towards the shutoff point with NO loss of efficiency at your nominal operating point.
4) In high static head systems, which is most pump applications, it is more difficult to save energy with a VFD.
5) A savings calculator is available from the manufacturer of VFD's, which does not take into account the natural characteristics from the pumps curve.



I would like this to be my last post in response to you because I have better things to do than to prove the obvious. This was written so that a farmer could understand it but, it is technically accurate. It is just hard to see the truth through VFD colored glasses.

We could all be more green, if manufactures didn't try so hard to keep us in the red.

 

[nightfox1925]

I am amazed with the dedication of both gepman and valvecrazy in support of their respective claims. I do wish though that these highly technical discussion won't just be left with bigger questions to spectators like myself and hope that both gentlemen would work each other and do the necessary actual tests...who knows, this may lead to a very important revelation and improvements...(sigh)

GO PLACIDLY, AMIDST THE NOISE AND HASTE-Desiderata

 

[waross]

I hear you nightfox1925.
I was thinking the same when I suggested;

Can either or both of you set up a test to measure the energy consumption (KWHr) to raise a given volume to a given head? Filling an elevated tank or wasting the discharge out an elevated pipe if the volume can be accurately measured would be the most acceptable test.
Then, apply in turn, a VFD and valves to reduce the flow 20%, 30% or 50% and measure the total KWHr required to pump the same quantity against the same head.

I want to know the KWHr to lift a given volume to a given head with each system.
I think that everyone will agree that a well designed system running at full output will probably not benefit from a VFD.
however we want to know what happens when the system is asked to perform at 50% capacity and the choice is between a VFD and a valve.
We would like to see a comparison test showing
HEAD x MASS / KWHr
Thanks fellows.

 

[Valvecrazy]

I hope gepman doesn’t take it personally. I have had this argument with many worthy adversaries before. It may take a few years but, I nearly always hear, “you know you were right”.
Quotes like;
“We haven’t touched that pump system with the valves controlling it in 14 years. The VFD system next to it is always needing maintenance, and the drive has been replaced 3 times.”
“The pump system with the Drive has used considerably less energy, not because it is more efficient but, because it has been down a lot of the time.”

Gepman has done an excellent job with the math. I would not argue with his math for a minute. Don’t just do the math though, test which one works best in the real world. Doing the math (especially using the Handy Calculators), is probably going to get you to try a VFD first. When you have had as many problems with the VFD as you can stand, try a valve. You may not get quite as good in energy savings as you did with the VFD but, having all those problems go away is worth a lot in itself. If a valve controlled system last two or three times longer than the VFD system, that has to be added back into the energy savings calculation, and can make much more difference than a few points of efficiency. The best tests are the ones you do yourself, because it takes time to make these decisions. If it works best for you in the long run, then it has won the test. Just don’t let the math keep you from trying a valve. A good centrifugal impeller does have some natural power reducing qualities, that can easily be taken advantage of, by using a control valve.

I have a couple of 5 HP, 150 GPM pump system demonstrators that can easily be switched back and forth between drive control and valve control. Of course I have chosen pumps with steep Brake Horse Power curves. They show basically the same power consumption with VFD as with the Valve. One of them stays in my shop in Texas, as I have learned this is something that is easier to show than explain. I also have several smaller and larger pumps in the test stand. They are always available for anyone who wants to come to Texas. The other VFD/Valve demonstrator will be at the World Ag Expo show in Tulare, California on February 12,13,14, Pavilion B, Booth 2128. I will not be there myself but, several qualified people will be. I would like to invite Gepman and anyone else who wants to see a side by side demonstration to come. Being as it is in California, maybe some of you can make it.

My thanks to Gepman and everyone else for this interesting discussion. I am afraid now you must make your own decisions. I only ask that you realize you have other options. BTW, the valve goes on the discharge side of the pump.

 

[Skogsgurra]

I said, about half-way down this thread, that this is one of the most interesting and valuable threads ever going in the motor and controls forum. After that, gepman and Valvecrazy have both added a wealth of information, which probably needs several days to absorb, understand, verify and relate to. I am sorry that I do not have that time now either. But doing that is high on my to-do list.

I think that this has been a great discussion. I still cannot say what is right and what is wrong. But I am sure that I will have a closer look at the non-VFD alternatives before deciding blindly that a VFD is needed in pump applications next time.

I make my living out of VFD problems. Bearing problems, EMI, harmonics and also plain fires and explosions in VFDs. Keeping a working knowledge is also a problem with VFDs. It is not unusual that even the manufacturer's HQ has a problem finding support people for drives being 10+ years old.

So, even if VFDs are wonderful devices from an automation point of view, they also do bring problems. Problems that a lowly valve seldom has.

Having said that, I need to add that many VFDs run without problems for many years. It is when problems occur unexpectedly that the situation gets critical.

Thanks to all for a great discussion. I am sure we haven't heard the last word from the participants. This theme merits an own thread next time.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[gepman]

I guess valvecrazy will only be a few booths down at the World Ag Expo from where the company that I work for will be giving out $1250 (or $0.08 per kWh saved on an annual basis) to any farmer that wants to install a VFD on a well pump. I will stop by your booth. I hope that your Texas company is not quite like that other Texas energy company, Enron. I also invite you to apply for a utility rebate for your energy saving valves, however calculations instead of anecdotal testimonials will be required.

As valvecrazy says the math is correct so the problem will be in his demonstration device. Please post a P&ID, BOM, and layout drawings so I can tell you what is wrong with your setup. I will need to know the pump curve, the valve Cv, VFD model number, motor model number, measurement equipment model numbers, etc.

I have advised many farmers NOT to install VFDs because the static head/total head percentage was too high and/or the amount of hours that the pump ran was not sufficient to justify a VFD. As I have said previously VFDs save energy by reducing the pressure at the given flow so if the pressure does not change there can be no savings.

See you in Tulare, maybe we can have a drink!

 

[nightfox1925]

And I hope after the Tulare exhibit, since I won't be able to avail myself for a visit, both/either gentlemen would come back and post new thread of what technically transpired in these demontrations so it may also help us make up our minds as applicable. Many thanks to everybody!!!

GO PLACIDLY, AMIDST THE NOISE AND HASTE-Desiderata

 

[sibeen]

It's been quite a few years since I've visited this forum and to find a discussion like this on my return is a joyfull thing; finding that I remembered my password and user name was more of a miracle

What a great thread. Not having ever been involved with motors or pumps much of it goes whoooosh over my head, but after 20 odd years in the UPS industry the harmonic problems are something I can get my head around.

I see that it has been suggested to use a Dy transformer at the front end of some of the drives to increase the pulse number to 12.

I would suggest that this is taken a step further by the use of different phase shift transformers.

If for example you used different phase shifts of (in degrees) +7.5, +15, +22.5, +30, 0, -7.5, -15, -22.5 and -30, then you efectively end up with a 54 pulse rectifier - 6 pulses x 9 different phase shifts.

Theoretically, if the drives were balanced in load across these 9 different phase shifts, then the input current harmonics would be down to about 2.4% compared to a 6 pulse input which is approximately 27%.

I'm also very interested in the great valve vs drive shootout and I hope the experimental results are posted here.

 

[gepman]

Valvecrazy
I have set up a test at the SCE Ag Pump Demonstration unit at AgTAC, which is across the street from the Ag Expo. Be there at 1:30PM, February 14. See

http://www.sce.com/RebatesandSavings/EnergyCenters/AGTAC/exhibitsanddisplays/

for exact location.

This is their standard VFD versus valve test unit which they show to all the farmers who want to see it. Come with a nametag that says "Valvecrazy" and I will recognize you.

 

[Valvecrazy]

Gepman, I would love to get to meet you and see your demonstration. However, as I said earlier, I will not get to come to California for this show. We have several going on at the same time so, only my guys from the west cost will be at that show. I will make sure they come by and see you.

I know that the power consumption of throttling a pump, being the same as the power consumption of a VFD, is something most people have to see to believe. That is why I will have a 5 HP demonstrator at Booth B 2128. However, I am attaching a curve of the pump that is on my demonstrator. You can see that at 200 GPM it is pulling a 5 HP load, then when throttled to 20 GPM the power required drops to a 2 HP load. I don't know why some people cannot see what I am talking about from simply looking at a pump curve.

As you said earlier;
"As I have said previously VFDs save energy by reducing the pressure at the given flow so if the pressure does not change there can be no savings."

Therefore, when maintaining a constant pressure or TDH, which is the case with almost every system I work with, there can be no savings with a VFD.

It is all about choosing the right pump. A VFD can make a badly designed pump use a little less energy but, it can never do better than a properly chosen pump.

Then there are the things that don't show up on a pump curve, and are hard to add back into the equation.

1) Parasitic losses of the drive when running and when in the standby mode
2) Environment control for the Drive itself
3) Heat losses from active or passive filters
4) Efficiency loss from stray voltage for everything else in the area
5) Shorter life of the equipment with VFD than with ATL controls
6) And now extra tariffs on the electric bill for VFD induced harmonics

If throttling means 5HP - 3HP = 2HP
Then VFD means 5HP - 3HP + (1,2,3,4,5,6) = ?????????

 

[peebee]

Back to the original question, regarding the use of phase-shift transformers to reduce harmonics, check out

http://www.mirusinternational.com

. They make various configurations of harmonic-mitigating transformers which do exactly what you're looking for.

From a quick GoodSearch on "harmonic mitigating transfomer", it appears that Square-D and other players have also gotten into this game now. Click here for more:

http://www.goodsearch.com/Search.aspx?Keywords=harmonic+mitigating+transformer

or Google it here:

http://www.google.com/search?q=harmonic+mitigating+transformer

 

[gepman]

As promised here is the report (at least my report) from the World Ag Expo and the VFD vs. Valve shootout. I have attached a picture of the valvecrazy's test setup.

The test setup consists of a pool or reservoir from which eight solenoid controlled nozzles spray up into the air and back into the pool (similar to an irrigation sprinkler). The water in the pool is brought into the suction of the pump in the lower left of the picture. The pump discharge goes through a check valve and then through the red pilot actuated diaphragm valve. The pilot pressure for the diaphragm valve is on the discharge of the diaphragm valve and senses the pressure on the discharge of the valve. This pilot signal can be adjusted with the small pressure regulator located behind the red valve which has a black top and a small red tag on it. This regulator has been adjusted so that the discharge pressure on the valve remains at a constant 30 psig. The water then goes through a wye strainer, has a pressure gauge to indicate pressure, then goes to a distribution header which distributes the water to the eight nozzles. The nozzle solenoids are separately controlled by eight switches so that anywhere from 0 to 8 nozzles can be open. The enclosure with the red switch is a four way switch for "off", "FVNR", "VFD", and "TEST". To the left of the enclosure is an ammeter display for the FVNR circuit, to the right of the enclosure is the VFD with a HIM display showing rpm, pressure, and amps. There is also a pressure transducer located near the pressure tank which provides input to the VFD for speed control.

The test to show that the valve uses less energy than the VFD is reading amps as displayed by the two devices above and NOT power. The first test done showed that the VFD used more amps than the valve. I then realized that the red diaphragm valve was still controlling in the circuit so I made a third run while disabling the pilot signal (by shutting of a valve in the pilot line) which is also shown:

Nozzles 1 2 3 4 5 6 7 8
1 VFD 6.3A 6.7A 7.2A 7.8A 8.3A 8.9A 9.5A 10.1A
2 Valve 3.8A 4.8A 5.9A 7.0A 8.0A 9.2A 10.0A 10.8A
3 VFD 5.6A 5.9A 6.3A 6.8A 7.2A 7.9A 8.5A 9.3A

Somewhat unsatisfied with these results I went to my car and retrieved my Amprobe 41PQ which measures power (one measurement at a time so to get total power I needed two measurements) and harmonics. The VFD run below is done with the diaphragm valve disabled so that it cannot modulate. Both measurements were taken at the same place in the circuit at the power input to the enclosure. There was no flowmeter to check flow rates.

Nozzles 1 4 8
4 VFD 1354W 1945W 2760W
5 Valve 1710W 2230W 2927W

I was surprised that the VFD used less power with all of the nozzles open but I then realized that the diaphragm valve was restricting the flow even with all nozzles open. The rpm display for the VFD was 3147 for 1 nozzle, 3180 rpm for 4 nozzles, and 3278 rpm for 8 nozzles. Full speed on the motor is 3450 rpm.

We were unable to run a test at the SCE AgTAC test center since they had a water problem with their cistern that stores the water for pump tests.

 

[Valvecrazy]

Thank You Gepman!! I appreciate you taking the time to run this test. I am glad you were able to do testing on our system since you could not get your system working. I imagine you were as surprised as I was the first time I realized how little difference there is in power consumption between Valve and Drive control.

Eight nozzles is all our system has for output. These do not let out quite enough flow to get to BEP as you can see from the Drive never getting the motor to 3450 RPM. Of course we all know that at BEP, the power consumption would be less with Across The Line than with the Drive, as there would be no parasitic losses. Therefore, we know the Drive would not be as efficient at BEP, and I did not think it important to get the pump to full output or BEP for this test.

Reducing the flow down to 8 nozzles, 4 nozzles, and 1 nozzle, you can see that there is only a couple hundred watts difference between valve control and VFD. However, I purposely did not use any filters to keep the Drive looking as good as possible. There is no line filter as this is trade show equipment, and I do not care if harmonics interfere with other equipment in the area. I also do not care if the harmonics produced with my VFD cause other equipment in the area to be less efficient but, that power loss should be taken into account.

I did not use a load filter on this system as it is for trade shows only. This motor only runs about 10 days a year at 8 hours a day so, it should be able to handle the voltage spikes for a while with no more hours of use than it is made to run. It also has a very very short wire from Drive to motor, which reduces the reflective wave.

However, you did not mention the measure of harmonics we were producing with the VFD. As with the title to this thread we all know harmonics are a problem. So in a real life situation I would need to add a line filter and/or a load filter as well. When you add back in the heat losses from the filters that are needed, the couple of hundred watts you show giving the advantage to the VFD, will no longer be there.

Just adding the filters needed would about equalize the power consumption between Valve control and Drive control. Then if you add in the parasitic losses from the Drive using power even when the pump is not running, and the losses of efficiency to other equipment in the area from "dirty power", Valve control looks even better.

All that being said, we can go back to the actual numbers of this test without adding parasitic losses, filter losses, and losses to other equipment in the area. Even with the couple hundred watts difference in the best case scenario for the Drive, the payoff of the Drive itself is almost non existent. If you consider the cost of the Drive, and shorter life of the Drive and equipment compared to running Across The Line, payoff would take many years, if possible at all.

I should mention all the other problems that exist as side effects of Drive control but, this is already a long thread. I will say that you do not have to worry about these other side effects, when you realize that the Drive is not really saving energy, and therefore should not be used in these type applications.

Back to the topic of this thread, to eliminate harmonics, the best way is to not cause them in the first place, in systems where a Drive has no power savings benefits.

 

[LionelHutz]

I'm just going to add a few comments;

It looks to me that you had about a 3hp, 208V motor there. This motor showed 167W saving running with all 8 nozzles open.

OK, the system had 8 nozzles to simulate a sprinkler or irrigation system. I would believe that in the real world you will very likely not have the same type of system perfectly designed so that you can run the motor at full speed with the sprinklers open and get the desired operating pressure. You would still either have to always valve throttle the pump or speed control it to get the system to work at the desired pressure.

So, now, scale this test system up to a real world application and you would maybe use a 30hp motor. So, the VFD would save 1670W of power compared to a valve when irrigating with the system running all 8 nozzles. Say the system runs 4 hours a day. That is 6680Wh of energy savings a day. Now assume that hydro costs about 16 cents per kWh this is about $1.07 saving per day.

Of course, the above numbers will change depending on how many nozzles are used when the system is operating.

So, looking at this example it is seems the VFD uses less energy but it really doesn't amount to much and the potential savings really can not be used to justify the added cost of the VFD. It appears the valve is probably the better solution for this particular system.

However, I don't really take the data measured using a test setup built by someone who is promoting valves over VFD's as proof that applying a VFD on a pump is always wrong.

 

[Valvecrazy]

"However, I don't really take the data measured using a test setup built by someone who is promoting valves over VFD's as proof that applying a VFD on a pump is always wrong."

Except for the above statement, thank you LionelHutz for your comments. Your scale up to 30 HP is a good idea and appreciated. However, it will be even more important for a 30 HP to have line and/or load filters. The heat losses of the filters will take even more away from the $1.07 per day the VFD is "saving".

I also disagree that a pump in the real would never run at BEP. Most of the systems I work with run at BEP most of the time, and only need varied flow rates part of the time. This makes Valve control look even better still.

As for the test being set up to favor Valve control, it is not. Actually it is set up to favor the Drive, as I added no filters and never run it at BEP. Gepman went all through the test set up and describes it very accurately. If there had been a "man behind the curtain pulling strings", he would have easily seen it. It is a very straight forward pump setup.

All of this test data is not really needed. I go back to the fact that all this information is in the pump curve. If you know how to read a pump curve and pick a pump with good horse power characteristics at low flow, actual kilowatt per gallon testing is not needed to see that Valve control can be just as efficient as Drive control.

See the attached pump curve again. Thanks to all!!

 

[Marke]

Asa matter of interest, what was the power consumption without the valve or the VSD?

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[Valvecrazy]

With or without a valve, the power consumption will follow the horse power curve. Regulating the pressure with a valve allows you to always have the same pressure on the sprinklers or nozzles no matter if you are running just one, or many at a time. Without the valve, the sprinklers or nozzles would have more pressure than needed, which would overshoot and mist when you were running too few at a time. However, the power consumption is still dependant on the flow, and follows the curve. It is "counter intuitive" that power consumption decreases as the flow is restricted but, it really happens no matter how you restrict the flow.

 

[LionelHutz]

Yes Valvecrazy, we understand that the power consumption will go down as the flow goes down. Understanding that the power consumption is related to the work being done is not a difficult concept.

gepman provided some good info showing cases where a VFD can save energy and is worthwhile. This example is not proof that a VFD is always the best solution. Similarily, you give an example showing that a valve is a very good solution but that is not proof that a valve is always the best solution. So, as like most things in life, each application has to be evaluated to determine the best system to install.

What I did find disappointing was that your test jig used ammeters to show the power used in the different configurations. For me, this alone is a pretty big red flag. Also, see my above paragraph. Just because you built a system that can demonstrate that a valve is a good solution does not mean a valve is always a good solution.

 

[waross]

Hello Valvecrazy;
We don't always get to design a system from the start. We sometimes have to modify existing equipment and do the best we can with equipment chosen by someone else.
Can you or someone else give us a guideline as to what to look for in a pump curve? What curve profile will work equally well with valves or VFDs and are there any pump curve profiles that will benefit from VFDs?
If throttling the pump output develops a fairly large pressure drop across the control valve is this an indication that the savings with a VFD may be more than shown with your test rig?
Thanks

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