Synchronous Reactance Controller for Energy Reduction 8

[ibkenb]

I am an electric utility power engineer and have been asked to evaluate a "Synchronous Reactance Controller" that promises our customer up to 10 % energy savings. Personnally, I think it is snake oil, but in fairness I would like to know if anyone has evaluated such a system and would share their results.

The device promises VAR correction which lowers the KVA demand and our associated charges. However, it also promises to filter the harmonics out of the supply voltage and to balance the three phase voltages. I am sure there are some efficiency savings with the harmonic improvements, but I do not know how to quantify them.

I cannot obtain information on the internals of the device and since it is connected in a shunt (parallel) I am at a loss as to how it may balance loads/voltage.

 

[gordonl]

It's impossible to evaluate a "Black Box", but the best I can see it being is a per Phase static VAR compensator. This could provide harmonic filtering, and power factor correction. There would be energy savings from the reduced harmonics, but also from the improved power factor, less current equals less line and transformer losses.

As for voltage balancing it could conceivably balance voltage by unequally correcting the power factor per phase, ie improve PF on low phases only, but this would work against the energy savings due to improved power factor.

The energy savings due to improved power factor I figure would be less then 1%, the only large energy savings I can think of is the improved efficiency of motors due to the improvement voltage balance. A post in the electric motors forum would yield a good estimate I beleive.

As for the ability of this box to balance voltage, it would depend on the kVAR rating relative to the transformer feeding the system. (If my theory is right about the per phase VAR compensator)

My gut feeling is there 10% is heavily relying on the motor issue stated.

 

[electricpete]

I know absolutely nothing about such devices.

It seems possible that a shunt device could help balance voltages by removing vars from the high voltage phase and injecting then into the low voltage phase. I can picture some static devices that act equivalent to a variable capacitor on one phase and a variable inductor on the other phase.

Likewise... to balance the real power loads, the device needs to remove power from one phase and inject on another phase. Presumably the amounts removed and injected are equal so the device doesn't have to generate any power. As far as picturing what those devices would look like that's a little tougher... but they can do some clever things with electronics.

I'm assuming the box must of course have voltage and current inputs as well to compute the power flows.

 

[busbar]

Isn’t a ‘synchronous reactance controller’ the same as a synchronous condenser; i.e., an uncoupled synchronous motor with excitation control of reactive power generation? The claimed balance/harmonic reduction/apparent power ‘savings’ could conceivably be obtained with such a device in an essentially brute-force manner, roughly similar to application of a zigzag autotransformer close to the loads of a 4-wire wye system.

It might be revealing to ask for a copy of their spreadsheet file of the analysis of your firm’s rate schedules to more accurately refine or confirm the ‘10%’ declaration.

It may be time to ask for some peer-reviewed papers, third-party test reports [and verifiable customer testimonials] on the claimed technology advance. If the assertions are to be substantiated, you would think the vendor would happily comply and would have offered such information with the original pitch. Should the information be deemed proprietary, how can anyone make an evaluation of positive or negative effects the device has on the customer’s and utility’s electrical systems? It also seems like they should be willing to provide NRTL listing reports as with most all low-voltage electrical assemblies, consistent with the mandate of local inspection authorities. Unless, of course, it is true cold fusion [once again.]

 

[peterb]

Sounds suspiciously like this post - thread238-21978.

 

[ibkenb]

Thanks to all who have responded to my posting.

In the area of efficiency, do any of you know any method or rule of thumb for quatifying the energy savings with harmonic filtering?

It seems that once the circuit passes through a VFD, any harmonic improvement would be moot and the only gains would be the VFD's reaction to the harmonic improvement .... if any.

 

[gordonl]

Harmonics in a system cause most components to have increased losses, ie - cables, transformers, motors. I beleive this would be less than 1% of total load.

 

[electricuwe]

To me everyone trying to sell a black-box claiming such features without being able or willing to explain how it works is very suspicous .

In the past there have been several examples where such companies fooled lots of people.

 

[jbartos]

Suggestion: It may be a good idea to see actual figures for the power distribution and loads for power factors, voltage drops and harmonic distortions. Some consumers may be noticeably inefficient, and energy savings can be accomplished. However, the analytical and measurement approaches are prudent and shall be performed first.

 

[thespaceman]

Energy savings are typically kWh reductions. Billing savings can come from power factor correction with a kVA based rate. Your true utility savings depends on the utilities' metering. Is it reading real power at the fundamental (60hz) frequency only? Is it reading true or total displaced power factor?
Overall heat can be reduced by voltage balancing and pf correction, but real energy bill reductions depend on the utility rate and metering.
Real power is real power ... a kW is a kW ... how can this device make a 50-hp motor look like a 45-hp motor from a kW view point?

 

[electricuwe]

We already discussed the topic of power measurement in the following thread:
thread237-20884

But without final result.

Are there utilities where the demand charge is based on Urms*Irms (which would mean to include distortion power) or
is the billing based only on the fundamental components ?

 

[jbartos]

Suggestion: Hypothetically,
Rx|I|**2=Pwattloss, say a loss power. Now
I=Ir+jIim
Sva=V x I*= Pwatt+jQvar, in VA, (the consumer total power draw)
Pwatt is paid for via meter.
If Iim is compensated and is >> Ir
the |Icompensated|=Ir << |I|
Pwattlosscompensated << Pwattloss, (This means savings!)
Now, what is needed is to see actual figures how much of savings is being made.

 

[busbar]

Filipski and Arseneau at the Canadian NRC have studied kVA/kVAh measurement for power systems to a great degree and it is used in Canada to bill for kVA demand (to what extent I'm not sure.)

With inexpensive, high-performance analog-to-digital conversion, it is a simple measurement/calculation based on per-phase Vrms · Irms, that inherently folds power factor and harmonic content into a simple rate structure.

GE says: “The Data Acquisition Platform (DAP) chip contains six independent, fully integrated analog to digital converters, one for each current and voltage signal. The converters are continuous time, delta-sigma analog to digital converters digitizing each signal 1.68 million times per second. Each input signal has a dedicated converter, multiplexing is not used. The converter over sampling rate of 512 provides approximately 3280 complete sets of decimated samples per second. For a full three element meter, 6 sets of 16-bit samples are processed, providing more than 54 decimated sample sets per line cycle.”

 

[electricuwe]

Busbar,

with a data aquisition as described in your last post, power and apparent power according to every deinition can be measured, depending on the algorithm which is used for calculation.

To base the billing on the product of Irms and Urms would lead to the following:

The utility would sell much more demand that it has generating capability. It would charge both the customer injecting harmonic power and the one suffering from harmonics (e.g. one using capacitor banks)

The customer may be responibsle for the distrorted current but usually only for a small portion of the volatge distortion.

To bill a customer for injecting harmonics is not an alternative to limit harmonics by applying standards like IEEE519 since depending on the harmonic order there are technical limits beyond not only additional losses are generated but equipment is destroyed.

Although it's my business to design power electronic equipment sold all around the world I have no reliable source claiming that a demand bill is based on Irms*Urms.

 

[stevenal]

Electricuwe

My utility charges industrial customers on energy use and billing demand. Billing demand reflects 15 min demand (kW) multiplied by the power factor penalty. The power factor penalty is (.95-average pf)+1 if the average pf is less than .95. Average power factor is calculated from kW-hrs and kVAr-hrs used during the billing period. I believe this billing method is fairly common.

The meters used do not filter out the non-fundamental quantities, but I cannnot claim they are true rms either. The meters meet ANSI standards, which says this about harmonics (C12.1-2000):

A.2.11.4 Variations in frequency
Frequency variations in a modem power system under normal operating conditions are insignificant. Any inaccuracies that might result from variations that occur are entirely negligible. However, the presence of voltage harmonics or current harmonics created by nonlinear loads may cause measurable inaccuracies. In the vast majority of metering installations, the accuracy is still within ±2%. Cases of severe harmonic distortion must be analyzed on an individual basis.

So I think the answer to your question is that distortion power is not intentionally excluded from measurement.

 

[busbar]

electricuwe—you mention that the “utility would sell much more demand that it has generating capability.” Can you elaborate on that?

 

[jbartos]

Suggestion: The consumer would be better off to be paying for the Pwatt consumption, since the harmonics may come from neighboring consumers, and a certain amount of the power factor may come from the utility, e.g. from the overhead (pole) transformer.

 

[electricuwe]

Stevenal,

the way of billing you describe is quite similar to the way its done in Europe.

The classic device for metering electicity, the ferraris disc does not include distortion power since it integrates the product of u(t)*i(t) ( or using a phase shifted voltage for measureing reactive power).

So to my mind the correct statement is:
distortion power is not intentionally included

Electronic devices like the one described by busbar could either exclude distrortion power or include distortion power depending wether multiplication of current and voltage takes place before integration or after integration.

 

[electricuwe]

busbar, let me give a simple example:

Imagine a utility with a 225 kVA generator supplying a customer with 100kW of induction motors and power factor correction to PF=0,95. The customer will have to pay a demand charge for 100 kW an no power factor penalty.

Now a second customer with 100 kW of VSDs is connected, injecting lots of harmonics, with a true power factor lamba of 0,8. Since a lot of the harmonics will be supplied by the power factor correction capacitors ot the first customer his power factor will be reduced, lets assume to 0,9.

Billing based on Irms*Urms will lead to the following result:

The utility could bill the two customers for a demand of more than 225 kVA (100 kW/0,9+100kw/0,8=230 kVA)

The first customer will not only have reduced lifetime of his power factor correction capacitors and increased losses in his induction motors. He will also have to pay an additional power factor penalty for that!

 

[stevenal]

electricuwe

You are correct about electo-mechanical kVArh meters: fundamental voltage is lagged by 90 degrees with a phase shifting transformer so that kWh is actually measured and read as kVarh. Without doing the analysis, I suspect that voltage harmonics don't necessarily shift by 90. Supply systems are pretty stiff and voltage harmonics are limited by standards, so I believe the error is negligible as the ANSI standard states. We are now using solid state meters for these customers. I suspect that these measure kVAh directly using the product of the two integrals.