Harmonic Study for 40-60% VFD loads

[controlnovice]

We'll be installing a new batch process where 40-60% of the load will be VFD driven.

We've installed VFDs before (just 1-4 per site) and usually just put a load and/or line reactor without much study.

Now that there will be so many (10 - 14, largest 100hp), is a harmonic study warrented? What should we expect to see from a study?

______________________________________________________________________________
This is normally the space where people post something insightful.

 

[mls1]

It's probably not a bad idea given the high percentage of harmonic producing equipment you'll be putting in. A study should look at IEEE 519 requirements and provide you with a complete harmonic characterization of your system. This looks at reactive and resistive elements and determines resonant conditions. For example, long feeder cables can create a parallel LC circuit with the supply transformer inductance. This may have a resonant point that corresponds to one of the harmonic currents from the VFDs. If it's bad enough, even a relatively small harmonic current can cause excessive voltage distortion.

It can also provide recommendation for diversifying the harmonic offenders, for example using transformers with different phase shifts to reduce cumulative effects, or properly tuned filters (i.e. the fifth and seventh may not be your biggest problem).

 

[EddyWirbelstrom]

Section 10.6 of 'IEEE Recommended Practice for Industrial and Commercial Power System Analysis' recommend a harmonic study when power factor correction capacitors are present in a system where 20% or more of the total load is comprised of converters or other harmonic-generating equipment.

A harmonic loadflow study requires information on the magnitude of the orders of harmonic current injected into the system by the non-linear loads.
VFD suppliers publish the per-unit magnitude ( based on VFD rating ) of harmonic current orders injected into a power system.
Does anyone have experience with how system impedance and VFD utilization affects the magnitude of current injected into the power system by the VFD ?

 

[ozmosis]

controlnovice
yes, you should certainly look into a harmonic study. There are a number of factors, and it also depends on where you are in the electrical world as different countries have different regulations.
All of the regulations around the world (IEEE519, G5/4 etc) are planning regulations. The term 'planning' is deemed a process you go through before you install non-linear loads.
If your PCC (Point of Common Coupling, i.e. the point where you connect to a public network)is the same as where you are saying your non-linear load is 40~60% of total load, then somebody somewhere could stop you from connecting any further and start demanding you do something to minimise the THD levels.
Looking at Current Distortion:
The current distortion limits of IEEE 519, expressed as a percent of the demand current, control the amount of harmonic current each user may impose on the utility system. The level of harmonic current imposed on a system is not a measure of risk of disturbance to loads on the system. It is the harmonic voltage generated as the harmonic currents flow through the system impedance that imposes a disturbance risk to the loads on the system. Applying current distortion limits within a user's own network is therefore meaningless. The key factor is the amount of harmonic voltage distortion produced.

* The IEEE current distortion limit value depends on the system voltage and the ratio of the short circuit current to the demand load current. If the ratio is larger, the limits become higher. Increasing the short circuit current will increase the ratio. When the ratio exceeds various IEEE thresholds the limit values are increased. Decreasing the load demand current will also raise the ratio however this will cause the calculated value to increase which may counter the desired effect (see the next point).
* The calculated value of harmonic current distortion is expressed as a percent of the demand load current. If the demand load current is increased, then the calculated value of current distortion will be decreased. If the demand load current is increased too much, then the IEEE limit value will be decreased due to a reduction in the ratio of short circuit current to demand load current (see the first point).
* The value of demand load current is the sum of the drive motor loads plus the other loads connected to bus 2 and if used bus 3. Adjusting the amount of other loads that are connected to these buses will change the demand load current.
* The harmonic current distortion is a measure used by the utility to limit the amount of harmonic current a user may impose on the utility and is not a measure of distortion risk to loads connected to the system. For this reason, the limits should only apply to the IEEE defined point of common coupling. The point of common coupling is defined as the point where the user connects to the utility where other users are also connected. Applying the Current Distortion Limits within a plant is not meaningful and in some cases not practical. For instance, if an internal plant transformer supplies only drives, then the demand load current at that point will be only the drive load current. If the drives were to produce 20% fifth harmonic current as an example, then the calculated value of 5th harmonic current distortion would also be 20% which is beyond the IEEE 519 limits for the PCC.
It is the harmonic voltage at a given point in the supply network that determines the risk of disturbance to loads at that point. The voltage distortion limits of IEEE 519, expressed as a percent of the fundamental voltage, control the amount of voltage distortion allowed at the point of common coupling.

Voltage Distortion:
* Harmonic Voltage Distortion is caused when harmonic current flows through the impedance's of the power system and causes harmonic voltage drops. The harmonic voltages distort the fundamental sine wave voltage and if excessive, can cause other equipment connected to the voltage bus to malfunction. The amount of voltage distortion is a measure of the distortion risk to loads connected to the system.

* Ohm's Law calculates the harmonic voltage as the product of the harmonic current times the impedance it flows through. In order to reduce the harmonic voltage distortion it is necessary to reduce the amount of harmonic current flowing or to reduce the impedance it flows through. Using SED2 drives will reduce the amount of harmonic currents generated. Adding line reactors can further reduce the amount of harmonic current generated by a drive.
* Using lower impedance transformers will reduce the impedance and result in lower harmonic voltage distortion.

So, back to my point of "Planning". Your VFD supplier should have tools available to allow calculations of the THID and THVD. This will give you an idea of where you stand. If the VFD's are already installed, then depending on where you are in the network, I advise you to do some measurement and determine if there is a high level of distortion before somebody else does.
mnewman
Your system impedance has everything to do with the voltage distortion.