DISCLAIMER: What this Page is NOT About
This page presents much of the theory behind paralleling portable generators. It is stated up front that this page is NOT A COOKBOOK BUILD YOUR OWN. While the principles behind the practice are simple, there are too many ways to go wrong with this when you build your own. You do not want to be burning out your well pump all because of a power outage and your inattention to details. In addition, there are always typographical errors in all publications which includes this one. However, with the knowledge explained in this page, you will gain the simple understanding you need to figure out for yourself what you want to do. But if you feel vague on what is going on…don’t do it.
Motivation for this Page
The author hunted around on the internet for information on theory behind paralleling portable generators and found virtually nothing explicitly devoted to that topic. Information can be gained by looking for knowledge on generic motors and generators that is relevant to paralleling portable generators but the discussion is mainly academic and very theoretical.
The purpose of this web page is to present the gut-level theory behind paralleling portable generators without going into a deep discussion of motor and generator armature physics and geometry.
The author recommends that for paralleling generators, you buy generators that are designed for and include optional paralleling kits. This takes the guesswork out of the process. It becomes plug and chug and you are kept from shooting yourself in the foot.
Limitations of Manufacturer Generators Designed for Paralleling
But you accept the limitations of manufactured paralleling in doing so. The primary limitation is that the manufactured kits and generators allow for paralleling only up to two generators. With a home-brew arrangement, there is no limit on how many portable generators may be run simultaneously in parallel other than how many you allow for in your build design.
Let’s Talk About the Questions
We have two or more generators that are designed to operate at the same frequency (60 Hz or 50 Hz) and the same peak amplitude. Any portable generators in such a configuration MUST meet these cooperative specifications. But no two generators, even of the same make and model, will be perfectly identical. There are manufacturing tolerances and different operating histories to consider accounting for differences. Therefore, even if fine-tuned to run at the same frequency, there will be minor differences in frequency and amplitude. The obvious question is, what happens when you parallel two or more such generators?
But consider an expanded question. We already know about matching frequency and amplitude. What happens when you parallel two generators that differ in make, model, and even wattage capacity? Can you parallel, for example, a Honda 10 kW 60 Hz generator with a Generac 2 kW 60 Hz generator? Answering this question leads us into the following discussion.
What Must Match
When paralleling two generators there are three fundamental requirements that must be met.
- Frequency — Both generators must be designed to produce a nominal 50 Hz or 60 Hz. Typically, they will be set from the factory to operate a couple of Hertz above the nominal so that under load they will statistically deliver the nominal 60 Hz. However, improved performance can be realized if any two generators that are to be paired are fine-tuned to produce the same frequency under no-load.
- Magnitude Voltage — Both generators must produce roughly the same magnitude voltage out. Again, as with frequency, performance will be enhanced if the no-load voltages can be fine-tuned before paralleling.
- Amperage — It is recommended to match Ampere capabilities of generators to be paralleled. This is not a big bite to swallow since most generators commonly used to power an entire household grid are all rated for 30 Ampere outputs. Carefully note, however, that this does not limit wattage ratings. Wattage ratings can differ as described below.
- Capacity or Wattage Ratings — You may use different capacity models in parallel, so we are not talking about matching the power ratings. However, the two (or more) models will share the load at a level equal to the lowest capacity generator. One generator can be 10 kW and another 6 kW, and they will share the loading fine. However, the maximum load they will supply will be 12 kW or 2X the smallest generator capacity. Carrying this to a silly extreme, you could parallel 10 kW and 2 kW generators but together, they will only supply 4 kW. It therefore makes sense to match wattage ratings between shared generators, but there is no requirement to do so.
Requirements Commonly Misunderstood
One requirement is commonly misunderstood–Capacity or a matching Wattage. When hunting around on the internet for paralleling generators you will find conflicting reports on Wattage capacity. If you think this one through you can be confident that you have a proper understanding.
The following discussion applies to armature driven portable generators as opposed to “inverter” generators. When mixing an inverter into the mix there is a different ball game. An “armature” generator generates potential by rotating a set of wires through a magnetic field. A gasoline/butane/diesel motor turns the armature. An inverter uses a DC battery to “chop” an AC power source.
Think basic electronics: what pushes energy? It is potential that pushes energy. Two generators operating in parallel will produce roughly the same magnitude voltage, but not exactly. This aspect of performance can at least be optimized if the no-load voltages of the two generators can be fine-tuned. The generator pushing more will see the other generator as a miniscule portion of its load creating tiny circulating currents. Circulating currents are a loss but cannot be completely eliminated, there will be some, though often irrelevant. But this loss therefore causes the load (the household grid) to see one voltage. One generator will supply slightly more of the current going into the main load than the other generator.
What is Synchronization?
Synchronization is where we consider the θ of the equation
Theta, for the present case, represents a phase difference in radians. But for illustration purposes, you can think of theta in terms of degrees.
Consider the illustration given in Figure 1. Here we have two sine waves separated by a θ of 180o. Notice that both VA and VB cross through 180o together but from different paths. To understand how this plays out in real life, consider a part of the illustration’s history not shown. What happened before time zero? Suppose both phases were switched from two dedicated switches. Switch A was closed first and then after a length of time, Switch B was closed. They are like runners in a field and track race where each started at different times but run at the same speed (frequency). For simplicity of analytics, we call this a “phase difference.” Do not let a silly term like “phase” intimidate you. Phase is a simplification for your benefit.
Two Generators Side-by-Side
Let us assume that two generators are already synchronized but not yet connected to run in parallel. Because no two generators (even of the same make and model) will be exactly the same, the two will tend to drift apart in phase while each maintains its own consistent frequency. For example, one would like to run at (by chance) 60.0000 Hz and the other at 60.219877653 Hz with no loads. They are both designed to run at a generic 60.0 Hz under a 90% load but this is not a perfect world. At any one particular instant in time, both are in perfect synchronization with a zero phase difference. Looking again at Figure 1, the two phases A and B would be superimposed over each other with perfection when there is a zero phase difference. Shown on an oscilloscope in transient operation, we would see the two phases slowly drift apart and then back in phase as a continuous process. The process will continue as long as the two generators are running and being so observed.
What is the Significance of Zero Phase To Us?
Magic things are available to us if we can positively identify any one instance of zero phase difference in a transient display. A zero phase difference is the Holy Grail for us. With a zero phase difference, the algebraic addition of the signals is 100% positive. Neither subtracts from the other and both work together in harmony to meet demands of the load.
Worded another way, when there is any phase difference between the two, one of the generators will appear to the other generator as a second load. That generator, for that particular instant in time, will think that it is servicing two loads. In our horror, when a phase difference is 90 degrees, the supply generator will probably be trying with all its might to meet the “demands” of the second generator leaving almost nothing for the intended load. We are almost looking at blown circuit breakers and possibly an event that is not beneficial to both generators and appliances in the house. Probably, a good single word to describe such a configuration is “uncool.” It goes without saying that uncool is not good. More will be said about this in the section where we discuss what can go wrong.
In Parallel and Side-by-Side
Let us continue the above discussion but now with the two generators hooked in parallel but with switches to allow or disallow engagement to the grid.
Apply the above discussion to Figure 2 with the switches open. But now 120 V lamps have been added. We choose lamps rated for 120 V because the potential differences are between two like phases. The lamp will exist between phase A of generators 1 and 2.
There is no applicability between phases A and B of generator 1 or generator 2. If we were to jump across phases A and B of generator 1, this would require a 240 V-rated lamp.
Looking at Figure 1, what happens to a potential difference between phases A and B when they are in perfect synchronization? Clearly, there is no potential difference because they are synchronized with a zero theta phase difference. The lamp cannot glow, it is dark (the Holy Grail).
Let us now suppose that one phase has drifted slightly. There is now a potential difference between the two phases. That voltage difference may be enough for the lamp to begin glowing. But the phases continue to drift becoming 90o apart, drivng the lamp to its maximum brightness. This is a maximum voltage. As the signals continue to drift, the potential difference becomes less until a zero phase difference and a dark lamp.
What Does the Lamp Therefore Mean?
The lamp is your indication of phase synchronization. You want to flip the switch when the phases are aligned so that the two generators will not know that anything has changed. You watch the lamp and observe how it continually drifts in and drifts out. You are then able to positively identity an instance of zero phase and electrically engage the two generators.
But the Lamp is Blinking, not Drifting?
If the phases are drifting in and out so fast that it appears as a flicker, you should consider changing something. It is imperative that the two generators run at roughly the same frequency before synchronization. If you work out the math, with only 1 Hz difference between the two generators you will have only a fraction of one second to make the connection.
If you were to throw the switch with the light flickering, you might miss the sweet spot enough to cause a serious energy exchange between the two generators and trip the generator’s circuit breaker. While this is not a very bad thing, it most certainly is not beneficial to the generator’s longevity and should be avoided if possible. Most likely, one of the generators is already under load and that is what should be changed. It is, of course, possible to actually cause damage so it is worth spending time to get it right.
I’ve Thrown the Switch. Now What?
You have thrown the switch and the two generators are now running in parallel radically increasing the power available to the load. Everything looks (and probably is) fine. But the two generators have not forgotten that we do not live in a perfect world. One will still begin to drift relative to the other, though too small to easily measure. When that happens, the rules continue to apply. One of the generators will want to drift ahead of the other or vise-a-versa but the sinking of one generator will be the pull of the other forcing them to remain in synchronization.
Drifting? What Does That Do?
This machine attempted drifting creates a potential difference within what is supposed to be one electrical phase. While each generator has a phase A and B, this is now a battle, for example, between generator 1’s phase A working against generator 2’s phase A. The two generators will trade consuming power between themselves which is called “circulating currents.” While this is a genuine loss of efficiency, it is minimal because the two keep each other in check keeping potential differences almost unmeasurable.
For this reason, it is beneficial if the two generators can be adjusted for equal frequencies and voltage under no load when they are not paralleled. If you can find the technical instructions for your generators, it will be a worthwhile exercise in adjusting them for equality under no load. But don’t blindly set them for 60 Hz. They are likely set a little bit above 60 Hz to maintain that margin, typically 63 Hz. The generators will check themselves with regard to frequency but voltage differences will contribute to overall lost efficiency.
There is an additional component to consider in circulating currents. That is the third harmonic. The third harmonic is simply a characteristic of any one particular generator that is intrinsically attached to its design. This is not necessarily a bad thing but does take away from overall efficiency in the parallel operation. When looking for generators to purchase, pay special attention to generators that are promoted as having a low THD (total harmonic distortion). The third harmonic will be totally eliminated by design on these models.
Drifting Kept in Check
The drifting is being kept in check by the generators themselves. Therefore, the generators are free to share the load demands effectively increasing the load capacity the paralleled generators are able to deliver.
In order to visualize the theoretical bi-modal nature of the portable generator, we imagine a generator whose armature can be turned by applying AC mains from the national electric grid or by a small gas engine. In this way we can better comprehend what is going on inside the generator-motor.