21
\$\begingroup\$

I have been monitoring the voltage in my own home (Mexico 120V 60Hz). I have been graphing the values every minute and I'm starting to notice that during the morning voltage gets up to 127, and during the day (4pm) it can go as low as 118V — why does this happen?

The effect is quite similar to when I turn on the microwave for short periods, however these variations from 118V to 127V happen slowly during the 24 hrs of the day.

enter image description here

\$\endgroup\$
  • 4
    \$\begingroup\$ Someone explained to me that it's like when water pressure drops when too many taps are turned on (in the day perhaps) and it is okay with only a few taps running (in the night). Made sense. \$\endgroup\$ – Sohail Nov 23 '15 at 6:33
  • 2
    \$\begingroup\$ This question reminds me of the case of the electricity grid in the UK where operators watch TV to detect when a popular TV show goes to advertisement and housewives all over the country collectively turn on their electric kettles: youtube.com/watch?v=UTM2Ck6XWHg \$\endgroup\$ – slebetman Nov 23 '15 at 10:01
  • 1
    \$\begingroup\$ As a side note, if running the microwave results in significant voltage loss - for instance, maybe less than 105v, the entire time it is running - you may want to check the exterior portions of your electric service for rodent damage. In South Texas we have a problem occasionally with squirrels chewing on the neutral line, creating a high resistance connection. \$\endgroup\$ – Sean Boddy Nov 23 '15 at 16:40
17
\$\begingroup\$

TL,DR; What you are seeing is primarily the result of reactive load droop sharing, with secondary loading voltage drops in the transformers and lines.

When generators run in parallel like they do to supply national grids, there are two mechanisms at play that largely go unnoticed by the local population.

Perhaps the most critical one is speed droop load sharing. It's actually pretty clever and fairly simple - every generator in parallel is programmed to have the same, predictable slow-down from no load to full load. This way if something large starts on the grid, all generators respond roughly equally, sharing the load in proportion to their full load ratings.

There is a bit more to this, of course. On a big grid, some generators like nuclear plants will be base load stations, and will hardly respond at all to these changes. On the other extreme are peaking stations like natural gas turbines, which can easily react very quickly to truly large shifts in grid loading. This kind of a set up makes the best use of both resources.

But this speed regulation trick doesn't help with reactive loading. In brief for those uninitiated, reactive load consists essentially of those back-and-forth exchanges of current between capacitive and inductive loads that do not result in the delivery of real power.

But we can use a similar trick by using a voltage droop regulator on the output of every generator on the grid. From no-load to full-load, every generator will have roughly the same drop in output voltage. People live and work in daily cycles, and as a whole bunch of inductive, motor operated loads start up, the grid voltage comes down a bit to keep things balanced. This usually hits hardest in the early evening, with a noticeable bump around breakfast time (usually).

These two systems work in concert to ensure that no single unit on the system ever absorbs an amount of real or reactive load that it can't handle (at least until something really unexpected happened). Of course, additional losses are incurred in steps between you and the grid, but I cannot give you exact proportions without detailed equipment information.

\$\endgroup\$
  • \$\begingroup\$ the voltage droop is due to increased current - it doesn't have to be reactive - resistive load will do it too. It doesn't happen 'to keep things balanced' - it's just Ohm's law. I don't think that generators are programmed as you describe. The generators stay in synch because they're all on the same grid and frequency drops due to loading (same as revs drop in your car when the load increases). You can increase the input power to compensate but at some point the generators have no more to give and the operators let the frequency drop a little. \$\endgroup\$ – Transistor Nov 23 '15 at 19:09
  • \$\begingroup\$ @transistor, I know for a fact that generators have this characteristic voltage droop, because I have seen it in action and manipulated it. Changing the voltage regulation setpoints on generators in parallel results in a change in apparent power with no change in real loading due to shifting the reactive load, and this is handled with a voltage droop feature. Stations on the line all try to balance reactive load because they don't get paid for supplying KVAR, just KW. \$\endgroup\$ – Sean Boddy Nov 23 '15 at 20:29
  • \$\begingroup\$ Thanks for the feedback. It prompted some research and I found an excellent article on esrdc.mit.edu/library/ESRDC_library/VR_parallel.pdf. \$\endgroup\$ – Transistor Nov 25 '15 at 19:53
6
\$\begingroup\$

Your power utility will generally try to generate at constant voltage. The power distribution network of power lines and transformers will have some resistance and this will cause voltage drop along the network. The voltage drop will be proportional to the current drawn and this varies through the day. From your voltage monitoring you should be able to determine periods of high use in your locality. (This my be not be so easy if the utility company switches on and off peak-time generators.)

The other thing that may happen is that the grid frequency may drop slightly from 60 Hz (in your area) during peak loading. If this occurs then the utility company has to run slightly faster than 60 Hz during periods of light load to make up for this otherwise any mains-powered clocks will lose time. That is, they will track the mains frequency and lose and gain time during the day but on average they maintain their accuracy.

\$\endgroup\$
  • 1
    \$\begingroup\$ Clocks that rely on grid frequency are rare enough these days that many utilities no longer directly worry about time error correction (keeping the average frequency at precisely 60). They usually focus more on statistics like ACE (Area Control Error) that indicate whether they are generating the amounts of power they have agreed to produce. \$\endgroup\$ – Carl Kevinson Nov 23 '15 at 22:38
3
\$\begingroup\$

Most likely your utility does not perfectly regulate voltage. The voltage is 120 V nominal; in practice devices would be designed to tolerate a range of 110-130 V or more. If you happen to be near a large industrial load (factory) that could explain some of the variation. Some power plants are easier to regulate than others.

There are effects that will cause a local voltage drop within your own home, but the variations you are seeing are greater than that would produce unless you have serious wiring problems.

\$\endgroup\$
  • 4
    \$\begingroup\$ The largest component in a residential area is likely to be air conditioning. It is cool in the morning and hot in the afternoon, plus everyone is getting home from work. \$\endgroup\$ – Austin Nov 23 '15 at 6:04
1
\$\begingroup\$

The simplest answer is that the power demand is less when people are asleep. Less power demand means less current demand. Less current demand causes lower transmission loses, this allows the voltage to go higher (towards its "no load" value).

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.