If everyone turned on their A/C at the same time at night, would it stall the grid?

Status
Not open for further replies.

Electric-Light

Senior Member
Say just for the sake of doing it, everyone in US and Canada adjust the thermostat's clock accurately and set their program to flip the A/C on for 10 minutes right at 8AM GMT sharp (12AM PST to 3AM EST). Would it brown out the grid?
 

iwire

Moderator
Staff member
Location
Massachusetts
:blink:

This has nothing to do with LEDs.

Are you feeling OK?:huh:



I would think yes, it might even crash the grid. Of course I have absolutely no training on this and am just guessing.:dunce:
 

mivey

Senior Member
You would see a voltage drop and lights would dim, at least on many systems. As protection schemes reacted, some voltages would rise. Depending on the degree of synchronization (and it is not just about the thermostat time) you could have everything from lights dimming to a cascading failure of large parts of the grid. I highly doubt we would loose the entire grid.

Factors reducing the impact include things that vary like:
system impedance
protection and isolation zones and settings
A/C load percentage of zone load
device protection
time zones
daily routines
clock accuracy
willingness to synchronize

just to name a few.
 

kwired

Electron manager
Location
NE Nebraska
In areas that are primarily residential areas - it may kick out local overcurrent devices - then again there is possibly even more load introduced when they try to close the device again.

On the rural system I am powered from air conditioning load is nothing compared to the irrigation load in the summer months. Try starting thousands of motors all at one time that most seem to average in the 60 to 100 HP range, compared to only hundreds of 2 to 5 ton AC compressors that will be trying to start at that same time.
 

iwire

Moderator
Staff member
Location
Massachusetts
You would see a voltage drop and lights would dim, at least on many systems. As protection schemes reacted, some voltages would rise. Depending on the degree of synchronization (and it is not just about the thermostat time) you could have everything from lights dimming to a cascading failure of large parts of the grid. I highly doubt we would loose the entire grid.

Factors reducing the impact include things that vary like:
system impedance
protection and isolation zones and settings
A/C load percentage of zone load
device protection
time zones
daily routines
clock accuracy
willingness to synchronize

just to name a few.

For the purposes of this thread maybe we could ignore all those delays? Let's say we time sync using the GPS system time signal. :)
 

Electric-Light

Senior Member
You would see a voltage drop and lights would dim, at least on many systems. As protection schemes reacted, some voltages would rise. Depending on the degree of synchronization (and it is not just about the thermostat time) you could have everything from lights dimming to a cascading failure of large parts of the grid. I highly doubt we would loose the entire grid.

Degree of synchronization being every programmable thermostat in nation put on schedule to turn on the A/C at 8AM GMT (during off peak) and assume the clock has been adjusted the day before so that sigma is no more than about +/- 15 seconds. My understanding is that mechanical systems(boiler, water vanes, what not) of power stations can not vary generation capacity quickly.

Let's say Sunday morning at 8AM GMT.
 
Last edited:

mivey

Senior Member
For the purposes of this thread maybe we could ignore all those delays? Let's say we time sync using the GPS system time signal. :)
You could have everything from lights dimming to a cascading failure of large parts of the grid. I highly doubt we would loose the entire grid.
 

Electric-Light

Senior Member
In areas that are primarily residential areas - it may kick out local overcurrent devices - then again there is possibly even more load introduced when they try to close the device again.

On the rural system I am powered from air conditioning load is nothing compared to the irrigation load in the summer months. Try starting thousands of motors all at one time that most seem to average in the 60 to 100 HP range, compared to only hundreds of 2 to 5 ton AC compressors that will be trying to start at that same time.

But it's not at Sunday at 12AM PST and 3AM EST (8AM GMT) which would be like putting a car in 4th gear at idle and punching the gas hoping you don't stall it. In that time frame, I'd think the national grid as a whole is closer to idle.

If you go from 0 to 2,500W load on a 5,000W generator, it would bog down and lose frequency for a bit as the throttle catches up. But, what if you do the same to a 2,500MW steam driven generator?
 
Last edited:

iwire

Moderator
Staff member
Location
Massachusetts
You could have everything from lights dimming to a cascading failure of large parts of the grid. I highly doubt we would loose the entire grid.

I am not disputing you but am curious how the grid could sustain that?

It seems there would be no time to ramp up.

Normally the grid would not even have to cover the running load of every AC in the US at the same time and here we are talking six to eight times that load are we not?



Softer gentler.
 

Barbqranch

Senior Member
Location
Arcata, CA
Occupation
Plant maintenance electrician Semi-retired
Do I have to open all my windows at once to match this condition, or just one or two? (That is the only air conditioning most anyone has, here on the North Coast of CA:p)
 

iwire

Moderator
Staff member
Location
Massachusetts
This makes me remember that at a commissioning one time the company had to do across the line starts of four - 500 hp motors (not at the same time) each driving very large fans at a facility located in a remote area of RI (well as remote as one can be in a state that size)

It was enough of a disturbance that a power company rep drove out to the site to verbally assault our foreman for some time.

Normally the motors where VFD controlled but for commissioning we had to use the bypass to start them. These fans are critical to the facility and they wanted to make sure a VFD failure would not side line them
 

jim dungar

Moderator
Staff member
Location
Wisconsin
Occupation
PE (Retired) - Power Systems
My understanding is that mechanical systems(boiler, water vanes, what not) of power stations can not vary generation capacity quickly.

While this is true, I am not aware of any large scale effect on the 'national grid' simply due to an sudden or even instantaneous demand for power, however localized outages and fairly widespread brownout conditions are well documented.
 

mivey

Senior Member
Degree of synchronization being every programmable thermostat in nation put on schedule to turn on the A/C at 8AM GMT (during off peak) and assume the clock has been adjusted the day before so that sigma is no more than about +/- 15 seconds. My understanding is that mechanical systems(boiler, water vanes, what not) of power stations can not vary generation capacity quickly.
What varies first is things like the system voltage and spinning loads (like grid-tied motors) and frequency. Then high-speed relays kick in and other things start coming in. This is at the sub-second level. After that, things start to happen at the seconds level and a lot of generation can react that quickly. Other generation takes minutes. The problem is controlling the oscillation as relays, loads, and generation react and trying to limit the range of cascading failures.

In a simplistic view, with a massive load addition, frequency relays will dump loads on parts of the system and the A/C systems on the rest of the system will be left to start up.
 

mivey

Senior Member
It was enough of a disturbance that a power company rep drove out to the site to verbally assault our foreman for some time.
If it is any comfort, you were further down the food chain. His manager had probably already been verbally assaulted by other customers and/or system operators. Then it flowed down hill to your foreman.
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
I can't imagine that adding a hugh instantaneous load like that would not have at least as much effect as the loss of transmission lines that caused the 2003 north east black out.

Even the rapid automatic load shedding by the grid control system could result in overspeed trips for the generators that were trying to supply the load when the protective system dropped the load.
 

Electric-Light

Senior Member
What varies first is things like the system voltage and spinning loads (like grid-tied motors) and frequency. Then high-speed relays kick in and other things start coming in. This is at the sub-second level. After that, things start to happen at the seconds level and a lot of generation can react that quickly. Other generation takes minutes. The problem is controlling the oscillation as relays, loads, and generation react and trying to limit the range of cascading failures.

In a simplistic view, with a massive load addition, frequency relays will dump loads on parts of the system and the A/C systems on the rest of the system will be left to start up.

How low is the Hz allowed to fall to reduce motor kW load in order to avoid a black out?
Would you say customer elected PoCo addressable power switches are of much value in system reliability improvement?
 

mivey

Senior Member
I can't imagine that adding a hugh instantaneous load like that would not have at least as much effect as the loss of transmission lines that caused the 2003 north east black out.

Even the rapid automatic load shedding by the grid control system could result in overspeed trips for the generators that were trying to supply the load when the protective system dropped the load.
Supposedly we have addressed some of those oscillation issues and cascading failures. Time will tell.
 

mivey

Senior Member
How low is the Hz allowed to fall to reduce motor kW load in order to avoid a black out?
There are different levels. I would have to go look as I can't remember if it is 0.2 Hz or 0.7 Hz but it is something like that. The duration is also a factor.
Would you say customer elected PoCo addressable power switches are of much value in system reliability improvement?
For this scenario, after a period of time they could be, and certainly for restoration. The first load dropping reactions happen at the primary voltage level.

In general, we would rather fix the system to handle normal load variance rather than drop load to improve general reliability. Load management is a different discussion.
 
Status
Not open for further replies.
Top