Generator operating frequencies. 27

[HamburgerHelper]

I have thought about this for awhile and I don't understand it. So, let's say you have generation spread out over a large region and there is a disturbance in the system that causes one of the generators to be at a lower or high frequency than the rest of the generators in the system and the controls don't work to bring that generator back up to normal frequency. What happens? I have a hard time understanding this because in my mind if a generator is operating at a different frequency than the rest of the system, that generator or island around the generator is effectively isolated from the rest of the system from a power flow perspective. The rest of the grid is going to try to motor or add generation to it as the phase angle of the different frequency generation slips around the rest of the grid. I just have a hard time grasping why a generator can operate for example at 59 hz while the rest of the grid is humming along at 60 hz.

 

[waross]

Think of a two horse hitch. One horse is traveling at 8 miles per hour. The other horse is traveling at 8.2 miles per hour.
If there is 2 feet of swing in the double tree, how far can they go before the faster horse is dragging the other horse?
The faster horse can only get a few feet ahead of the other horse until there is a wreck.

Or consider an old power skid with two diesel engines geared together at the same ratio to drive one output shaft.
One engine is running at 1800 RPM, the other engine is running at 1750 RPM.
Something's gotta break.

Two generators are running in parallel. They can only get a few mechanical or electrical degrees out of step before bad things happen.
If one generator is running at 60 Hz and the other generator is running at 59 Hz, there will be the equivalent of a phase to phase fault once a second. For a few seconds anyway.

Two parallel generators will be running at exactly the same frequency. One may be a few degrees advanced ahead of the other, rather than matching position exactly.
Then, if the sine waves are superimposed one sine wave will be a few degrees out of step with the other, but at the same frequency.
The generator that is leading will be producing a greater share of the power. That is how loading between generators is controlled.
But, at a 30 degree difference between generators running at the same frequency, breakers will be tripping.
I have seen operators close a synchronizing breaker with a 30 degree difference (faulty synchroscope connections).
The breaker trips free instantly. On that size and type of breaker an instantaneous trip is caused by 10 times rated current.
30/360 = 12, so 1/12 of a cycle out of step is enough to start the light show.
When the grid frequency changes slightly, all the sets change together. The slower responding sets may lag a few degrees while the faster responding sets pick up a greater share of the load.
We are talking about a few percent of a cycle offset at the same frequency.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Mbrooke]

"Two parallel generators will be running at exactly the same frequency. One may be a few degrees advanced ahead of the other, rather than matching position exactly.
Then, if the sine waves are superimposed one sine wave will be a few degrees out of step with the other, but at the same frequency"


Perhaps, but what if these were miles apart? Won't the frequency be slightly different at each generator?

 

[marks1080]

One system frequency.

 

[marks1080]

Sorry... should have added:

IN A STABLE SYSTEM

 

[Mbrooke]

Ok- noted.

 

[cranky108]

Scotty, I agree to a point.

The most sensitive units might not be the fastest at overall reaction. I should have said most sensitive controls and reaction timing.

 

[DanEE]

davebeach quiteThe phase angle variation produces a quasi-stable situation. The angle variations need to be accounted for but units with differing angles can all be treated as having the same frequency. As long as nothing goes unstable the frequency of the various units keep together even while individual units rock back and forth relative to the average frequency.

I've also been told that power companies will sometimes bring generators online at a phase angle to correct for power factor in the distribution network.

In regard to several posts regarding phase angle measurements throughout the network, with the extreme accuracy of the reference clocks used to drive GPS satellite timing, GPS reference receivers are commonly used for many different types of precision measurement and control e.g. not only for locking RF transmitters and receivers to precise frequency control, but also far more demanding locking into phase synchronized, geographically distributed coherent radar systems, and of interest to this thread, phase angle measurement at various points in the power distribution network.

Although this article points out a possible vulnerability that could occur with GPS signal spoofing...

http://gpsworld.com/wirelessinfrastructuregoing-against-time-13278/

I can't imagine a power company using a GPS signal directly as opposed to using it as a long term corrective factor to some multiple number of independent on site high precision frequency reference standards and would think these systems would be alarmed to the hilt for any irregularities. Even basic antenna hardware is available as a countermeasure to the possibility.

http://gpsworld.com/orolia-gpsgnss-passive-anti-jam-antenna-offers-horizon-blocking/

Precision reference standards have been around for a long time..

http://www.virhistory.com/navy/freq-equip.htm

 

[HamburgerHelper]

The more I read about frequency control, the more it becomes clear that generation collectively reacts very quickly to correct for frequency deviation. If you happen to be off by 0.1 hz, you have at most 5 seconds for other generation to increase their power output to prevent you from slipping poles. I imagine that if you have a strong disturbance like losing a unit, everything nearby is told immediately to go balls to the wall to ramp up generation. I suppose any lost in frequency also helps with reducing the load imbalance.

What are the ramp rates of various forms of generation?



This I thought was kind funny. Kosovo has been running area imbalances and has been slowing down the clocks in europe this year by as much as 6 minutes between January and march.

http://fortune.com/2018/03/08/kosovo-serbia-energy-grid-frequency/

 

[waross]

Power companies increase the energy into the prime mover to increase the power (kW) output. This may be more steam in a steam turbine, more water in a hydro unit or more fuel in a gas turbine.
The increased power input moves the output a small angle ahead of the other units.
Utilities increase excitation or voltage setting to correct power factor.
Everything runs at the same frequency, regardless of distance.
I have a short SUV. On a straight road, the front wheels turn at the same miles per hour as the back wheels.
Today I just bought a new truck. It is very long (33 feet overall). How long does it have to be before the front wheels can run at a different MPH than the back wheels? (I left it at the frame shop. It will be shortened to a more usable length.)
On the way home, I saw a turnpike double; a semi-trailer with a 53 foot trailer pulling another 40 foot trailer.
The total overall length was probably over 100 feet. Is that long enough that the front wheels can run at a different MPH than the rearmost wheels?
Sometimes size, or distance for that matter, doesn't matter.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[davidbeach]

Sometimes, though, the left wheels turn a bit slower than the right wheels while at other times the right wheels turn a bit slower than the left wheels. Most of the time they're pretty much the same speed and even themselves out in the long run.

 

[waross]

grin.
And to be fair, I have seen and heard of a couple of cases where the front and back wheels did turn at quite different speeds.
It was after a mechanic installed the wrong ratio differential in one end of a four wheel drive vehicle.
Oh and once when a Wiggle Wagon somehow got the front axle in high range with a rear axle still in low range.
Of course it only lasted for a few feet on a hard surface before disaster struck.
I don't know.- Maybe something like parallel generators running at different frequencies, for a very short time (counted in cycles, on the fingers of one hand) before disaster strikes.

Wiggle Wagon.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[HamburgerHelper]

I read up on frequency generation and talked with a guy familiar with ramp rates. The numbers he was giving me was around 5 to 50 MW per minute I believe. That means when you have a unit trip out, I guess, you are relying heavily on rotational inertia to hold up the system before you can get your generation up to pick up the MW deficit. And because the ramp rates are so slow, the deficit has to be picked up by a whole bunch of units to prevent whatever region or generator from going unstable. I suppose that when your system starts slowing down in frequency, the load decreases as well and helps lessen the problem kind of like how low voltage ,too, helps reduce load.

I thought this was kind of relevant but Kosovo has been slowing down the European grid by causing energy imbalances.

https://arstechnica.com/tech-policy...olved-lost-6-minutes-returned-to-oven-clocks/

 

[waross]

Search this site for "droop".
In a generation system, one and only one set will be the swing set.
The swing set sets the frequency.
All other sets will be running in "droop" mode.
Droop mode allows the sets to follow the lead of the swing set.
Droop mode also allows the grid to continue to operate should the swing set be lost.
There will be frequency deviations with load changes until the swing set is restored to service.
These changes will typically be less than 1 Hz. but in the event of the complete loss of load on a set the frequency deviation will typically be limited to about 3 Hz. That is the limit for a drop from full load to no load. Most load changes will result in much less frequency deviation.
I was system engineer for a time for a very small island utility. We initially had a combined capacity of 2.2 MW with five sets.
The system has always been run in droop mode without a swing set.
A sudden, heavy load may pull the frequency down on the whole grid system, not just on the closest sets.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[cranky108]

On the grid, not all units can pickup/increase generation. Those are called fixed, or renewables. Other units will run below the max capability for what is called spin. Other units will be hot, but remain off-line (or additional spin) for a quick start.

Yes with a decline in frequency, the load will also decline some.

It's the concern of generation ramp rates that renewables are disliked in the utility industry.

 

[bacon4life]

Frequency events are MUCH slower than voltage changes. For example, the image below shows large frequency excursion on Feb 17th. Note that the time scale is in seconds, with a time to reach the minimum frequency of about 6 seconds. The faults that typically initiate events like this usually take less than 0.06 seconds (~4 cycles). The average rate of frequency change for this event was about 0.03 percent per second.

DanEE-I can certainly imagine utilities being a decade behind in responding to the thread of spoofed GPS signals. Although atomic clocks were available, they are an order or two in magnitude more expensive. There are an awful lot of GPS receivers that were installed for informational purposes like time stamping relay event reports. As the GPS signals are hooked to other things like "information" PMU's or provided SNTP signals, it can be hard to tell which step should trigger the upgrade to an atomic clock.

 

[ScottyUK]

HamburgerHelper -

Loading rates are very dependent on the type of prime mover and in the case of steam sets also the capability of the boiler plant. The fastest loading rates are the hydro schemes - for example the big pumped storage scheme at Dinorwig, Wales can go from sync idle to 1.7GW in less than 20 seconds (all six machines pre-synchronised and spinning in air), or from standstill to full load in just over a minute per machine. No thermal plants can match that kind of loading rate, certainly on a percentage basis.

 

[crshears]

I think this is possible if the condition is spread across a very wide range... Ie Quebec say 60Hz, Ontario at 59.87...

This may not be well-known, but [other than for one exception which I will explain later] the Hydro-Québec / Trans/Énergie system is only connected asynchronously to the rest of the Eastern Interconnection, i.e. outside of Ottawa, Ontario via the AC-DC-AC converters at the Outaouais station, between Québec and the state of New York at Chateauguay, etc.

This is not to say that synchronous connection has never been made, but it has always been inadvertent and typically will not last much more than five minutes.

The worst case scenario is that, because the connection is unstable, the flow through the tying circuit oscillates ever larger, eventually crescendo-ing to an automatic trip on overload. The other much more preferable scenario is that the inadvertent paralleling of the two systems is recognized and caught by a power system controller who independently separates the two systems as the flow through the tie point pass through zero. Experience has shown this to be the much less impactive option.

The exception mentioned above is explained in part of an e-mail I received:

"The northern Vermont local grid is divided into sub-grids that can be connected to either the Quebec or Vermont power system without interrupting service. The line 1400 is energized from Stanstead and loads are connected at Newport. VELCO regulates deliveries from HQ by switching loads between the Québec and the New England systems. This regulation mode is commonly referred to by us as "block loading".

"Since the HQ and NEPOOL systems are operated asynchronous, loads must be connected to either one or the other, but not both simultaneously (except during switching). The actual transfers are accomplished by an automatic device when HQ and NEPOOL systems momentarily drift into phase. This transfer of loads between HQ and NEPOOL without power interruption is commonly referred to as a "quick switch."

"Quick switches are accomplished by momentarily connecting the load being transferred to both sources "in phase" and then disconnecting one of the sources before the systems drift out of phase. Once initiated, it may take several minutes for the two systems to come into phase and the actual switching (connecting and disconnecting) to take place."



CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[waross]

Thanks for sharing that information CR. That certainly is a special case.
To avoid confusion can we describe this as a tie between two different systems and not parallel generators on the same system?

This is not to say that synchronous connection has never been made, but it has always been inadvertent and typically will not last much more than five minutes.

The worst case scenario is that, because the connection is unstable, the flow through the tying circuit oscillates ever larger, eventually crescendo-ing to an automatic trip on overload.

From my observations of generators being paralleled out of phase (Miswired synchroscope) I would suggest that when the two systems drift about 20 or 30 degrees out of phase the automatic trip will happen.
At different frequencies, it won't take long to get 20 degrees or 1/18 of a cycle out of step.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[HamburgerHelper]

I don't think 20-30 degree is necessarily too much. I have seen neighboring utility tie relays set to block reclosing at 35 degrees out of sync. That utility had a strong grid but you would only see large angles on long tapped lines that were open ended. The current surge due to angular differences is going to be dependent on angle difference between both sources and each end's thevenin impedances to their sources. If you have two generators near each other without much impedance between them, you probably want them closed in close so you don't have a lot of power swinging back and forth.

I thought this was interesting but up in Minnesota, MISO had the issue that they couldn't close in a line due to the angular differences they saw while their system was loaded. They had to sit and wait for a period where the system load decreased enough to allow them to close in the line.

 

[crshears]

Correct, Bill; this is a tie between systems...but the bigger the systems the closer the slip frequency between them typically is before synchronization occurs or is even attempted, most commonly in the order of < 0.01 Hz.

As noted earlier in this thread, once the two systems are tied, their frequency is the same; what causes problems from that point forward is the "stringy-ness" of the HQ system and the subsequent response differential between its units on one side of the tying circuit and the "infinite bus" of the Eastern Interconnection on the other, the combination of which leads to the excessive phase angles you are describing.

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]