# Consistency in AC source frequency.

#### firebreather

Joined May 27, 2013
19
I have always wondered how the generators at hydro-electric power stations maintain a constant speed of rotation to produce a constant frequency of supply to the nation .

My point is that frequency depends on speed. So if , at a time , there is more inflow of water from the reservoir/river/falls, so that the generator turbines rotate at a higher speed , thus increasing the frequency of power supply.

A solution is to convert the a.c generated to d.c using diodes ! and filter capacitors ! and then reconvert it to a.c using LC circuit or others, which have a rock-solid frequency.

This is almost impractical, so then how is this problem handled ?

#### wayneh

Joined Sep 9, 2010
17,153

#### WBahn

Joined Mar 31, 2012
26,398
How do you maintain a constant speed in your car even as you go up a slope or down a slope? How does your cruise control do it?

Imagine a motor-driven valve to throttle the water going through the turbing. Turbine starts spinning too fast, close the valve a little bit. Spinning too slow, open th valve a little bit.

At one dam that we took a tour of they actually had a flyball governer sticking up out of the turbine casing.

#### PackratKing

Joined Jul 13, 2008
847
I read the article in the link you put up, not to mention a gazillion other times, and I still can't wrap my head around 50 Hz...

The circle, is based on 360° the clock is divided into 60 minute hours, and each minute is 60 seconds...

Except for the 100 minute clock often used in employment timekeeping,

50 Hz just doesn't make sense

#### LDC3

Joined Apr 27, 2013
924
I read the article in the link you put up, not to mention a gazillion other times, and I still can't wrap my head around 50 Hz...

The circle, is based on 360° the clock is divided into 60 minute hours, and each minute is 60 seconds...

Except for the 100 minute clock often used in employment timekeeping,

50 Hz just doesn't make sense
Instead of using degrees, they could have used a unit called grads. There are 400 grads in 360º.
BTW, who uses the 100 minute clock. I haven't heard of it before.

#### matty204359

Joined Apr 6, 2011
105
Instead of using degrees, they could have used a unit called grads. There are 400 grads in 360º.
BTW, who uses the 100 minute clock. I haven't heard of it before.
when I look at my pay stubs they give times like 9.5 hours(9 hours and 30 minutes) or 7.7 hours(7 hours 42 minutes). its basically hours plus a percent of an hour. its for accounting so they don't have to convert when multiplying #of hours by your hourly rate or your over time rate or you holiday pay rate.

#### WBahn

Joined Mar 31, 2012
26,398
A solution is to convert the a.c generated to d.c using diodes ! and filter capacitors ! and then reconvert it to a.c using LC circuit or others, which have a rock-solid frequency.
Just how "rock solid" would that LC circuit be? You would probably be hard pressed to get it to better than 0.1%. Is you used a quartz crystal you might expect 10ppm for a good one. If you use a oven-controlled crystal oscillator you might get a bit better than 1ppm. I may be off on some of these, but I don't think by a whole lot.

So let's see what 1ppm gets you.

How many seconds are there in a year?

(365.25day/yr)(24 hr/day)(60min/hr)(60s/min)=31557600s/yr

So something that is 1ppm would be expected to be up to half a minute off after one year. A cheap clock that uses the power line will be unlikely to be off more than a second.

This is done by commanding the generators to run at a speed that reduces the error between a clock that uses the AC frequency as it's time base and one that uses an atomic clock, such as GPS or WWV.

#### WBahn

Joined Mar 31, 2012
26,398
I read the article in the link you put up, not to mention a gazillion other times, and I still can't wrap my head around 50 Hz...

The circle, is based on 360° the clock is divided into 60 minute hours, and each minute is 60 seconds...

Except for the 100 minute clock often used in employment timekeeping,

50 Hz just doesn't make sense
What doesn't make sense about it. Count 50 cycles and increment your seconds counter. Count 60 seconds and increment your minutes counter. Count 60 minutes and increment your hour counter. Count 24 hours and increment you day counter.

If you want to use a mechanical clock, use a synchronous motor that turns at 50Hz (or 3000rpm). Use a 3000:1 reduction to run the second hand and a 60:1 reduction from that to run the minute hand and a 12:1 reduction from that to run the hour hand (or 24:1 if it is a 24-hour clock).

Where does a circuit having 360° play into this anywhere.

#### studiot

Joined Nov 9, 2007
4,998
I have always wondered how the generators at hydro-electric power stations maintain a constant speed of rotation to produce a constant frequency of supply to the nation .

My point is that frequency depends on speed. So if , at a time , there is more inflow of water from the reservoir/river/falls, so that the generator turbines rotate at a higher speed , thus increasing the frequency of power supply.

A solution is to convert the a.c generated to d.c using diodes ! and filter capacitors ! and then reconvert it to a.c using LC circuit or others, which have a rock-solid frequency.

This is almost impractical, so then how is this problem handled ?
We seem to have wandered from the question.

The speed of the turbine is determined not by the flow but by the pressure head.

The pressure head is set for a given installation by the height between the intake from the dam or reservoir and the bed of the turbine. This is of course constant. There is a small variation in head due to the different pressure losses down the pipe at different flows, but this is small.

Does this help?

#### kubeek

Joined Sep 20, 2005
5,758
There is a small variation in head due to the different pressure losses down the pipe at different flows, but this is small.
The water level is not very constant, I´d guess something like +/-2m in a 60m reservoir.
But anyway, the main thing about the grid is that all the generators over the continent are tied together and act as one entity, so as one generator tries to run a bit faster the others counteract it, this way the frequency of the grid as a whole can be controlled quite precisely.
If there is more power demand than supply then the frequency starts to drop from the standard 50Hz and vice versa, so by throtling some of the powerplants you can regulate the frequency.

#### studiot

Joined Nov 9, 2007
4,998
The water level is not very constant, I´d guess something like +/-2m in a 60m reservoir.
So ?

What matters is the height of the takeoff point above the generator floor, not the surface height in the reservoir.
That is why pipes are led down the valley as far as possible to the generator floor.

But yes, you are correct there are electrical network considerations/arrangements as well, including the synchronisation effect you describe.

I was just trying to keep it simple.

#### Ramussons

Joined May 3, 2013
1,062
. . . .

The speed of the turbine is determined not by the flow but by the pressure head.

. . . . .

Does this help?
Not by Pressure Head alone. The speed of any turbine is determined by Both Head AND Flow. It so happens that the "head" in Power Stations is more or less constant and the hence speed of the turbine is maintained by adjusting valves to control the flow.

studiot What matters is the height of the takeoff point above the generator floor, not the surface height in the reservoir. That is why pipes are led down the valley as far as possible to the generator floor.

Sorry, studiot, that's incorrect. The Head is the water level of the Reservoir ABOVE the Turbine. The whole pipeline (called a Penstock) is an integral part of the water head AND that's why power stations are located as far as possible way down valleys to get as high a head as possible.

Ramesh

Last edited:

#### kubeek

Joined Sep 20, 2005
5,758
What matters is the height of the takeoff point above the generator floor, not the surface height in the reservoir.
Why doesn´t the surface height matter? With given height from turbine to surface the hydrostatic pressure on the turbine will be the same no matter how deep the takeoff point is below the surface. Or am I missing something?

#### studiot

Joined Nov 9, 2007
4,998
You have a large fixed head as I described plus a small variable head or overpressure which can be throttled at the turbine house to provide a measure of fine control. It is important the the variation of reservoir level is small compared to the fixed head.

#### PackratKing

Joined Jul 13, 2008
847
Where does a circuit having 360° play into this anywhere.
A circle has 360° , an hour 60 minutes, a minute, 60 seconds
the reference to a circle having 360, is to show that all three represent one complete revolution in time. No conversion math required...

Though it may jive with the Schmetric system, I will stick to what I have learned, that works and makes some sense... Not to labor the point, I do my electronic math, based on Ohms' Law... Thevenin, imho, is a blistering idiot.

Long ago, my instructor in Navy Electronics, taught not to try to wrap your head around things on the order of Gigahertz frequencies... or distances represented by multiple billions of light years referred to in items located in the boundless universe... just graph one cycle, and take for granted that things do move that fast, or are that far away.

This bone of contention reflects the practice of putting both S.A.E. and Metric fasteners on one car chassis, and obligates me, a mechanic, to maintain two sets of tools... Utterly ridiculous, to the point of maddening!

Last edited:

#### LDC3

Joined Apr 27, 2013
924
This bone of contention reflects the practice of putting both S.A.E. and Metric fasteners on one car chassis, and obligates me, a mechanic, to maintain two sets of tools... Utterly ridiculous, to the point of maddening!

I would like to hang a few automotive engineers...
I think where the parts were assembled determine if the heads are S.A.E. or metric. I believe that the United States is the only country that manufactures using S.A.E. standards. You can blame congress for not conforming to the world view.

#### studiot

Joined Nov 9, 2007
4,998
This bone of contention reflects the practice of putting both S.A.E. and Metric fasteners on one car chassis, and obligates me, a mechanic, to maintain two sets of tools... Utterly ridiculous, to the point of maddening!
In World War II a bigger glitch occurred when the British sent drawings of aircraft and other components over to be manufactured in the US.
The British inch was slightly different from the US inch and -suprise suprise - the results were misfits.

#### wayneh

Joined Sep 9, 2010
17,153
A circle has 360° , an hour 60 minutes, a minute, 60 seconds
And all of those values are completely arbitrary. They were established long ago because the Byzantines (I think it was) were obsessed with the number 60. There is no intrinsic, physical reason for these, it's purely a human thing. Like our bias toward using base 10 because we have 10 fingers.

#### wayneh

Joined Sep 9, 2010
17,153
Sorry, studiot, that's incorrect.
+1
Head pressure does not care where the takeoff point is. The inlet can even be far below the outlet, as long as the water surface is above the outlet.

Pumping water from a pipe in the deepest hole in the ocean to a point 1 foot above sea level is identical to pumping from sea level to +1 ft. (Neglecting piping losses.)

Pressure and flow in a dynamic system are a bit like the proverbial chicken and egg. Cause and effect not easily separated.

#### PackratKing

Joined Jul 13, 2008
847
And all of those values are completely arbitrary... .
You'll get no argument from me on that point...!!!

Studiot... You ain't just a whistlin' dixie on that one !! I have run onto Brit. specs a number of times in career... wot a mess !!!