Frequency standard

I'm afraid Phil is right. As the load increase beyond what can be supplied by the machines already online, another unit is brought onto the grid. Normally some units produce less than their full capacity to give what is called "spinning reserve". The spinning reserve on the grid is normally at least equal to the largest unit on the grid so that if a machine trips off there is enough reserve capacity on the grid to cover it. When demand exceeds the total capacity of the grid you get "brown outs". This is done by a system controller switching off an area for a while to reduce grid load. It's simply a matter of operating a switch which causes a huge circuit breaker to open and cuts of power to a suburb or small town. Brown-outs are rotated to different areas so as to reduce inconvenience to customers, but power utitiltys usually have pre-arranged contracts in place to cut off non-essential loads first. When a unit does trip off accidently, the grid frequency may drop a few hertz for a few seconds until the spinning reserve picks up the load, but you wouldn't be able to forecast a problem in advance, you would only know of the event afterwards when the frequency has dipped. In general, bigger grids are more stable.

Elmo

"Suzy" "Phil Allison"

** I will not answer dumb questions or engage in pointless chat.

The folk behind that site have NO credibility and have given NO proof of their claims.

The idea put forward that individual appliances ought to dance to the tune of tiny variations in the AC frequency is

UTTERLY LUNATIC !!!

A typical f****it greenie idea.

Like banning incandescent bulbs and transformer plug packs.

....... Phil

Balancing a load is always difficult, like driving a car at exactly the speed limit, you will vary somewhere around the desired setpoint. Generators use PID controllers to maintain correct load but there will still be variations due to time lag of the system and over-correction by the controller. On a big grid these variations will be smaller.

Slightly off topic but interesting, the grid load on a typical week day has a sharp peak at exactly 6PM.

Elmo

I agree with Phil on this one, the idea is lunacy. The web interfaced frequency meter is neat though, although it's not even calibrated:

"We have not yet callibrated the meter, so the number you can see may differ slightly from the real grid frequency by a constant small amount. The reason why we show so many decimal places in the frequency is to ensure you can see it change. The changes (the movements of the needle) should be quite accurate. There will be a slight delay in the signal because it takes at least two seconds to send the data over the internet. Disclaimer: Please don't rely on our grid meter for any industrial or other applications. We can't guarantee accuracy or reliability!"

Dave.

"Suzy" wrote

But isn't it true, Phil, that more demand than supply will drag down alternator shaft speed and thus frequency, and vice-versa?

***** Oh dear,haven't you ever heard of a governor and what is designed to do?? Perhaps your email address says it all!

Brian G.

Yes, that's what I meant. BTW, how do you know when the sharp peak is (if not by looking at a drop in frequency)?

For Suzys purposes it doesn't have to be calibrated, she's more concerned with fluctuations and differentials than the actual frequency.

Elmo

"Suzy" = not@valid

** ROTFL !!!

Better go learn about ohms law - honey ...

...... Phil

What a pompous answer! It is true and is not "dumb" as you assert Americanly. But I put it to you in good faith as I thought I detected a smidgeon of real knowledge and was willing to be corrected if you had a point to make. More and more you show about 1% of knowledge mixed with 99% or baseless abuse.

Have you any credibility with this displayed inability to enter into a rational discussion. How do you set yourself up as judge of the credibility of others? I wonder what your qualifications and proven experience are?

Sound quite plausible to me. Wouldn't bother with a toaster though...

Back to normal

I agree with you about that, but no doubt to you (o mighty one) it is "pointless chat". What a pity you have no tolerance for mere mortals.

Suzy

But still hasn't told us *why*. I any case I do hope she has a good plan to properly filter and average out noise and other fluctuations from her "measurements".

Dave.

I work at a power station, and on our internal internet system we have a load graph. We can bring up load data going back several years. It's only updated every ten minutes. I haven't looked at the frequency recordings for ages but I'll have a look when I go back to work.

The load generally rises during the day, drops off a bit about 5 pm and then peaks at 6pm as everyone switches the tv on to watch the news over a cup of tea. On a hot day the the load doesn't drop off but keeps rising about 5 pm. I blame this on daylight saving, people leave a cool air conditioned workplace and come home to a hot house and put their air conditioners on flat out to cool the house down. If daylight saving was an hour opposite you wouldn't be coming home at the hottest time of the day. It's a way of passing cooling costs onto the worker.

Elmo

Hey Suzy what about using a phase locked loop? Since you're mainly looking at frequency shifts of a faily short duration a phase locked loop should be able to give you an indication of the frequency shift if you have it heavily damped.

Elmo

Any info on how NSW is going for summer capacity?

My 2c is that I escaped a number of outages this summer as the heat didn't happen. So I am wondering if we are going to cop them next summer (if the heat comes back)?

I remember trying to measure variations in mains frequency 30 odd years ago.

All I can say is you will need a very accurate measuring kit.

While I can't remember what the accuracy of the xtal we used was, I do remember we could measure _no_ variation from 50Hz at any time. the result was a big surprise to us at the time, I do remember that.

By and large the 50Hz mains could be used as a frequency standard in it's own right, both short and long term.

I really doubt that total load affects frequency in any meaningful way as you apparently believe from your website.

But measure away, who am I to stop a seeker of knowledge! :)

Ross

"RMD"

** Can be done very easily - if you have a stable sine generator with fine tune and a basic CRO and frequency counter.

Just set up a lissajous pattern, with a ratio of 20 to 1, on the CRO creen - then tweak the audio gen frequency to get the pattern stationary. Read the generator frequency off the counter.

When it is 1000 Hz, the AC supply is 50 Hz.

When it is 1005Hz, the AC supply is 50.25 Hz.

** Ditto.

Bout as useful as reading tea leaves.

However, an AC voltmeter will tell you if the load in your local area is unusually heavy or light at some given time.

...... Phil

Any cheap frequency meter with reciprocal measurement will be able to measure it just fine.

Dave.

Great! I'll take much more notice of your replies than that strange strange person PA!

Yes, but I'm interested in going one better than the website, and actually calibrating. That was the reason for my original post about dividing down from crystal oscillators of eg 2 MHz

How do you plan on "calibrating" your crystal oscillator? "Calibration" usually = "traceability".

What is the end purpose of all this measurement anyhow?

Dave.