Mains frequency monitoring

Forums › Home Workshop DIY › Mains frequency monitoring

jamesrt

1378 posts

Uber Geek

ID Verified

Trusted

Lifetime subscriber

#230728 10-Mar-2018 13:01

I was reading an article on The Register about mains powered clocks in Europe losing time; the gist of the article is that the 50Hz frequency of the mains was drifting, causing clocks to be inaccurate.

It got me to wondering if there is any easy way of measuring the mains frequency and/or drift here. Basically, we bought a new electric (mains powered) clock for our bedroom which sometimes keeps very good time and sometimes becomes wildly inaccurate up to gaining 3 minutes/day.

Just wondering if there was a similar cause; and how to find out.

Not that I could do much about it anyway, I guess...

1 | 2

15209 posts

Uber Geek

Lifetime subscriber

#1972330 10-Mar-2018 13:10

There is a guy with a running monitoring project:

https://www.paulmonigatti.com/projects/grid/

Says actual voltage will vary but frequency results are valid for the entire north island. Data available.

Backblaze

Backblaze Unlimited Backup. World’s easiest cloud backup. Get peace of mind knowing your files are backed up securely in the cloud (affiliate link).

Aredwood

3885 posts

Uber Geek

#1972519 10-Mar-2018 19:54

In NZ, Transpower maintains the frequency between 45.5 and 50.5Hz. If the frequency goes outside of those values, they issue an excursion notice. Sometimes they don't issue one for an entire month, meaning the frequency stayed between those values for the entire month.

https://www.transpower.co.nz/system-operator/operational-information/excursion-notices

Although I don't think they aim to maintain a set number of cycles per day though.

Edit.

Maybe your clock is being affected by harmonics. Is it an early digital clock or a mechanical mains clock?

jamesrt

1378 posts

Uber Geek

ID Verified

Trusted

Lifetime subscriber

#1972522 10-Mar-2018 20:02

Aredwood: Maybe your clock is being affected by harmonics. Is it an early digital clock or a mechanical mains clock?

New Digital; already returned the first one as I thought it was faulty; the second also does it, but not as bad now. Purchased from JayCar, so not expecting it to be rubbish.

ObidiahSlope

260 posts

Ultimate Geek

#1972529 10-Mar-2018 21:26

My understanding of this issue is from before the generator - Transpower split so may be out of date.

A commitment is made to keep the Hz count accurate over a 24 hour period. There are often load variations in the grid which can slow down or speed up generators thus affecting the frequency before load control equipment adjusts. Because the variations are both up and down over a 24 hour period they usually even out. However a variation forecast is made approaching midnight and the system is sped up or slowed down if necessary to achieve an accurate 24 hour Hz count.

Obsequious hypocrite

richms

26406 posts

Uber Geek

Trusted

Subscriber

#1972532 10-Mar-2018 21:33

Someone I know who has been involved with the grid a long time ago in the UK said that the change from resistive loads to more complex constant wattage loads has made all the old tricks of controlling consumption by dropping the voltage levels not work anymore, in the past voltage sagged, resistive loads drew less current, things stabalized. Also with frequncy, motors would slow and therefore take less power, and things stabilized.

Now as voltage drops, everything takes more current to keep wattage constant, and motors are all inverter driven so the incoming frequency is irrelevent to a degree. Voltage drops, current goes up, losses go up, voltage drops more etc.

Richard rich.ms

jamesrt

1378 posts

Uber Geek

ID Verified

Trusted

Lifetime subscriber

#1972534 10-Mar-2018 21:36

Thanks for the replies; I'm mostly just curious.

I live in an area of new development; I can see a good half-dozen houses being built outside my window at the moment (well, not literally, as it's dark; but you get the idea!); and there is another huge patch of houses further up the hill.

I have no idea if the increasing (and fluctuating) load in the local area could have an impact or not. And I'll bet that a number of the local power tools being used are quite "noisy" too; albeit they should be filtered to some extent by RCDs, etc?? (Not really my area!)

Technofreak

6104 posts

Uber Geek

Trusted

#1972547 10-Mar-2018 22:34

jamesrt:

Thanks for the replies; I'm mostly just curious.

I live in an area of new development; I can see a good half-dozen houses being built outside my window at the moment (well, not literally, as it's dark; but you get the idea!); and there is another huge patch of houses further up the hill.

I have no idea if the increasing (and fluctuating) load in the local area could have an impact or not. And I'll bet that a number of the local power tools being used are quite "noisy" too; albeit they should be filtered to some extent by RCDs, etc?? (Not really my area!)

If your clock varies by as much as you say it does, I'd say it's faulty. In my experience the 230 V AC frequency is pretty accurate. It has to be otherwise the various generators would be "fighting each other". They all have to be at the same point on the AC wave form. If one slows down they all have to slow down to keep in sync with each other. To give an example of how critical it is, when a generator at any power station is spun up and brought on line it has to be synced up with the national grid prior to it being connected.

A few years ago a generator at the New Plymouth power station was synced 180 degrees out of phase and when it was connected to the national grid there was a rather large shudder as the generator armature regained the correct position relative to the AC waveform on the national grid. The forces were so large the shaft driving the generator was bent and the generator had to be shut down. I believe it was never repaired.

I wonder that the clock might convert the AC to low voltage DC to power the mechanism and the 50 Hz is not being used to regulate the time at all.

It also possible there is some spurious noise causing your problem. RCD's do not usually have any filtering properties.

Sony Xperia XA2 running Sailfish OS. https://sailfishos.org The true independent open source mobile OS
Samsung Galaxy Tab S6
Dell Inspiron 14z i5

Bung

5439 posts

Uber Geek

Subscriber

#1972576 10-Mar-2018 23:40

From the link in gzt's post even with an event like the Christchurch earthquake the frequency variation was only 0.5Hz. The usual instantaneous variation is +/-0.05Hz. That wouldn't explain sudden gains of minutes. Looking for similar cases it seems that some cheap clocks that use the mains frequency as a reference can be affected by electrical noise, lighting control signals etc

webwat

2036 posts

Uber Geek

Trusted

#1972591 11-Mar-2018 00:46

"Electrical noise" would be harmonics or feedback from capacitive loads such as phone/battery chargers, computers and TVs, quite of lot of the devices we use around the home these days. Modern computer power supplies aren't as bad as they used to be, but no idea how bad all the plug-packs we tend to use are. Wouldn't have thought its that significant in a home situation but who knows.

Time to find a new industry!

Aredwood

3885 posts

Uber Geek

#1972875 11-Mar-2018 22:28

Light dimmers are also really bad for electrical noise. And some cheap washing machines also use a variant of the light dimmer circuit to vary the motor speed on the spin cycle.

neb

8906 posts

Uber Geek

Trusted

Lifetime subscriber

#1972893 11-Mar-2018 23:18

jamesrt:
Basically, we bought a new electric (mains powered) clock for our bedroom which sometimes keeps very good time and sometimes becomes wildly inaccurate up to gaining 3 minutes/day.

Is there any heavy electrical equipment being operated nearby? Something like an electronically-regulated motor can introduce quite a bit of hash onto the power lines, caused by the interaction between the motor's commutator and the triac in the speed regulator. This would result in some devices seeing a higher frequency than 50Hz, the exact pseudo-frequency depending on how they detect the mains frequency (zero-crossing, peak detection) and the nature of the noise. The fact that it's gaining time rather than gaining or losing indicates that it's something like this that's the problem.

neb

8906 posts

Uber Geek

Trusted

Lifetime subscriber

#1972894 11-Mar-2018 23:22

jamesrt:
I live in an area of new development; I can see a good half-dozen houses being built outside my window at the moment (well, not literally, as it's dark; but you get the idea!); and there is another huge patch of houses further up the hill.

In that case I'd suspect even more strongly that it's mains hash due to electric motors. The clock isn't faulty, it's measuring the mains frequency as best it can, but what it's seeing isn't a 50Hz sinewave but who knows what sort of signal.

I'm surprised they're still selling clocks that use the mains as a timebase, given how cheap and readily available crystal regulators are.

Tinkerisk

3060 posts

Uber Geek

#1972911 12-Mar-2018 00:21

I wonder if there are still clocks using the power lines as frequency generator. I'd tought they all are at least quartz based (or sometimes the atomic reference clock signals are spreaded out via radio transmission directly to the wall clock receivers) and/or via NTP.

https://teara.govt.nz/en/video/6708/atomic-clocks

- NET: FTTH, OPNsense, 10G backbone, GWN APs, ipPBX
- SRV: HA server cluster, 0.1PB storage capacity on premise
- IoT: thread, zigbee, tasmota, BidCoS, LoRa, WX suite, IR
- 3D: two 3D printers, 3D scanner, CNC router, laser cutter

elbrownos

109 posts

Master Geek

#1972934 12-Mar-2018 08:36

Frequency Keeping is one of several ancillary services that Transpower pay generators to provide.

https://www.transpower.co.nz/system-operator/electricity-market/ancillary-services-overview

The frequency keeper must maintain the time error to 5 seconds or less, and return the error to zero at least once per day.

I worked at Mercury for a couple of years and we used Maraetai II power station for frequency keeping, which meant the machines wore out more rapidly than others due to continuously varying their output.

pctek

807 posts

Ultimate Geek
Inactive user

#1973452 12-Mar-2018 18:31

Really? I have a few clocks, all are wind with a key and the newest is 100 yrs old.

I never rely on them to the second, never mind minutes....isn't that hat the net is for?

1 | 2