How variable is your mains voltage?

[MLXXX]

On 14/11/2018 at 1:35 PM, rmpfyf said:

If accurate there's a little bit of stepping outside normal operating frequency band on the low end.

The dips to just below 49.85Hz were of short duration. (I gather that the standard may not concern itself with dips of that magnitude, if they are short-lived.)

 

Using a synchronous motor for a record turntable

 

If using the mains to run a record player turntable with a synchronous motor, a dip to 49.85Hz would represent a slowing down by 0.15/50 = 0.3%.  In music, a semitone is a change in frequency by a factor of 2^(1/12), or 5.94%. So the discrepancy is about 1/20th of a semitone, or about 5 cents. [See http://www.sengpielaudio.com/calculator-centsratio.htm for precise calculations.]

 

That is a very small difference, but might just be detectable in a carefully controlled test under good conditions by some subjects in certain frequency ranges. [See https://en.wikipedia.org/wiki/Cent_(music)#Human_perception.]  That sensitivity to a 5 cent pitch change would be for a change occurring within seconds: human memory for very fine gradations in pitch doesn't last at all long.

 

I strongly suspect that occasional mains grid frequency wobbles up to +- 0.15Hz relative to 50Hz would go unnoticed when listening to a vinyl record with a turntable driven by a synchronous motor.*

 

* Edit. Clarification: a synchronous motor actually synced to the mains, not as in some turntables, a synchronous motor synced to an internally generated and stable frequency.

 

Edited November 15, 2018 by MLXXX
I've found on review that my initial post was correctly calculated, so I've reinstated the wording!

[Guest rmpfyf]

4 minutes ago, MLXXX said:

The dips to just below 49.85Hz were of short duration. (I gather that the standard may not concern itself with dips of that magnitude, if they are short-lived.)

 

Using a synchronous motor for a record turntable

If using the mains to run a record player turntable with a synchronous motor, a dip to 49.85Hz would represent a slowing down by 0.15/50 = 0.3%.  In music, a semitone is a change in frequency by a factor of 2^(1/12), or 5.94%. So the discrepancy is about 1/20th of a semitone, or about 5 cents. [See http://www.sengpielaudio.com/calculator-centsratio.htm for precise calculations.]

 

That is a very small difference, but might just be detectable in a carefully controlled test under good conditions by some subjects in certain frequency ranges. [See https://en.wikipedia.org/wiki/Cent_(music)#Human_perception.]  That sensitivity to a 5 cent pitch change would be for a change occurring within seconds: human memory for very fine gradations in pitch doesn't last at all long.

 

I strongly suspect that occasional mains grid frequency wobbles up to +- 0.15Hz relative to 50Hz would go unnoticed when listening to a vinyl record with a turntable driven by a synchronous motor.

 

It's enough to elicit a reaction from ancillary services, so it doesn't go unnoticed. 

 

I'd wager a few good guitarists would have something to say about 5 cents being unnoticeable  

[MLXXX]

3 hours ago, rmpfyf said:

 

It's enough to elicit a reaction from ancillary services, so it doesn't go unnoticed. 

 

I'd wager a few good guitarists would have something to say about 5 cents being unnoticeable  

All frequencies would be affected for the duration of the grid frequency wobble so it wouldn't be as if the bottom E string of the guitar was out by 5 cents relative to the A string. The A string also would have shifted by 5 cents.

 

Re the benchmarks for grid frequency, I gather they're based on frequency averaged over a period of time, although I don't know what that time is.  It would be in relevant standards/guidelines somewhere. If the frequency dipped to 49.83Hz for 10 seconds, and in the next 10 seconds averaged 49.87Hz, and then continued above 49.85Hz, would that require a special exception report as failure to meet a benchmark?  (I could understand it triggering action to prevent a further drop in frequency.)

Edited November 14, 2018 by MLXXX

[Ittaku]

48 minutes ago, Eggcup The Daft said:

I guess (without searching to see if it's already covered) that the next question is, what effect does the variation have on an audio system?

 

I guess the worst case is a European product shipped "as is" to Australia expecting 220V and then receiving 250V+. What changes are likely to happen?

I already mentioned at least one manufacturer specifically makes their amps for Australia as their 220V version blows caps when used here.

[MLXXX]

21 minutes ago, Ittaku said:

I already mentioned at least one manufacturer specifically makes their amps for Australia as their 220V version blows caps when used here.

I see Papua New Guinea is still a 240V country. A few others are too, at least according to this webpage: https://en.wikipedia.org/wiki/Mains_electricity_by_country#Table_of_mains_voltages,_frequencies,_and_plugs

 

A manufacturer is asking for trouble sending a 220V rated amplifier to a country that is in course of changing from a former standard of 240V to a current standard of 230V, unless the engineering for the amplifier has allowed for large variations in supply voltage.

 

Years ago, designers of equipment would typically specify significantly higher voltage ratings for power supply filters than were expected to be necessary. This gave a large safety margin. You could expect that particularly to be the case for medical grade and military grade equipment.

 

Edited November 14, 2018 by MLXXX

[Guest Eggcup The Daft]

36 minutes ago, Ittaku said:

I already mentioned at least one manufacturer specifically makes their amps for Australia as their 220V version blows caps when used here.

My question was about changes to the sound of the system. Obviously, no sound is a pretty major change though

[Guest Eggcup The Daft]

1 hour ago, rmpfyf said:

 

It's enough to elicit a reaction from ancillary services, so it doesn't go unnoticed. 

 

I'd wager a few good guitarists would have something to say about 5 cents being unnoticeable  

When one string is out rather than all of them, certainly...

[Addicted to music]

3 hours ago, rmpfyf said:

 

PV is relevant as its sync'd to grid. It's not as though there's an AC voltage for PV systems and another for everything else (line drop asides). You're right about the preferred band, though that's not an upper limit. The current local voltage down signal occurs at 255VAC as per AS/NZS 4777.

 

Equipment is excepted to run beyond 253VAC for short periods of time (<=1%). No equipment I've tested or developed goes into arrested development the second things drift over 253VAC. The limits literally represent the 99th and 1st percentiles of supply voltage. As stated earlier, it's a probability distribution based Standard, not a hard and fast rule. For reference, a bog-standard EDMI smart meter (revenue grade, used often in parts of Australia) works between 45-65Hz, 176-276VAC... and they'll withstand up to 290VAC!

 

Most LED globes will survive supply requirements just fine. So will most equipment - tested to do as much in development and market homologation. If supply is at 250VAC+ for sustained periods of time, different story.

 

Not uncommon to see upper voltage alarm limits in industrial equipment (set by equipment manufacturers) at 258VAC (or a little more - basically the top band + room + accuracy). 

Here a sniping of a manual on the equipment I work on..  Where does it say that it’s expected to work over 253?    Nowhere!  So you saying that you have tested most equipment is just an assumption that “oh it works if we apply ove 253 and it’s still working!  The thing is it’s working right now straight after the test!   And it’s very obvious that that AS is just an assumption and nothing else.

 

 

[Addicted to music]

Lest say those SMPSU kept blow up....   manufacturer is going to ask what was the input voltage?  If I said to  253V and climbing....  No guess as to they’re going to say next....  “fu...ck of f “.   Don’t wanna know!   

 

 

1 hour ago, Ittaku said:

I already mentioned at least one manufacturer specifically makes their amps for Australia as their 220V version blows caps when used here.

Yes, 

 

it depends on the voltage rating of the caps used and what secondary voltage to that winding is in reference to the “ input voltage”.  And that input voltage is normally taken as 240V 

[MLXXX]

50 minutes ago, Addicted to music said:

Here a sniping of a manual on the equipment I work on..  Where does it say that it’s expected to work over 253?

The voltage range quoted of 220 - 240 V for Europe and Asia Pacific would be the ranges of the nominal mains voltages. 

 

Some countries have a nominal voltage of 220V, some 230V, and some 240V.   For example:

 

Gibraltar	C, G		240 V	50 Hz
Greece	C, F		230 V	50 Hz
Greenland	C, E, F, K		220 V	50 Hz

 

 

I've extracted the above from the table at  https://en.wikipedia.org/wiki/Mains_electricity_by_country#Table_of_mains_voltages,_frequencies,_and_plugs

 

Of course unless a person uses a laboratory-grade precision AC power supply, the actual voltage at the power point will not be exactly 240.00V, 230.00V, or 220.00V.

 

A question to be asked is: what is the permissible range of voltage in the various countries?  The designer of the equipment will need to have regard to the worst case extremes, which presumably would be the highest voltage permitted in any of the nominally 240V countries, and the lowest voltage permitted in any of the nominally 220V countries.

[davewantsmoore]

3 hours ago, rmpfyf said:

I'd wager a few good guitarists would have something to say about 5 cents being unnoticeable  

Sure, but that's typically between two seperate vibrating strings .... and it's audible because you can hear the beat frequencies.

 

That's quite different to the whole audio shifting up and down in pitch by 5 cents.....  a lot of things would need to conspire to make that audible.    Unless you were a freak with super-perfect pitch.... but then those people are freaked out by most recordings (even ones which don't vary like we're talking about)   

[vivianbl]

A perennial topic, at least in recent years,  on audiophile forums, is the purported benefit of running a separate power circuit (or even several power circuits) from the home switchboard to the room where the hi-fi equipment is located. It is said that one advantage of this is to  minimize voltage drop.  That raises the question, "What is the mains voltage to begin with?".
 
Unless you generate (or for a select group of very keen audiophiles these days, "regenerate") your own power, the voltage at your switchboard can be expected to cover a pretty wide range in the course of a 24 hour period.  In additional to the variations that have always existed in mains supply voltage, arising from mismatches between supply and demand , a big factor in the summer months these days is the direct local injection of electricity into the grid from other homes in your neighbourhood that are running a mains inverter powered with solar panels.
 
In the last few days, the mains voltage at my home in the inner norther suburbs of Brisbane, during the daylight hours, has covered the range 239V (approx) to 250V (approx).
 
Below is a graph of the variation for the 13th November 2018, as provided by a Sunny Boy inverter located immediately adjacent to the house switchboard. Readings were captured via bluetooth approximately 4 times a minute. [The readings were consistently about 2 volts higher than displayed on a Watts Clever device (model EW-AUS5001) plugged into a power point near the switchboard.]
 
What approximate readings are other forum members experiencing?
 
Does anyone have a mains supply that stays within a tighter range than mine does?  As can be seen below, I've recently had a variation range of about 10 volts during the day. 
 
 

Over the winter months the daily voltage ranged from 225v to 245v. When I checked over the past few days the readings during the day varied from 236v to 252v. But observe that it is mostly now in the high 240s.

I’m getting solar system installed next week. Concerned about the system shutting down if if voltage goes higher and also damage to equipment not just audio.

[Guest rmpfyf]

3 hours ago, MLXXX said:

All frequencies would be affected for the duration of the grid frequency wobble so it wouldn't be as if the bottom E string of the guitar was out by 5 cents relative to the A string. The A string also would have shifted by 5 cents.

 

Re the benchmarks for grid frequency, I gather they're based on frequency averaged over a period of time, although I don't know what that time is.  It would be in relevant standards/guidelines somewhere. If the frequency dipped to 49.83Hz for 10 seconds, and in the next 10 seconds averaged 49.87Hz, and then continued above 49.85Hz, would that require a special exception report as failure to meet a benchmark?  (I could understand it triggering action to prevent a further drop in frequency.)

 

Really depends what you're listening to, would point out frequency shifts are quite fast also so the effect over a tune of a rock-steady supply would probably be audible. 

 

Grid dynamics for frequency regulation are complex - there are six markets for contingency events (6 seconds, 60 seconds 5 minutes both in raise and lower for a total of six markets), the fast response stuff can be broadly automatic - if it dipped as you suggested people with generating sources would have already contributed to have arrested the slide. 

 

More critical generally to get frequency right than voltage. 

 

This doc is useful http://energylive.aemo.com.au/-/media/Files/Electricity/NEM/Security_and_Reliability/Power-system-requirements.pdf and also goes into good bits on services and where they apply - see Fig 3 near the end.

 

2 hours ago, Eggcup The Daft said:

When one string is out rather than all of them, certainly...

 

There are guitarists out there that measure their fingerwork in cents. It's an incredible skill.

 

2 hours ago, Addicted to music said:

Here a sniping of a manual on the equipment I work on..  Where does it say that it’s expected to work over 253?    Nowhere!  So you saying that you have tested most equipment is just an assumption that “oh it works if we apply ove 253 and it’s still working!  The thing is it’s working right now straight after the test!   And it’s very obvious that that AS is just an assumption and nothing else.

 

Right where it states 220-240VAC mains supply - that infers it'll work at voltages consistent with a mains supply specified with a nominal voltage of 230VAC.

 

They're sold to run on a 230VAC nominal grid. The grid will never be at 230VAC all the time. It'll be around it most of the time, it'll have extremes. Yes, I used to work in test and development, and have had equipment on grid simulators - the job certainly isn't done (in bringing anything to market) without testing for what mains conditions can be expected. And those conditions are a broad set.

 

A breaker on the device you mentioned conforming to IEC IEC60947-2 would need to cover 85% of the minimum nominal and 110% of the maximum, so would need to be good for 187-264VAC.

 

This doesn't mean that power is likely to be supplied at 253VAC or over consistently, or that equipment should be expected to survive as much. Equipment on a 230VAC grid should be able to support a range of voltages above and below it for intermittent periods without failure. Australian Standards are no different to standards anywhere else in the world - and if you sell a product into a market, you design for the standards of that market - to do any less is negligent. I would stress that outside of these limits whilst the grid can be there, there's no requirement for an appliance to work - it's allowed to shut down and protect itself, and the grid is under obligations to take corrective action in these instances. 

 

Put another way I've never heard of, and would never expect to hear of, a piece of equipment designed to run strictly at one exact AC voltage. 

 

This doesn't mean the grid is perfect - out-of-specification performance is inevitable and there's recompense for those affected (it helps to have data) - in which case the Standards are invaluable, as they provide a benchmark for expectations. The Australian Standards involved are far from an assumption - quite reasonable and realistic, actually.

 

1 hour ago, MLXXX said:

A question to be asked is: what is the permissible range of voltage in the various countries?.

 

Depends on the problems inherent to the region. Bits of the US design for 5% variation, and in some regions where power is a bit sh***y it's known to have some appliances 'brownout rated'... in short, 25% under nominal. Eep.

 

Unlike Australia, not all countries have national standards for this sort of thing.

 

20 minutes ago, vivianbl said:

Over the winter months the daily voltage ranged from 225v to 245v. When I checked over the past few days the readings during the day varied from 236v to 252v. But observe that it is mostly now in the high 240s.

I’m getting solar system installed next week. Concerned about the system shutting down if if voltage goes higher and also damage to equipment not just audio.

 

 

It'll limit before it shuts down... and you can always do what @Assisi did.

[wanta911]

4 hours ago, MLXXX said:

I see Papua New Guinea is still a 240V country.

 

And a certain part of Australia is still on 240V -6%/+6% as standard.

 

We're a bit slow over here 

[MLXXX]

1 hour ago, rmpfyf said:

Really depends what you're listening to, would point out frequency shifts are quite fast also so the effect over a tune of a rock-steady supply would probably be audible. 

It appears to be a variation potentially around the threshold of perception, but to my mind it's close enough to being of practical consequence that to be on the safe side,  if I were very keen on listening to a lot of vinyl discs (I'm not!), I'd choose a turntable motor governed by a crystal controlled clock, and not one governed by the instantaneous mains frequency.

 

2 hours ago, rmpfyf said:

 Really depends what you're listening to, would point out frequency shifts are quite fast also

Well certainly with emergencies the change in frequency can be substantial, and rapid, judging from this diagram that appears in a pdf you referred me to:

 

 

 

A drop to 49.5Hz would be 1% change relative to 50.0Hz, or 17 cents.  If that drop took place in just 5 or 6 seconds it is much more likely to be audible than a drop of only 5 cents. 

 

Here are two files: one where the frequency of the sine wave stays at 500Hz and one where it drops to  495Hz over 5 seconds. 

 

In no particular order they are:

FILE A

FILE B

 

Fellow forum members, don't worry in the least if the two files sound the same to your ears!  It is not easy to hear a change of 17% of a semitone, if the drift takes place over several seconds.

 

[MLXXX]

 

3 hours ago, vivianbl said:

 

Over the winter months the daily voltage ranged from 225v to 245v. When I checked over the past few days the readings during the day varied from 236v to 252v. But observe that it is mostly now in the high 240s.

I’m getting solar system installed next week. Concerned about the system shutting down if if voltage goes higher and also damage to equipment not just audio.

 

The risk of damage to equipment is low. It's not long ago that Australia was nominally a 240V country, and parts of Australia are still in transition to a nominal 230V.  I note that 252V is only 5% above 240V. Appliance are designed to operate on voltages somewhat more or somewhat less than the nameplate nominal voltage for the appliance.  There is a nuisance factor with incandescent bulbs if a household still uses them: they tend to need to be replaced more often.

 

 

Power limiting function

 

Some inverters in use today in Australia comply with the "Grid protection requirements" in Part 3 of the 2005 version of AS 4777.

 

An inverter you're having installed now would I presume need to be be certified under the 2015 version of AS 4777.2.  You could expect it to have a sophisticated power limiting function to reduce its output gradually if the grid voltage in your neighbourhood drifts high; and ultimately to shut down your inverter in response to sustained excessive voltage.

 

Something that hasn't been mentioned to date is deferring using heavy load appliances for the purpose of pulling your switchboard voltage low. Your inverter senses the mains voltage where it connects to your house power, normally at the switchboard. At my place the electric hot water system is arranged to turn on during the middle of the day. I find when the heating element is on, the voltage at the switchboard drops by around 2.4V.  That could be just enough to make the difference between a modern inverter throttling back its output, or not.

[Guest rmpfyf]

7 hours ago, MLXXX said:

Well certainly with emergencies the change in frequency can be substantial, and rapid, judging from this diagram that appears in a pdf you referred me to:

Yes - SA blackouts were a good example of this.

 

6 hours ago, MLXXX said:

An inverter you're having installed now would I presume need to be be certified under the 2015 version of AS 4777.2. 

100%.

 

6 hours ago, MLXXX said:

At my place the electric hot water system is arranged to turn on during the middle of the day. I find when the heating element is on, the voltage at the switchboard drops by around 2.4V.  That could be just enough to make the difference between a modern inverter throttling back its output, or not.

This is something everyone with resistive hot water tanks should do - SolarEdge, Fronius and others have or are coming out with kits that will manage this with variable power too. Payback is very fast, and the grid will thank you

 

Should also be stressed that some inverters are capable of providing reactive power where so configured and controlled. The 'controlled' bit is not widely done yet but if implemented it's a very good thing.

[Assisi]

I was asked the question by @MLXXX in the SolarIQ thread as to whether I had reported the matter of high supply voltage to my house.  I had had a discussion awhile back but did not pursue the matter as I had the EdgeIQ ( previously called  SolarIQ).  The question prompted me to pursue the matter with the provider with a request for a check with logging.  Within hours the response was no logging required as the smart meter indicated high voltage at all hours of the day for me plus neighbours.  This will be rectified at the transformer within 5 days.  If this does not solve the problem more works will be done until it is fixed!

 

In accordance with the "Victorian Distribution Code" the Provider Powercor is obligated to rectify the matter.  If equipment is damaged because of high over voltage outside the limits , there is a a claim form on the providers Web site to request compensation.  Because of the smart meter the provider knows the supply voltage and I am told that they generally do not argue the matter.  They want to please the and make sure the customer has a good experience.

 

Right now the supply voltage is 258V

John

Edited November 15, 2018 by Assisi

[Assisi]

This graph was emailed to me from Powercor to explain why there was no need to undertake the logging.  Voltage is all over the place.  I assume that everything in the yellow band is outside the limit of 253V

 

 

John

[andyr]

8 minutes ago, Assisi said:

This graph was emailed to me from Powercor to explain why there was no need to undertake the logging.  Voltage is all over the place.  I assume that everything in the yellow band is outside the limit of 253V

 

 

John

 

How did you get hold of that graph, John.  I'd be interested in seeing how my mains is behaving.

 

Andy

 

[Assisi]

Andy

My provider is Powercor.  When I asked them to do the logging the

Technical Officer Supply Quality
Reliability Delivery

 Contacted me to say that he had "looked at" my smart meter and those of my neighbours.  He could see that there definitely was an issue.  The graph is what he could see and he emailed it to me.  He was very helpful as I have found Powercor to be the few time I have needed to make contact .  They really want to please the customer.   I assume every Victorian provider could  provide you with the same information.  Contact your provider.

John

Edited November 15, 2018 by Assisi

[Guest Eggcup The Daft]

1 hour ago, Assisi said:

Andy

My provider is Powercor.  When I asked them to do the logging the

Technical Officer Supply Quality
Reliability Delivery

 Contacted me to say that he had "looked at" my smart meter and those of my neighbours.  He could see that there definitely was an issue.  The graph is what he could see and he emailed it to me.  He was very helpful as I have found Powercor to be the few time I have needed to make contact .  They really want to please the customer.   I assume every Victorian provider could  provide you with the same information.  Contact your provider.

John

I can't comment for Victoria, but in NSW it appears that you can have problems with complaints if you are with a different provider to the one that supplies the meter. We get encouraged to shop around for the cheapest deal but there can be consequences...

[Assisi]

45 minutes ago, Eggcup The Daft said:

I can't comment for Victoria, but in NSW it appears that you can have problems with complaints if you are with a different provider to the one that supplies the meter. We get encouraged to shop around for the cheapest deal but there can be consequences...

@Eggcup The Daft

 

My meter has Powercor printed on it.  It is Powercor that tells the biller what to charge me including the payment for the meter.  Powercor is primary to me in the context of supply and that is who I contact if there is any problem.

[Guest rmpfyf]

4 hours ago, andyr said:

 

How did you get hold of that graph, John.  I'd be interested in seeing how my mains is behaving.

 

Andy

 

 

You need a smart meter; voltage is inferred at a minimum. In VIC the standard is as such https://www.energy.vic.gov.au/__data/assets/pdf_file/0024/43890/Minimum-AMI-Functionality-Specification-Victoria.pdf

 

Probably easier to outfit a third party solution. If you've got solar, get yourself a Reposit box. 

[mwml15]

As I got power regulator that it will shutdown when it stay high voltage too long. It might not be good for the av equipment n TV but that one way to tell there an issue with high voltage.