Voltage Complaint

[bernardgreen]

so we are constantly having to deal with PV users complaining when the system shuts down.

When you say you have to deal with them how much do you do. I would say you need only tell them that their equipment is out of spec and therefore non compliant and for the safety of others must be disconnected from the network. If there are frequent incidents of over volting then I suggest the DNO service to the house is dis-connected and let them live ( electrically ) on ( under ) their own power station.

I wonder if the feed in voltage should be set lower so the PV only feeds into the network when there is a real local need to "top up" the network, ( as shown by low network voltage ).

 

[westie101]

When you say you have to deal with them how much do you do

If the volts at the meter are out of limits with the PV system switched off we will tap the local transformer down.
If under the same conditions we are within limits we just inform the customer of this and do nothing.

 

[stillp]

I haven't got the link at home, but the actual requirement is for the voltage to be within limits for greater than 95% of the time measured at 10 minute intervals over 7 days.
So spot checks can only be used as an indication and must be on calibrated instruments.

Isn't it even worse, in that the measurement is the average over each 10 minute period?

 

[mfarrow]

Its OK if you take the spot reading, rather than the average.

 

[ban-all-sheds]

heating appliances work quicker, they consume less energy

How does that work?

 

[JohnW2]

Isn't it even worse, in that the measurement is the average over each 10 minute period?

Indeed - assuming that's what it means. I suppose that's easily the most likely and most obvious interpretation of "measurent interval" (per the factsheet westie posted). An average over 10 mins could obviously hide a multitude of short-term horrors.

If you ever have a moment, could you perhaps have a look at BS EN 50160:2000 and see if you can confirm that its a 10 min average and also see if it says anything else about any other 'limits' on voltage (or its measurement). As I wrote last night, it would seem a bit odd if the supply were allowed to go 'as high or as low as it liked' for up to 5% of the time, without any 'absolute' limit.

Kind Regards, John

 

[bernardgreen]

heating appliances work quicker, they consume less energy

How does that work?

In the case of a kettle the faster heating means less time for heat to escape from the kettle and be wasted.

In a water heater with perfect thermal insulation there would be no energy saving if the supply voltage was increased.

 

[JohnW2]

heating appliances work quicker, they consume less energy

How does that work?

Not terribly well, in most situations, I'd say. The argument one sometimes hears is that if one is, say, boiling a kettle, the faster one bolis it, the less will be the heat losses (which are essentially 'per time' losses dependent on kettle-ambient temp differential). I somehow doubt it makes much difference - and in many/most situations, it's probably not even a relevent argument.

Kind Regards, John

 

[ban-all-sheds]

It won't make any significant difference whatsoever.

 

[stillp]

If you ever have a moment, could you perhaps have a look at BS EN 50160:2000

No, but I could look at the latest version!

Under normal operating conditions:
− during each period of one week 95 % of the 10 min mean r.m.s. values of the supply voltage shall
be within the range of Un ± 10 %; and
− all 10 min mean r.m.s. values of the supply voltage shall be within the range of Un + 10 % / - 15 %

The standard makes the point that networks are designed on a probabilistic basis, so although the voltage could in theory go to any value for some of the time, it is statistically unlikely. However any large deviation would presumable be covered by the provisions dealing with spikes, surges, swell, or interruptions.

 

[stillp]

Just for further clarification, there is a draft application guide to 50160, which will be TR 50422:2013 when published. it states that "A balance needs to be found, and in the case of this Standard, the 10-minutes-averaging time proposed in the 503 measurement standard EN 61000-4-30 is used.The Standard referred to is 50160.

 

[JohnW2]

Under normal operating conditions:
− during each period of one week 95 % of the 10 min mean r.m.s. values of the supply voltage shall
be within the range of Un ± 10 %; and
− all 10 min mean r.m.s. values of the supply voltage shall be within the range of Un + 10 % / - 15 %

Many thanks. Well, that goes quite a long way towards what I was asking for ('absolute limits'), except that they are talking about 10-minute means. In fact, as I see it, at this level of 10-min means, it's not even very probabilistic - they require 100% of such means to be between -15% and +10%, the only really probabilistic bit (at the 10-min average level)being that 5% may be between -10% and -15%.

The standard makes the point that networks are designed on a probabilistic basis, so although the voltage could in theory go to any value for some of the time, it is statistically unlikely.

As above, this is really mainly relevant within the 10-min averaging periods. I don't doubt that what they say is probably right - but to support it, and certainly to quantify 'statistically unlikely', one would need to know something about the distribution of voltage within each 10-minute period. With just an average, it's impossible to draw any probabilistic conclusions.

However any large deviation would presumable be covered by the provisions dealing with spikes, surges, swell, or interruptions.

That's true, but I suspect there is a bit of a 'gap' between those (usually) extremely brief happenings and a full 10-minute period.

Kind Regards, John

 

[stillp]

"NOTE Application: for the purpose of this standard, the duration of a voltage swell is from 10 ms up to and including 1 min."

 

[JohnW2]

Power Quality
Have a read, http://www.reo.co.uk/files/power_quality_fact_sheets_1_-_10.pdf[/QUOTE]
Many thanks - very interesting. Just one question/confusion ... The factsheet, entitled "UK Power Supply" says that the -6%/+10% 'limits' were changed to ±10% of 1st Jan 2003, and the Standard seems to also assume ±10%. I thought the UK was still working to the -6%/+10% - can your clarify?

Kind Regards, John

 

[JohnW2]

"NOTE Application: for the purpose of this standard, the duration of a voltage swell is from 10 ms up to and including 1 min."

OK. Thanks. That goes a fair way to filling the 'gap'. Are there (different) limits for such a 'swell'? One probable problem is that I suspect the standard DNO monitoring equipment may well only determine and report 10-min averages, without knowing or reporting anything about any 'swells' (or 'dips') which may have occurred during the averaging period - so I'm not sure that 'swells' would necessarily be detected by such monitoring.

Kind Regards, John