the 1% or not and network voltages

[edexlab]

1% volt drop as we all know is the advised recomendation, personally if I'm doing a job I follow advice like this, I'd rather not go back to a customer and say oops got it wrong sorry!
however the firm I sub to have decided long runs on the larger systems we're doing will be ok with upto 5% volt drops , for example one of these jobs is a 50kw with a 90 metre run and I expect them to have problems but they are sure it'll be ok

What I would like to know is has anyone else done the same and did they experience problems, I know the theory with voltage rises (not an expert) but is this a problem in reality

With regard to Network voltages I have been talking to some DNO guys who are upgrading the transformer on a job (where the customer has 3 systems a 10kw existing and a 40kw/10kw installed by us on sites literally across the road)

They were complaining that Pv is a problem for them as it is causing lower voltages on the grid, I thought at first they meant this was because they are having to tap down the transformers but they said no this doesn't help but the lowering of voltages occurs anyway due to the localised generation and having a lower network voltage meant that some more rural properties were experiencing problems

So now I'm a bit confused as everything I've read points to voltage rises where microgeneration exists and at first I thought maybe it was some kind of wind up but I spoke to another guy there who had the similar ideas , I am aware that these guys are not degree level engineers but I'd be a bit concerned if they did'nt know the correct facts
So can anyone clarify this for me is the above true or not? any info would be appreciated especially from the Engineers on the forum

 

[Gavin A]

I asked power-one about this, and they said there's no problem with going higher than 1% volt drop providing that the grid voltage is relatively low to start with, and the inverter won't therefore be tripping out if it's needing to push out at proportionally higher voltages.

Only problem with this will come if the local grid voltage is forced higher either by the DNO, or by the addition of other PV systems in the area in future, at which point your systems will trip our before the other systems on the local grid.

 

[Worcester]

@edexlab, I've been talkling at length with a number of Western Power Network planners about Grid Voltages, and they say it is as you suspect, micro-generation / renewables gneration is causing higher voltages (when it's generating) this is causing all sorts of problems. Apparently the central transformers can't reverse feed generated power so it has to be consumed locally . If there are lots of generating plants that tends to push up the grid voltage, so the engineers have had to go round and tap DOWN the end transformers for a lot of people, to LOWER their voltage. Now take a dull and wet day with no solar pv generation (we'll ingnore the wind effect for the moment) and all of a sudden you've potentially got customers complaining of LOW voltage, so in a round-about way both are right!

The solution to the higher voltage drop is simple, put a G59/G83 relay as appropriate at the incoming grid point and either use remote signalling to cut out the inverter if on a shared cable, or simply apply at the grid connection piont. There you are, you saved £x on cable, and spent £y on a G83/G59 relay and witness testing.

If they run with >1% voltage drop and the site is required to use G59 there is a VERY good chance you'll get the inverters tripping if you rely on the inverter controls. We've seen this very problem in practice, even where we'd designed for 1% drop, the high incoming voltage was at the top end, so instead of swapping the transformer at their cost (they had approved a 100kWp system..) as it was already on the first tap, the DNO "kindly" let us apply G83 settings to the inverters and everything was fine therafter.

 

[edexlab]

Worcester
thanks for the answer above , I'm amazed that they are doing this as they have also had problems on another job where the VD was around 3 % and this resulted in the transformer being tapped down and as you say the low voltage was indeed a problem more so during the winter months ie increased loading on network etc
I'm not familiar with the use of G59 relays as WPD have accepted the systems without separate relays, inverters have the G59 certification and I can see how positioning the relay at the origin can have some advantages but how does it solve the volt drop issue?
do you use the relay to provide the required G59 settings/protection and change the parameters on the inverters to allow the higher voltage that will be present while the system is generating as this seems the only way this would work ?

 

[edexlab]

Also I have been talking to another Electrician about some of these problems and he suggested installing a separate private transformer to step the voltage down slightly at the incoming mains but I suspect that these would have to be either adjustable or custom built and as with the central inverters be able to reverse feed? to allow export
I would also imagine that this would probably be an unnecessary headache and as expensive as fitting the correct size cable
any thoughts on whether this idea is feasible.

 

[Worcester]

You got it right on the first reply, the inverters are allowed to 'over voltage' locally - you can set any parameter you like on most inverters with the correct authorisation from themanufacturer, and then at the DNO conenction point, the relay is providing the right level of protection, so it gets over the local losses.
Just bear in mind it's not just voltage drop, these are real losses, so how much is 2% a year worth to the customer , over 25 years, did they build that into their performance calulations they submitted to the client


If you put a transformner between the inverter and the DNO conenction, you'll now need a seperate relay / certification for the transformer as you are no longer conencting the inverter direct to the grid. We have this excat situation where we have installed whole site Voltage Optimisation, it goes between the main incoming and the main DB (these DB's tend to be about 2m high and 4-6m long ) so then we have to fit a G59/G83 relay upstream of the VO transformer and then remote relays to cut out / in the Inverters.

Same as when you use the EMMA with EC (Export Control nee GVS) you have to tell the DNO that you are limiting the export to the grid with a specific device.

 

[edexlab]

Good solution but i have a couple more questions

The setup you are talking about with Voltage optimisation is this a specific requirement of the DNO that you use this method of separate relays controlling each inverter which are then controlled by the G59 relay or is it just that it was impractical to branch off for a separate PV DB due to lack of space in the plant room and you're doing this simply to allow the Pv circuits to be connected into the existing DB 's

Also have you installed any systems with back up generators onsite ?

The firm I'm subbing to have had panels built which remotely controls the pv systems and switches them off when there is a loss of grid power to avoid damage to the generator , and I've been thinking about alternatives to this in the event this set up can't be used for some reason.

 

[Worcester]

Specific DNO requirement - yes / no - it's the only way to meet the G59 regulations, - you don't need one relay per inverter, you just need a relay that cuts the power to the PV distribution board (we always put a dedicated DB in when there is more than one inverter), sorry if I confused you above.

Yes we have quite a few sites where there is a stand by generator, most of the time though it only feeds key essentials on the site and the rest is isolated, whichever way it is done the generator should have an automatic isolation / islanding switch so that it can't reverse feed to the grid. By connecting and/or monitoring 'upstream' of that point so far we have had no problems. If you connected the PV downstream of the islanding point then you'd have to once again have you G59 detection in place upstream. If there is no islanding then the DNO will through a wobbler, as the client will not have authority to connect such a generator to the grid.

 

[edexlab]

The Generators are controlled by an automatic changeover switch so there is no risk of backfeeding to the grid
I'm currently doing 3 x >20 kw / 1x 30kw for a client and all have generator backup for each building covering everything, although on one which has its own new supply we are connecting upstream of the generator
The firm I'm doing the job for are using the G59 settings (firmware updated) on Tripowers which as far as I know has been approved by the local DNO although I can't say for certain that the DNO knows about the generator back up , but I'm going to ask the question to see if they were aware of this as I had no part in the application.
I'm sure that as long as it does switch over and become an Island with no backfeed then the DNO isn't concerned as it then becomes a private supply ,the panel with remote control over the PV is being used because it isn't practical to connect upstream of the generator and we need a way to switch the Pv off when the Generator starts up.