AC Voltage Drop 250m run

[Worcester]

We've got a site where we're installing a 50kWp (3 phase) system on a barn and then feeding the main buildings 250m away.

I've just gone through the BRB, and allowing for both the operating temperature and load factor, I got results of 95mm2 Cu and 120mm2 Al - this seems a bit on the large side!

Clearly 1% is nonsense, the Guide (2.3.1) says in this case don't use 1% and refer to the BRB.

Only other figures I can find in the BRB, are 3% and 5% which seem a bit high, and gives cables sizes much smaller and now down to current carrying capability.

Any one got a better answer? (and why)

Thx

 

[BruceB]

In other commercial situations, with substantial loads over those sort of distances, some designers even go above the 3% and 5% figures you are quoting depending on the characteristics of the load and the normal voltage at the origin of the installation.

For a logical answer, I would consider 2 arguments:
- first for various sizes of cable consider the energy lost in the cable and compare the value of that energy (generation, export and avoidance of purchase) over 25 years with the cost of the thicker cable. That should make it clear whether thicker cable is economically sensible.
- second, consider the actual site conditions, the voltage at the origin, the voltage rise for various cable sizes between the origin and the inverter(s). If thin cables are putting the voltage at the inverter too near 264V then no good. (You have to judge what 'too near' is by the design of the inverter and the stability of the supply locally)

Of course the first argument can be got around partly (for the generation element) by putting the generation meter at the far end rather than the building end of the cable. Decide whether you and/or your customer are wearing a green head or an economic head.

And finally when doing the sums, from your description, as you are connecting to an existing building it gets that FIT rate rather than the standalone rate.

Regards
Bruce

 

[Worcester]

@BruceB

Thanks for that. This is definitely an economic decision, my concern was to ensure a high enough output from the inverter at the load end, so that the inverters don't trip out and thus maximising output, brain wasn't working when thinking that the inverter will raise it's output to cope!


So need to ensure that the 'seen' voltage by the inverter allowing for the drop is within the limits of actual voltages delivered at load end PLUS losses (rise). We're looking at 3 x SMA SB17000TL's which has a grid voltage spec of 160-280V which is a pretty wide band.

So allowing for worst case scenario of 253V at Load end, and 264V inverter cut out gives me an 11 Volt drop or 4.17%, which brings it down to 35mm2 of Cu ... Will have to see if Steptoe has got some of that and colour the ends up appropriately.

Will definitely put the meter at the inverter end . However since the site will consume about 75% of what will be generated (you'll only ever get 100% if continuous base load during generation time is > max output capability) we want to minimise the losses.

 

[paulsamuel1984]

Worcester,

The 264V inverter cut out that you mentioned, is that the G83/1 requirement? Is it the same values for the G59/2 certification?

I thought under the G59/2 settings, the inverter tripped out at 253 after 1 second?

 

[BruceB]

G59 has a 2 stage requirement, G83 only 1. The 2nd stage of G59 is the same as the G83 requirement.


Regards
Bruce

 

[whlphil]

Whats the values for the 2 stage? I seem to remember reading the g59 cert for the tripowers and it mentions 253v but reading the same cert for Solarmax MT's that read 264v is that the two separate levels with long and short cutoff?

 

[JD6400]

Hi Worcster , what amperage are you looking at and is buried at all , also i assume it is obviously 4/5 core ?

 

[BruceB]

Whats the values for the 2 stage? I seem to remember reading the g59 cert for the tripowers and it mentions 253v but reading the same cert for Solarmax MT's that read 264v is that the two separate levels with long and short cutoff?

That is confusing:

SMA has a 2 stage overvoltage cut-off for G59 which is 230+10%=253V and 230+15%=264.5V
Solarmax uses 240+10%=264V and 240+15%=276V

They cannot both be right as I read it.
My own reading of G59/2 is that SMA have it correct, but I am willing to be corrected by someone who knows better.

Regards
Bruce

 

[paulsamuel1984]

BruceB

If I was looking at a worst case, like Worcester of 253V at load end, I have 11 volts before inverter cuts out at 264V.

But if I am looking at 253 at the load end, and the 1st stage cut out under G59 is 253 volts, then a 2.5% volt drop for example could cause issues.

Correct my if I am wrong in my assumptions, I understand the 2 stage requirement but I am not sure what I should be referring to when determining if the volt drop is acceptable.

 

[BruceB]

I have only read the book, not actually done a G59 install. All my >16A per phase installs have been G83/1 Stage 2. So I cannot give you an answer from personal experience. But it would appear to me that the Stage 1 over-voltage cutoff under G59 from the paragraph I posted above is 253V for 1s and the Stage 2 over-voltage cutoff is 264.5V for 0.5s. That seems potentially a good reason for having a separate G59 protection device near the service head, rather than way down the line at the inverter(s), and leaving the inverters on G83 settings from the over-voltage perspective.

Post #2 in this thread gives my view on how to decide if voltage drop is acceptable.

Regards
Bruce

 

[paulsamuel1984]

BruceB,

Thanks for the information, I just want to get it straight in my head before I proceed.

If, as a worst case the supply voltage was 253 (+10%), the inverter output would be this value plus the required voltage to negate the volt drop. So for example if volt drop on cable to the grid was 2%, then the inverter output would be 258V.

This would trip out under G59/2 but not under G83/1.

If I run a voltage recorder for 2 weeks at the load end, where I am connecting the PV, I can obtain the load voltage and work out the supply voltage based on the potential volt drop. This would assist in determining if the existing cables are acceptable and the inverters will not trip out.

Your alternative for a separate G59 protection device closer to the supply point is very useful, thanks for this.

 

[BruceB]

That sounds OK, but I am a bit wary of saying anything that you are going to rely on as I have not seen the actual set up. Have you got any locals you can ask for a second opinion? There are others on this board with plenty of G59 experience who would be better placed to comment.
Regards
Bruce

 

[paulsamuel1984]

BruceB,

I appreciate that, I have to go and see one of my electrical sub contractors later and I want to run this by him. I wanted to seek some other opinions also and this forum never seems to let me down when I have a query!

Thanks again, if anyone else has any other thoughts, please feel free to let me know.

Regards