firmware change to Aurora PVI-3.6 OUT-D to remove 3600 watt "ceiling"

[Sunshine777]

I have an 4kWp system (16 250w Sanyo panels) installed in late 2011 with the subject inverter. Somewhat to my surprise I discovered that the inverter is "throttled" not output more than 3600 watts. I understand a firmware upgrade can be installed to remove this limit. That brings me to my questions:

1. I assume the DNO (SSE) needs to be notified and to agree to the change (going from G83 to G59)?
2. Is there anyone here who has or is authorised to make the FW change? I'm located in SE Hants.

I suppose a valid question is what real gain I will see if the inverter can peak at 4000 watts - hard to quantify in advance. But I do notice that even on hazy days I can hit the 3600 watt peak so can't help but think the panels have more to give much of the time.

 

[Gary K]

My 3.84kWp can hit the 3.81 limit of my inverter but only when the sun appears after a period of shade, and only until the panels warm up. I doubt for the same reason, you can manage a steady reading over 3.6kW in constant sun so would think the gain would be very minimal.

 

[Sunshine777]

My 3.84kWp can hit the 3.81 limit of my inverter but only when the sun appears after a period of shade, and only until the panels warm up. I doubt for the same reason, you can manage a steady reading over 3.6kW in constant sun so would think the gain would be very minimal.

Sorry but I don't agree. On many days the clipping in the output graph is very visible.

 

[Gary K]

Fair enough,could you turn a string off and see what the steady output from 8 panels is before making a decision?

 

[Gavin A]

the limiting will be a lot less than 1% of generation across the year - we install 5kWp systems on 3.68kW limited AC output inverters, and they come in at just under 1% modelled loss of output due to the limiting over the year, as a comparison.

The Sanyo panels will have higher peak outputs due to their 1/3 lower losses at higher panel temperatures, so they would naturally peak above 3.68kW AC output more of the time than a standard mono/ poly panel, but I still doubt it's going to have more than a 0.3-0.5% impact across the year.

Also, virtually all of that gain would not be used in the house, just exported, so the gain is only in the feed in tariff side of things, not energy savings in the house.

There's also no guarantee the DNO will allow this, we've had several similar requests refused in recent months to the point that we've largely given up asking the question being as it's less than a 1% loss anyway for most of the systems we want to install.

 

[Sunshine777]

Here's a graph from a day in May of this year. Admittedly May is a month with typically clearer skies and cooler temps but it serves to highlight the issue.

 

[Sunshine777]

Fair enough,could you turn a string off and see what the steady output from 8 panels is before making a decision?

Excellent lateral thinking - will try that on the next suitable day.

 

[The Solar King]

You also have to take into account the cost of G59 application. You will never get your money back.

 

[The Solar King]

Just on Gavins's Post. I have modelled a 6kW west facing system using a larger G83 limited inverter. This is in Scotland, but even upping output by 10% above prediction showed no more than 30kWh loss per year due to clipping. The benefits in improved output during the rest of the year far outweigh the minor losses in May June and July (which may not even occur).

People get stuck up their backsides over FIT categories and can fail to look at the bigger picture of what is possible.

I wouldn't worry too much about any losses you are experiencing for the very reasons Gavin gives.

 

[Gavin A]

Not so sure about an actual west facing system, the east west splits are definitely ok at those sorts of levels or potentially a little higher.

 

[Sunshine777]

Excellent lateral thinking - will try that on the next suitable day.

OK, turned off a string today for half an hour and watched the output. During sustained sunshine the one string had an output of 2000 watts, peaking at 2030 or thereabouts but staying above 2000 watts for at least 10 minutes till the next cloud came over. Ambient temp is about 20 and panel temp is about 42 - there's a light breeze as well.