Expanding Existing Solar without MCS

[Jinglish]

Hello,

 

I’ve just had 10 x 410W panels fitted to my SE facing roof. I’ve got a Solax Boost G4 3.68kW inverter and this has been professionally installed with MCS certificate and G98 DNO. I’m on Octopus outgoing.

 

I want to expand my system by fitting a further 8 panels to another SolaxBoost G4 3.68kw on a SW facing roof so that we have generation throughout the whole day rather than mostly in the mornings. However I want to fit this expansion myself….

 

So my questions are:

 

1) Since both devices qualify for G99 fast track I can do the DNO myself but do I need to provide an MCS certificate afterwards?

 

2) If the answer to Q1 is no then since I’m already exporting to Octopus then I don’t feel that I need another MCS or to tell them i’ve expanded. I mean how would they know? Even if I go over 3.68kW the DNO is happy so who cares?

 

J

[JohnMo]

Don't need MCS for anything solar.

 

Unless specifically asked by a supplier, keep quiet. It may get complicated otherwise.

[Jinglish]

That’s what I thought. But do I do a G99 fast track or keep quiet from DNO also?

Edited May 14 by Jinglish

[JohnMo]

16 minutes ago, Jinglish said:

But do I do a G99 fast track or keep quiet from DNO also?

You know there is only one way to answer that on an open forum, so why ask?

[Jinglish]

Fast track it is then 😉 

[Dillsue]

Keep things straight with your DNO as it helps manage the grid for everyone's benefit.

 

No need for MCS.

 

The OFGEM guidance for FIT payers requires them to actively monitor for unexpected over generation. You're not gonna be getting FIT payments but Octopus should already have mechanisms in place for monitoring unexpected FIT Generation so whether they apply those mechanisms to other export schemes is anyone's guess. Even if the did detect over generation they might not be bothered if youre not on a government regulated export scheme(FIT/SEG)?? The risk you run is they pull the plug on your outgoing payments. How big that risk is is anyone's guess!

[Jinglish]

Thanks that’s what my plan was, notify DNO but not Octopus.
 

What information do I need to provide once the G-99 SGI-3 work is completed?

Edited May 15 by Jinglish

[Dillsue]

Just had a look at the spec for your inverter and it allows 200% oversizing on the DC input so theoretically you can hook up all your panels to the existing inverter so no DNO involvement and extended generation through the day. I didn't see if there was a max limit on each MPPT input so check with manufacturer if you can put 4kw on one and 3.2 on the other. 

[Jinglish]

I get what you’re saying however the current inverter nearly maxes out close to 3.6kW at noonish with the 10 panels I have so adding more would clip my generation at the overlap. Plus the other facing roof is actually on a detached garage so wiring would be very difficult and cost more to use the same inverter even if i replaced it with a larger capacity and two MPPTs.

 

[Dillsue]

If there's installation constraints to hooking up extra panels to the existing inverter then only you can weigh those up versus a G99 application, part P sign off and cost of a second inverter. We have a ground mounted 3kw array which is hooked up via a buried SWA cable which is straight forward to install to the inverter inside the garage.

 

For the clipping, assuming you've not got batteries, do you need more/can you use more during a summers day than 3.68kw for the few weeks a year that you'd likely be clipping? With split arrays you can get a combined generation forecast from PVGIS by downloading hourly forecasts for each array and summing them in a spread sheet to see what the peak generation for each hour is. You can then see how many hours/days you'd be clipping in reality. Our second system has 8.5kw of panels spread between SE and NW to give day round generation and peak generation is 5.2kw. The 5.2kw forecast rarely happens and its probably even rarer for us to use the peak for the relatively short time it's there.

[Jinglish]

You make sound points. Just to double check your suggestion I expanded my new proposed array from 8 to 10 panels (as I can actually fit 10). Took a clearer day, and inverted the current SE profile to roughly represent the proposed SW roof; overlayed them which yields the graph below. The overlap area in red would exceed the 3.68kW rating which you can see is a significant area to miss out on. I know this is under optimal conditions which is a lower occurrence rate in the UK but I managed to get a new same model inverter for £200. Looking at S/H panels I found 10 x 400W for £200 then brackets and fitting/wiring myself will cost about another £300. So looking at £700 total to expand up to the SGI-3 limit.

 

I have a solar iboost which will heat the hot water for surplus generation before selling it to the grid for 15p

 

I’ll put all this into OpenSolar software and see what that yields.

 

I don’t have a battery as I think in a few years time V2G will become more mainstream. I’ve got an EV but using Intelligent Go until bidirectional chargers support CCS2 and the motor industry allows this use. Basically I wouldn’t see a return in a house battery by the time V2G is mainstream.

 

 

[Dillsue]

I'd be cautious about assuming Octopus will accept your export where there's a mix of generation from MCS and non MCS systems- they told me they wouldn't accept ours!

 

IF the red bit on your graph is the actual clipped bit then you need to work out how many kwh that equates to then multiply by the number of days you realistically think clipping will happen multiplied by your unit import rate. You'll then see how much you'll likely loose and how that compares to the the cost of adding a second system. As PVGIS is based on historical figures you'd likely be more accurate using its prediction rather than as8ngle day snap shot from a few days ago??

[Jinglish]

Thanks. That was the who point in this post to take caution and ask others. How would they know if I did a non-MCS expansion? Did you tell them or did they discover yours?

 

The likely hood of me generating over 3.68kw which isn’t self-consumed is probably quite low. This is due to Solar iboost and possibly my EV. I guess worst case is I can fit a form of limit of 3.68kw back to the grid so I wouldn’t benefit selling back at 15p over 3.68kw but gives me self consumption headroom up to 7.36kW.

 

Yes correct about the red clipping, this was just a quick mock up in paint. I’ll run it through the MGD 003 when I get more time but your feedback is noted.

[Dillsue]

I asked Octopus before even applying for Outgoing and after a few weeks back and forth they said no, but may do in the future. They are trialling accepting non MCS export but theres a few hoops to jump through to get accepted.

 

I applied for Outgoing 6 months later at the end of last year and supplied only details of our 4kw MCS system thinking I could sneak in the non MCS generation..... they haven't processed the application! Not sure why as they've not been in touch but I'm collating export figures to see if its really worth giving up our FIT deemed export before I chase them up

[JohnMo]

1 hour ago, Dillsue said:

but theres a few hoops to jump through to get accepted.

And a bunch of money to pay them upfront, which would take umpteen years to get back, no point. Plus I believe if it's not a new build you need the structural engineer report to certify your roof is suitable etc.

 

 

[Dillsue]

2 hours ago, JohnMo said:

And a bunch of money to pay them upfront, which would take umpteen years to get back, no point. Plus I believe if it's not a new build you need the structural engineer report to certify your roof is suitable etc.

 

 

Yep and building control sign off and part P sign off. I reckon a grand all in

[JohnMo]

 

 

8 minutes ago, Dillsue said:

Yep and building control sign off and part P sign off. I reckon a grand all in

So about 7,000kWh export required if you can get 14.5p or 22,000kWh if you get 4.5p to break even - Really not worth the effort or hassle. The first figure is close to 1.5 years full production being exported, which never occur. This month so far, just about zero exported.

[Jinglish]

Thank you for your advice; on reflection and crunching some numbers I’m going to wire in the extra 10 panels to the unused second MPPT on my existing 3.68kW inverter (it can handle the over sizing) All i need to do is run two 4mm MC4 cables from the garage to the main dwelling approx 40m. The overlap area is not worth chasing for the extra cost and headache. Reckon I can do this for less than £500!!.

[sharpener]

57 minutes ago, Jinglish said:

All i need to do is run two 4mm MC4 cables

 

Might be a good idea to check the voltage drop. IIRC when I had 3.2 kW added to my second roof only 20m distant I got them to wire as two separate 4mm^2 strings to get voltage drop below the recommended 3%.

 

Victron calculator here suggests 6mm^2, however it now seems to lack the volt drop calculation I remember using but other apps are available.

[JohnMo]

1 hour ago, Jinglish said:

All i need to do is run two 4mm MC4 cables from the garage to the main dwelling approx 40m

6mm2 at least, but you really need to do calcs though. Armoured cable may be better as well with MC4 connections.

[Jinglish]

So my inverter details are on the left, min voltage is 40V

 

My new expanded is 10 stringed array in series to creates the left Vmp (Optimal Operating Voltage) against worst case ambient temperature of 30.8C gives me 287V

Current in series is approx. 10.5A

 

Typical 4mm2 cable is 0.005ohmes per m so with a cable run of 40m (round trip of 80m) I make it a 3.57v drop if I upgrade to 6mm2 (0.003ohms/m) this reduces to 2.38v

using this caculator: https://midsummerenergy.co.uk/voltage-drop-calculator

 

What is an acceptable voltage drop? (I guess green text is good and orange text is border line) Appears to be 3% as previously suggested

 

So again thanks all for the help 6mm2 seems to be the way forward as I'm likely less than 40m anyway probably more towards 30m

Edited May 17 by Jinglish
Found voltage drop answer

[Jinglish]

So I’ve gone full circle.

 

I’ve decided to keep the second 3.6kw Solax and install it in the garage. 9 x 410W panels will fit on the  West facing roof. Giving me the option in the future to fit another 9 on the East facing roof to max that inverter out throughout the whole day.

 

The existing Solax in the main dwelling with 10 East facing panels can be expanded with an additional 3-4 South facing panels in the future.

 

So for now I will have 19 panels with scope to expand to 32 panels split across 2 inverters across different directions. 

 

I’ve applied for G99 fast track and supplied the SLD so just awaiting approval before I fit.

[Jinglish]

To help others expand their system themselves it was very easy. If you hunt around on selling sites you can grab unbelievable bargains I got the following all brand new (except panels)

Solax G4 boost 3.6kW invertor £200

BYD 10 x S/H 410W panels £150

Clenergy rails, hooks, clamps, screws etc.. £234

Electrical Cabelling, isolators, fusebox, MC4, ducting etc... £217

 

TOTAL £801

 

I used free OpenSolar software to do the design and SLD which was approved by National Grid via G99-SGI3 (aka fast track) process.

 

This expansion adds another 4.1kW peak (estimated 2,617kwh per year) to the existing 4.1kW bringing me to a total of 8.2kW peak but clipped to 7.36kW (estimated 5289kW per year)

 

Octopus have no need to be told I have expanded, I'm already on their outgoing plan (15p) + intelligent go from the original system's MCS.

 

Going by the OpenSolar calculation this predicts a payback of the £801 in around 1.5-2 years

[Mattg4321]

On 16/05/2024 at 15:27, JohnMo said:

And a bunch of money to pay them upfront, which would take umpteen years to get back, no point. Plus I believe if it's not a new build you need the structural engineer report to certify your roof is suitable etc.

 

 

I’m looking into this currently and am 99% sure the 3 things you need are DNO approval, part P building control notification and electrical installation certificate. 
 

People seem to be getting confused with the building control notification bit and thinking its related to the roof structure, but it’s actually the part p notification. 
 

It’s £250, with £150 being refunded if they reject you. At 15p a unit exported, if you have a good amount of excess generation that you can’t self consume it should pay back quite quickly depending how much exactly you’re sending to the grid for free. 

[Dillsue]

12 hours ago, Mattg4321 said:

I’m looking into this currently and am 99% sure the 3 things you need are DNO approval, part P building control notification and electrical installation certificate. 
 

People seem to be getting confused with the building control notification bit and thinking its related to the roof structure, but it’s actually the part p notification. 

Are you sure about the building control approval? Using a part P spark dispenses with the need for the householder(Octopus customer) to get involved with BC, so why would Octopus ask for BC approval in respect of electrical works? The part P spark issues the installation cert so no BC involvement for the customer.

 

As you're adding load to the roof you'd need BC approval for that, or an exemption process similar to using a part P spark. I beleive the MCS process involves structural assessment of the roof and negates the need for BC approval. If you go the non MCS route, who verifies the adequacy of the roof if its not BC approval??