Solar Upgrade AGAIN

[tszaboo]

I agree that Australia does not have "net metering" as the US define it, but Australia reuses the term and that Wikipedia article tries to shoehorn all the different ways the term is used into a single thing. So trying to use the term "net metering" in an international forum/context immediately fails to convey any real meaning unless you want to constantly refer to the specific jurisdiction/rules/explanation.

... p.s. "net metering" that is 1:1 time shifting is broken, it doesn't price storage. For all the free market bluster of the US they sure like to add bizzare bureaucratic distortions to a free market (electricity, corn, steel, etc).

Net metering as a term makes sense, you just have to define the period of it.
It can be yearly, when your net production over a year is substracted from your net consumption.
Or 15 minutes based, which is the typical "momentary" measurement, as far as I know, they will not go lower time resolution than that. This is often called Time of Use Net metering.

[Someone]

I agree that Australia does not have "net metering" as the US define it, but Australia reuses the term and that Wikipedia article tries to shoehorn all the different ways the term is used into a single thing. So trying to use the term "net metering" in an international forum/context immediately fails to convey any real meaning unless you want to constantly refer to the specific jurisdiction/rules/explanation.

... p.s. "net metering" that is 1:1 time shifting is broken, it doesn't price storage. For all the free market bluster of the US they sure like to add bizzare bureaucratic distortions to a free market (electricity, corn, steel, etc).

Net metering as a term makes sense, you just have to define the period of it.
It can be yearly, when your net production over a year is substracted from your net consumption.
Or 15 minutes based, which is the typical "momentary" measurement, as far as I know, they will not go lower time resolution than that. This is often called Time of Use Net metering.

Except the smart meters in Australia (thus likely most of the world since these arent unique regional designs) are instantaneous, and report accumulated incoming and outgoing power (as the single meter sees it) separately over the requested billing/accounting duration. "Net Metering" is a holdover from mechanical/dumb metering days when there was no practical option to accumulate the two independently (hence stupid things like fully gross metered installations, with no self consumption). The 15 common minute reporting period is not a pre-summed import+export total, but two separate values accumulated in separate counters/fields. Even the consumer material explains this:
https://www.solarquotes.com.au/good-solar-guide/net-metering/

[EEVblog]

I agree that Australia does not have "net metering" as the US define it, but Australia reuses the term and that Wikipedia article tries to shoehorn all the different ways the term is used into a single thing. So trying to use the term "net metering" in an international forum/context immediately fails to convey any real meaning unless you want to constantly refer to the specific jurisdiction/rules/explanation.

... p.s. "net metering" that is 1:1 time shifting is broken, it doesn't price storage. For all the free market bluster of the US they sure like to add bizzare bureaucratic distortions to a free market (electricity, corn, steel, etc).

Net metering as a term makes sense, you just have to define the period of it.
It can be yearly, when your net production over a year is substracted from your net consumption.
Or 15 minutes based, which is the typical "momentary" measurement, as far as I know, they will not go lower time resolution than that. This is often called Time of Use Net metering.

Except the smart meters in Australia (thus likely most of the world since these arent unique regional designs) are instantaneous, and report accumulated incoming and outgoing power (as the single meter sees it) separately over the requested billing/accounting duration. "Net Metering" is a holdover from mechanical/dumb metering days when there was no practical option to accumulate the two independently (hence stupid things like fully gross metered installations, with no self consumption). The 15 common minute reporting period is not a pre-summed import+export total, but two separate values accumulated in separate counters/fields. Even the consumer material explains this:
https://www.solarquotes.com.au/good-solar-guide/net-metering/

I've still got dumb net metering.

[EEVblog]

So it looks like I can get Enphase IQ7A's for $180ea, that's not bad.
The new IQ8A's are $235ea
No idea what panels I'll get, as LG are goneski, but I'd happily buy old stock if anyone has any.

[bdunham7]

So it looks like I can get Enphase IQ7A's for $180ea, that's not bad.
The new IQ8A's are $235ea

AUD$ ?  That would be pretty cheap IIRC. 

[EEVblog]

So it looks like I can get Enphase IQ7A's for $180ea, that's not bad.
The new IQ8A's are $235ea

AUD$ ?  That would be pretty cheap IIRC.

Yep, AUD. Seems quite cheap. Cheaper if I buy 20

[richard.cs]

Net metering as a term makes sense, you just have to define the period of it.
It can be yearly, when your net production over a year is substracted from your net consumption.
Or 15 minutes based, which is the typical "momentary" measurement, as far as I know, they will not go lower time resolution than that. This is often called Time of Use Net metering.

On a short enough timescale all metering is net metering. i.e. if you export for a few cycles and import for a few cycles alternately the meter measures net. EU meters at least have a 3600 J buffer that you can freely borrow from the grid and return. I have a controller for a water heater that works that way, integrates energy use and does bang-bang control of the heater to stay within a window about 1 kJ wide giving zero metered import/export without actually needing to modulate the heater power. On a mechanical meter it rocks the wheel backwards and forwards without incrementing the counter.

Edit: fixed typo

[EEVblog]

Just realised that my eixsting Enphase system has two Q relays and doesn't seem to have room for any more.
They are 4.8KVA rated each, and each one currently only has 7 x 295VA IQ7+ inverters on them.
https://www4.enphase.com/sites/default/files/downloads/support/Q-Relay-DS-EN-AU.pdf
That leaves me 2735VA capacity left over for each Q relay.
If I get the IQ7A's at 366VA each that gives me an extra 7 panels per relay in theory.
But I think there is some derating going on here because my existing system needed two Q relays when the total VA was only 14x295VA=4130VA
So I'm not sure how that works.
I can of course install an additional box with an extra Q relay if needed, it's just, messy.

[fourtytwo42]

Net metering as a term makes sense, you just have to define the period of it.
It can be yearly, when your net production over a year is substracted from your net consumption.
Or 15 minutes based, which is the typical "momentary" measurement, as far as I know, they will not go lower time resolution than that. This is often called Time of Use Net metering.

On a short enough timescale all metering is net metering. i.e. if you export for a few cycles and import for a few cycles alternately the meter measures net. EU meters at least have a 3600 J buffer that you can freely borrow from the grid and return. I have a controller for a water heater that works that way, integrates energy use and does bang-bang control of the heater to stay within a window about 1 kJ wide giving zero metered import/export without actually needing to modulate the heater power. On a mechanical meter it rocks the wheel backwards and forwards without incrementing the counter.

Edit: fixed typo

I think this is highly dependant upon the exact model of smart meter involved, I have heard several stories where this method typically used by grid frequency burst controllers does not work and newer more expensive high frequency pwm techniques have had to be used instead.

[richard.cs]

I think this is highly dependant upon the exact model of smart meter involved, I have heard several stories where this method typically used by grid frequency burst controllers does not work and newer more expensive high frequency pwm techniques have had to be used instead.

A few years ago I went digging around in the metering standards trying to work out what would happen when my meter got upgraded and it certainly appeared that there was a deliberate 1 Wh deadspot mandated. I think it was referred to as an "anti creep" feature.

Edit, a very quick search found this, which suggests most but not all meters implement it: https://learn.openenergymonitor.org/pv-diversion/background/meters

[rowlandcrew]

I have just gone through an incremental upgrade to try an ensure no grid draw when charging the electric car while powering all the normal stuff.  Some observations seem generally applicable:
1) supposedly, and verifiable, the panels to purchase should have bypass diodes so shading issues are limited to the shaded panels.
2) for most system sizes, likely the cheapest way through would be 400V strings going into a reasonable multiple inverters, including possibly buying broken or used solaredge inverters (refer to mppt comment below).
3) for 2022, buy used 400 watt panels.
4) If you find used solaredge optimizers and want to try panel level mppt, some Pxxx "excess" box lots are cheap, there is no need to be eco-system tied to solaredge inverters. Just buy either a solaredge SMI or the SE1000key to effectively turn off the proprietary connection and use whatever inverter you want.
5) Given the cost of the above, it seems hard to justify microinverters, but if they can be found cheap, then go for them.
6) Most optimal to have an installed solar system that is of the type you want (grid tie, or whatever) because increments  or mods you can do yourself far more efficiently than subcontracting, with only the oversignt you want.
7) the recent film capacitor debate is relevant. It appears common to use an output inductor and an X2 cap to limit noise and improve PF. My experience is the cap suffers from incremental "burn in" (I tried epcos). At least for solaredge I have not seen any semiconductor burn out in failed inverters, just caps and mov's.  Onboard Inverter monitoring appears hyper sensitive to inappropriate LC filtering, probably because of utility-regulator requirements, which may happenstance improve semiconductor survivability.  The solaredge method is to quickly open internal relays.  See photo for example of a "burn in" event that was managed by the solaredge computer.  the dropout of one of my 3 paralleled inverters lasts for about 1 minute, does not reboot the inverter, has occured 5 times, once a day, since install of a new X2 cap a week ago.
8)  I would like to see some continuous scope monitoring of X2 caps with captured self healing events to compare. The data must be out there somewhere.   

[EEVblog]

I have just gone through an incremental upgrade to try an ensure no grid draw when charging the electric car while powering all the normal stuff.  Some observations seem generally applicable:
1) supposedly, and verifiable, the panels to purchase should have bypass diodes so shading issues are limited to the shaded panels.
5) Given the cost of the above, it seems hard to justify microinverters, but if they can be found cheap, then go for them.

My issue is that I already have the Enphase microinverters, so from a system standpoint it makes sense to pay a bit more and standardise on them instead of getting yet another string inverter system and yet another monitor and data collection system etc.

Debating whether or not to just buy the parts myself and install it all myself, and then get an electrican to do the final mains hookup. Or just pay more and have it all done in a day as a system job.

[bdunham7]

My issue is that I already have the Enphase microinverters, so from a system standpoint it makes sense to pay a bit more and standardise on them instead of getting yet another string inverter system and yet another monitor and data collection system etc.

Two additional minor things to think about:

1) My ten-year old system has apparently developed an issue with one panel or microinverter.  I haven't figured it out yet, but it is likely to be a failing inverter or (less likely) a rat has chewed a cable or there is a bad connection.  Until I get time and reasonable weather to get up there and have a look, my system is down to 29/30ths of capacity.  So I'll get to it when I get to it.  Not all Enphase systems are like this. Some, like yours, do have single points  of total failure like the relays, but panels, inverters and the Envoy hub aren't in that group.

2) The custom inverter cabling on the Enphase setup is a significant additional expense, so even if you already have the Envoy, etc, you have to figure in the wiring on a per-panel basis.  Did you get a quote for that?

[rowlandcrew]

"My issue is that I already have the Enphase microinverters, so from a system standpoint it makes sense to pay a bit more and standardise on them instead of getting yet another string inverter system and yet another monitor and data collection system etc."

Sounds like you prefer uniform circuit layout, but I wonder how much this is so as a function of you spending extra dollars for uniformity.  As for multiple systems, I simply added the incrementally added AC production loop through my data logger. yes this gives a sum, but it is immediately apparent if one of the inverters is down or, with my one second time resolution I can see one inverters' capacitor healing.  My main benefit to a data logger is managing home use times. while not as fun as seeing all the technical details, the focus quickly becomes power flow cost. 

Your "starting" solar asset is whatever is installed (remaining) after raising the roof (after dismantling parts of your solar system).  Your valuation of this is different than an installer because you care about additional issues.  Paying someone to do the work leans toward uniformity because installers only like uniformity.  this installer preference is to their benefit, not yours.  The equation for cost is personal-- a sum of:   your time, clock time, opportunistic parts bargains, fun, the benefit of last minute upgrade opportunities, where to put battery, better installation practices that contractors dont care for.....   

I did not want to buy additional conduit, monitoring, etc.  But I did need more racking, some conduit, as a minimum to get to the nearest tie-in point.  The minimalist plan chosen was to slip in three more wires into the nearest installed infrastructure.  this did save parts, time, money, for me.  As far as utility connections, if they even looked, they only see a little more power showing up.  Example marginal stuff to add would be larger conduit with spare wires that can carry DC or AC (for later use), better access to the hardware (layout) because at some point something will die.  One item I did not go for was the newer spec solar wire, because of cost (copper, diameter, and later planned upgrade).  they have 1500V wire that opens the future for HV panels and longer strings, both of which are destined to reduce total cost.

If you are raising your roof, this sounds like larger scale modifications will have marginal costs.  With a small roof, doubling each panel power output (trade old for new panels) is the most effective way to reduce power purchase up to the point where you produce and sell twice what you buy ($sell/$buy).  As the utility reduces payments, this requires rebalancing the cost for a battery against using the grid as your battery. 

[NiHaoMike]

Just realised that my eixsting Enphase system has two Q relays and doesn't seem to have room for any more.
They are 4.8KVA rated each, and each one currently only has 7 x 295VA IQ7+ inverters on them.
https://www4.enphase.com/sites/default/files/downloads/support/Q-Relay-DS-EN-AU.pdf
That leaves me 2735VA capacity left over for each Q relay.
If I get the IQ7A's at 366VA each that gives me an extra 7 panels per relay in theory.
But I think there is some derating going on here because my existing system needed two Q relays when the total VA was only 14x295VA=4130VA
So I'm not sure how that works.
I can of course install an additional box with an extra Q relay if needed, it's just, messy.

Does the Q relay do something with the current flowing through it (e.g. measure it) or could you wire in a contactor to switch the additional strings?

[EEVblog]

"My issue is that I already have the Enphase microinverters, so from a system standpoint it makes sense to pay a bit more and standardise on them instead of getting yet another string inverter system and yet another monitor and data collection system etc."

Sounds like you prefer uniform circuit layout, but I wonder how much this is so as a function of you spending extra dollars for uniformity.  As for multiple systems, I simply added the incrementally added AC production loop through my data logger. yes this gives a sum, but it is immediately apparent if one of the inverters is down or, with my one second time resolution I can see one inverters' capacitor healing.  My main benefit to a data logger is managing home use times. while not as fun as seeing all the technical details, the focus quickly becomes power flow cost. 

[EEVblog]

Does the Q relay do something with the current flowing through it (e.g. measure it) or could you wire in a contactor to switch the additional strings?

It's just there to isolate the system. Legal requirement. Each one has some spare current handling capacity.

[NiHaoMike]

I was thinking to have the Q relay switch some contactors and then use the contactors to switch as many strings as you want.

[EEVblog]

Change of plans.
4 big panels (e.g. 450W-500W) on the pergola roof out the back going into a Fronius inverter to replace the existing SunnyBoy.
Will need a two string inverter, one the existing 12x250W panels, and the new 4x500W-ish panels. The existing SunnyBoy is maxed out at 3kW already even though it has another available string input.
Will give us a nice peak power boost during the day to charge the EV year round without too much cost and impact the visual appeal of the front roof.

Don't know about mounting the panels on the pergola colourbond roof sheets though? Anyone done this?
Could even have a go at mounting them myself maybe.

[EEVblog]

Someone on Twitter pointed me to these corregated solar rail mounts:
https://legacy.k2-systems.com/en-AU/pv-mounting-systems/singlerail-system

Looks good, I think I'll DIY these on the pergola.

[Ed.Kloonk]

Someone on Twitter pointed me to these corregated solar rail mounts:
https://legacy.k2-systems.com/en-AU/pv-mounting-systems/singlerail-system

Looks good, I think I'll DIY these on the pergola.

They are in QLD. I don't think there's very much in the way of suppliers in NSW.

[EEVblog]

Someone on Twitter pointed me to these corregated solar rail mounts:
https://legacy.k2-systems.com/en-AU/pv-mounting-systems/singlerail-system

Looks good, I think I'll DIY these on the pergola.

They are in QLD. I don't think there's very much in the way of suppliers in NSW.

They don't ship?

UPDATE: i tried to register for the software and it's only for installers it seems, not for private use. Seems I can't buy their stuff 

[Ed.Kloonk]

Someone on Twitter pointed me to these corregated solar rail mounts:
https://legacy.k2-systems.com/en-AU/pv-mounting-systems/singlerail-system

Looks good, I think I'll DIY these on the pergola.

They are in QLD. I don't think there's very much in the way of suppliers in NSW.

They don't ship?

I prefer to pick up where possible esp when shipping costs are involved or am dubious of the delivery ability. I'm sure they ship, and I'd be interested to know if they do small quantity.

Edit. Ah, yeah. The mob near Campbelltown told me the same last year. They just aren't interested.

[EEVblog]

I found this Fronius Primo GEN24 that does a 3kW small scale backup that would be ideal for what I had in mind of backing up just the fridges and maybe light and some other things in case of power failure. Without the hassle of having a full house AC backup system.
Would support both my existing 3kW string and the new 4x500W pergola system.

https://www.fronius.com/en/solar-energy/installers-partners/products-solutions/residential-energy-solutions/gen24plus-inverter-with-flexible-backup-power#anc_keyfeatures

[rowlandcrew]

here is a scaleable microgrid isolation device that fundamentally is a directional contactor to install at the service entrance.  the idea is to have some battery and be able to defeat utility curtailment of your solar production, yet you can suck as much grid power as you want.