That makes a lot more sense (and makes me much more interested).
I didn’t see that in the manual but if you’ve got a link to the supplemental tech brief, I’d be interested to check it out.
Hoymiles HYS-7.6LV-USG1 Hybrid Inverter
- Thread starter ncsolarelectric
- Start date
I didn’t see that in the manual but if you’ve got a link to the supplemental tech brief, I’d be interested to check it out.
NEM2 only compensated excess generation at annual true-up based on NSCR ($0.08 this year).
They credit you the positive NEM balance or (annual net export - annual net import) X NSCR, whichever is lower.
zanydroid
Solar Wizard
I am net consumer so NSCR doesn’t apply. Since net consumer/producer is defined by KWh it benefits the net consumer to maximize the dollar value of the export.
Also PCE credits positive NEM balance for consumers (and I think producers too but that doesn’t sound sustainable), so even if I was $0 annual balance (generation only not T&D) it’s still better to maximize the dollar value of export bc I get a check back.
The 3-5% will probably end up being maybe $15-20 per year but I want it ?
zanydroid
Solar Wizard
Ok. I think there are going to be a lot of NEM2 folks going to net consumer from stuff like adding HP or EV. With 10kWh ESS at $5K (excl labor) in 1-2 years it should break even in under 5 years.
I don’t know where the break even point is for more solar on NEM3 vs storage + arbitrage on NEM2. Probably never for heatpump users.
I may have misinterpreted what you meant by ‘arbitrage.’ I took that to mean charge when prices are low and export when they are high.
If you are charging from grid when prices are low and self-consuming, I would not have called that arbitrage (but perhaps that’s just me).
Also, many of those NEM2 customers already planned for increased consumption including HPs and EVs when they sized their system, not to mention they have their 1KW expansion to add when needed…
‘Capture before export and self-consumption’ is going to be a lot easier and less costly than trying to boost export credits during peak period. And as I stated earlier, once one is in the position of being a net generator, there will be pretty much no value in boosting peak period export credits (since that all gets washed away by NSCR).
zanydroid
Solar Wizard
I just meant only sell at max rate, and take the rate plan with biggest spread between max and min.
It's definitely far from 100% certain that planning was done. Just anecdotally there's a few people asking about expansion here per year, and a few on Reddit, because they didn't plan ahead or it was too far in the future at the time of original installation to make sense. Also depends on the sophistication of the customer and the discussion with the installer. Also there could have been a massive design flaw. 1kW may or may not be enough, depends on how good the remaining area for installing solar panels is. In ideal conditions (5 hours per day) it would get close. But the ideal positions may have already been taken up by existing panels.
I'm pretty sure no amount of expansion would be able to cover both EV and HP on my property. IIRC my math says 9 year break even with 18kpv @$10K (DIY install), so substituting the 18kpv with a $4000 cheaper inverter would be a significant difference. For a turnkey install it will require a lot more cost drops or unfavorable POCO rate changes to make sense.
The math is very different from net producer (which I personally have not analyzed in depth b/c it's not important for me, but I did enough to decide that it's pointless as a net produce)
The primary thing making this easier/harder is that the 9540 hardware is not available so you cannot export to the extent needed by the above planning without being detected. The secondary thing making this easier/harder is that the necessary 1741 Power Control System variant (Export Only (outback description of what these are)) and export sell configuration may not be supported by the inverter (which is only a programming thing, but it is still a dealbreaker). The 18kpv definitely has this capability. And note this is a capability of PowerWall, so it's deemed useful enough to be in the most mainstream of inverters.
From PowerWall docs (and talking to neighbors)
All MID panels I'm familiar with require additional CT's in backup-panel configuration to be placed on the grid input to display grid consumption, support export control/limits and support TOU management. Here is the SolaX manual, see last page, additional CT's. There are connectors for these additional CT's in the inverter and/or BI (MID panel).
FWIW, MID panels are rapidly evolving into smart panels with multiple relay-controlled smart circuits for AC-coupled PV, generator, load shedding and load diversion Fortress Power Avalon ESS.
zanydroid
Solar Wizard
What does MID stand for?
Microgrid Interconnection Device, MID Device aka automatic transfer switch or anti-islanding relay. Not sure who used the term first, probably Enphase. Hybrid AIO and backup inverters all the way back to Trane and Xantrex XW have this relay between the grid line and the backup line.
For users needing higher-powered inverters or the freedom to upgrade / parallelize, the centralized power hub / MID / 2-box solution is attractive.
But for smaller power levels of ~10kW or less with no interest to add more, all that stuff can be put directly in with the inverter itself…
Enphase had to go that route because they integrated their (small) inverters in with their batteries (and needed to put a switched Autotransformer somewhere in the system).
The concept would s great to promote / facilitate modularity except they are all walled-garden (Enphase being the best (worst) example).
The single-box solution where hybrid inverter itself is integrated directly into MID will be more cost-effective for modest inverter sizes (with little interest in parallelizing).
zanydroid
Solar Wizard
How much do MIDs usually cost? Trying to understand the cost tradeoffs between various built-in transfer switch and CLP.
The ideal case is for the inverter to have both internal and external options for flexibility to reconfigure.
Being satisfied with internal only with no remote option implies that you are very confident that you don't need an upgrade path, or you are willing to switch inverters as needed.
Remote only option probably adds a good chunk of cost and definitely a lot of bulk. I don't know how much MIDs usually cost, so I just googled the Growatt 240V ATS one, the one with a neutral forming AT. That is $300 but is also bare bones.
I don't think that's the case for all inverters. IIUC SolArk 12Ks and XW, when stacked, don't use a MID. 12K uses their internal relays in a coordinated way, and XW uses a dry contact for a remote relay (though I guess they could also coordinate the relays together, I think having a single relay controlled directly by stacking leader, rather than telling the followers to control independent relays, feels more robust)
Your right on, if you need a 12 kW backup inverter like Sola-Ark (up to 15 kW), Lux Power or EG4-18kvp and if your grid and backup lines are close to the main panel, this will be sufficient.
- You have losses for grid-tied operation in the lines from the main panel to the inverter back into the backup panel
- If you need to AC-couple micro-inverters like IQ8's via the gen port, then you loose the gen port.
- If you want to stack up another 12k inverter you'll better off starting with a MID panel in the first pace.
- How about an inverter bypass switch? Not needed with a MID panel
- Do you want to have a few smart circuits for load shedding or load diversion?
zanydroid
Solar Wizard
Growatt also does this with their MIN architecture, and used a full sized inverter. The SPH architecture does not need the AT. I believe the SPH also has a CLP port directly on it (manual isn't out yet, I think I teased that from a spec sheet)
Not sure if I mentioned the other inverters using this architecture in this thread, apologies for repeating: dcbel (current), solaredge (at least in the past, you can buy their autotransformers still). It helps with reducing the product development costs for split-phase specific equipment.
Anywhere from $800 (Fox ESS), $1600 for Franklin aGate, $2000 for Enphase IQ System controller to $3200 (Schneider SE BCS).
But here is the catch: you cannot mix and match MID panels! You need to buy the whole package.
Would like to see a generic MID panel with open MID and PCS control standards.
zanydroid
Solar Wizard
I hope those are for smart panel functionality. Otherwise that is some blatant "get you on the add-on feature cost in the walled garden" cash grab.
Wow, Schneider went from letting you put the functionality together yourself to $3200. That's confidence.
Gotcha - I would have called that ‘time-shift + common sense’ reserving ‘arbitrage’ for those trying to ‘make money off of the grid’ by exporting battery power during peak hours, but whatever.
Yeah, so what gets interesting for those folks is, if they add an approved hybrid for time-shift and self-consumption and then they add a bit of new DC-coupled solar to the available MPPTs to charge their battery with extra energy to offset overnight load (but not to export), do they need to get permission from their utility for those added panels?
Of course, if you are maxed out and still cannot generate more than your consume annually, you want all your consumption when rates are low and all of your export to happen when rates are high.
But you’ll still be better-off if on any given day you either exported or imported, but never both…
Sounds like we are in perfect agreement then.
Detected how?
If you go through a full day with zero export until your battery is full and then export to sundown when you switch to being zero import until the next morning when the cycle starts anew, can that be ‘detected’?
If you go through a full day with zero export without any export and the continue into the night with zero import until your battery is drained and you start importing to offset load until sunrise, can that be ‘detected’?
Could not follow the link - can you provide Cliff Notes version of the concept / capability?
If it refers to exporting to grid from energy stored in battery, I’m coming to the conclusion th as t thee as t is not terribly important for modest -sized / modest-cost systems.
As I said, walked-garden.
That would be an absolute game-changer (and a reason to get one).
Yes! That's why I won't consider the Schneider BCS, you can do this with an external contactor. But it is recommended for stacking more than 2 XW Pro's, otherwise you'll have relay arcing and stucked MID relays. The 200A rated relays in Sol-Ark and Lux Power are rated for 200A carry, not necessarily for 200 A make or break. The 200 A relays in MID panels are fully rated 200A make and break, almost all MID panels I know use the Gruner 200A latching relay, from Powerwall 2 up to Franklin WH aGate.