micro-inverters or optimizers when shading is major issue ?

[RShackleford]

SPS works with the new -41 inverters if you're using the TS4-Fs. You did need a 'work around' with the -40 since the TS4s had an external power source.

If you look at the link I posted (post #43) for a -41 inverter, it states "Note that using TS4 optimizers will disable the Secure Power Supply functionality of the inverter."

[nwdiver]

Maybe it's better than that. The SMA inverters (I'd probably use something like https://www.wholesalesolar.com/29317...us-41-inverter) have three MPPT inputs, so I'd probably do 3 strings of 5-6 panels each. Or maybe it's worse, would each string need 4 unshaded panels, so I'd really be better off with only one string ?

The other thing is, apparently the "secure power" thing does not work if you have optimizers - period. Because of rapid shutdown (and I'm subject to 2017 NEC).

SPS works with the new -41 inverters if you're using the TS4-Fs. You did need a 'work around' with the -40 since the TS4s had an external power source.

[RShackleford]

I really don't think you 'need' optimizers. For example, if you have a 12 panel string so long as at least ~4 panels are unshaded ...

Maybe it's better than that. The SMA inverters (I'd probably use something like https://www.wholesalesolar.com/29317...us-41-inverter) have three MPPT inputs, so I'd probably do 3 strings of 5-6 panels each. Or maybe it's worse, would each string need 4 unshaded panels, so I'd really be better off with only one string ?

The other thing is, apparently the "secure power" thing does not work if you have optimizers - period. Because of rapid shutdown (and I'm subject to 2017 NEC).

[nwdiver]

This is the hard part to determine.
When is adding more panels cheaper than optimizing...

I also recently learned that this is not only about ending up with the same power, but also the stress put on the panels bypass diodes. When the panel is shaded enough that they activate it stresses the diode, and these seem to have a limited life.
Optimizers are supposed to extend the life of the bypass diodes by limiting when they will activate.

I could not find concrete data on what is an acceptable amount of activation before it causes failure. Is it just a percentage in the year or a daily percentage... Like if I have 6 month of the year were they will activate 20% of the day production time (overall 10% of the year), is that a cause for failure...

Bypass diodes are solid state and if the J-Box is located in an area that it gets sufficient cooling (an even bigger problem for optimizers) then there's no 'stress' when it's activated. There's no reason a bypass diode couldn't be 'activated' for the life of the panel. It's also current that causes heat not voltage. The current in the panel doesn't change with an optimizer. If a cell group is shaded in a panel the bypass diode will be activated with or w/o an optimizer and be exposed to roughly the same amount of 'stress'.

The increase in output for optimizers will be <5%. With panel costs <$0.50/w you can definitely add ~5% more watts for $30 per panel. Even if you add racking and labor you're still <$1/w so... that's 30w on a ~300w panel for ~10%. IMO the only time optimizers make economic sense is when you're dealing with multiple angles and/or you're trying to squeeze every kWh out of a limited amount of space.

[scrambler]

This is the hard part to determine.
When is adding more panels cheaper than optimizing...

I also recently learned that this is not only about ending up with the same power, but also the stress put on the panels bypass diodes. When the panel is shaded enough that they activate it stresses the diode, and these seem to have a limited life.
Optimizers are supposed to extend the life of the bypass diodes by limiting when they will activate.

I could not find concrete data on what is an acceptable amount of activation before it causes failure. Is it just a percentage in the year or a daily percentage... Like if I have 6 month of the year were they will activate 20% of the day production time (overall 10% of the year), is that a cause for failure...

[nwdiver]

Because of my shading issues, I need optimizers anyhow, and was planning on Tigo (not wanting to be locked into SolarEdge's "ecosystem").
Not sure how the Tigo ones handle rapid shutdown though.

I really don't think you 'need' optimizers. For example, if you have a 12 panel string so long as at least ~4 panels are unshaded to provide sufficient voltage to the inverter an unoptimized SMA string inverter will perform within ~5% of an optimized SE inverter. IMO the additional cost isn't worth the improved efficiency. Just spend the money you would have spent on optimizers on more solar or more powerful panels and you'll get more kWh/$.

The new SMA inverters also come with 'sunspec' integrated. So you can just add a TS4-F to each panel to comply with NEC 2017. These are ~$30ea compared to ~$60ea for optimizers.

[scrambler]

For me it seems reliability between SE and Tigo is the biggest differentiating factor as efficiency difference (either way) will be minor.
With Tigo you are unlikely to have any issue at all, with SE you will most likely have some...

[ButchDeal]

My understanding is that the Tigo optimizers will work with most any string inverter, is this wrong ?

SMA optimizers and rapid shutdown modules (no optimization) will work with any string inverter but they would require additional equipment to handle monitoring and shutdown etc.
Note that the SMA monitoring kit has a much shorter warranty for some reason too.

Are you saying Tigo won't handle shading issues as well as SE ? This article would seem to contradict both those:

https://solarity.cz/blog/solaredge-vs-tigo/

That Article doesn't say much about which optimizes better...


SolarEdge has module level MPPT with buck& Boost voltage control,
SMA has module level buck and string level MPPT.
SolarEdge is more efficient as well.

SMA charges for full monitoring which SolarEdge includes.

[RShackleford]

Tigo would be the SMA ecosystem though.
They do handle NEC 2017 rapid shutdown but are not as efficient of an optimizer system, and monitoring is weak.
The Bimodal solution from SMA is expensive, and cumbersome like most AC coupled systems.

My understanding is that the Tigo optimizers will work with most any string inverter, is this wrong ?
Are you saying Tigo won't handle shading issues as well as SE ? This article would seem to contradict both those:

https://solarity.cz/blog/solaredge-vs-tigo/

[ButchDeal]

Because of my shading issues, I need optimizers anyhow, and was planning on Tigo (not wanting to be locked into SolarEdge's "ecosystem").
Not sure how the Tigo ones handle rapid shutdown though.

Tigo would be the SMA ecosystem though.
They do handle NEC 2017 rapid shutdown but are not as efficient of an optimizer system, and monitoring is weak.
The Bimodal solution from SMA is expensive, and cumbersome like most AC coupled systems.

TIGO was purchased by SMA a few years ago and loosely integrated in.

[RShackleford]

Also if you have NEC 2017 rapid shutdown requirements you would have to add the optimizers from SMA (Tigo) to get that, adding cost to the build.

Because of my shading issues, I need optimizers anyhow, and was planning on Tigo (not wanting to be locked into SolarEdge's "ecosystem").
Not sure how the Tigo ones handle rapid shutdown though.

[ButchDeal]

For my goal of getting a grid-tied system up and running, without excessive cost, with an eye towards multi-mode upgrade in the future, it seems like a good stopgap measure. Those SMA string inverters (with the "secure power") look like they're roughly as inexpensive and as good as any other string inverter.

Having to switch the outlet on sure beats my current grid-out option, which is a 1kw inverter I connect to Nissan Leaf, and then connect to the main panel (using an interlock). Both are supplying 120vac, so I can only feed one side/pole/phase of the panel so need to move wires/breakers to all my essential loads are on that side. Or just feed both sides, leave the 240vac loads off (wouldn't have enough power to run them anyhow).

you can't legally wire the SMA outlet into your house. It has to be a single dedicated (dual) outlet. You have to switch it on manually which disable normal grid operation of the inverter. You would have to also switch it back to get the inverter to work with the grid. You would have to plug in your loads to this single outlet outside your home with extension cords as well.

Also if you have NEC 2017 rapid shutdown requirements you would have to add the optimizers from SMA (Tigo) to get that, adding cost to the build.

[RShackleford]

Yes, it's manual. And if insolation drops below load it will cut out but it's by far the cheapest add-on option available for a grid-tied system. WAY better than nothing. I'm sure there were a lot of people in PG&E territory that would have appreciated 1 outlet to keep the fridge cold... laptops, cell phones and iPads charged.

For my goal of getting a grid-tied system up and running, without excessive cost, with an eye towards multi-mode upgrade in the future, it seems like a good stopgap measure. Those SMA string inverters (with the "secure power") look like they're roughly as inexpensive and as good as any other string inverter.

Having to switch the outlet on sure beats my current grid-out option, which is a 1kw inverter I connect to Nissan Leaf, and then connect to the main panel (using an interlock). Both are supplying 120vac, so I can only feed one side/pole/phase of the panel so need to move wires/breakers to all my essential loads are on that side. Or just feed both sides, leave the 240vac loads off (wouldn't have enough power to run them anyhow). Of course, other than the fridge and cable modem, my main essential loads are some lighting; so I may be better off just plugging the Leaf's L1 charging cord into the SMA during the day, and connecting the Leaf to the main panel as I've described. In no case am I going to be able to run my 1/2HP well pump, that's the rub.

I actually started a thread asking about my goal, but it got response from only one guy, who seems like a bit of a SolarEdge "brand ambassador".
https://www.solarpaneltalk.com/forum...able-to-hybrid

[nwdiver]

Last I heard, it was a duplex outlet, and you had to manually switch it ON after a grid failure. And possibly after each cloud goes past.

Yes, it's manual. And if insolation drops below load it will cut out but it's by far the cheapest add-on option available for a grid-tied system. WAY better than nothing. I'm sure there were a lot of people in PG&E territory that would have appreciated 1 outlet to keep the fridge cold... laptops, cell phones and iPads charged.

[Mike90250]

Depending on how important this is to you SMA has a feature with their inverters they call Secure Power Supply (SPS). It provides 1 outlet with up to 2kW off-grid provided there is sufficient sunlight. Won't backup your whole house but it can keep the fridge cold and all your batteries charged. It can even charge an EV. Slow L1 but it's better than nothing.

Last I heard, it was a duplex outlet, and you had to manually switch it ON after a grid failure. And possibly after each cloud goes past.