Use only 1 o 2 strings of a 3 string grid tied inverter.

[josepr]

The price difference between a 3.8KW and 6kw grid tied string inverter is $150. Initially I want to install 3kw of DC. According to PVWatts that is enough to cover my current needs. But in the future, I want to transition to all electric appliances (water heater, dryer, stove)(as they become old or require replacement). It is safe (do not affects the inverter or void the warranty) to install just 1 or 2 strings of a 3 strings inverter? So that in the future I only need to buy panels and cable to expand the capacity. Also, in this jurisdiction you do not require additional permits if you maintain your AC size. You do not need permit and the Electric Company permission to add, replace or remove panels if your AC capacity remains the same.

[SunEagle]

Depending on if the grid tie inverter efficiency is not bad with half the panel wattage I would go with the 6kW unit.

Remember the NEC states the circuit breaker size for the AC connection is based on the inverter kW rating and not the panel wattage.

[Mike90250]

If you add electric appliances of the type you describe, you are likely to require a Main Panel upgrade to handle the extra load, unless your house is already prewired .

[foo1bar]

It is safe (do not affects the inverter or void the warranty) to install just 1 or 2 strings of a 3 strings inverter?

Probably.

I'd look up the datasheets and installation instructions for the particular models and check.

For example, both the Solaredge 3800H and 6000H look like they would both work with a single string of 10 300W panels. I don't think the 6000H can do 3 strings, but you probably wouldn't need 3 strings for only 6kw (you could do 2 strings of 12 of 300W panels (7200W DC) and I think it'd be within spec for the SE6000H.

Depending on where you are, you may need "rapid shutdown" - which often makes Solaredge systems a good solution from an economic view.

Good luck

[solarix]

There is so little difference in price between small inverters and big ones, we almost always just go right to the 7.7kW size for every body. Yes you can just do two strings initially leaving the third for future expansion. All I ever hear is "more power", "more power". We are askied every time we work on an old system if it can be upgraded to a larger size. Much cheaper to allow for expansion initially than down the road. Some day you are going to want an electric car, plan for charging it of solar now....

[Ampster]

My Permission to Operate specifies both the capacity of the panels and the inverter. That is PG&E in California. You would have to look into the details with you power provider. I agree with Solarix that planning for more capacity is less expensive now than changing out or adding an inverter later.

To me it is a financial hedging decision that can affect your ROI minimally.. The small extra cost mentioned by Solarix is trivial in the long term.

EDIT: Upon rereading the OPs original post it appears that once the inverter size is approved by the power company, there is no further approval necessary to add additional panels. That adds further weight to the decision to go with a bigger inverter.

[J.P.M.]

My Permission to Operate specifies both the capacity of the panels and the inverter. That is PG&E in California. You would have to look into the details with you power provider. I agree with Solarix that planning for more capacity is less expensive now than changing out or adding an inverter later.

To me it is a financial hedging decision that can affect your ROI minimally.. The small extra cost mentioned by Solarix is trivial in the long term.

As long as it's excess capacity in the inverter, FWIW, I'd agree that's a potentially wise choice to design for some excess inverter capacity, mostly because the increased cost of a larger inverter is not much.

Adding, say, 50% size to an inverter however is not as costly as increasing the array size by 50 %. I'd highlight the idea that if maximizing system cost effectiveness is one design goal, and excess capacity being usually among the biggest factors that can kill that array cost effectiveness, it needs careful consideration and balancing among the other design goals.

[solarix]

Even if you never add array to an oversized inverter, the inverter's lifespan will probably be much longer because it is working well below its capacity. That alone could justify the marginal increase in upfront cost. There is only a $218 difference in my wholesale costs between a 3kW SunnyBoy and a 7.7kW Sunnyboy.... That is less than half the cost of their extended warranty.

[J.P.M.]

Even if you never add array to an oversized inverter, the inverter's lifespan will probably be much longer because it is working well below its capacity. That alone could justify the marginal increase in upfront cost. There is only a $218 difference in my wholesale costs between a 3kW SunnyBoy and a 7.7kW Sunnyboy.... That is less than half the cost of their extended warranty.

Q to an experienced installer: Besides the additional waste heat and probably higher equipment temps it generates, what other effects do you think might shorten the life of an inverter as a result of being operated close(r) to its rated output for longer periods of time ?

Thanx in advance for any info.

Respectfully,

[solarix]

I don't have any specific data to evaluate the advantage of an oversized inverter on lifespan, but the conventional wisdom is that elevated temperature is bad for electronics in general. Definitely true that the fan is going to run a lot more on a fully loaded inverter. These are muffin fans like used in desktop computers and if the fan bearings die, the fan dies, and then the inverter really gets hot and has to derate its output to protect itself.

[J.P.M.]

I don't have any specific data to evaluate the advantage of an oversized inverter on lifespan, but the conventional wisdom is that elevated temperature is bad for electronics in general. Definitely true that the fan is going to run a lot more on a fully loaded inverter. These are muffin fans like used in desktop computers and if the fan bearings die, the fan dies, and then the inverter really gets hot and has to derate its output to protect itself.

Thank you.

I've been singing the keep electronics cool song longer than I've been designing heat transfer equipment, etc. Just thought you might have some insight with respect to PV equipment that was unique.

FWIW, I did some heat transfer work a few years ago on cooling inverters, particularly nat. convection cooled types, and confirmed on my system, that showed adding about 10 CFM of effective forced air flow (not free air fan speed) per kW of inverter capacity will decrease the temp. diff. between a nat. convection cooled inverter's heat sink temp. and the surrounding ambient air temp. by about 1/3, but getting effective air flow to/around the heat sink and keeping it clean is the interesting design rub. It's also probably not something most inverter mfgs. would find to be in their best interests.

Thanx again.

J.P.M.

[josepr]

Thank you.

I've been singing the keep electronics cool song longer than I've been designing heat transfer equipment, etc. Just thought you might have some insight with respect to PV equipment that was unique.

FWIW, I did some heat transfer work a few years ago on cooling inverters, particularly nat. convection cooled types, and confirmed on my system, that showed adding about 10 CFM of effective forced air flow (not free air fan speed) per kW of inverter capacity will decrease the temp. diff. between a nat. convection cooled inverter's heat sink temp. and the surrounding ambient air temp. by about 1/3, but getting effective air flow to/around the heat sink and keeping it clean is the interesting design rub. It's also probably not something most inverter mfgs. would find to be in their best interests.

Thanx again.

J.P.M.

The reason I chose the 6kw vs the 7.7kw is that the 6kw do not has a fan and the 7.7 has a fan. I was thinking the 6kw would be more reliable (no fan failure). But maybe having a fan is a good thing after all.

[J.P.M.]

The reason I chose the 6kw vs the 7.7kw is that the 6kw do not has a fan and the 7.7 has a fan. I was thinking the 6kw would be more reliable (no fan failure). But maybe having a fan is a good thing after all.

My guess is if the unit was designed with forced cooling, it can't hurt. Also, for what was probably a relatively small increase in price the larger inverter's overcapacity may be worth it. Besides, fans aren't that complicated and a known tech, and so the probability of failure is probably smaller than that of some more sophisticated equipment.