Anyone paralled Enphase inputs (multiple inverters, single DC feed)?

SolarGeek

I'm curious if anyone has tried connecting more than one Enphase to the same panel array? I'm aware they're designed for one inverter per approximately 230 watts of panel/array and that's the only code approved application.

But if one has a 24V off-grid system that's already wired for a single DC feed, what happens if you try to use multiple Enphase grid-ties running from the same feed (say 4 of them on a 900 watt 24V array)?

I know many are doing just that with the eBay small grid-tie units and they seem to get along with each other and collectively produce the expected output. I'm not sure the no-name Chinese units even implement any sort of real MPPT algorithm (despite their claims) which might help them in this regard.

I would think MPPT might an issue with the Enphase units as each is going to be independently moving the operating point around in search of max power. And if unit A is moving it up while unit B is trying to move it down, the MPPT logic could get confused. But it's possible they'll hunt around until they're all reasonably happy making only small changes (as the real max power point is the same for all of them).

MPPT aside, with the array being essentially a constant current source, the inverters should share the DC feed nicely. Each will take what it needs until the voltage drops too low and then back off a bit. So I'm guessing it would work unless the MPPT "hunting" problem seriously degrades the efficiency, or worse, one or more shut down sensing an error condition.

And, for those of you thinking this is a really bad idea, are there any other high quality and reasonably priced small grid-tie options for a small 24V 900W system (30Vmp) that also needs to work off-grid with a 24V battery bank? I'm not aware of any besides the un-listed no-name Asian units on eBay. And from all I gather those are a *worse* option.

It would be nice if Enphase had some DIY friendly agency listed competition designed for indoor use where the inverters are not baked between hot panels and a hot roof. I realize that requires dealing with all the DC wiring code issues, but count me among the skeptical the Enphase units will really last 15+ years on a hot roof. And if they start failing in large numbers in say 5 years Enphase could easily close their doors making that 15 year warranty worthless. Few companies (and the venture capital firms and/or stock markets that fund them) are in it for the long haul these days. But that topic has been discussed already. My point is they're likely to last much longer if they're running in a cooler environment so that's another advantage of what I'd like to do.

So has anyone tried paralleling the inputs of multiple Enphase units on a single DC feed?

BB.

Re: Anyone paralled Enphase inputs (multiple inverters, single DC feed)?

I don't think it will work well (but I have never tested anything like that--so I have no proof).

Current sources share current (when in parallel) just fine when driving a battery or single MPPT based device.

However, current sources driving multiple non-voltage sink loads--I don't think it will.

The folks that design and build MPPT type solar charge controllers all warn that using two or more MPPT charge controllers on a single array will not work well at all (the MPPT search algorithms will not converge on the optimum Vmp*Imp point with multiple independent controllers on one array)... I would guess that warning would be pretty much the same reasons for multiple GT inverters sharing a single array.

-Bill

SolarGeek

Re: Anyone paralled Enphase inputs (multiple inverters, single DC feed)?

Thanks Bill. I agree it likely comes down to the MPPT. Depending on the Enphase implementation, they may or may not converge properly. It's obviously a fairly expensive experiment to find out, which is why I'm hoping someone here has at least tried it?

My only other options are not very attractive:

- Use a single Enphase and waste 75% of the array output.

- Re-wire everything for a combined AC on-grid system and some sort of DC change over scheme for battery charging off-grid. But you need 4 discrete panels/arrays for the Enphase connection, but want them all in parallel for the off-grid connection. That's not easy to do in anything resembling a code-legal way without manually re-wiring the array each time you switch.

- Use one or more of the eBay GT units which have a reputation for blowing up regularly.

- Wait until some company (hopefully) comes out with an agency listed grid-tie inverter that will work in this application.

As I said, there are plenty of people using the eBay GT units in parallel on a single feed, and they work reasonably well. There are youtube videos showing the proof with the DC current and voltage from the array being monitored and the AC output of multiple GT's shown. The systems typically run around 85% efficiency which isn't awful and certainly beats the ~20% I'll get with a single Enphase.

But, as I said, it's possible the eBay GT units don't really do true MPPT. So that might help them play nice together when connected in parallel to the same array.

BB.

Re: Anyone paralled Enphase inputs (multiple inverters, single DC feed)?

Some of the older/simpler MPPT algorithms are just 80% of Voc... That could share well.

But it is the marginal conditions (sun rise, sun set, broken clouds, etc.) that can mess MPPT controllers up.

You are hitting the doughnut hole of solar--Those with GT systems that cannot run their systems if the utility is down (typically Varray of 200-600 VDC) and the folks with off-grid systems and utility power that would like not to waste power the 99% of the time the grid is running (15-150 VDC).

There are a few ways around the problem--but they all require extra/different hardware (connecting a GT Inverter to an Off-Grid TSW inverter and back driving power--aka AC Coupled--through the OG inverter to charge the battery bank when off-grid--some are untested/unsupported and others are real products like from SMA and possibly the Xantrex XW system) or getting a Hybrid Inverter System (like the Xantrex/Schneider XW system) that can do GT connections from a 24 or 48 volt battery bank.

But small module wise--I think you see what can be done...

-Bill

Windsun

Re: Anyone paralled Enphase inputs (multiple inverters, single DC feed)?

Bad idea - it would be the same as if you tried to connect charge controllers up that way - they will never be perfectly balanced and some will have a much higher load than others.

SolarGeek

Re: Anyone paralled Enphase inputs (multiple inverters, single DC feed)?

BB. wrote: »

You are hitting the doughnut hole of solar... a Hybrid Inverter System (like the Xantrex/Schneider XW system) that can do GT connections from a 24 or 48 volt battery bank.
-Bill

Thanks BB... the smallest Xantrex XW is 4KW and around $3000 so that's not going to pay for itself in my lifetime. Being green is nice, but not when it costs thousands out of pocket you'll never get back.

You are correct, I'm really in the doughnut hole. Honestly, I should consult to one of the solar manufactures to help bring a DIY-friendly agency listed micro-inverter to market and give Enphase some serious competition. I'm kind of shocked that nobody else has done it yet. Their business model is about 80% right but they've totally blown it on that last 20%.

Niche markets (i.e. doughnut holes) can be very profitable when the big guys ignore them. In this case, it wouldn't be hard to also eat into a big chunk of Enphase's more mainstream marketshare if you had a cost competitive agency approved micro inverter that was more DIY friendly and versatile. Someone needs to show both Enphase and the Asian offshore guys they both have it wrong.

BB.

Re: Anyone paralled Enphase inputs (multiple inverters, single DC feed)?

Trying to share power from a common large array between Grid Tied and Off Grid uses (which is what I think you are trying to do) for small to mid-sized systems is difficult.

One of the reasons why we try to get people to characterize their loads and define what it is they want from the system (save money, emergency power, off grid power, "going green", etc.).

It is very difficult (and usually expensive) to reconfigure an existing solar system into something else (larger, GT to Hybrid, etc.).

Even though I love the idea of Hybrid (GT + Off Grid)--The simple fact for my situation is that plain vanilla Grid Tie plus a small / quiet backup genset with a week or so of fuel is still a better solution than any Hybrid system I can think of (basically, the battery bank keeps killing the solution--expenses of battery bank, replacement back costs, extra system losses and costs, non-portability, etc.) all keep me going back to my current configuration (GT + Honda eu2000i + transfer switch).

My power is reliable (enough) that it is not worth the expense--And my "Armageddon" is the "Big One" (large earthquake). Having my "emergency" electrical system tied to my home and a ton of batteries inside that home anchors me here (when there is a real risk that my home could be taken down by an earthquake/authorities can force evacuation even if my home is OK, etc.).

When in practice the generator weighs only 50 lbs, is portable (can take to my in-laws if their power goes out) and the extra stored fuel can either be used for the generator or for the car to bug-out of town to someplace with services.

And I just have to recycle the gasoline to my car once a year and refill (with fuel stabilizer).

-Bill

SolarGeek

Re: Anyone paralled Enphase inputs (multiple inverters, single DC feed)?

BB. wrote: »

(basically, the battery bank keeps killing the solution--expenses of battery bank, replacement back costs, extra system losses and costs, non-portability, etc.
-Bill

I understand but I already *have* the battery bank, dc wiring, disconnect, etc. So, for me, the only piece missing is to get some benefit when the grid power is on.

I agree those starting from scratch should carefully evaluate the long-term costs of various options. But, for me, I already have the entire off-grid system and I'm only looking for cost effective ways to leverage it when the power is on.

dgsloan

Re: Anyone paralled Enphase inputs (multiple inverters, single DC feed)?

After reading through the post I thought that I would add a few things that I have tried. While waiting for spring so I could run a 220v cable underground for my Enphase inverters I bought a few Gorilla Inverters (plugin) as I like to experiment. Of course I wanted to charge my EBike batteries (2kw) directly but once charged use "The Juice" for the house. Interesting part about the inverters that I bought was that half worked at below 5 celsius and and half did not. Since the 1kw array was only putting out 650w at that time of year I decided to parallel the panels to less inverters (GT). This worked so I tied the ones that worked with the high temperature ones together (parallel inverters) with a diode to direct the current to the low temperature inverter for half the parallel power - rather if the temperature rises above 5 degrees then both inverters would process the power. When all inverters were working I pulled the AC plug on half of them to see whether half the inverters would handle the power. The result was that the power production increased by 20 percent. I thought that this was maybe a sweat spot so I tested it over several power ranges and found the same results. My conclusion was that these inverters can be paralleled but at a cost of efficiency due to their MPPT algorithm. I trust the Enphase inverters will be fairly efficient over a wide temperature range but would not trust paralleling them - Best bet would be to split the panels off to each inverter and when your batteries are charged (use diodes to combine to the battery charger) then get a relay to enable the AC which you could trigger in several ways - such as if you are thinking of the dump load route then the dump load would trip the relay.