Safety and Legality of DIY Small 300 Watt Grid Tied Inverter

[russ]

So, here's a seller on Amazon that claims this system is UL certified and that you can get a 30% federal tax credit. Could there be any truth to this?

http://www.amazon.com/Readymade-Inve.../dp/B006R7K8MA

Shyster sight - Really doesn't claim it is UL listed - only drops in the two letters UL which is totally meaningless.

Sites like this must sell garbage to themselves for a while to get a good rating - phony as can be!

The brand name is something no one has ever heard of - probably manufactured in the most backward parts of China.

The warranty would most likely be totally useless. Peakbagger explained about the tax credit.

Not nice people!

[inetdog]

There just isnt a way to get around the electrical code when it comes to connecting to the grid. Now if you want to set up a micro grid and not connect to a utility, you can do anything you want in areas that dont require inspections but be aware that if someone gets hurt, its just another reason why an insurance company doesnt have to pay.

And the seller is, in a way, completely honest about it. Step 2 of the two step process:
"2. Plug the cord into your 110V wall socketPlease follow local regulatory electrical codes for installation. "
Anywhere in the US where the National Electrical Code is enforced, that local code requirement is that you not plug it in. "So sorry that it does not work, but you must not have followed the second step correctly....."

And the weak claim claim for UL status applies only to the panels. "90 Watt Solar Panel,Add upto 6 panels - 540Watts, UL certified." (Usually UL status is "listed", not certified.) Which to maintain whatever certification they might actually have must be installed according to NEC and the manufacturer's instructions.

[russ]

And the seller is, in a way, completely honest about it. Step 2 of the two step process:
"2. Plug the cord into your 110V wall socketPlease follow local regulatory electrical codes for installation. "
Anywhere in the US where the National Electrical Code is enforced, that local code requirement is that you not plug it in. "So sorry that it does not work, but you must not have followed the second step correctly....."

And the weak claim claim for UL status applies only to the panels. "90 Watt Solar Panel,Add upto 6 panels - 540Watts, UL certified." (Usually UL status is "listed", not certified.) Which to maintain whatever certification they might actually have must be installed according to NEC and the manufacturer's instructions.

Unless the wording ''UL listed'' is used it has zero meaning.

The seller is just covering their butt - mix a little truth with the lie and it becomes more believable.

What the seller is suggesting is not legal in most civilized countries - ones with electrical codes. Here I could use it with no problem and I still don't want it.

[Ian S]

There just isnt a way to get around the electrical code when it comes to connecting to the grid.

Here's their "out" in that regard. From the description:

Please follow local regulatory electrical codes for installation.

I would also note that the inverter looks a lot like the typical $20 one you buy for use in your car. Wonder if it comes with an optional cigarette lighter plug. LOL.

[Mike90250]

To add fuel to the fire, generally, these small GT inverters, are not very efficient, so your panel generates 200W of power, but only 100w (50%) gets delivered. Real GT inverters run around 94% efficient.

[Sunking]

To add fuel to the fire, generally, these small GT inverters, are not very efficient, so your panel generates 200W of power, but only 100w (50%) gets delivered. Real GT inverters run around 94% efficient.

What is really sad is people are so gullible and want to believe anything Green Terrorist will tell them. Last time I checked these Plug In inverters cost around $200, throw in a 250 watt panel and you have $500 tied up. People fail to use simple math and common sense. These things generate roughly 5-cents of electricity per day if even that. $500 / $0.05-day = 10,000 days or 27 years to get you money back or 0% ROI. If I invest $500 today, I expect to collect $1300 in 27 years or a 5% ROI.

[RussN9ZP]

What is really sad is people are so gullible and want to believe anything Green Terrorist will tell them. Last time I checked these Plug In inverters cost around $200, throw in a 250 watt panel and you have $500 tied up. People fail to use simple math and common sense. These things generate roughly 5-cents of electricity per day if even that. $500 / $0.05-day = 10,000 days or 27 years to get you money back or 0% ROI. If I invest $500 today, I expect to collect $1300 in 27 years or a 5% ROI.

Thats an unfortunate bi-product of the internetz. There is just as much bad info out here as there is good, Common sense applies at this point.

[inetdog]

Update: New review added to the Ensupra listing on Amazon.

It gets better. Now we have a review which warns that the plug-in installation must be done properly and gives a much more laborious and detailed still illegal way to do it.....

"This review is from: Readymade Solar Power Kit- 500Watt Grid Tie Inverter with 90Watt Solar Panel, Prewired and Configured. Do It Yourself (DIY) Solar; UL ; 5-years Warranty
EDIT: This system is a great system and completely legal. IF WIRED THE CORRECT WAY. This is only to be plugged into an ISOLATED, SEPARATE circuit, without any other loads on it for it to be legal and not harm any systems. Also I would recommend using an isolation transformer like this Tripp Lite IS500 500W Isolation Transformer 4 outlet 6ft Cord to protect your system."

What was omitted is that not only does the separate circuit have to have no other loads on it, it must also not be connected to the grid.
In that case the inverter will never turn on and thus will be much less likely to catch fire. (Although the input-side circuitry could still fail....)

[MPaulHolmes]

Your answer is trivial and does nothing to respond to my observation/question that we can and should do better. I am a degreed engineer from MIT. I've seen the connection from my first legal micro-inverter grid-tied solar panel installation. It's nothing more than a mechanical switch, several breakers and wiring, + to +, - to - and ground to ground. The reality is that there are large utilities and an antiquated building department structure entrenched in this country that are unable or unwilling to see this sort of thing happen. California utilities were required by law to pay for excess electricity generation several years ago. They dragged their feet and took almost 2 years to implement the payback system for the consumer. In the end they pay only a fraction of the value of the power people like myself put into the grid. They are making money off of my excess power, fortunately I sized my system to minimize the excess. So back to the original observation. The wiring issue is simple as I see it given that a three prong plug has all the correct polarity/wiring built in to it's design. As far as the safety issue of power in the system when there is a power loss, it would seem to me that a device that kills power in a similar manner to a GFI would be able to sense when power from the grid goes down and shuts itself off would work. Let's see some well thought out responses.

I made an inverter that will also be able to be grid tie. I haven't used it for grid tie though. The anti-islanding feature is trivial. If your microcontroller is already monitoring grid AC voltage/AC current with the A/D converter, it's just a matter of doing something when the AC voltage feedback deviates from the sine wave for whatever amount of time you want. You could easily disable the inverter in microseconds. And to put that feature in hardware, you just need a comparator and some NANDs (which adds like $1.00 to the cost. haha). For example, so it's not dependent on the microcontroller staying functional, you have the AC signal go through a small transformer, then the output stepped down, and centered around 2.5v control DC voltage. Then, for the hardware protection that's independent of the microcontroller, just rectify the new low voltage AC waveform, add a filter capacitor, and a resistor that is constantly sucking on the output power. If the input of the small AC transformer ever stops being AC, the node where that sucking resistor is will quickly fall. Feed that voltage into a comparator, constantly comparing the voltage to whatever point you want. As soon as the comparator output changes state, it triggers a NAND latch circuit that disabled the pwm output of all the IGBTs. It might sound confusing, but is actually a very simple, inexpensive, reliable circuit.

[inetdog]

it's just a matter of doing something when the AC voltage feedback deviates from the sine wave for whatever amount of time you want.

All well and good, except both theory and experiment have shown that certain combinations of reactive and resonant loads can allow a GTI with just a simple waveform comparator to continue powering the load when the grid is disconnected. Not nearly as bad as driving voltage back into the grid wiring, but still not allowed by code.

[MPaulHolmes]

All well and good, except both theory and experiment have shown that certain combinations of reactive and resonant loads can allow a GTI with just a simple waveform comparator to continue powering the load when the grid is disconnected. Not nearly as bad as driving voltage back into the grid wiring, but still not allowed by code.

Not a waveform comparator but an lm293 comparator. If the grid is down the wont be a sine wave on the output of the step down transformer that feeds the a/d converter.

[russ]

Not a waveform comparator but an lm293 comparator. If the grid is down the wont be a sine wave on the output of the step down transformer that feeds the a/d converter.

No big deal - whatever you cook up won't be legal to connect in to the grid with.

[inetdog]

Not a waveform comparator but an lm293 comparator. If the grid is down the wont be a sine wave on the output of the step down transformer that feeds the a/d converter.

The problem with your approach is that it assumes that the reference waveform will go away when the grid goes away.
Both theory and experiment show that with just the wrong loading the simple GTI will continue to drive and provide its own reference waveform.

It is more complex than this, but imagine a parallel resonant load composed of inductive and capacitive components such that the resonant frequency is close to 60 Hz.
That "tank" circuit will continue to oscillate at 60Hz while the GTI pumps power into it to sustain the oscillation.

In practice you also have to do something like introduce deliberate distortion or pulses into the GTI output. Then make sure that the current distortion is swallowed up by the near zero impedance of the grid, and no voltage distortion results. Whatever you do also has to react within a specified number of cycles.

[MPaulHolmes]

The problem with your approach is that it assumes that the reference waveform will go away when the grid goes away.
Both theory and experiment show that with just the wrong loading the simple GTI will continue to drive and provide its own reference waveform.

It is more complex than this, but imagine a parallel resonant load composed of inductive and capacitive components such that the resonant frequency is close to 60 Hz.
That "tank" circuit will continue to oscillate at 60Hz while the GTI pumps power into it to sustain the oscillation.

In practice you also have to do something like introduce deliberate distortion or pulses into the GTI output. Then make sure that the current distortion is swallowed up by the near zero impedance of the grid, and no voltage distortion results. Whatever you do also has to react within a specified number of cycles.

If the grid is down, theres no ac wave coming in. That woild resilt in drastically diffwrwnt current feedback for various positions on the wave. If the grid is down from a bbecause tje line is now open, wouldnt the current feedback be different from a functional grid? If there was a short causing a down grid, that would also have different feedback. I have ac current waveform, dc current wavwform, ac voltage and dc voltage moment by moment. How on earth could something as dramatic as no grid be undetectable given that feedback?

[Sunking]

If the grid is down, theres no ac wave coming in. That woild resilt in drastically diffwrwnt current feedback for various positions on the wave. If the grid is down from a bbecause tje line is now open, wouldnt the current feedback be different from a functional grid? If there was a short causing a down grid, that would also have different feedback. I have ac current waveform, dc current wavwform, ac voltage and dc voltage moment by moment. How on earth could something as dramatic as no grid be undetectable given that feedback?

If you loose power, it is because the POCO has opened your line somewhere up stream. All your neighbors and anyone else connected to your section will be in parallel with you trying to feedback and supply them power. POCO gets a real chapped ass when you try to do that. So much so they will file criminal charges against you.