Author Topic: Grid Tie, Oh My  (Read 580 times)

1 Member and 2 Guests are viewing this topic.

OperaHouse

  • Hero Member
  • *****
  • Posts: 1066
  • Country: us
Grid Tie, Oh My
« on: October 24, 2017, 07:56:44 AM »
Just a couple weeks before closing camp I bought six Trina 280W grid tie panels. Didn't
really need them, but heard rumors of future price increases and wanted to power more
things at camp without worrying about peak sun. With limited space I also needed to get
rid of older odd shaped 12V that didn't do much in the system and reduced packing density.

At the time of leaving, those panels only powered the dishwasher heating element and added
a little extra watts in the morning to get the fridge going sooner.  I had bought an oil
finned heater to take the bedroom up a couple degrees now that it was fall.  That experiment
was inconclusive as my wife opened the windows every day.

I asked the installer I bought the panels from if he had any equipment that was broken. He
gave me 16 enphase M215 grid tie inverters that had been hit by lightning.  Had hopes that
I could figure a common failure, dig into the potting and fix a few. Maybe come up with some
interesting application for them. Nastiest potting ever, couldn't even salvage the boost
transformer. Could only salvage the MC4 connectors from them.

Grid tie is one of those things I have been thinking about. You see those cheap non certified
units on ebay (and the rebuild kits for them). At least you can rebuild them. Seen some make
it yourself articles for them. One just bumped power at the peak of each cycle with a square
wave and another only bumped power on the positive half of each cycle. The assumption was any
motor would average out the power.

Random trolling of youtube the other day came up with this guy powering his house off line
with three 1,000W Chinese grid tie inverters. A horrible video, but most videos have ten seconds
of actual useful information.  Grid tie inverters have to see some power in order to operate.
He flashed up a scope image of a rather decent sine wave.  Then he flashed the waveform of
the inverter only. It was a MSW square wave!  Obviously a single kick is enough to start it.
Maybe the the three inverters fed off each other once they started each cycle. This is
something worth investigating.

The last week at camp I picked up a LG front load clothes washer. Their direct drive motor
uses only 280 W according to the service manual, easy enough for a small inverter. I was thinking
about feeding the raw panel string DC buss directly into an inverter and avoiding the battery.
Had hopes of running the washer directly off DC, but the washer's VFD doubles the incoming
voltage. Some kind of inverter would have to be used. The heater takes 1,000W and that could be
run directly off DC buss. Grid tie with a 400W MSW offers all kinds of possibilities. The grid
tie raw DC buss could still be powering my water heaters and the washing machine at the same time.
I'm seeing all new possibilities for integration and it still won't be conventional.




Bruce S

  • Global Moderator
  • Super Hero Member Plus
  • *****
  • Posts: 4574
  • Country: us
  • USA
Re: Grid Tie, Oh My
« Reply #1 on: October 24, 2017, 10:38:00 AM »
I haven't been able to push myself into getting one of those grid-tie inverters to play with.

Since these need to "see" grid to start, would it be possible to "start" them with one of the APC or similar UPSs that will start even with AC power?
The 1500XL we have will start without AC all-the while beeping, and the wave looks like a MSW , but maybe, just maybe this is enough to start those grid-tie units?

Bruce S
A kind word often goes unsaid BUT never goes unheard

OperaHouse

  • Hero Member
  • *****
  • Posts: 1066
  • Country: us
Re: Grid Tie, Oh My
« Reply #2 on: October 24, 2017, 12:49:29 PM »
Stay tuned, we will find out. I just bought a 350W one on ebay CHEAP.  If a single unit will sync with a MSW inverter that will be great.  Given a dual conversion, efficiency is reported to be about 70%.  The real interesting thing will be what happens to the waveform when the consumed wattage exceeds what the grid tie can produce and it relies more heavily on the MSW inverter.  Still, it presents some interesting opportunities  to get power from the panels over a distance.  From looking at the videos I'll have to mod these units with bigger capacitors.  Everything needs more capacitors!

george65

  • Sr. Member
  • ****
  • Posts: 409
  • Country: au
Re: Grid Tie, Oh My
« Reply #3 on: October 24, 2017, 06:39:26 PM »

I have read of using grid tie inverters off grid and controllers to facilitate that. not been able to find how to actually do it.
If one could set something up so something like a 2-5Kw inverter as installed on legit solar systems that could be triggered to run off grid, that would be extremely handy!

What do these grid tie inverters actually look for?  Mine appear to look at the AC voltage, frequency  RISO ( whatever that is? ) and what I think is earth leakage.  Being anti islanding they also cut back feed in a millisecond if the mains disappears.

joestue

  • Hero Member
  • *****
  • Posts: 1419
Re: Grid Tie, Oh My
« Reply #4 on: October 24, 2017, 07:06:45 PM »
any good sine wave inverter should be enough to fool them. they look for voltage every half cycle, that's one of the reasons why the power factor is less than 100%, more like 99.5%. note that the power factor and power factor due to harmonics are called the same thing but they are different problems. the grid tie inverter is in fact in phase with the line, but because it doesn't generate any power for a small fraction of the sine wave, you get the harmonic content that drops the "power factor" to about 99%. there is also residual pwm ripple that makes it through.

i would be very supprised if a msw inverter could work, for a couple reasons. one being the msw wave has no voltage at the "start" of each sine wave cycle, secondly most msw inverters can't back feed the battery they run off of.

a msw inverter feeding an LC resonant tank (could be as simple as an induction motor with enough capacitance across the line, or it could be a synchronous motor, such as a petrol generator) with an inductor between the LC filter or the generator, or induction motor, the inductor is needed because msw inverters can't tolerate any direct capacitive load. then you could use a cheap msw inverter to make a sine wave of a fixed frequency, and that should fool the grid tie inverter.

but since you can't backfeed the battery, the grid tie inverters may shut down due to overvoltage if you don't have a sufficient load to draw from them, as they push the line volts up too high.

OperaHouse

  • Hero Member
  • *****
  • Posts: 1066
  • Country: us
Re: Grid Tie, Oh My
« Reply #5 on: October 27, 2017, 08:44:18 AM »
You have to have guts to plug one of these in!  Got my grid tie inverter arrived yesterday. 
It was the typical ebay "Works, just missing a screw." It was in like new condition and rattled.
Never did find out what that was. No obvious shorts or burnt components.  Someone tried to
fix it, but they never got as far as lifting the board.  Almost a virgin. Notice long black
vertical blobs from lower left of the transformer and fan.  This is where a 4,700uf 50V
capacitor should be. Reports are these blow all the time since they are no bigger than your
pinkie. They tried surface soldering some wires from the board to a replacement cap and insulating
them with this goop. That cap was nowhere to be found. Pretty bad job of soldering.

Not shown in the photo, hard to see even with the naked eye, is a #30 wire grounding the copper
shield on the transformer. Wire this small is hard to use in manufacturing. Cost is not likely the
reason. It probably acts like a fuse should insulation break down from a lightning surge on the
power line.

I just bought 400 electrolytic power capacitors.  I'm waiting for those to show up. They are triple
the physical size and 1/4 the capacitance.  Two in parallel will perform much better than what
was in there.  It will take a couple of days anyway to get the courage to plug this into the grid.


OperaHouse

  • Hero Member
  • *****
  • Posts: 1066
  • Country: us
Re: Grid Tie, Oh My
« Reply #6 on: October 27, 2017, 01:27:53 PM »
Well that wasn't fun.  I temped in a capacitor and hooked it up to the grid with my XANTREX HPD30-10.  Seemed to be operating normally. At about 29V and 5.9A the power supply started buzzing and then dies.  Must have sent something nasty back down the line.  You can short this supply forever at 10A and not do any damage.

OperaHouse

  • Hero Member
  • *****
  • Posts: 1066
  • Country: us
Re: Grid Tie, Oh My
« Reply #7 on: November 22, 2017, 03:48:18 PM »
Got my two SUNIVA 280W solar panels today.  Been buying up broken grid tie inverters and sine wave inverters.  Now I can start doing experiments with water heating and feeding power to the house when energy use is sufficient.  Don't want to set off any alarms by feeding energy to the grid.  Don't know how smart my meter is, but it is internet connected. Will be developing a grid tie controller to switch the GTI on and off.  With the number of cheap GTI out there that have gone bad, they don't seem like a good investment for most people.