Author Topic: Using the Grid for a Reference Signal to GTI  (Read 6154 times)

0 Members and 1 Guest are viewing this topic.

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
Using the Grid for a Reference Signal to GTI
« on: April 24, 2014, 04:28:56 PM »
I'd like to wire a battery-less grid-tie inverter (GTI) into my barn, but in a way that it does not put any energy back into the grid, and only powers my local loads and dump loads, and only when the sun is shining (no energy storage).
I have grid connection to the barn, but I do not want to use it to power my loads.
I do not want to do AC coupling, because of the cost of another battery-based inverter, and of course the battery bank, I don't need there.

So, I need a device that would hook up between the grid and this GTI, and would only provide a voltage/frequency reference waveform for the GTI to monitor and operate from. This device would couple the GTI to the grid in a way that only one-directional flow of energy is possible, from the grid to the GTI.

Looking from the grid side, this device would look like a normal load, and it would have a fairly large impedance so that it is only a small load to the grid, say 50 W or less.

Looking from the GTI side, this device would need to have very large impedance, and some other "check valve", to disallow the GTI to run any energy into it and into the grid (all inverter current needs to go to my local/dump loads connected between this device and the GTI). At 60 Hz this probably means a very large resistor or inductor, or a very small capacitor, or some combination of these.

Without regard to this set up violating or not any grid-tie rules/code:
Does anyone know of a simple device that may be used to generate a reference signal for the GTI using the grid, preferably some off-the-shelf unit?
How about any other simple ways to accomplish what I am trying to do, using the grid since it already contains a clean reference waveform and no additional waveform generator is needed?

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: Using the Grid for a Reference Signal to GTI
« Reply #1 on: April 24, 2014, 04:58:29 PM »
so you have excess energy, you want to dump it into AC loads whenever that energy is available. but you also want it to be clean power, synced to the grid.
honestly i don't see a need to do this.
a standalone inverter should work just as well.

the easiest way i can think of is find an APC pure sine wave UPS.
feed about 90 volts ac from the grid (use a 10w isolation transformer) into the UPS.
the voltage will be too low to boost, so it will run off batteries (this is where you will need to supply dc at the voltages it can handle. however, the output frequency will remain synchronized to the grid from the isolation transformer. when the grid voltage is in the range of 120vac, it will transfer back over to grid power, or buck, or boost it.
My wife says I'm not just a different colored rubik's cube, i am a rubik's knot in a cage.

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
Re: Using the Grid for a Reference Signal to GTI
« Reply #2 on: April 24, 2014, 05:36:08 PM »
thanks Joe,

I got a nice GTI for real cheap, and it's in the right power range (I do not have a standalone inverter in this power range, and my loads in the barn do not need storage/batteries).
Also, I DO NOT need power synced to the grid, I just need a UL1741-compatible volt/freq reference signal (at any phase in relation to the grid) to enable this GTI and keep it going. I want to generate this reference signal from the grid.

This UPS idea sounds promising, even though it needs a small battery inside the UPS, but I am not sure I understand how this would take me from 90V to 120V, etc.

Anyway, given that I do not need to be synced to the grid, can you explain this idea in simpler terms if it still applies, or change it for no grid-syncing needed?

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
Re: Using the Grid for a Reference Signal to GTI
« Reply #3 on: April 24, 2014, 05:42:58 PM »
sorry, I missed the frequency-wise syncing in your post, I thought you meant phase-wise syncing.

Anyway, can you still explain this UPS idea in simpler terms?

mab

  • Sr. Member
  • ****
  • Posts: 428
  • Country: wales
Re: Using the Grid for a Reference Signal to GTI
« Reply #4 on: April 24, 2014, 05:48:00 PM »
the closest I can come to what you want would be to use a device like an Immersun (this is a UK spec device but I assume there might be something equivalent where you are):-

http://www.immersun.co.uk/

then with your GTI connected to the grid and feeding in a.c, the Immersun monitors the power flow at your meter and diverts any surplus to your dumpload (immersion, or other resistive load) so you don't export more than a few watts.

problem is, the cost of such a device is probably a significant portion of what you'd pay for a stand-alone inverter anyway.

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
Re: Using the Grid for a Reference Signal to GTI
« Reply #5 on: April 24, 2014, 06:11:38 PM »
thanks mab,

I looked at immersun, and it is expensive. Besides, I do not want to export ANYTHING into the grid, this would violate the grid-tie rules. I just want to use a token amount of grid power to make my reference, only when the grid is up.

BTW, my main load in this case is a water well pump which can run at any time to slowly fill the tank (and it will run in a battery-less system without problems, we've tried it), and some dump loads (an AC motor and heaters) that can be switched in/out manually if I know I have excess energy, and if I need them.

Mary B

  • Administrator
  • SuperHero Member
  • *****
  • Posts: 3169
Re: Using the Grid for a Reference Signal to GTI
« Reply #6 on: April 24, 2014, 07:18:14 PM »
Use a stand alone inverter and  a transfer switch, if the transfer switch has power it goes to inverter, lose inverter it drops back to grid. What is in the barn that is frequency sensitive?

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: Using the Grid for a Reference Signal to GTI
« Reply #7 on: April 24, 2014, 11:52:27 PM »
Mary he's trying to use the GTI he already has as a stand alone inverter, i missed this bit from the first post.

one way to do it might be to wire up a single or three phase VFD to deliver 120volts at 60 hz. (you'll need an LC filter)
this will form a virtual grid.

a vfd is going to waste 30 or so watts I would suspect, depending on a number of things. the LC filter on the output, the size of the IGBT brick inside it as well, and what generation it is.

my mind isn't working right now, i'm not sure if you need a vfd with a brake and an external resistor (a hot water heater element is fine)
My wife says I'm not just a different colored rubik's cube, i am a rubik's knot in a cage.

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
Re: Using the Grid for a Reference Signal to GTI
« Reply #8 on: April 25, 2014, 03:07:03 PM »
Yes, Joe's got it, I am trying to use a GTI entirely off-grid as far as the energy flow, except to use the grid to generate a reference signal the GTI needs to operate.
Mary, there is nothing in the barn that's frequency-sensitive, just the GTI that needs a stable line frequency (and voltage) reference to be able to function.
There are several reasons for me to want to do so where I am at, I don't want to bore anyone with details, but hopefully everyone can benefit from a good solution to this generic problem.

I like Joe's solution to create a VIRTUAL GRID using a VFD (or some other device). I don't know much about these VFD rascals, so I'll need to get up to speed.
But in general, a virtual grid is what I need, something that the real grid will see as a small load (just like any other load you have in your house), and that my devices (like the GTI) will see as low-power voltage source, and a stable voltage/frequency reference signal to operate from.
And, this being a VFD, I may also get a chance to control my GTI with it, by deviating the frequency from nominal to exceed the UL1741 limits (as AC-coupled systems generally do).

Also, this way I am not setting myself up for an accusation by the power company that I have endangered or contaminated the real grid in any way (I am only pulling small amount of energy out of the grid just like any load would, and putting nothing into the grid).

Thanks Joe.

Mary B

  • Administrator
  • SuperHero Member
  • *****
  • Posts: 3169
Re: Using the Grid for a Reference Signal to GTI
« Reply #9 on: April 25, 2014, 03:43:05 PM »
I don't know how you can isolate the VFD enough to protect it from reverse current flow...

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
Re: Using the Grid for a Reference Signal to GTI
« Reply #10 on: April 25, 2014, 05:17:20 PM »
YES, I was thinking about it, same thing happens in a battery-based inverter that is AC-coupled to a GTI.

But, first I need to learn enough about VFDs to be able to understand what options I may have.

SparWeb

  • Global Moderator
  • Super Hero Member Plus
  • *****
  • Posts: 5452
  • Country: ca
    • Wind Turbine Project Field Notes
Re: Using the Grid for a Reference Signal to GTI
« Reply #11 on: April 27, 2014, 12:31:33 PM »
This guy has a way with inverters:   http://www.anotherpower.com/board/index.php/topic,902.0.html
He visits this site occasionally but perhaps not recently, as he likely would have mentioned something with respect to your question by now.
That particular thread may or may not suit your goal, but Mr. Oz has come up with many similarly interesting mods to electronics before.
No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
System spec: 135w BP multicrystalline panels, Xantrex C40, DIY 10ft (3m) diameter wind turbine, Tri-Star TS60, 800AH x 24V AGM Battery, Xantrex SW4024
www.sparweb.ca

Mary B

  • Administrator
  • SuperHero Member
  • *****
  • Posts: 3169
Re: Using the Grid for a Reference Signal to GTI
« Reply #12 on: April 27, 2014, 03:54:02 PM »
Might be easiest to hack the inverter and find the grid reference line and apply 60hz there via a stable oscillator.

boB

  • Sr. Member
  • ****
  • Posts: 389
  • Country: us
    • boB
Re: Using the Grid for a Reference Signal to GTI
« Reply #13 on: April 27, 2014, 06:23:25 PM »
Might be easiest to hack the inverter and find the grid reference line and apply 60hz there via a stable oscillator.


Yes, and while you're at it, make it voltage source instead of current source... (if it isn't already VS.)

Mary B

  • Administrator
  • SuperHero Member
  • *****
  • Posts: 3169
Re: Using the Grid for a Reference Signal to GTI
« Reply #14 on: April 28, 2014, 02:32:27 AM »
I would think the line to the control electronics is voltage and not current sourced... TTL electronics probably.

DamonHD

  • Administrator
  • Super Hero Member Plus
  • *****
  • Posts: 4125
  • Country: gb
    • Earth Notes
Re: Using the Grid for a Reference Signal to GTI
« Reply #15 on: April 28, 2014, 06:59:51 AM »
But surely that sense line would be analogue of some sort, eg zero-cross detection, rather than a rather crappy asymmetric TTL threshold?

Rgds

Damon
Podcast: https://www.earth.org.uk/SECTION_podcast.html

@DamonHD@mastodon.social

DanG

  • Hero Member
  • *****
  • Posts: 1122
  • Country: us
  • 35 miles east of Lake Okeechobee
Re: Using the Grid for a Reference Signal to GTI
« Reply #16 on: April 28, 2014, 11:04:08 AM »
Home Power magazine had a featured system four or so years ago that was primarily 120V - and fed a 240V circuit through a transformer  - the transformer outputs went to a grid-active 240V inverter that picked when it sensed loads downstream from well pump etc.. So... the 120V inverter(s) were supplying a fraction of the 240V load (or vice-versus). If you'd plow through back issues you'll find the schematic, I'm forgetting whether the 240V unit was on constant standby or initialized only on demand. Something like this could work out for basic grid-spoofing, have the transformer inline between signal inverter and power unit that would help ballast they system.

Mary B

  • Administrator
  • SuperHero Member
  • *****
  • Posts: 3169
Re: Using the Grid for a Reference Signal to GTI
« Reply #17 on: April 28, 2014, 03:42:34 PM »
I would have to find a schematic Damon... one I did see used crappy old TTL to detect zero crossing and output a reference waveform for the driver board and also a grid good signal. Was pretty old though.

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
Re: Using the Grid for a Reference Signal to GTI
« Reply #18 on: April 28, 2014, 04:07:58 PM »
thank you all,

I've seen the posts from Oztules, he is years ahead of me as far as being able to modify an inverter. Plus, I am not sure I need an inverter to do this job.

The VSD/VFD idea Joe suggested sounds good and generic. But I am still concerned that the GTI's power would be back-fed through it and into the grid, unless some form of dynamic braking or regeneration was used. I am still trying to understand this. These VSDs are expensive, but they would be a good generic solution to emulate the grid and enable the GTIs, so this may be $$$ well spent.

About the grid reference line, I was under the impression that a GTI samples the grid per UL1741 via the same connection as the one it uses to deliver power to the grid. So, there is no separate grid reference line, just a software or hardware "switch/diplexer" inside a GTI that either pushes the power out, or listens for the condition of the grid.
If it's true then it would be hard to fool a GTI via a stable 60 Hz reference (and it also needs a voltage reference), because you'd have to deal with the GTI's firmware and its timing.
But, I've never disected a GTI so I don't know if this would be possible and how. Besides I'd like to stay away from modifying the GTI itself, and focus on spoofing it instead.

Anyway, we've got a few options on the table, and I'll slowly work through them.
In the meantime let's have some more if possible, because this relates to a very generic issue of using a GTI in an off-grid system, regardless of the source of the reference signal it needs to operate (a VSD tied to the grid, an independent circuit that simulates a low-power version of the grid, another AC-coupled inverter, etc).

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: Using the Grid for a Reference Signal to GTI
« Reply #19 on: April 28, 2014, 04:32:47 PM »
Cheap VFDs can't back feed the grid, they are fed with a diode block.

The problem with a vfd is lack of an lc filter. you need to make one. i suggest two microwave oven primary coils as an air core inductor, and a 10 or 20 uF motor run capacitor. one inductor on each phase, not both in series, and put them side by side to increase the inductance, oriented properly of course.

The second problem with a VFD is the output voltage regulation. it needs to be very "loose". you'll get a little bit of flop with the 2 volts of conduction loss in the igbts, but beyond +/-5 volts, you are going to be sending power back into or pulling it out of the dc bus. which is fine, because its probably harmonic current flowing round in circles.

If you want to experiment with this, I highly recommend using a variable transformer to couple the vfd to the GTI, you should be able to push power one way or the other by changing the voltage ratio. watch the dc bus voltage with a volt meter, it will rise when power starts flowing back into the VFD. which will consume about 30 watts perhaps, so you'll have some space just before the brake transistor turns on, but not much.

you should be able to feed your loads with the VFD (from the mains/grid), or the GTI, or some measure of both by varying the auto transformer.

the VFD will need a brake resistor, and provided it is sufficiently stable, the GTI should be able to dump its full load into the VFD, which is going to go back into the dc bus, which is going to get dumped into the brake resistor via pwm, and it should be able to do this before the voltage rises to the point the GTI trips off due to mains overvoltage.
My wife says I'm not just a different colored rubik's cube, i am a rubik's knot in a cage.

Mary B

  • Administrator
  • SuperHero Member
  • *****
  • Posts: 3169
Re: Using the Grid for a Reference Signal to GTI
« Reply #20 on: April 29, 2014, 02:36:55 AM »
Depending on the inverter and how smart it is I am guessing a small micro controller of some sort does the housekeeping functions. So it would read good grid and tell the inverter to run, it would also take care of matching the GTI to grid frequency and making sure it stays there.

thank you all,

I've seen the posts from Oztules, he is years ahead of me as far as being able to modify an inverter. Plus, I am not sure I need an inverter to do this job.

The VSD/VFD idea Joe suggested sounds good and generic. But I am still concerned that the GTI's power would be back-fed through it and into the grid, unless some form of dynamic braking or regeneration was used. I am still trying to understand this. These VSDs are expensive, but they would be a good generic solution to emulate the grid and enable the GTIs, so this may be $$$ well spent.

About the grid reference line, I was under the impression that a GTI samples the grid per UL1741 via the same connection as the one it uses to deliver power to the grid. So, there is no separate grid reference line, just a software or hardware "switch/diplexer" inside a GTI that either pushes the power out, or listens for the condition of the grid.
If it's true then it would be hard to fool a GTI via a stable 60 Hz reference (and it also needs a voltage reference), because you'd have to deal with the GTI's firmware and its timing.
But, I've never disected a GTI so I don't know if this would be possible and how. Besides I'd like to stay away from modifying the GTI itself, and focus on spoofing it instead.

Anyway, we've got a few options on the table, and I'll slowly work through them.
In the meantime let's have some more if possible, because this relates to a very generic issue of using a GTI in an off-grid system, regardless of the source of the reference signal it needs to operate (a VSD tied to the grid, an independent circuit that simulates a low-power version of the grid, another AC-coupled inverter, etc).