Author Topic: BONDING TWO INVERTERS AT NEUTRAL  (Read 17503 times)

0 Members and 1 Guest are viewing this topic.

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
BONDING TWO INVERTERS AT NEUTRAL
« on: June 18, 2016, 06:25:54 PM »
This came up in the context of converting a house from grid-powered to inverter-powered.

I want to connect two UNSYNCHRONIZED single-phase 120Vac inverters to power this house, without making any changes to the house's standard AC distribution panel (like adding a second and separate Neutral bus bar into the panel).
These are MSW inverters with clean rectangular voltage waveform at no-load (duty cycle = 0.75), same argument should be valid for PSW inverters.

All branch circuits in the house have their own Neutral return wires (no branch circuits with common-Neutral).

I don't need to generate a 240Vac split-/single-phase supply in the house, just the 120Vac single-phase supply for all branch circuits.

The Neutral bus bar in the panel may or may not be bonded to Earth ground (I have it floating now, later may bond it to ground).

Each inverter has 2-wire AC output: Hot (black) and Neutral (white).
I assume standard H-bridge AC output, without any output filters (because of MSW).
I also assume these inverters cannot tolerate power back-feeding through them, and would be damaged if that happened.
For now I am not worried about the Earth/chassis ground (green) connections at inverters and at the panel.

I've verified that each inverter can operate with their Neutral wire bonded to ground (by connecting the scope's ground clip to the inverter's Neutral). However, the factory tells me that these inverters are designed to operate with floating Neutral.

The planned connections are:
Connect the Neutrals of both inverters together at the panel's Neutral bus bar.
Connect the Neutrals of all branch circuits (L1 and L2) in the house to the same Neutral bus bar in the panel.
Connect the Hot of inverter #1 to the circuit breaker feeding the L1 side of the panel.
Connect the Hot of inverter #2 to the circuit breaker feeding the L2 side of the panel.

So, the only connection made between these two inverters is between their Neutrals. The Hots of both inverters are NOT connected.
The inverters may be turned on at random times, this will result in any possible phase difference between them (0 to 360 deg).

The 120 Vac L1 loads would be operated off inverter #1, and the 120Vac L2 loads off inverter #2.
No 180 deg phase difference between L1 and L2 at the panel, but the independent L1 and L2 loads don't care.

Anyone see any problems with connecting the inverters' Neutrals together this way at the distribution panel, with the Neutral bus bar floating?
What if these connected inverter Neutrals, and the panel's Neutral bus bar, were to be bonded to ground?
Or, maybe the inverters' Neutrals HAVE TO be held at the same absolute potential (bonded to ground at a single point through the Neutral bus bar) for this system to function without blowing the inverter(s)?

DamonHD

  • Administrator
  • Super Hero Member Plus
  • *****
  • Posts: 4125
  • Country: gb
    • Earth Notes
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #1 on: June 19, 2016, 05:18:31 AM »
My instinct is that that would be a disaster: if those neutrals are designed to float then they are not intended to be connected to anything else.

Even if they floated in the same direction relative to the same battery terminal (say 55V) my guess would be that you'd get destructive circulating currents.

Someone should be along soon who actually knows what they are doing.

Rgds

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

@DamonHD@mastodon.social

thirteen

  • Hero Member
  • *****
  • Posts: 980
  • Country: us
  • Single going totally off grid 1,1, 2013
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #2 on: June 19, 2016, 12:12:24 PM »
To me it would be worth a phone call to the manufacturer. I would also draw a wiring diagram of which inverter does what and label certain outlets at the point of use.  Every electrical box I have installed has a wiring diagram of which wire goes where. Saves problems when remodeling or changing things in the future. 13
MntMnROY 13

mab

  • Sr. Member
  • ****
  • Posts: 428
  • Country: wales
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #3 on: June 19, 2016, 02:58:42 PM »
if the inverters operate (individually) with a grounded neutral, then they shouldn't care that there's another inverter neutral connected. If you're using an inverter or generator as a mobil power source to a single appliance it's recommended to leave the output floating; but if you're supplying multiple appliances in a house then the N-G bond is required (at lease here in the UK), so I'd go with the bond.

my worry would be compliance with regs in regard to adding N-G bonds in the house in parallel with the grid supply (if you switch back to grid).

dnix71

  • SuperHero Member
  • ******
  • Posts: 2513
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #4 on: June 19, 2016, 04:08:03 PM »
Won't work unless the inverters have a single time source.
I don't recommend this one, but it admits the limits of neutral sharing.
http://www.ebay.com/itm/6000W-LF-Split-Phase-Pure-Sine-Wave-Power-Inverter-DC12V-AC220V-110V-LCD-Charge-/321873624566
Note the fine print says you can have L1 plus N and/or L2 plus N, but you cannot share N as L1 N L2.

That would mean you would need separate neutrals busses for each side of the panel and no shared neutrals past the panel. That would also mean you would need separate grounds for each side as well in case of a short.

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #5 on: June 19, 2016, 04:37:33 PM »
Won't work unless the inverters have a single time source.
I don't recommend this one, but it admits the limits of neutral sharing.
http://www.ebay.com/itm/6000W-LF-Split-Phase-Pure-Sine-Wave-Power-Inverter-DC12V-AC220V-110V-LCD-Charge-/321873624566
Note the fine print says you can have L1 plus N and/or L2 plus N, but you cannot share N as L1 N L2.

That would mean you would need separate neutrals busses for each side of the panel and no shared neutrals past the panel. That would also mean you would need separate grounds for each side as well in case of a short.


i have no idea what they tried to say in that ebay posting.

but as Mab says, if the inverter doesn't care about the neutral bonded to ground, or floating, then you can connect two inverters together at the neutral and use two of them to feed two separate loads.

but you can't have any 220v loads due to no synchronization, and cannot connect them in parallel.

by definition if the inverter doesn't care about the output floating then you can swap the "hot" and the "neutral" if you wish.


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: BONDING TWO INVERTERS AT NEUTRAL
« Reply #6 on: June 19, 2016, 05:31:33 PM »
Separate grounds not only violate NEC code but they are a very bad lightning hazard as induced current causes one system to rise way higher in voltage than the other...

Won't work unless the inverters have a single time source.
I don't recommend this one, but it admits the limits of neutral sharing.
http://www.ebay.com/itm/6000W-LF-Split-Phase-Pure-Sine-Wave-Power-Inverter-DC12V-AC220V-110V-LCD-Charge-/321873624566
Note the fine print says you can have L1 plus N and/or L2 plus N, but you cannot share N as L1 N L2.

That would mean you would need separate neutrals busses for each side of the panel and no shared neutrals past the panel. That would also mean you would need separate grounds for each side as well in case of a short.

dnix71

  • SuperHero Member
  • ******
  • Posts: 2513
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #7 on: June 19, 2016, 05:41:13 PM »
I don't recommend that inverter or having separate grounds and neutrals if that violates code, but that is the only way to make a couple of cheap inverters work side by side in the same building - keep them separated.

To meet code you would probably need two panels, each fed by a single inverter and with completely separate wiring. Not practical or what you want to do in this case.

Almost all the msw inverters are floating neutral, which is a misnomer. It isn't neutral, there really is no neutral, both wires are hot and required to make a circuit, that's why you can't connect them together safely.

Floating neutral inverters are safer for portable equipment but never intended to supply house wiring.
« Last Edit: June 19, 2016, 05:45:21 PM by dnix71 »

OperaHouse

  • Hero Member
  • *****
  • Posts: 1308
  • Country: us
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #8 on: June 19, 2016, 05:43:12 PM »
Many MSW inverters are dual H bridge and the negative 12V supply shares the same common.  If you ground one of the outputs the battery becomes live.  If the battery is also grounded it will short out that output.  Two inverters connected to the same neutral, but not grounded will short out through the common battery negative.  Factory saying not to ground the neutral is a good indication that this is the case with your inverters.

You can connect a small lamp from N to negitive battery.  It will likely light up half brilliance indicating both outputs are live in reference to ground.  Two isolation transformers would work but that is an expensive route.

Johann

  • Newbie
  • *
  • Posts: 46
  • Country: us
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #9 on: June 19, 2016, 11:59:58 PM »
Many cheap MSW inverters have 2 hot wires and no actual neutral. Both hot wires have 60 volts each and the ground on many of them is not an actual ground either.
Some of the cheap MSW inverters may have a wired ground and if you use a neutral ( one hot wire) and bound it with the neutral of the house and you use the ground system, then you would make a short between the neutral and ground.
If you do not use the ground system, then there is still a possibility that your battery bank becomes hot with 60 volt AC if you got your battery bank hooked up to ground rod, same would be if your panels are grounded to the ground rod.

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #10 on: June 20, 2016, 03:07:43 PM »
Thank you all for the insightful comments.

On its face it looks like a disaster, but when you think about it, it may be possible when both Neutrals are bonded to ground. But this makes it sensitive to any accidental un-bonding errors.

Look at the schematic below, this is what I am trying to do in simple form.
I do not see any closed circuits whereby one inverter may be affecting the other on the AC side (but I am not showing the DC side, and both inverters work off the same battery bank - no problem with this as I've been running like this for years powering two separate AC circuits off these two inverters).
* 2 inv.pdf (27.11 kB - downloaded 308 times.)
The instantaneous voltages:
V1 =/ V2 in general
Vn1 =/ Vn2 in general (unless N bonded to ground, then Vn1 = Vn2 = 0 with respect to ground)

These are decent quality MSW inverters, with the V-waveform selected to match the PSW parameters, so I'd like to theorize as much as possible before actually turning the switch on for this whole thing.

I've talked to the factory a while ago, and they were going to do a hardware test using their own inverters, but I haven't heard back from them.

This house will be powered by inverters only, no going back and forth between inverters and the grid. For the generator I have a different setup.

I agree there is really no true neutral in this inverter (just the two H-bridge AC output terminals), but I refer to it as Neutral because the wire is white (the other wire is black).

I still need to think about your points made about shorting through the battery or the battery becoming hot, so initially my plan is to proceed conservatively and install two separate Neutral bus bars. Only after the measurements show no potential differences between the two bars, and no other problems at the battery, making all connections at one common Neutral bar.

Even if the testing shows that one Neutral bar will not work, these measurements should give interesting insight how this circuit works, given these specific inverters.

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #11 on: June 20, 2016, 03:11:50 PM »
OK, the "2 inv.pdf" schematic image does not show up as picture in the post, but it can be downloaded to look at.

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #12 on: June 20, 2016, 03:34:41 PM »
Thinking about the problems at the battery bank some more, these are HF inverters with the isolation transformer in their DC-DC converter.
This may prevent any ground faults, current loops, or voltage potential biases on the AC side from feeding back into the battery bank.

Johann

  • Newbie
  • *
  • Posts: 46
  • Country: us
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #13 on: June 20, 2016, 09:24:14 PM »
NOTICE: I changed the link so it will play the right video this time. last link was the wrong video.

Jack,
here is a youtube video you may want to watch.

Sorry folks, I deleted the original/last youtube link I posted yesterday, it was the wrong link/video.

Let's see if I can re-post the right link this time.
Here is the right youtube video.
https://www.youtube.com/watch?v=FjuzyAujdC4


« Last Edit: June 21, 2016, 04:14:06 PM by Johann »

OperaHouse

  • Hero Member
  • *****
  • Posts: 1308
  • Country: us
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #14 on: June 21, 2016, 10:50:46 AM »
What you say sounds very logical, but it is not true.  I have about a dozen HF inverters.  I buy them broken and repair them.  They all share a common battery negative in the H bridge common.  Did you try my  lamp test to negative?

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #15 on: June 21, 2016, 08:34:27 PM »
Thanks Johann, these are great videos, and the fellow seems really knowledgeable.

Seems like everything here hinges on the internal design of the inverter, not as much on the outside circuit, and that's tough to determine in general.

I took some measurements around my inverters, got some interesting results:

First, there is no ohm-meter continuity between any of the 3 AC output pins (Hot, N, G) and any of the 2 battery terminals (+ -), probably because of the isolation transformer in-between these two.
Also, the AC Neutral is not bonded to Ground internally.
The AC Ground is connected to the inverter chassis ground.
I believe the fellow in the video was firm on inverter destruction if there is AC Ground to battery (-) continuity (and when the inverter's Neutral is grounded). This implies no isolation transformer in the inverter.
So, it looks like the back half of the inverter (after the isolation transformer) is isolated from the DC inputs and floating, and this should allow for bonding of either the Hot or the Neutral to Ground ??
Maybe that's why my inverters can operate with their Neutral grounded.

Second, I measured with my cheap voltmeter AC Neutral to battery (-) voltages on a running inverter (like in the lamp test).
On both the 1000 Vdc and 750 Vac ranges I got some large numbers (about 1800 V) that I can't believe, or over-range (I do not have a true RMS voltmeter, and this is a MSW inverter).
I did not connect a lamp between these two for the fear of blowing something in the inverter, given the results above.

Any comments on this?

9856-0
V L1 =/ V L2 in general

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #16 on: June 21, 2016, 09:29:33 PM »
it works, if the inverter N output can be grounded to battery negative or positive.

as others have said if there is ~55vac between N and battery negative (or positive) then it won't work.


connecting two inverters together at the neutral will be a smoke test if it doesn't work... try a lightbulb first.

digital meters are useless in sufficiently high noise environments, try finding an analog meter. though its discouraging your inverters produce that much noise.
« Last Edit: June 21, 2016, 09:38:59 PM by joestue »
My wife says I'm not just a different colored rubik's cube, i am a rubik's knot in a cage.

OperaHouse

  • Hero Member
  • *****
  • Posts: 1308
  • Country: us
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #17 on: June 22, 2016, 10:48:44 AM »
You seem stuck on this idea.  So, good luck.    I remember one of the inverters had a 100K resistor connected from the ground pin to the frame.  It is not an isolation transformer, it is a boost transformer.  They have a common shared with the negative battery so they can power the micro or PWM chip.  They do this to save money.  More expensive sine wave inverters do float the H bridge and allow the output to be connected to ground.  Certainly nothing Harbor Freight sells can do that.
« Last Edit: June 22, 2016, 12:42:38 PM by OperaHouse »

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #18 on: June 22, 2016, 08:08:51 PM »
OK, with this OH's explanation I begin to see what's going on.

I did the lamp test, connected a 25W 120Vac lamp between the battery (-) and the AC Neutral, the filament did not light up at all (the lamp burns bright when plugged directly into this MSW inverter).
Same for the battery (+) to AC Neutral lamp connection, the filament is cold.
So, what does this mean in the context of connecting the Neutrals and bonding them to ground?

BTW, these inverters are PowerBright ML2300-24, have you worked with these to see how they are built?

Also, OH, to elaborate on your last post.
I agree it's a step-up transformer to the DC-link voltage, but I thought it also provided galvanic isolation.
Are you saying the common shared connection you describe in the cheap inverters somehow makes it a boost auto-transformer? Maybe post a schematic?

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #19 on: June 22, 2016, 09:33:18 PM »
OK, with this OH's explanation I begin to see what's going on.

I did the lamp test, connected a 25W 120Vac lamp between the battery (-) and the AC Neutral, the filament did not light up at all (the lamp burns bright when plugged directly into this MSW inverter).
Same for the battery (+) to AC Neutral lamp connection, the filament is cold.
So, what does this mean in the context of connecting the Neutrals and bonding them to ground?

That is excellent, you should have no problems connecting them as you plan to. you do not have to ground the battery negative or positive post, but you might as well.
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: BONDING TWO INVERTERS AT NEUTRAL
« Reply #20 on: June 25, 2016, 01:51:40 PM »
I guess we may not hear from OperaHouse about his comments.

Like he was saying before, this inverter seems to have two separate and symmetric channels (on the R side of the pic below).
The battery inputs feed each channel in parallel (the fat wires).
Looks like there may be a 2-transformer boost stage in each channel ??
Each channel has 10 switches (maybe 2 H-bridges + a chopper ??)


I still plan to proceed conservatively, first wire a temporary separate Neutral bus bar and take some measurements before connecting these inverters' Neutrals. Will see what happens, it may take some time but I'll post the result when/if done.

BTW, even though these PowerBright inverters may happen to be designed ok for this function, I am not recommending them because of their poor quality. Mine seem shoddy construction, they break down frequently for various reasons, and are hard to repair.

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #21 on: June 25, 2016, 11:35:18 PM »
The heatsinked fets in the above photo might be a two phase push-pull topology, but they could be a two phase forward converter.

The H bridge mosfets are on the other side of the board, the red wire with fiberglass sleeve insulation connects them to the inductors located next to the center electrolytic capacitor.

the 8 H  bridge fets are bent over and flat against the back of the aluminum extrusion. the 4 screws holding them are visible.
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: BONDING TWO INVERTERS AT NEUTRAL
« Reply #22 on: June 29, 2016, 02:36:33 PM »
Thanks Joe, like you said the H-bridge mosfets are on the back side of the board tucked into pretty tight space.
That reminded me that I need to cool both sides of this board (I had to modify this inverter's cooling to manual after the fans auto-control crapped out).

fiyr

  • Newbie
  • *
  • Posts: 1
  • Country: us
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #23 on: February 04, 2017, 04:55:07 PM »
Sorry to resurrect an old thread but what was the outcome of this?

Was this tested?

joestue

  • Hero Member
  • *****
  • Posts: 1759
  • Country: 00
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #24 on: February 05, 2017, 04:09:30 PM »
my guess is that it works, as in post #18 the lightbulb test showed no voltage between input and output.

what kind of inverters do you have?
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: BONDING TWO INVERTERS AT NEUTRAL
« Reply #25 on: February 10, 2017, 02:29:02 PM »
fiyr sent me a message, I'll post here so that all can benefit.

His configuration is similar to mine, except that he has PSW inverters, and a separate battery for each inverter.
He also verified that his inverters run OK with their Neutrals grounded.

Regarding the PSW, my thinking is that the main difference between my MSW and his PSW is the gate modulation of the H-bridge. Here we've talked about the internal wiring inside the inverter, which should be unrelated to the gate modulation.

Regarding the batteries, my DC+ and DC- terminals of both inverters are connected, his are not. Looks like his configuration may be safer since it does not allow any current loops between the inverters through the DC wiring, in case of the AC leakage to the DC side (which may vary as the inverters operate with varying loads), or for other reasons.

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #26 on: April 24, 2017, 01:59:06 PM »
I am still not able to verify on actual hardware our speculations about connecting the two inverter neutrals, and to ground.
My second PowerBright ML2300-24 inverter went bad recently.
The first one exploded almost all of its MOSFETs in the DC-DC converter section, after about 3 years of fairly gentle use. It scorched the main board pretty bad, so it's probably not repairable.
The PowerBright company has refused any assistance with my repairing their inverter.

In trying to fix the second unit I made a quick and simple schematic, see below.
This may help shed more light on why this inverter has no problem with grounding its ACout/Neutral, or may tolerate connecting the two inverter neutrals together and to ground.
The H-bridge MOSFETs are stuck to the board upside down, so I cannot read the part numbers, and I am only guessing at the location of the source and the drain in this schematic (the gate can be identified by a control circuit connected to it).

This is somewhat off topic but if anyone has experience with repairing inverters, I'd appreciate a push in the right direction. The symptoms on my second failed unit are:
It now puts out the MSW AC waveform with timing and duty cycle as before, but with the amplitude that dropped from about 140V to about 90V.
The voltage on the DC link capacitor (between the DC-DC converter and the H-bridge) is only 95V (should be around 140V).
It looks like the H-bridge section and its control works just fine, but the problem is with the DC-DC converter section, like in my first inverter (but I don't see any exploded MOSFETs here).
There are four 50A fuses in the DC-DC section, and two blow immediately on connect to the 24V battery bank.
This happened after about 2 years of use.
Jack

OperaHouse

  • Hero Member
  • *****
  • Posts: 1308
  • Country: us
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #27 on: April 24, 2017, 03:35:41 PM »
A good test for any FETs is to check the two outside pins of the FET for resistance.  Most often the gate will short.  For the boost converter, place an incandescent lamp for the appropriate supply voltage in series with the power.  The FET are usually connected to power all the time and you can easily determine which boost inverters have failed.  There may be seven or so transformers.   If the lamp doesn't light, The driver transistors for the FET gates may have failed.  The HV sections are usually in parallel, but I have one where each one adds 35V in series.  That 95V may be the case.  The output H bridge FET can be tested for shorts with an ohm meter and it sounds like you already did that.

jack11

  • Jr. Member
  • **
  • Posts: 95
  • Country: us
Re: BONDING TWO INVERTERS AT NEUTRAL
« Reply #28 on: April 25, 2017, 01:13:17 PM »
Thank you, I'll run these tests to see what happens.

The other option I thought of (for this PowerBright inverter and for others that fail in similar ways) would be to trash the DC-DC section entirely, feed the high voltage DC generated elsewhere directly into the link capacitor, and let the H-bridge work on it. But in doing this I'd need to make sure to retain the power supply to the H-bridge control section which now works off the 24Vdc input.

Sounds like a lot of work for a defective piece of crap, but this inverter generates the final AC waveform with harmonics that happen to be just right for a project I have.