Author Topic: An inverter question  (Read 1787 times)

0 Members and 1 Guest are viewing this topic.

Capt Slog

  • Sr. Member
  • ****
  • Posts: 378
An inverter question
« on: October 25, 2008, 09:34:36 AM »
There have been a few posts lately about wiring inverters.


I've seen it said that the earth pin of the inverter output socket is attached to the centre tap of the transformer in the unit, and the problems this causes in wiring the inverter to existing circuits, due to floating neutrals and how ordinary grid mains is tied neutral to ground.


As it had been my intention for some time to do such a wiring project, and I have the inverter that I wanted to use, I thought I'd better check the insides of it.  On opening it I found that there was NO earth connection at all, which surprised me rather.


It occurs to me that if I want to use this inverter for a lighting circuit I would want to add a ground for safety to protect against the cables being accidentally cut or a metal fitting becoming live.  I can't see that this affects the operation of the inverter in any way, or am I missing something?


I'm thinking on the lines that the earth only comes into play in the event of emergency, and then fuses blow shutting the whole thing down.  


Thanks

« Last Edit: October 25, 2008, 09:34:36 AM by (unknown) »

Flux

  • Super Hero Member Plus
  • *******
  • Posts: 6275
Re: An inverter question
« Reply #1 on: October 25, 2008, 06:54:10 AM »
This is something of a minefield and I am not very keen to get involved.


I have no real idea of the inverters available in N America so I can only comment on what I have found in the UK.


Here our supply is derived from a 3 phase 415v system with neutral grounded. One side of the 230v is grounded ( neutral ) at the supply transformer. The phase leads are live at 230v with reference to neutral.


I haven't had a great deal of experience with inverters ( playing as a hobby I can't afford them so I only see what I have fitted for others). All the cheap modified square wave things use a H bridge to chop the high voltage dc into a series of square pulses. The dc is isolated by the converter transformer from the battery supply and the ac output is completely isolated from the input and the case of the inverter which has a ground connection. These seem to be aimed largely at the car and camping market and they suggest you ground the inverter case. The load is assumed to float and the inverter transformer should be more than adequate to provide safe isolation. You can't get a shock from the output unless you touch both leads which in most cases should be impossible. Whether you can convince the stupid safety law makers it's safe is another matter. I think there is some special case for isolated systems for portable generators but I keep clear of the legislation.


Most of these inverters will run perfectly happily with one of the output leads grounded so it you tie it to a normal house wiring system it will still work. The neutral and ground are then linked just as for a grid supply. There is absolutely no way that any of the small inverters will blow a fuse and if you want fault protection then you need to use a RCD breaker that trips on differential current. Even most small engine driven generators don't have enough fault current to blow fuses and again they need similar treatment.


The only sine wave inverters I have met also had isolated output and a ground to the metal case. The filter circuits are invariably grounded to the inverter case and you should ground it to keep RF interference at bay. The Ones I have met are perfectly happy to let you ground one output lead.


There are older generations of inverter that use transformers on the output but I haven't met any for a long time although some of the high price units may do so ( I haven't played with them).


The world is geared to the grid system and most of the legislation is based on it, you open a minefield with isolated devices but personally I think they are safer but convince the man!


I think things are rather different in N America, the 110v system works the same way but the 220v is not from a 3 phase but from a split single phase with the centre point grounded. If you try to use a 220v inverter there is a problem in that the mid point of the inverter output gets grounded. If it is completely isolated then it will work on 220v and you can ground the case for the interference suppression. There is no way you can take any 110v load from it and all loads must be 220v unless resistive and completely balanced about the 110v.


If there is any form of internal connection from the inverter output to elsewhere then there will be a fry up.


I have never met a 60Hz 220v inverter so I have no idea if it is different from the 230v 50Hz ones.


I try to avoid this area like the plague. I have seen things installed by those who should know better that look to be complying with all the legislation bur some of it has been pretty dangerous if you stop to think about it from first principles rather than just blindly follow grid practice. What I do and what I tell others to do may differ.


Flux

« Last Edit: October 25, 2008, 06:54:10 AM by (unknown) »

dnix71

  • SuperHero Member
  • ******
  • Posts: 2513
Re: An inverter question
« Reply #2 on: October 25, 2008, 11:02:35 AM »
If this is a unit made for the US, then there are two types, divided by output wattage.

>5,000 watts and <5,000 watts.


Those over 5k watts MUST be wired like a house. Small pin hot to neutral and ground, large pin is neutral. Neutral and ground are bonded together. The ground is a redundant neutral, for safety.


The inverters <5,000 watts are almost all floating neutral, which is safer for a portable device, but floating neutral inverters MUST NOT EVER be connected to house wiring that is grid tied.


On a floating neutral inverter/generator, the exciter circuit looks for a electrical connection between the large and small slots (which would be hot and neutral if this was in a house).


Ground is isolated from both slots, and is intended to be connected by suitable cable from the frame of the inverter to the ground. The ground in this case is intended to be a low impedance path away from the inverter operator to minimize shock in case something fails.


You cannot get shocked by a floating neutral inverter except by touching both outlet slots at the same time (normally). There are special exceptions to this rule. People have died by using floating neutral inverters on the bed of a pickup truck and not grounding the bed of the pickup truck. The US Park Service has issued rules encouraging its employees to not use floating neutral inverters/generators, for this reason.


If you connect a floating neutral inverter to a house that is correctly wired for the grid, you will very quickly destroy the inverter because the neutral and ground of a house wire are bonded and the output of the inverter will be sent to earth.

« Last Edit: October 25, 2008, 11:02:35 AM by (unknown) »

dnix71

  • SuperHero Member
  • ******
  • Posts: 2513
Re: An inverter question
« Reply #3 on: October 25, 2008, 11:23:05 AM »
Using a floating neutral inverter/generator in the bed of a pickup truck is dangerous because if something fails in the inverter, even if you ground the bed of the truck, you are still standing on the ground. You can also have the truck battery float to high voltage from an inverter circuit failure and the frame of the truck then becomes high voltage.
« Last Edit: October 25, 2008, 11:23:05 AM by (unknown) »

zeusmorg

  • Sr. Member
  • ****
  • Posts: 364
Re: An inverter question
« Reply #4 on: October 25, 2008, 01:01:17 PM »
  Hmm, maybe there's a way to make this simple, it it has a plug in it for plugging in appliances, it WON'T be useful to try to run anything in a house circuit off it.


 I can't understand why people try to "get off cheap" when it comes to an inverter.


 If it is small it won't run much anyway so why take the time and effort trying to convert it to a use it wasn't intended for?


 Usually i find the time invested attempting to make something do what it wasn't designed for will cost you more and time and effort than just buying appropriate equipment.


 Most of this equipment comes with operating instructions. I doubt any of it says "plug this into your house wiring, it'll be ok"


 Usually the smallest circuit in a house(single circuit) is 120v @ 15a and this is 1800 watts. Most of these small inverters will not even supply that.


 I have found that there are many posts and threads over the past year or so dealing with this issue.

« Last Edit: October 25, 2008, 01:01:17 PM by (unknown) »

Capt Slog

  • Sr. Member
  • ****
  • Posts: 378
Re: An inverter question
« Reply #5 on: October 26, 2008, 02:41:51 AM »
Just to clarify, I'm inthe UK and the inverter is very small; just 150w, 300w peak.


I know that sounds ridiculously small, but it's advantage is that when it runs with no load, it's using very little power.


The circuit I wanted to run from it is a few lights and these are CFLs.  I've checked the inverter with CFLs and it runs them witout any problems.  Because the CFLs are such low wattage, I can run quite a few (upto the 150w max) but I realise that it's not exact in that I can't run 15 x 10W.


Thanks for all the replies so far, a lot of very helpful stuff in there that has nade ne think about the issues.


.

« Last Edit: October 26, 2008, 02:41:51 AM by (unknown) »

fungus

  • Sr. Member
  • ****
  • Posts: 448
Re: An inverter question
« Reply #6 on: October 27, 2008, 03:05:23 AM »
This is how I understand it:

On the small inverters, usually when you get to the high voltage side, they have a center tap at ground, and take the positive or negative pulses off each side of this .. usually this is perfectly fine .. Although usually in America (I think, probably other places too), the neutral is bonded to the ground at the service entrance to the house .. so when you switch your inverter on, the neutral is connected directly to the ground, shorting out the inverter on one half of each cycle, ruining it ..
« Last Edit: October 27, 2008, 03:05:23 AM by (unknown) »

zeusmorg

  • Sr. Member
  • ****
  • Posts: 364
Re: An inverter question
« Reply #7 on: October 27, 2008, 04:43:32 AM »
 Ok, the safest way to do this would be forget about breaking out existing wiring, and run an all new feed line. Totally eliminating any connections to your existing service.


 In some wiring, you'll find that an electrician( or an amateur electrician) has connected the ground and neutral together somewhere other than at the breaker box. Tricks like grounding the EMT boxes.. attempting to make existing wiring work with a 3 plug outlet, where originally it was a 2.. sometimes finding connections like that can be very time consuming, because of where the wiring has been run. (or a box sealed up behind drywall.)


 Also if the inverter doesn't have proper circuit breaker protection or fuses these would need to be added.


 I'm not saying it can't be safely done, it is just that unless you understand how to test to make sure there are NO connections, hooking up a "floating ground" inverter to an existing system can make things smoke.


 If you are determined to use existing wiring, you would have to disconnect the existing run  at the breaker box,(hot feed, neutral and safety ground) then test those to make sure that there is no live feed, or connection to the ground or neutral.


 In other words PROCEED AT YOUR OWN RISK.

« Last Edit: October 27, 2008, 04:43:32 AM by (unknown) »

Flux

  • Super Hero Member Plus
  • *******
  • Posts: 6275
Re: An inverter question
« Reply #8 on: October 27, 2008, 01:00:40 PM »
I don't want to take this much further but all the cheap inverters I have seen don't use a centre tapped transformer, they use a H bridge output ( may be exceptions)


I think the problem is with the method of sensing the ac voltage and feeding it back to the control scheme. If this was completely isolated you could ground one side of the output. I suspect it is not entirely isolated and if you have a condition where the dc input is grounded and you ground one side of the output you defeat the feedback and destroy it. This is also likely to be the reason for accidents involving shock from what should be a completely isolated supply. If the feedback is not isolated from battery negative then a ground on battery negative could make one output lead live otherwise you couldn't get hurt unless you touched both leads.


I agree that it is better to use these cheap things for isolated loads and not try to tie them to grounded neutrals. The better sine wave small inverters I have met are truly isolated and you can ground one output lead and the maker says that you can do so.


Flux

« Last Edit: October 27, 2008, 01:00:40 PM by (unknown) »

Clifford

  • Full Member
  • ***
  • Posts: 158
Re: An inverter question
« Reply #9 on: October 27, 2008, 02:07:03 PM »
Most of the cheap inverters don't like both a negative ground at the battery, as well as a neutral ground on the output.  They will work fine with one or the other, but not both.


There is a section in the current National Electric Code in the USA about ungrounded DC systems.  Limited to under 48V, I think.


If you use an ungrounded 12V system, you are supposed to have two pole switches, and a few other upgrades.

« Last Edit: October 27, 2008, 02:07:03 PM by (unknown) »