ELI can the meter damage UPS and like?

[ericmark]

Reading the very scanty instructions for the 3KW inverter it should close down if overloaded. But I am not so sure when a ELI meter puts a load on the system to measure ELI. Can this damage the inverter?

All UPS systems must be the same, but at the time I worked with inverters and UPS no one really considered the testing with a ELI meter so not sure what the results would be?

Also since power out is so dependent on battery state, although it may pass all tests with fully charged batteries, once they are reaching the cut off point can they still open the protective device? Or would it shut down on electronics and re-energise each time fault is removed?

The RCD is also a worry with the wave form produced will they trip? Especially again when voltage may be low?

This article does talk about UPS but really it does not come under 17th Edition anyway. However I would still like to see some protection. Others must do the same, and I would hope there is not need to re-invent the wheel and some one has done it all before?

I have looked at 551.4.3.3 and the R ≤ 50/Ia would be satisfied if R1 + R2 is less than 1.9 ohms which I am sure it is.

However seems prudent to ask in case there is something silly I have missed.

 

[tim west]

Some years ago I was trying to explain to some guys at work that had put an RCD on the output of a dimmer circuit that they were wasting their time doing so as it wouldn't trip, so in experimentation we decided to try testing the trip times at different output levels.

Anything under 80% output level the trip didnt work and until about 95% were the trip times too slow, it seemed that when the waveform was close to a sine wave (as close as you can get with phase angle dimmers) that the trip would actually work to spec.

There are sine wave dimmers on the market so there's no reason not to assume that a UPS manufacturer may incorporate the same technology to produce a sine wave output as opposed to a chopped waveform in the future, till then beware.

As for protective devices such as fuses or breakers the device, when protecting the output circuit wiring it will act on the basis of power and heat produced through it, surely design can take into account the temperature rating of any stipulated output cable to be used and offer a device that only needs to trip if the same cable is being heated outside of its spec?
In other words if the UPS isnt supplying enough power to cause a heating problem with an overload then there isnt a need to disconnect or make the output cable spec one of a fire resistant material that can withstand higher operating temperatures to allow for when the protective device is no longer operating within it's spec.

 

[NotHimAgain]

Eric you are mixing many things here .

First your references to UPS systems in general and the article on 'commercial' systems. The 16th Edition of BS 7671 has included requirements for such systems since 1997. The EWR has effectively had such requirements since 1989, and no doubt the Factories Act also implied that such systems must be safe. So I don't see why the authors of the article think that the 17th Edition has changed things that much.

Now to your inverter - I take it this is the unit that your son will be using on his boat. As you know, the boat installation is not covered by BS 7671. The actual regulations that should be applied would depend on a number of factors, such as: length; use; etc.

You might choose from BS EN 60092 part 507 if the length is 24 metres or more and it is just a pleasure craft, or it could be ISO 13297, and of course there is also the RCD (Recreational Craft Directive 94/25/EC), the list goes on. Obtaining all these standards might cost more than the boat .

In so far as these standards address the issue of small inverters, they generally treat them as small generators. For example, BS 7671 includes 'Generating sets' with 'Electrochemical accumulators' as their power source.

So the question is why do you want to attack the unit with an Earth loop Impedance test instrument - i.e does the inverter operate with an earth reference, or is it just a floating system (effectively an IT system).

 

[ericmark]

The modified sine wave

means the peak voltage is lower than with a normal sine wave and of course since there is no slop wave form clipping devices like thyristors will not work. There is a warning with the inverter that some meters do not read true RMS and will report a low voltage as a result.

However I was in discussion over RCD tripping where there is a volt drop and it was pointed out that most RCD's are rated to trip at half the rated voltage. However when I realised how many different types of RCD are available not just active and passive one realises some may work OK and others may fail or trip. Likely a suck it and see. Once we are sure there is not an internal earth we will bond some thing to earth. So will be a TN-S system.

With luck someone will have used RCD's on inverters before and can tell me best type and I do not think I will damage inverter with RCD tester.

Over current is more of a worry. As the batteries discharge there will be a curve as to power drawn to inverter shut down so although the 16A MCB may trip with fully charged batteries as the capacity drops it will reach a point where the auto shut down will proceed the MCB operating and the information as to if it latches or will auto re-start is not given.

One idea is to use an active RCD and hope any auto shut down would then trip the RCD and prevent the supply being returned under fault conditions.

NotHimAgain is right it's for my son's narrow boat 57 foot. And yes I realise BS7671 does not count but not really worried about regulations all I want is for it to be safe.

My hope is someone will say the XYZ RCD is what you need it's designed for this use.

The idea is to connect a 16A socket to the inverter and to route so when the shore line is not connected to the 16A plug at stern this can be plugged in instead. Simple method of ensuring never can shore and inverter power be connected at the same time. So the same RCD and MCB will be used for shore and inverter power.

I have never tried to use active RCD's with anything more than a single socket outlet so I don't know what the problems may be as to tripping? My daughter has one fitted for lawn mower which she uses for the microwave now and that does not seem to trip. But may be different on a marina?

Thank you for your input any other ideas welcome. Maybe I am worrying over nothing I have found many houses with no RCD and over the years the earth has degraded to near zero where pipes have become plastic yet up to date the injures as a result have been very few. But my daughter-in-law is a plumber so a little wary!

 

[NotHimAgain]

You haven't answered my question - does the inverter operate with an earth reference - if not there is little point in considering RCDs as there would be no earth path for any fault current to flow back to the source.

You might get an RCD to trip if there were any capacitive coupling but this is hardly reliable. As I suggested above, consider this in the same way you would a small generator. Now in some cases you would want to earth reference these, and in others you wouldn't usually bother. It depends upon the risk factors.

With regard to RCDs - active devices use a difference amplifier and this needs a power supply. Passive devices use a precision relay and this does not need a separate power source. Most RCBOs are active types - JPEL/64 have such faith in them that they effectively ask you to forget that they are RCDs and suggest that circuit design ensures that they work as mcbs - why else would they be included in Table 41.3 .

This subject is still under consideration by the great and the good, but there are concerns about performance under fault conditions on systems subject to voltage reduction during a fault - for that read any system with a high PFC - such as most of ours in the UK , or systems with limited capacity - such as your inverter .

Will it be safe - well it depends upon the risk factors. For example - if you only use it to supply class II equipment the risks would be low.

The use you are suggesting might lead you down the road of ensuring that you have an earth reference - this would generally mean connecting the inverter neutral to earth - but you must check that out. I would then use a double pole passive RCD - I cannot guarantee that it would work with that output waveform but, IMO, its your best shot.

 

[Spark123]

However I was in discussion over RCD tripping where there is a volt drop and it was pointed out that most RCD's are rated to trip at half the rated voltage.

Bog standard RCDs trip on Current, not voltage.
Some which have a functional earth lead require a voltage to operate, iirc this is down to 50v.
The fact that you are using a modified sin wave may have a dramatic effect on the tripping characteristics as an RCD relies on the induced magnetic flux from the current in the coils to operate. As the required induced magnetic flux into the sense coil is proportional to the rate of change of currents in the live coils any DC injected into the RCD may lock it.
Saying that some RCDs don't use torroids like they used to and some specifically say they are suitable for AC and pulsating DC.

 

[ericmark]

Thanks from what you say I need type A not type AC that is a good start.

Yes there will be an earth connection but not sure where. Need to connect up and see if already connected within inverter before connecting any external links.

However a search for Active Twin pole Type A RCD does not find much and so many don't say if Type A or Type AC on adverts.

But first I need to find what is already fitted.

 

[Spark123]

As NotHimAgain states above, if your system is not "Terra" as in the source is not connected to earth an RCD will not be effective.
I'm not sure if "active" is the correct terminology as to me, active means it holds in and trips on mains failure whereas passive don't trip when the supply fails.
An RCD with electronics within, such as most RCBOs and some RCCBs (think MEM are one manufacturer) may refuse to reset on a non earthed supply.

 

[NotHimAgain]

I agree the terminology active and passive is poor.

I am using active to describe RCDs with an active element i.e. the difference amplifier. I am using passive for those RCDs that still use a precision relay.

The former requires a supply for the amplifier, and the latter does not as the relay is driven by the small voltage difference that is induced due to the difference in line and neutral currents.

There is concern over the 'active' type particularly when used in high fault level systems. These systems can suffer from voltage collapse during the high current flows produced in a fault. This concern is noted in BS 61009

From BS 61009
4.1.2.2 Not opening automatically in case of failure of the line voltage:
a) Able to trip in case of a hazardous situation (e.g. due to an earth fault), arising on
failure of the line voltage (}additional requirements are under consideration~);
b) Not able to trip in case of a hazardous situation (e.g. due to an earth fault), arising on
failure of line voltage.
} Note deleted~

 

[Spark123]

The old 16th edn used to require (or did it say preferably?) RCBOs (BS(EN)61009) to operate on the MCB element whearas reading the 17th edn regs, this appears to have been dropped and the RCD element can be used. I can't imagine many normal sparks having access to nor reading documents such as BS(EN)61009 tbh so if it is hidden in there why isn't it in the 17th edn regs, or am I missing it?

 

[NotHimAgain]

I doubt that the 16th edition actually 'required' it. Like the 17th it just just alluded to it by the fact that the BS EN 61009 devices were included in 41B (now 41.3).

It is a problem that only really affects countries with high earth fault levels (mainly northern Europe, north America etc). It has/is being addressed by RCD manufacturers but it still causes some concern.

It could also affect very small systems such as small inverters and generators.

It is not the sort of thing that usually bothers most electricians as, like many things, it is not widely discussed.

I cannot point to lots of bodies caused by this problem so it is not something that should concern people too much until we hear otherwise .

 

[RF Lighting]

I'm no UPS expert, but AFAIK, mine works by allowing 230V to go straight through it when the mains is healthy using the circuit CPC as normal, and only switches onto the inverted battery supply when the mains fails.

Presumably the invertor supply would be an earth free isolated supply?

Have I got that all wrong?

(sorry for the hijack)

 

[NotHimAgain]

Ericmark's link points to an article about the way they work. It is basically as you have said, but you cannot assume that the system would be 'earth free'. This may be true on small units, but it is unlikely, and indeed undesirable, on large systems feeding a number of outlets.

Much of the problem with these units is lack of information - some manufacturers are very good and will give you chapter and verse, others say little. It can be very difficult if you are doing a PIR - you may need a get out clause .

 

[RF Lighting]

Mine is only a dinky little one.

It never even entered my mind about commercial systems, where all that equipment with high CPC currents and no earth really wouldn't be a good mix

I'll shut the door on the way out

 

[Spark123]

How can you have high earth currents in an electrically separate system