Inverters sine or not

[oldade]

Hi

I have a question about inverters.

1 Is it true that the pure sine wave invertes are the least efisant ?

2 Would it be worth while having two inverters and two 240 circuits one clean (signwave) and one dirty IE (non signwave)

from a efishancey point of view.

I am about to start a newbuild and it would be quite easy to install two 240 circuits.

What do we all think ?

I know I have two spelling mistakes I am sure some one will enjoy correcting them.

Thanks

ADE

Edited April 12, 2008 by oldade

[NB Alnwick]

Hi

I have a question about inverters.

1 Is it true that the pure sine wave invertes are the least efisant ?

2 Would it be worth while having two inverters and two 240 circuits one clean (signwave) and one dirty IE (non signwave)

from a efishancey point of view.

I am about to start a newbuild and it would be quite easy to install two 240 circuits.

What do we all think ?

I know I have two spelling mistakes I am sure some one will enjoy correcting them.

Thanks

ADE

 

Our old Sterling 1800W modified Sine wave inverter does everything we need - we have various 240v appliances - the only major items that we do not have are an electric washing machine and an electric dish washer.

[Keeping Up]

Yes sine wave inverters are less efficient. But I solve it differently, still with two separate 240v circuits. I have a small sine-wave inverter for low-power items (TV, phone-charger, etc) and a large sine-wave inverter for high-power itemas (hoover, toaster, microwave, spin-dryer, hair dryer, etc).

 

The large inverter is very inefficient but switches itself off completely when it's not being used. The small inverter is also inefficient in perecentage terms but the power involved is so small that I don't bother to switch it off unless the battery situation is critical.

[sueb]

I charge my phone & run my computer on 12v.

Sue

[Ex- Member]

I charge my phone & run my computer on 12v.

Sue

 

Yeah us too and, TV, dab radio, lap top and mobiles, can't see the point of using an inverter to go up to 240 then back down to 12

[chris w]

Hi

I have a question about inverters.

1 Is it true that the pure sine wave invertes are the least efisant ?

2 Would it be worth while having two inverters and two 240 circuits one clean (signwave) and one dirty IE (non signwave)

from a efishancey point of view.

ADE

 

You shouldn't think of pure and quasi inverters as giving a "clean" and a "dirty" mains output. A pure sinewave inverter will have about 2% harmonic distortion (a measure of its "cleaness" if you will). The figure for quasi sinewave inverters is about 6% so the difference is not huge. The problem is only with appliances that need a very pure sinewave because they specifically rely on the wave shape. Examples of this are motor control circuits and some timers like washing machines motors and microwaves with digital timers.

 

A quasi sinewave inverter will use about an additional 1A to run itself. A pure sinewave inverter will use around 5-7A to run itself so the pure sinewave version is certainly more inefficient. It is also more expensive but will run all devices within its power capability. A quasi sinewave should run virtually all devices other than those highlighted above.

 

Chris

[Keeping Up]

Depends on the size of the inverter too. Our small 300W sinewave inverter takes only 0.7 amps to run itself - but our big one (which isn't really an inverter at all because it is really a motor driving an alternator) takes 10 Amps to run itself but only when it actually IS running.

[chris w]

Be aware too that some pure sinewave inverters go into a "standby" mode with no load on them. This is the figure quoted as the "quiescent" current. However, when on load, the "quiescent" current is much higher.

 

Chris

[dor]

A quasi sinewave inverter will use about an additional 1A to run itself. A pure sinewave inverter will use around 5-7A to run itself so the pure sinewave version is certainly more inefficient.

Chris, I respect your obviously considerable knowledge about things electrical and electronic, but what you have just said is crap.

 

You have often gone on about the huge inefficiency of PSW inverters, but I wonder wher eyou draw your experience from.

 

I've been fitting out my boat now for several months, and use a number of 230V tools. I turn on my inverter when I go to the boat, partly so the power is there when I need it, but mostly so I can run my DAB radio and listen to Planet Rock. My inverter, when running just my DAB radio, draws 1.1 amps and this is an 1800 Watt Pure Sine Wave inverter. So I am interested to know where you get this figure of 5 to 7 amps for a PSW inverter to run itself.

 

Just like in your previous pronouncements on how much power 12V fridges consume - which was in complete variance to my, and others, practical experience - I think things have moved on. Your figures are true for equipment a few years ago, but I think things are a bit more efficient now.

[chris w]

Chris, I respect your obviously considerable knowledge about things electrical and electronic, but what you have just said is crap.

 

You have often gone on about the huge inefficiency of PSW inverters, but I wonder wher eyou draw your experience from.

 

I've been fitting out my boat now for several months, and use a number of 230V tools. I turn on my inverter when I go to the boat, partly so the power is there when I need it, but mostly so I can run my DAB radio and listen to Planet Rock. My inverter, when running just my DAB radio, draws 1.1 amps and this is an 1800 Watt Pure Sine Wave inverter. So I am interested to know where you get this figure of 5 to 7 amps for a PSW inverter to run itself.

 

Just like in your previous pronouncements on how much power 12V fridges consume - which was in complete variance to my, and others, practical experience - I think things have moved on. Your figures are true for equipment a few years ago, but I think things are a bit more efficient now.

 

I can't speak for your particular inverter because I don't know the model/manufacturer. Which one is it?

 

As for 12v fridges, they will consume around 40-50AH per day - I've said it before and I'll say it again. If you purport to have one that consumes significantly less than that, I would find that hard to believe. Who are these "others" who also seem to have low AH 12v fridges. From my recollection, "others" experience seems to concur with mine (and the manufacturers' data sheets).

 

Chris

[dor]

Chris, my inverter is a Powermaster 1800.

 

As I've stated before, my fridge consumes an average of about 1.2 - 1.4 amps, once it is down to temperature. This is for a 150 litre (i.e. domestic-size) cabinet. Rather than being at odds with other people, several others have reported similar results with their fridges. The manufacturer's data sheet actually gives an average consumption of 0.9 amps which I would find hard to believe, but that would be in their ideal test conditions rather than the real situation of a room temperature boat.

[Gibbo]

You shouldn't think of pure and quasi inverters as giving a "clean" and a "dirty" mains output. A pure sinewave inverter will have about 2% harmonic distortion (a measure of its "cleaness" if you will). The figure for quasi sinewave inverters is about 6% so the difference is not huge.

 

Er.....NO!!!!!!!!!!!.... Try 30% to 40% THD for MSW

 

The problem is only with appliances that need a very pure sinewave because they specifically rely on the wave shape. Examples of this are motor control circuits and some timers like washing machines motors and microwaves with digital timers.

 

A quasi sinewave inverter will use about an additional 1A to run itself. A pure sinewave inverter will use around 5-7A to run itself so the pure sinewave version is certainly more inefficient.

 

Again..... er.... No.

 

Try 2 to 4 amps for 2000 watt unit.

 

Gibbo

[smileypete]

Fridges are usually rated at 25°C with no door opening.

 

To say they will consume xx every day is a bit laughable, as it's so highly dependent on ambient temperature.

 

Even my D rated mains fridge consumes 0.4kWh/day in winter, and the door gets opened quite a lot.

 

cheers,

Pete.

Edited April 13, 2008 by smileypete

[Gibbo]

I just measured the current draw on a few inverters.

 

1. Statpower 12 volt 1500 Watt modified sinewave. This is the unit that 90% of 1500 Watt MSW inverters are clones of. No load current draw= 0.7 Amps.

 

2. Kotec 12 volt 1500 Watt pure sinewave inverter. Most often seen sold under the Sinergex brand name. The Rich Electric (badge sold by Sterling) is almost identical electronically so will be very similar. No load current draw (not in idle) = 1.7 Amps.

 

3. Mastervolt Mass 12/1200, 1200 Watt pure sinewave combi. Not in idle mode. Current draw = 0.9 Amps.

 

4 Mastervolt Mass 12/2000, 2000 Watt pure sinewave combi. Not in idle mode. Current draw = 1.1 Amps

 

Gibbo

[chris w]

Er.....NO!!!!!!!!!!!.... Try 30% to 40% THD for MSW

 

Gibbo

 

For a multistep quasi sinewave (ie: not a simple squarewave cheapo version) like the Sterling et al, I can send use the mathematical Fourier analysis of the waveform and the result is 6% THD. It's a mathematical fact not supposition. Even a simple squarewave would be better than 30-40% THD.

 

Chris

 

Chris, my inverter is a Powermaster 1800.

 

As I've stated before, my fridge consumes an average of about 1.2 - 1.4 amps, once it is down to temperature. This is for a 150 litre (i.e. domestic-size) cabinet. Rather than being at odds with other people, several others have reported similar results with their fridges. The manufacturer's data sheet actually gives an average consumption of 0.9 amps which I would find hard to believe, but that would be in their ideal test conditions rather than the real situation of a room temperature boat.

 

You can't quote an average without quoting a duty cycle - it's meaningless. I could have a fridge that draws 60A for 1/2 hour and switches off for the rest of the day. The average current is 1.25A !!

 

Also, how are you measuring the average current. Are you standing by the fridge noting the current and time throughout a 24 hour period?

 

Chris

[Gibbo]

For a multistep quasi sinewave (ie: not a simple squarewave cheapo version) like the Sterling et al, I can send use the mathematical Fourier analysis of the waveform and the result is 6% THD. It's a mathematical fact not supposition.

 

I think your calculator needs some new batteries.

 

A modified sinewave has 3 voltage levels. Assuming the horizontal sections are perfectly flat then the THD content varies between a minimum of 9% and a maximum of 51.3% depending upon the widths of each on period.

 

Line frequency MSW inverters have a variable on period and therefore the THD varies between 9% and 51.3%

 

High frequency MSW inverters have a fixed mark space ratio and this is *usually* fixed with the on time starting 22.5% degrees into the cycle. This results in a THD of 36.2%

 

Even a simple squarewave would be better than 30-40% THD.

 

A square has a harmonic distortion content of exactly 43.52%

 

It can be proved very simply via your usual route of sums.

 

Gibbo

[dor]

You can't quote an average without quoting a duty cycle - it's meaningless. I could have a fridge that draws 60A for 1/2 hour and switches off for the rest of the day. The average current is 1.25A !!

 

Also, how are you measuring the average current. Are you standing by the fridge noting the current and time throughout a 24 hour period?

 

It's simple enough Chris; I said my fridge draws an average of 1.2 - 1.4 amps. I also referred to the real situation in a room temperature boat. So, in case I need to explain it again, the figures for the current are typical average currents for the fridge in typical use on a narrowboat that is at a reasonable room temperature.

 

I've measured the duty cycle by linking a small electric clock into the circuit so I can get an average time on during a specific period. I've measured it over 24 hours and over a week whilst it is being used.

 

 

And I think Gibbo has confirmed the sort of current draws for a PSW inverter - a lot less than your 5 - 7 amps!

 

I don't want to pick a fight Chris, but you frequently state your theoretical figures which can be at significant odds with figures found by people in practical applications.

Edited April 13, 2008 by dor

[chris w]

Fridges are usually rated at 25°C with no door opening.

 

To say they will consume xx every day is a bit laughable, as it's so highly dependent on ambient temperature.

 

Even my D rated mains fridge consumes 0.4kWh/day in winter, and the door gets opened quite a lot.

 

cheers,

Pete.

 

It's hardly laughable, as we need to estimate a figure in order to ascertain the likely consumption. That's what engineering is about. There are many variables of course, but we have to make some assumptions and understand the result is based on those assumptions. Otherwise, we woudn't even know what ballpark we are in.

 

Your fridge, you say, consumes 0.4KWhr/day in winter (how are you measuring that?) which is 33AH per day in winter. One would therefore expect that the consumption would be significantly higher in summer.

 

Chris

 

I think your calculator needs some new batteries.

 

A modified sinewave has 3 voltage levels. Assuming the horizontal sections are perfectly flat then the THD content varies between a minimum of 9% and a maximum of 51.3% depending upon the widths of each on period.

 

Line frequency MSW inverters have a variable on period and therefore the THD varies between 9% and 51.3%

 

High frequency MSW inverters have a fixed mark space ratio and this is *usually* fixed with the on time starting 22.5% degrees into the cycle. This results in a THD of 36.2%

 

 

 

A square has a harmonic distortion content of exactly 43.52%

 

It can be proved very simply via your usual route of sums.

 

Gibbo

 

I'll send you the Fourier analysis.

 

Chris

[Nickhlx]

Would it not be more meaningful to measure the efficiencies of the various inverters at no load, 1/3rd load, 2/3rds load and rated max. load as power in against power out ?

 

Gibbo - have you time to do this for a couple of the more popular ones ? - it might be more real life than just off-load readings ?

 

Nick

[Gibbo]

I'll send you the Fourier analysis.

 

Chris

 

Try this instead.....

 

http://www.itee.uq.edu.au/~aupec/aupec99/liu399.pdf

 

Can you show me your maths that gives you the THD of a squarewave as being less than 30% and I'll try to work out where you've gone wrong. It is, of course, 43.52% as everyone knows.

 

 

Gibbo

 

 

 

Would it not be more meaningful to measure the efficiencies of the various inverters at no load,

 

That one is easy. I can quite categorically state here and now that the efficiency of every inverter available, at no load, is 0%. They are all much of a muchness

 

1/3rd load, 2/3rds load and rated max. load as power in against power out ?

 

Gibbo - have you time to do this for a couple of the more popular ones ? - it might be more real life than just off-load readings ?

 

They are all pretty equal to be honest. Only the no load drain varies so much which of course makes the efficiency lower at lower power levels but has less effect at high power levels.

 

The units I tested earlier today were just one old piece of junk kicking round the workshop. The others were here for a one off special for another dealer. They're packed away now so I can't run the tests.

 

In the case of modern ones if you take a MSW and assume it to be 90% efficient across the range then add in the no load drain you will be very close. For a sinewave do the same assuming 85%. If you graph these you'll see that the efficiency drops as the power drops because the no load drain becomes a bigger proportion of the input power.

 

Gibbo

[chris w]

Try this instead.....

 

http://www.itee.uq.edu.au/~aupec/aupec99/liu399.pdf

 

Can you show me your maths that gives you the THD of a squarewave as being less than 30% and I'll try to work out where you've gone wrong. It is, of course, 43.52% as everyone knows.

 

 

Gibbo

 

The MSW (multistep) sinewave is only around 6% THD (6.5% to be precise). The Fourier analysis is on its way.

 

Chris

Edited April 13, 2008 by chris w

[chris w]

Gibbo

 

This is the Fourier analysis of my Sterling 1800W MSW inverter's waveform. I haven’t gone into the theory behind Fourier Analysis as I assume you know it already and it would take several pages on here to explain it and still be unintelligible to anyone below degree level mathematics. But, as you must know, it is the standard method by which waveforms are analysed for THD.

 

 

 

 

The waveform for my Sterling 1800W inverter is shown above. Using standard Fourier analysis and the fact that the waveform has both half-wave and quarter wave symmetry, the integration is carried out over the period 0 to pi/2, with the result that:

 

b_n = (4/n x pi) {Asin(nβ) + (B-A)sin(nα)} for odd values of n only

 

This result has four variables, of which all could theoretically be varied to achieve minimum distortion. However, from the diagram, B=2A on the Sterling and so the values of α and β are optimised for minimum distortion.

 

With this restriction, evaluation of the Fourier coefficients shows that the minimum distortion is about 6.5% (-24 dB), and occurs at β = 0.42 x pi and α = 0.248 x pi. Further, the third harmonic is only about 0.17% (-55db) of the fundamental, meaning that even minimal low-pass filtering greatly reduces the fifth and higher order harmonics to leave a relatively clean sinewave.

 

Chris

[oldade]

Gibbo

 

This is the Fourier analysis of my Sterling 1800W MSW inverter's waveform. I haven’t gone into the theory behind Fourier Analysis as I assume you know it already and it would take several pages on here to explain it and still be unintelligible to anyone below degree level mathematics. But, as you must know, it is the standard method by which waveforms are analysed for THD.

 

 

 

 

The waveform for my Sterling 1800W inverter is shown above. Using standard Fourier analysis and the fact that the waveform has both half-wave and quarter wave symmetry, the integration is carried out over the period 0 to pi/2, with the result that:

 

b_n = (4/n x pi) {Asin(nβ) + (B-A)sin(nα)} for odd values of n only

 

This result has four variables, of which all could theoretically be varied to achieve minimum distortion. However, from the diagram, B=2A on the Sterling and so the values of α and β are optimised for minimum distortion.

 

With this restriction, evaluation of the Fourier coefficients shows that the minimum distortion is about 6.5% (-24 dB), and occurs at β = 0.42 x pi and α = 0.248 x pi. Further, the third harmonic is only about 0.17% (-55db) of the fundamental, meaning that even minimal low-pass filtering greatly reduces the fifth and higher order harmonics to leave a relatively clean sinewave.

 

Chris

 

 

Thanks

 

Thats to deep for me.

 

ADE

[dor]

Gosh Chris, I wish I was as clever as you.

[Gibbo]

Gibbo

 

This is the Fourier analysis of my Sterling 1800W MSW inverter's waveform. I haven’t gone into the theory behind Fourier Analysis as I assume you know it already and it would take several pages on here to explain it and still be unintelligible to anyone below degree level mathematics. But, as you must know, it is the standard method by which waveforms are analysed for THD.

 

 

 

 

The waveform for my Sterling 1800W inverter is shown above. Using standard Fourier analysis and the fact that the waveform has both half-wave and quarter wave symmetry, the integration is carried out over the period 0 to pi/2, with the result that:

 

b_n = (4/n x pi) {Asin(nβ) + (B-A)sin(nα)} for odd values of n only

 

This result has four variables, of which all could theoretically be varied to achieve minimum distortion. However, from the diagram, B=2A on the Sterling and so the values of α and β are optimised for minimum distortion.

 

With this restriction, evaluation of the Fourier coefficients shows that the minimum distortion is about 6.5% (-24 dB), and occurs at β = 0.42 x pi and α = 0.248 x pi. Further, the third harmonic is only about 0.17% (-55db) of the fundamental, meaning that even minimal low-pass filtering greatly reduces the fifth and higher order harmonics to leave a relatively clean sinewave.

 

Chris

 

Apart from the waveshap is COMPLETELY wrong.

 

You have assumed that there is a voltage level at zero volts: correct

You have assumed there is a voltage level at peak volts: correct

You have assumed there is a voltage level at peak/2 volts: incorrect.

 

That middle voltage level simply does not exist. There are 3 voltage levels on a MSW inverter. 0 volts, + peak volts and - peak volts. There is not one inbetween.

 

Now go try your sums again.

 

And where are your sums for a square wave having less than 30% THD?

 

Gibbo