Inverter Dying?

[rwidman]

Originally Posted by Auspicious

On this we can agree. I have yet to find a manufacturer that doesn't provide reasonable wire sizing guidance. Sometimes that means big wires and people get cranky about the wire costs. Some people lose sight of the need to count the wire in both directions (battery to inverter and inverter to battery). We are talking about significant current and heat will increase as the square of the current. Big wire is good. Proper crimped connections are good.

With any product that is sold over the counter to anyone with a credit card, there is the risk of improper installation and use. For some reason, many people will ignore the instructions or misread them. The of course, there are those who think they know better than the manufacturer how to install or use a product.

My neighbor tried to hook an inverter to his truck to power a hotplate. It didn't work. I tried really hard to convince him that the pair of ten foot long 12 gauge wires he used were not large enough but I couldn't get him to understand.

[Auspicious]

Originally Posted by rwidman

My neighbor tried to hook an inverter to his truck to power a hotplate. It didn't work. I tried really hard to convince him that the pair of ten foot long 12 gauge wires he used were not large enough but I couldn't get him to understand.

And therein lies the problem. Unfortunately posts on this thread just make the problem worse.

[transmitterdan]

Originally Posted by rwidman

But that's exactly what you're doing. Not the part about making sure the cables are capable of carrying the required current but the part about getting alternating current out of a battery or measuring alternating current on the conductors from the battery.



You are throwing technical terms around like you know something and you may in fact know something but you should know that pulsating DC is not "alternating current".

If you believe that the DC battery current drawn by a "good" sine wave inverter and the RMS current are identical then you are simply mistaken. And the reason the RMS value is higher is because there is "ripple" current of 120Hz, 240Hz and higher harmonics. By definition these are AC currents and are thus ignored by a DC meter. But the wire and connector are heated by this extra AC current. It's as much as 25% more heating.

Similarly, the maximum or peak current is much higher than the DC current. It is 60% or more higher than the DC meter shows. This must be taken into account when considering the voltage drop of a wire run.

The Fourier theory behind these facts is not "blowing smoke". But ignoring it can cause real smoke.

[Auspicious]

Originally Posted by transmitterdan

the reason the RMS value is higher is because there is "ripple" current of 120Hz, 240Hz and higher harmonics.

Are you sure you aren't simply losing sight of the inefficiencies in the inverter?

Switching supplies run at much higher frequencies than 50 or 60 Hz even while delivering pure(ish) sine waves at the desired delivery frequency.

[transmitterdan]

I am ignoring efficiency. Efficiency makes the situation a little worse than I have calculated but not enough to be significant. Just increase the DC current estimate by dividing the efficiency factor. A 1000W inverter at 80% efficiency draws about 100A DC at 12V (1200W).

I will repeat myself with no intention of blowing smoke. The 60 (or 50 in some countries) Hz AC current produced by the pure sine inverter comes from the battery. There is no where else to get it. There is no magic inside the inverter that can get around the laws of physics. The inverter draws rectified AC current (thus creating a DC component) from the battery. It inverts every other cycle of the current to turn this into pure AC current. It uses high frequency switching to accomplish this seemingly impossible feat.

And now someone will want to argue that I am an idiot. That's ok. If you don't believe me go and measure the battery current with AC and DC meters. There will be AC current and DC current coming from the battery. Then measure the peak current and you will be very surprised by how much higher it is than the DC meter shows.

[Coops]

If you "experts' cannot agree, what chance does an imbecile like me have of knowing what is believable? Eeeny meeny miny mo?

Coops.

[Terra Nova]

Not a chance.

[transmitterdan]

All,

I apologize for any insult that may have been perceived by my posts. It was not intended to be so. I know a lot of inverter installations that nearly burned boats and I am sure some "mysterious" electrical fires are due to improper inverter installations. This thread started due to an "almost" fire. This problem will get worse as more cruisers switch to larger battery banks and bigger inverters. Believe me that the current from the battery bank is higher than you would first think based on simple DC formulas. And lots of inverters don't have provision for proper sized wire which compounds the problem. Further, the availability of low priced high power inverters invites low cost choices for wire, fuses and terminations. I don't want to see or read about another death due to a boat fire which is why I persist. Hopefully some other marine experts gifted with the ability to explain better will be able to help convince.

[socaldmax]

Originally Posted by Coops

If you "experts' cannot agree, what chance does an imbecile like me have of knowing what is believable? Eeeny meeny miny mo?

Coops.

Just read Auspicious' posts. He's using the correct terminology.

I understand what transmitterdan is saying, and agree that I would increase wire size based on that, I just have a hard time with some of his terminology. Direct means it flows in one direction, alternating means it alternates direction, and those aren't "lay" definitions.

You can have AC superimposed on DC, but it's not the battery providing AC, it's the inverter inducing AC currents in the cables. If you connect an incandescent bulb to the same battery and the AC current is not present, then that's proof that the battery is not producing AC.

[socaldmax]

Originally Posted by transmitterdan

All,

I apologize for any insult that may have been perceived by my posts. It was not intended to be so. I know a lot of inverter installations that nearly burned boats and I am sure some "mysterious" electrical fires are due to improper inverter installations. This thread started due to an "almost" fire. This problem will get worse as more cruisers switch to larger battery banks and bigger inverters. Believe me that the current from the battery bank is higher than you would first think based on simple DC formulas. And lots of inverters don't have provision for proper sized wire which compounds the problem. Further, the availability of low priced high power inverters invites low cost choices for wire, fuses and terminations. I don't want to see or read about another death due to a boat fire which is why I persist. Hopefully some other marine experts gifted with the ability to explain better will be able to help convince.

I agree 100%, I am all for safety. There's no harm in using the shortest/thickest cables - even if they are way overkill based on the charts. I usually mount the inverter as close as possible to the batteries with a combined cable length of 5 ft or less. Yes, the cables are a little more expensive, but it's a one time expense that is worth the peace of mind.

Also, it's a good idea ot do a thermal survey with an IR thermometer on all connections in the system while under heavy load or charge. I've seen 3 transfer switches catch fire, and in all 3 cases, it was traceable back to an improperly tightened terminal screw. Get them as tight as you can with a screwdriver, it's amazing how fast they overheat and burn up, and it's always when you least expect it or when you need it most.

[transmitterdan]

The inverter doesn't produce current. The battery does that. The inverter creates a load resistance via high speed switching transistors that varies in a sinusoidal shape thus causing the battery to produce rectified sinusoidal current. The current into the inverter produced by the battery is not constant like it would be in a light bulb.

Here might be a way to convince you. Let's assume a 1000W resistive load such as a water heater element. At 120V RMS the current is 8.333A RMS. The peak voltage times the peak current is 2000W. At the moment when the inverter has to produce this 2000W of peak power where does it get the energy? Obviously the battery. The battery is 12V. The efficiency is 80% so the instantaneous power required from the battery is 2000/.8=2,500W. The current required to create 2,500W at 12V is about 208A which will be the peak battery current going into the inverter. The battery produces this current from it's stored energy. The inverter cannot store that much energy lest it have huge capacitor banks.

If you use simple DC formulas you might think the current is never higher than 1000/12/.8=104A. But you would be wrong. And that is why inverters sometimes cause fires.

[rwidman]

Originally Posted by Coops

If you "experts' cannot agree, what chance does an imbecile like me have of knowing what is believable? Eeeny meeny miny mo?

Coops.

Read the manufacturer's instructions and meet or exceed them. You'll be fine, assuming you understand them. If not, get professional help.

[rwidman]

Originally Posted by transmitterdan

..................... And now someone will want to argue that I am an idiot. That's ok. If you don't believe me go and measure the battery current with AC and DC meters. There will be AC current and DC current coming from the battery. Then measure the peak current and you will be very surprised by how much higher it is than the DC meter shows.

I won't argue that you're an idiot but I will argue that several of your statements and use of terminology are incorrect. I think at this point you are trying to backtrack to try and justify your previous statements rather than admitting that you misspoke.

[socaldmax]

Originally Posted by transmitterdan

The inverter doesn't produce current. The battery does that. The inverter creates a load resistance via high speed switching transistors that varies in a sinusoidal shape thus causing the battery to produce rectified sinusoidal current. The current into the inverter produced by the battery is not constant like it would be in a light bulb.

Here might be a way to convince you. Let's assume a 1000W resistive load such as a water heater element. At 120V RMS the current is 8.333A RMS. The peak voltage times the peak current is 2000W. At the moment when the inverter has to produce this 2000W of peak power where does it get the energy? Obviously the battery. The battery is 12V. The efficiency is 80% so the instantaneous power required from the battery is 2000/.8=2,500W. The current required to create 2,500W at 12V is about 208A which will be the peak battery current going into the inverter. The battery produces this current from it's stored energy. The inverter cannot store that much energy lest it have huge capacitor banks.

If you use simple DC formulas you might think the current is never higher than 1000/12/.8=104A. But you would be wrong. And that is why inverters sometimes cause fires.

The bold part is where I agree with you. Once a current is rectified, it is a direct current. However, like you pointed out, it is a varying sinusoidal current. Clamp on AC ammeters mistakenly take this varying current, with it's rising and falling magnetic fields, to be alternating, which it isn't.

In your earliest post, you stated something to the effect that the battery provided alternating current, and it simply does not. Everything else you've said about harmonics, the possibility of AC riding on DC current (ripple), proper sizing of cable, etc I agree with it all.

I simply cannot call varying amplitude DC current "AC." I've always used an o'scope to accurately measure DC voltages and any AC ripple present.

In your last example, you started off with a 1,000 w resistive load, then switched to a 2,000 watt peak load. It might be easier just to say "size the cable for the peak load of the inverter."

[irwin37]

I hate this, just getting sucked in.... If the input to the inverter is rectified than reversing the polarity should have no effect. Lets see you try it.

The input to the inverter is not rectified as this creates one more point of efficiency loss (Diodes) There are very large electrolytic capacitors across the inputs to mitigate the ripple current, is the ripple current AC, yes but the ripple is caused but the inverter switching high output loads with power being supplied by an imperfect current source.

By the way pure sign wave inverters do not switch at 60 or 50 hertz, the sign wave is created using pulse width modulation and switch at a much higher frequency. 50K to 100K cycles depending on manufacture.

This is getting really silly. For a good safe install follow the instructions and if you not sure call the manufacture.