Inverters?

I doubt that a modern white-goods heater is more than two-and-a-bit kW (~10 A). That seems a very standard Europe-wide limit point. Interested to hear of exceptions such as a 3 kW heater. The suggestions of avoiding electric heating are of course sensible. Modern washing machine don't need to do much water heating when they work at 30 degC and use so little water compared to old ones. Dishwashers of course do need some more heating, and it can't all be done by supplying hot water. Brochures stating energy use could be helpful: normally the kWh of a 'standard' program can be found, though without distinction of heat and motor load.


My main point to make is that inverters are not equal, and nor are loads, even if they are similar sorts of device. 


A really good inverter gives a fair approximation of a sinusoidal output voltage, regulates its magnitude to within a few volts of the rating over much of the rated load range, supplies up to its stated ratings for continuous and peak as long as provided with a strong enough input source. It also has reasonable safety with regard to the risks of its chassis, battery inputs and output-socket earth-terminal getting dangerous voltages between them; it might have some RCD-like function to detect leakage, or run its output as a floating system.  That sort of inverter comes at a cost: I believe it's some hundreds of pounds for something into the >2 kW category. I've only played with such things belonging to other people.  Someone doing a nice job of a boat or caravan should get such a device. Those that I've seen for sale from respectable places in Europe (not mail order from far away) have all so far been good in the above ways, except possibly the waveform if they're somewhat cheaper ones declared as not being "true sine".


Loads can also be very varied in what surges they take on start-up, and what voltage-dip from an inverter they tolerate during such surges.  If the inverter has a bad voltage waveform (cheaper types), typical modern IT-stuff like chargers and computers will run fine from all I've seen, but might reputedly have a shorter life with the higher stress on diodes and capacitors.  I'm interested to know more from Graham about sensitivity of different types of motor drive.  That comment made me wonder whether any manufacturer will 'approve' an inverter unless it's basically emulating a normal low-impedance, largely sinusoidal mains supply: I doubt the market of off-grid users is yet big enough to make it worth sounding permissive beyond "at your own risk"?  Anyway: even if the load is a "washing machine", they won't all be equal in their interaction with a given inverter.


As a little information about the other side of the spectrum, I'll mention a little about the inverters I've used, just for some sockets in my car. I don't care about the "power quality" or the happiness of the things plugged in, such as the ancient toaster by my feet in the winter, or passengers' chargers, or my laptop charger.  I just want as much power as I can get, at next to no cost. So mine is one of those highly dodgy eBay-mediated imports from the far east (Hong Kong I believe, in this case). I've imported a fair bit of eBay-mediated electrical equipment of various sorts, mainly through interest at seeing what it contains. It's seldom disappointing: i.e. there's usually something horrific, even if the basic function is actually very good. In the case of inverters (of which I've had several) I can say that the cheap ones fail on every point mentioned above. The output voltage is what's deceptively called "modified sine-wave" [see e.g. here], which basically means a square wave with another shorter one piggy-backing it. That's ok for many things, but may stress capacitors and diodes a bit more. It can make some loads noisy (audibly). The ratings are not true: my latest, from 2017, claimed 2000 W continuous and 4000 W peak.  It doesn't quite get to 1500 W for a second, even when applied in small added doses of resistive load with the input voltage held in the specified range. The output voltage drops to <200 V when coming up to its maximum output (of 1500 W: this is 1500 W *rated*, i.e. it's actually a good deal less as the voltage is below the intended). And I have to do some bonding of the chassis, output earth terminal and an input battery lead in order to be confident that a fault in the straight-onto-chassis semiconductors wouldn't cause a dangerous voltage. I mount an RCD on the side to protect the output before it emerges. The output isn't isolated, and can trip the RCD for faults on either pole. I would not want to run it in its supplied state. The supplied output socket was a universal type with big gaping holes and no connection of the earth terminal.  However, on the positive side this latest one cost £30 and free shipping, and after my modifications has served for 2½ years sitting on top of the battery in the engine compartment, partly shrouded in a freezer bag apart from the ends where the air goes through.

I doubt that a modern white-goods heater is more than two-and-a-bit kW (~10 A). That seems a very standard Europe-wide limit point. Interested to hear of exceptions such as a 3 kW heater. The suggestions of avoiding electric heating are of course sensible. Modern washing machine don't need to do much water heating when they work at 30 degC and use so little water compared to old ones. Dishwashers of course do need some more heating, and it can't all be done by supplying hot water. Brochures stating energy use could be helpful: normally the kWh of a 'standard' program can be found, though without distinction of heat and motor load.


My main point to make is that inverters are not equal, and nor are loads, even if they are similar sorts of device. 


A really good inverter gives a fair approximation of a sinusoidal output voltage, regulates its magnitude to within a few volts of the rating over much of the rated load range, supplies up to its stated ratings for continuous and peak as long as provided with a strong enough input source. It also has reasonable safety with regard to the risks of its chassis, battery inputs and output-socket earth-terminal getting dangerous voltages between them; it might have some RCD-like function to detect leakage, or run its output as a floating system.  That sort of inverter comes at a cost: I believe it's some hundreds of pounds for something into the >2 kW category. I've only played with such things belonging to other people.  Someone doing a nice job of a boat or caravan should get such a device. Those that I've seen for sale from respectable places in Europe (not mail order from far away) have all so far been good in the above ways, except possibly the waveform if they're somewhat cheaper ones declared as not being "true sine".


Loads can also be very varied in what surges they take on start-up, and what voltage-dip from an inverter they tolerate during such surges.  If the inverter has a bad voltage waveform (cheaper types), typical modern IT-stuff like chargers and computers will run fine from all I've seen, but might reputedly have a shorter life with the higher stress on diodes and capacitors.  I'm interested to know more from Graham about sensitivity of different types of motor drive.  That comment made me wonder whether any manufacturer will 'approve' an inverter unless it's basically emulating a normal low-impedance, largely sinusoidal mains supply: I doubt the market of off-grid users is yet big enough to make it worth sounding permissive beyond "at your own risk"?  Anyway: even if the load is a "washing machine", they won't all be equal in their interaction with a given inverter.


As a little information about the other side of the spectrum, I'll mention a little about the inverters I've used, just for some sockets in my car. I don't care about the "power quality" or the happiness of the things plugged in, such as the ancient toaster by my feet in the winter, or passengers' chargers, or my laptop charger.  I just want as much power as I can get, at next to no cost. So mine is one of those highly dodgy eBay-mediated imports from the far east (Hong Kong I believe, in this case). I've imported a fair bit of eBay-mediated electrical equipment of various sorts, mainly through interest at seeing what it contains. It's seldom disappointing: i.e. there's usually something horrific, even if the basic function is actually very good. In the case of inverters (of which I've had several) I can say that the cheap ones fail on every point mentioned above. The output voltage is what's deceptively called "modified sine-wave" [see e.g. here], which basically means a square wave with another shorter one piggy-backing it. That's ok for many things, but may stress capacitors and diodes a bit more. It can make some loads noisy (audibly). The ratings are not true: my latest, from 2017, claimed 2000 W continuous and 4000 W peak.  It doesn't quite get to 1500 W for a second, even when applied in small added doses of resistive load with the input voltage held in the specified range. The output voltage drops to <200 V when coming up to its maximum output (of 1500 W: this is 1500 W *rated*, i.e. it's actually a good deal less as the voltage is below the intended). And I have to do some bonding of the chassis, output earth terminal and an input battery lead in order to be confident that a fault in the straight-onto-chassis semiconductors wouldn't cause a dangerous voltage. I mount an RCD on the side to protect the output before it emerges. The output isn't isolated, and can trip the RCD for faults on either pole. I would not want to run it in its supplied state. The supplied output socket was a universal type with big gaping holes and no connection of the earth terminal.  However, on the positive side this latest one cost £30 and free shipping, and after my modifications has served for 2½ years sitting on top of the battery in the engine compartment, partly shrouded in a freezer bag apart from the ends where the air goes through.