Modified Sine Wave Converter

[Nick]

We have a TripLite modified sine wave converter, which will not allow us to charge our computers in the AC outlets. Is there something we could plug in that will convert the AC power to the pure sine wave our computers seem to want?

[booster]

Originally Posted by Nick

We have a TripLite modified sine wave converter, which will not allow us to charge our computers in the AC outlets. Is there something we could plug in that will convert the AC power to the pure sine wave our computers seem to want?

You will be much better off getting a small pure sine wave inverter that you can use for critical stuff. Just a DC connection needed and most will have a built in GFCI outlet on them. Many of us, including us, do that even with a big PSW inverter as they use less resting power than the large ones.

[Rockwood27]

I've never had an issue charging a computer from my TrippLite. Not all outlets are feed by the inverter. Try a different outlet.

[booster]

Originally Posted by Rockwood27

I've never had an issue charging a computer from my TrippLite. Not all outlets are feed by the inverter. Try a different outlet.

Some electronics will run on a MSW inverter, some won't, and IMO it probably isn't good for any of them. Quite a while ago we had an 8086 huge laptop that got wrecked by a MSW inverter in a car. We now always use PSW for all electronics.


In a Roadtrek it is true that all outlets aren't tied to the inverter, but if you are on shore power they all are running on shore power as there is a transfer switch in the inverter, so moving outlets doesn't change anything. If on inverter, the 2 or 3 inverter outlets are on inverter and the rest are dead.

[ikanode]

Something to consider is buying "car chargers" for you laptops. Meaning a charger made for your computer that plugs into a 12v cigarette lighter. These should be more efficient than the inverter route since there's one less conversion:
12 v to approx 18V vs
12V to 120V to approx 18 volt.

[Nic7320]

Almost all electronics can run on modified sine waves. The rectification inside most power supplies doesn't care what the waveshape is, as long as the peaks are high enough to top off the capacitors.

"Sensitive electronics" is mostly marketing hype to sell PSW inverters.

The appliances that *do* care about waveshape often have AC motors in them.
Add to that list anything designed decades ago with iron core transformers. The additional harmonic content of a MSW can heat up motors and transformers.

I've only found one induction cooktop that didn't like a MSW, for reasons unknown. So there are some devices that need pure sine waves.

That all said, MSW inverters are usually more efficient, since they don't have extra circuitry trying to make pretty little sine waves.

Victron seems to be an exception to this, as their Phoenix line of Pure Sine Wave inverters are probably the most efficient inverters you can buy.

[wny-pat]

Originally Posted by ikanode

Something to consider is buying "car chargers" for you laptops. Meaning a charger made for your computer that plugs into a 12v cigarette lighter. These should be more efficient than the inverter route since there's one less conversion:
12 v to approx 18V vs
12V to 120V to approx 18 volt.

That's what I do, and what I do with my CPAP breathing machine too. Just get the proper 12 vdc cord. If they operate in 12 volt, use it!!!

[GeorgeRa]

Originally Posted by Nic7320

Almost all electronics can run on modified sine waves. The rectification inside most power supplies doesn't care what the waveshape is, as long as the peaks are high enough to top off the capacitors.
"Sensitive electronics" is mostly marketing hype to sell PSW inverters.
......

It wasn’t a marketing hype when my reliable Nikon battery charger fried within a single charge on cheap inverter. Sine vs semi square discussions seems belong to the past, just like MPPT and PWM. Costs of decent units are getting closer and efficiency could be 30% higher. Higher costs, 2-3 times for sine vs modified square can be justified by some myself included. Since TI intro of sine wave chip reliability of the sine units is very high.

https://www.altestore.com/blog/2015/.../#.YEExoWhKguU

[booster]

Originally Posted by GeorgeRa

It wasn’t a marketing hype when my reliable Nikon battery charger fried within a single charge on cheap inverter. Sine vs semi square discussions seems belong to the past, just like MPPT and PWM. Costs of decent units are getting closer and efficiency could be 30% higher. Higher costs, 2-3 times for sine vs modified square can be justified by some myself included. Since TI intro of sine wave chip reliability of the sine units is very high.

https://www.altestore.com/blog/2015/.../#.YEExoWhKguU

Yep, yep, and yep. Rectifiers don't clean up dirty power unless they are specifically designed to do it, and not all electronics are made that way as it costs money to do. There is good reason why a Dometic microwave will trip out on the Tripplite even before it is making energy to cook, just from connecting it. I know because I did it. All you need to do in many cases is check the temp of the "brick" to a computer or other device when it is running on shore power vs MSW inverter and you will likely see a big difference if it is a dirty power MSW inverter. This is all energy and spikes that it has to take out, but usually can't get it all. A "car" charger from the OEM for the device will be full protected and either fix the bad power, or trip out on fault, but they are counting on the worst case being alternator dirty DC power, so not applicable to the AC input bricks.


On our Asus laptop, the AC charger that comes with it, when running off a good PSW inverter, runs almost 30* cooler than a 12DC dedicated generic charger, for instance.

[Nic7320]

Inverters using modified sine waves swing from +170 volts to zero, and then down to -170 volts, giving you something around 120 VAC RMS (the heating value, or the "root mean square" mathematical function of the waveform, and this is not a typo, 170 volts is the peak value).

A 120 VAC pure sinewave also swings from +170 volts to -170 volts, but in a sinusoidal wave. So both MSW and PSW subject your electrical components to the same electrical stresses of plus or minus 170 volts peak, so it's unlikely the waveshape had anything to do with an equipment failure.

But the design and quality of the inverter's circuitry can certainly create havoc and overvoltage stress. That's not from the basic waveshape, it's from poor design or manufacturing.

And then there's another issue that is often (or usually) overlooked; Most inverters provide two legs of power (hot and neutral) that "float," that is, the neutral is not bonded to ground. They assume that's handled elsewhere. In a home, code requires these two conductors get bonded together at the service entrance (i.e. the main fuse panel). But most DIY and even some pro inverter installers do not properly ground an inverter, or even know why the extra lug is there.

The problem with this is that high-frequency common-mode currents created by the MOSFETs inside the inverter will create a voltage difference between the floating power leads and chassis ground. And that accumulated charge builds up between neutral and ground -- and can kill any appliances that have electronic components that connect to chassis ground or electrical ground.

And this isn't just a problem with inverters. SOME portable generators don't bond neutral and ground either, because it really should be bonded elsewhere -- only ONCE, preferably at the power entrance. Otherwise your GFCIs will detect an imbalance and go nuts.

I found out my old 900 watt Sportsman generator was not bonded. But my 2 kW Generac generator was bonded inside the generator. So I carry a little pigtail adapter where neutral is tied to ground, for when I need it (mostly for the Sportsman gen.)

So where should it be bonded? RV panels don't always address this issue properly. And again, it should only be bonded once, so if you plug in at home, it's already bonded there (or it better be!). And if it's also bonded in your RV, then your GFCI will likely trip.

It just depends how and where the high frequency switching currents inside an inverter flow. They're sneaky, they're switching 500 to 2000 times faster than 60 cycles, and can find capacitive pathways that 60 cycles doesn't even know about.

Admittedly, it's a little bit complicated to understand how your RV electrical system was built. It would be good if RVs had a wiring diagram somewhere we could find it (like next to all the warning labels).

A simple ohmmeter check will tell you if your RV or inverter or generator is bonded. Turn everything off and disconnect from shore power. If the resistance between neutral pin and ground is less than a couple ohms, it's bonded (somewhere).

[booster]

I think you have the bonding misunderstood, and how the recommendations and codes actually read.


Paraphrasing the rule is simple. There is one, and only one, neutral/ground bond and it has to be at the power source, not anywhere else. A generator or inverter in an RV should be bonded when not on shore power and unbonded when on shore power, so should generators. That is the way it is supposed to be for safety, and the way the good auto bonding inverters and generators work. If you don't have such a system, there is some risk involved that is up for discussion to many people.


If you look at the waveform for most MSW inverters, that don't resemble the neat steps shown in most drawings for them. They are ragged and what most will do is only be close to stepped at very low, compared to rating, outputs. As they get nearer to rated, they turn in to square waves.



I agree with George and the link he provided, and based on past failures and temp readings, will not have a MSW running anything valuable. Why risk it when the PSW is not a risk to the equipment and also uses less energy. The cost difference is relatively small compared to one shortened device life, IMO.


We have also been moving all our DC powered adapters out and going with factory to the device AC sources and have found they use less total power or equal at worst, run cooler, and don't generate any interference. Most stuff, we run off a 120 watt Samlex PSW that has tiny parasitic and can easily be shut off besides.

[Nic7320]

"There is one, and only one, neutral/ground bond and it has to be at the power source, not anywhere else." implies a single bond at Hoover Dam power source would be sufficient. Or at least, on at the secondary winding of the main distribution transformer for a neighborhood.

But neutral to ground bonds are required at the main service entrance for each and every house. Not in any subpanels within a location, only the main service panel.

So if the house is already bonded, there is no reason to also bond at a home generator as well. But some portable generator manufacturers don't know how their equipment is going to be used, and consider their liability issues, so they bond it there anyway. And then some don't.

Now throw RVs and inverters into the mix. By its location next to a house or in an RV park, and lacking a ground rod, anything inside an RV is really a subpanel. But do RV manufacturers trust that our shore power pedestals are properly bonded upstream, or put a bonding strap in the RV just in case?

I'm itching to find out. When it stops raining, I'll go measure my RV.

Then when you add an inverter into an RV, (or -- any number of inverters that may be in one rig), bonding at each source would be a mess.

I know for a fact, Victron inverters provide an extra grounding lug tied to the NEMA 5-15 electrical safety ground, but they are not bonded to the neutral. Their output is truly "floating" source. My 2 kW AIMS inverter also does not bond, nor do the Walmart modified sinewave EverStart inverters. So based on these observations, it looks like most inverter manufacturers rely on neutral bonding to take place *somewhere* -- and not at the inverter source.

This can really be a problem when you have appliances that generate high frequency noise, such as inverter based air conditioners, or inverter based heat pumps, or solar power inverters. They have to provide a high frequency path to control their internally generated common-mode currents, since the bonding strap could be hundreds of feet away and invisible to high frequencies. If they do this well, the noise is suppressed. Otherwise, it will creare all sorts of problems with GFCIs.

I would question how your MSW inverter measurements were taken. If you didn't take a differential voltage measurement with an oscilloscope using two matched and properly compensated scope probes, those spikes you saw could be from ground bounce that exists where the probe ground reference point was, and not actually part of the output waveform. Or if you did do a differential measurement and your probes didn't have the same cable length, what you're seeing is the skew between the two probe signals superimposed on top of the waveform. Even placing a 10:1 capacitive divider probe in the wrong place can pick up erroneous signals. Long and short of it, is that it can be tricky to measure this accurately.

And while all this is very technical and useful to understand, a previous point is certainly true: Very good pure sinewave inverters are now available at reasonable cost. Even some Victron inverters can be had for about $100.00
(plus $45 or so for the bluetooth adapter, if you want it).

[booster]

Your stretching here, I think. Think about it as a van is just like any other remote building except mobile so it can't have the required ground rod like an out building. The inverter or generator has to float compared to earth, but not to itself or the ground which in an RV is normally the chassis. The bond has to be at the power source, be it generator, inverter, or power post because if the bond is in the van you can have the ground as a current carrying wire and if it goes open the chassis will go hot.


Of course when on shore power the post has to have a continuous to earth ground to be bonded and that is why you use a power checker when you plug in. It also checks for polarity which if reversed can make the chassis hot if neutral and ground are bonded because neutral would then be hot. The risk actually will get higher if you have an internal bond inside the van.



One and only one bond, at the power source (defined as the supply to any subpanel if you chose to) and that bond has to be at the source to the subpanel, not anywhere else in the circuit.



The "safety ground" lug on an inverter, generator, or outlets generally all go to the chassis, which on shore power will go to earth at the power pole, but when on generator or inverter the grounded chassis goes back to the bonded neutral at the genny or inverter to complete the circuit and keep the ground, and chassis, from going hot.


If you have a permanently wired and bonded inverter, or one that is left connected when on shore power, you have two bonds are can get a hot chassis because of it. Of course something has to go wrong for that to happen, like an open wire, but those are common in things that bounce down the road.


You can believe whatever you want, but that is what codes and requirements for mobile power say.


Same is true of MSW inverters, if you choose to use them on sensitive electronics have at it, but many of choose not to for a lot of reasons and experiences.

[Nic7320]

You're correct saying an RV has its own self-contained (and... earth-isolated) power system, and SHOULD have one neutral to chassis ground bond.

But then when you plug into shore power, you now have two bonds; a neutral to chassis ground bond inside the vehicle, and a shore power neutral to electrical ground bond at the distribution system. Let's ignore the earth rod for now as its there primarily for lighning protection.

That extra neutral bond path in the vehicle will be a problem. Some of the neutral current will flow through the vehicle's bond into the ground wire, and back to the shore power source neutral through its ground bond. As you know, the ground wire should not carry any current -- unless there's a serious electrical fault. Its job is primarily to protect us, not to carry any normal AC current.

So how can two topologies do two jobs with one configuration? It's no wonder people have trouble tripping GFCIs.

What is needed is a shore power inlet that disconnects the vehicle's internal neutral-ground bond connection when it detects shore power. "Power Defender" or EMS type units do test and disconnect the whole RV for any faults coming from the pedestal, but they cannot lift a neutral to ground bond inside the vehicle.

It seems likely many RVs have this issue. My guess is many simply do not bond and float their neutral, and occassionally damage some equipment.

[avanti]

Originally Posted by Nic7320

But then when you plug into shore power, you now have two bonds; a neutral to chassis ground bond inside the vehicle, and a shore power neutral to electrical ground bond at the distribution system.

It seems likely many RVs have this issue. My guess is many simply do not bond and float their neutral, and occassionally damage some equipment.

No! This is just wrong. Having either zero or more than one bonds is not permitted and must never be done.

Gensets and many larger inverters manage their bonding automatically. If you have an on-board source that doesn't do so, you need to use a transfer switch that is capable of switching the bond. There MUST always be exactly one bond, and no reputable upfitter would do otherwise.

[GeorgeRa]

In my van I have two inverters, the primary 1000W Magnum connected to a few outlets shared with shore with automatic bond switching and the 300 Morningstar with 2 separate outlets. The 300W unit is permanently bonded. These two circuits are fully independent.

[Nic7320]

Avanti, that makes perfect sense -- switch the bond OFF when connected to shore power, and ON when the generator is on with the generator's ATS.

I'll confirm that when i measure my shore power neutral to ground resistance.

But that means we should also do the same with our AC inverters. They need an ATS with bond switching. Is anyone doing this? (other than GeorgeRa?)

My homebrew Tesla battery box bonds neutral to ground inside the box, and then goes through a manual transfer switch which isolates the neutral and the bond when switched over to shore power. I had to think about that, so this was a worthwhile discussion. Thanks Avanti!

And I might have to step back on my MSW claim.

MSW is basically a 170 volt power supply feeding into an H-bridge that switches the legs back and forth. It should go like this: 0 volts, + 170 volts, 0 volts, -170 volts, and then keep repeating the sequence for more cycles. It shouldn't go above 170 volts peak, or below -170.

Any well designed MSW converter should also have "catch" diodes to keep the outputs within the power rails, but some real cheap units may not. And if they go to a high impedance "off state" for any appreciable amount of time between the 0, 170, 0 -170 volt pulses, the voltage can swing as far as it wants during this time -- until it encounters something to stop it. Thus could be really bad with inductive loads.

So that means there are clean PSW inverters, some good MSW inverters, and some real crappy MSW innverters. Unless you have an oscilloscope, you'll never know the difference between the last two.

Maybe I'll take a cheapo cigarette lighter inverter apart and look at the circuit.

[booster]

Originally Posted by Nic7320

Avanti, that makes perfect sense -- switch the bond OFF when connected to shore power, and ON when the generator is on with the generator's ATS.

But that means we should also do the same with our AC inverters. They need an ATS with bond switching. Is anyone doing this? (other than GeorgeRa?)

That is exactly what has been being said all along, and just what Avanti said. Internal to the van bondshave to be turned on and off or the power source totally disconnected/isolated like George does.


It was also mentioned in earlier posts that inverters and generators should have auto switching bonding built in to them to be safe. Lots of inverters and generators do, as was mentioned also. There a several discussions on bonding on the forum that include circuit drawings and at least one that shows how the auto bonding switching is wired inside a Magnum inverter/charger.



We have an autobonding Magnum 2000 watt PSW inverter/charger, and we also have two small 120 watt Samlex PSW inverters on standalone circuits, one to run the TV/entertainment stuff and one set up a device charging station. Both of the small inverters are (now) permanently bonded because they are isolated from the van wiring and ground, which is what George has also, I think. On them, about all the bonding does is protect you from a short in a device putting 110vAC on the case or whatever else that may go hot with short, if it has a grounded cord on it.


One thing I have been meaning to test, but haven't, is what happens if you have an isolated inverter and the hot somehow shorts to the van body when not on shore power so no earth ground? Since the inverter floats vs earth anyway, what voltage would be on the body of the van? Would the inverter just quit working with on a single leg connected? I have no idea what would happen or if it would be a hazard, but it is a thing that could happen, I think. I have seen several volts on the chassis of the van vs earth ground when I have measured it, when the big inverter is powering the whole van and it is using the chassis ground with it's bonded neutral.


Here is the Magnum bonding information from their manual

[Nic7320]

The main takeaway here is that we were talking about the same thing, but under different sets of circumstances. You have a well designed inverter with an ATS. Not all inverters or generators have bonding or an ATS.

This bonding issue is a serious problem.

My Forest River Solera 24S had an inverter in a past life, installed by *who knows who* -- maybe a clueless dealer or the previous owner. It had no ATS. But what they did was isolate a circuit string and power only the TV and a couple outlets. It had no electrical ground to chassis ground connection, nor any neutral-ground bond on that circuit at all. With proper isolation, a floating circuit can work, but it's not the safest way to do it. At least it wasn't a double bond.

But I did mention a TV. TVs introduce an antenna ground that is sometimes tied to chassis ground. Now it's no longer a well-isolated floating system. Both legs of AC power are floating, but now the ground is... somewhere uncontrolled, subject to capacitive coupling and common mode current injection. And the TV probably doesn't like that (unless they isolate all of its inputs; antenna, HDMI, AV inputs, etc. which is VERY unlikely).

In any case, floating grounds and unbonded neutrals leave all sorts of equipment at risk. Yet the market is flooded with inverters that do exactly that.

Both of my Victron Phoenix inverters (1200 W and 375 W) do not bond or have ATS. But their Multiplus line have an ATS.
The 2000 W AIMs inverter does not bond or have ATS.
Walmart inverters sold under the EverStart brand have no bonding or ATS.
One Generac 2000 watt generator I bought has bonding, but no ATS.
One Sportsman generator I bought has no bonding and no ATS.
The 3600 watt Onan in this rig has bonding and an ATS.

So it's a mixed bag. And only a few understand how to do it correctly, so a lot of equipment gets damaged in the process and maybe a few people.

If your isolated inverter "hot" wire shorts to case, it's no different than a bird on a wire; it doesn't have a return path if it's not connected and powering anything. But like I noted with my TV antenna ground, when loads are plugged in, it *might* find a return path and now your case is hot, relative to chassis ground. It could shock someone.

This said, there are thousands (or more likely millions?) of inverters out there without chassis grounds or neutral bonds. They rely on whatever insulation and isolation is in their circuitry or connected loads. It's a risk and now we have better practices. Remember, prior to the mid 1950s, even houses didn't have safety grounds.

[booster]

Bottom line is if you have an inverter that is not bonded, you need to bond it if it is on an independent circuit. If it is tied into the electrical system of the van you need to bond it only when it is inverting and make sure it is not bonded when you are on shore power.


The option that has been used by others recently with unbonded inverters is to add and automatic transfer switch between inverter output and shore power input and that switches at least the hot and neutral of both. It would be put after the shorepower/generator ATS if you have one. That will automatically disconnect the inverter when on shore power, even though it could still be generating 110v to the transfer switch. The ground, if not disconnected will be connected to earth ground at that point at the power post so can't carry any voltage on the chassis, and the inverter neutral and hot aren't connected to anything. I may be possible that really poor inverters that use +60 and -60 volts might have trouble with this though I think, but having the put 60v on the chassis would also be a problem with them.


Personally, I prefer the separate neutral bonding relay like is shown in the Magnum drawing. They have an also internal automatic transfer switch to take care of the hot and neutral switching between shore and and inverter. That is the most foolproof setup.



Again personally, I would not use and unbonded inverter wired into the system as having the bond makes them safer. If it is a standalone a permanent bond is fine as long as it is not one a independent circuit or has it own outlets that you plug in to.