Safety hazard w/ Growatt 5000ES (European version) when connected to American grid

[robby]

Man, I gave up. Do what ever you want.

One Week Later.....

 

[Tony Scott]

Man, I gave up. Do what ever you want.

That is a legitimate question, all these experience people and no one found a problem.

 

[Hedges]

That is a legitimate question, all these experience people and no one found a problem.

For things like house wiring, and cars, we consider not just how it behaves when things go right, but also when they go wrong.

Your car may have features like air bags that deploy in a crash, and a fuel pump that shuts off in a crash.
The car will run just fine if those features aren't functional.

Similarly, house wiring is designed with things like GFCI, AFCI, and grounding.
You could probably use appliances for years without ever observing if those systems were functional.

These things are meant to help keep you alive and uninjured when something goes wrong.
Some of us have knowledge and experience, and of course people who developed NEC do, to foresee hazards and how to mitigate them.
NEC and NFPA grew out of the insurance industry, because they had loss data indicating what often led to fires.

 

[robby]

My guess is that contrary to all of Signature Solars talk of a solution it's safe to assume that at this point they have already talked to one or more electricians in the USA and have not gotten an answer that fixes the problem.
They could ask Growatt to redesign the board but Growatt sales in the USA probably account for 1% of their total sales. Would it be worth it to them to put in a relay and other hardware in for such a small market.

I suspect that SS is busily looking for a true Isolation Transformer out of China for cheap money. That is going to be the only solution that is under their control and can solve the current problem. The new problem is that they are not really going to be able find one for a cheap price.

 

[Hedges]

Hard to compete with a high-frequency switch-mode inverter, which is cheap because it left out the transformer
One solution is 2x 120V high-frequency inverters, wired 2s. I think some brands offer that in a single box.

 

[robby]

Hard to compete with a high-frequency switch-mode inverter, which is cheap because it left out the transformer
One solution is 2x 120V high-frequency inverters, wired 2s. I think some brands offer that in a single box.

I think the Growatt 3000 does that using two Inverters and a data cable. Not sure why the data cable is needed since two Honda Inverter Generators will just phase shift until they match.

 

[Zwy]

I think the Growatt 3000 does that using two Inverters and a data cable. Not sure why the data cable is needed since two Honda Inverter Generators will just shift the phase until they match.

I looked but couldn't find where Honda Inverter gennys do this for 240V split phase. There is the parallel kit to parallel two 120V inverter gennys to provide twice the 120V watts, but I haven't seen anything for 240V split phase.

 

[robby]

I looked but couldn't find where Honda Inverter gennys do this for 240V split phase. There is the parallel kit to parallel two 120V inverter gennys to provide twice the 120V watts, but I haven't seen anything for 240V split phase.

That is what I was saying, that you can combine two 3000W Hondas and get 6000W at 120V.
I understand why the Growatts use a data cable for 240V but they also have a combined 6Kw 120V mode that still uses a data cable. Maybe it's because they are also getting power from the Grid.
https://www.growatt.es/upload/file/contents/2021/09/614478a7787e9.pdf

 

[Zwy]

That is what I was saying, that you can combine two 3000W Hondas and get 6000W at 120V.
I understand why the Growatts use a data cable for 240V but they also have a combined 6Kw 120V mode that still uses a data cable. Maybe it's because they are also getting power from the Grid.
https://www.growatt.es/upload/file/contents/2021/09/614478a7787e9.pdf

Ok, you are mixing things up here. You first stated the Honda genny's are providing phase shift which they aren't, they are merely paralleling the input on 120V single phase doubling the watts.

The data cable between 2 GW or MPP units is for 240V split phase operation where the phases oppose each other, just like the power that comes off the grid. In order to do that, the units would need to communicate with each other to create the phase shift. Many of these units can also output 3 phase, it's just a matter of communication between the units and phasing at the correct offset.

With the SPF3000, which I own one installed in my truck camper, the primary reason for the 2 units or more needing a parallel cable for 120V single phase operation shared between 2 units is load balancing, sine wave synchronization would be secondary if an output filter isn't present. I haven't looked at the internals for an output filter and I doubt there is one, much easier to just communicate between 2 units and synch the sine wave. The ports and logic board are already there for 240V split phase and 3 phase operation, it's just a matter of programming. This allows greater flexibility than the Honda genny. Honda inverter gennys are doing this crudely by running a common cable in the parallel kit between the units creating a common bus and using an output filter for sine wave synchronization.

Now, why someone would buy multiple SPF3000's and want to use the units for only 120V single phase operation is another question. Wouldn't really make sense considering there are much better choices out there. I would suppose someone might if they already had one unit and wanted to expand the available watts but the output wire size would need to be increased along with breakers. Wouldn't really make financial sense long term.

 

[kjswiley]

Is David Poz on this forum, we should have him verify his screw was removed on the box he opened. (Or anyone who has already opened the box voiding the warranty). I have a hard time blindly accepting that Chinese manufacturers have removed the screw for all units for 1 manufacturer rather than for all units in the region, how do they keep track of this? I believe it was done, but that is a lot of faith to put in hoping they didn't miss your unit, or it was somehow a unit destined for Europe which was redirected, etc. This makes me want to open my boxes and physically verify each unit for peace of mind, which of course would void any warranty.

edit: is verifying that there is no connectivity between ground and the marked Neutral (L2 for our use), either input or output as Struct mentioned sufficient instead of opening the boxes?

 

[robby]

Ok, you are mixing things up here. You first stated the Honda genny's are providing phase shift which they aren't, they are merely paralleling the input on 120V single phase doubling the watts.

Lets not derail the thread. When two hondas are running separately the sinewaves cannot line up perfectly, They must be aligning the phases.

 

[robby]

Is David Poz on this forum, we should have him verify his screw was removed on the box he opened. (Or anyone who has already opened the box voiding the warranty). I have a hard time blindly accepting that Chinese manufacturers have removed the screw for all units for 1 manufacturer rather than for all units in the region, how do they keep track of this? I believe it was done, but that is a lot of faith to put in hoping they didn't miss your unit, or it was somehow a unit destined for Europe which was redirected, etc. This makes me want to open my boxes and physically verify each unit for peace of mind, which of course would void any warranty.

edit: is verifying that there is no connectivity between ground and the marked Neutral (L2 for our use), either input or output as Struct mentioned sufficient instead of opening the boxes?

Poz knows stuff he is not saying. I think he had several of these issues and never said anything.
I was wondering why he keeps the Grid separate from his Growatts. He evidently just uses a transfer switch to move his house from the Inverters to putting it on the grid. BTW Whatever happened to his third Growatt? I guess I missed that video.

 

[Cheap 4-life]

Poz knows stuff he is not saying. I think he had several of these issues and never said anything.
I was wondering why he keeps the Grid separate from his Growatts. He evidently just uses a transfer switch to move his house from the Inverters to putting it on the grid. BTW Whatever happened to his third Growatt? I guess I missed that video.

He said he gave it to a friend. Actually made a video that showed him mounting it on a piece of plywood to transport

 

[MrM1]

Poz knows stuff he is not saying.

He POZitively does (couldn't resist)

 

[Zwy]

Lets not derail the thread. When two hondas are running separately the sinewaves cannot line up perfectly, They must be aligning the phases.

Well, you brought up the subject and made it sound like the parallel Hondas are providing split phase which I found hard to believe, this was for information for others, just so they don't go down that rabbit hole.

As for the sine wave synchronization, it is done with an output filter, nothing fancy about it.

 

[Shale MacGregor]

Oof, bought 3 of these and a Victron Inverter for off-grid. As far as European these do 50hz or 60hz, and I was originally sold on that by Victron Multiplus system that came as 240 50hz by default, but was offered for NA using the inverter and switching to 60hz mode. This was my first solar build so I guess I will need to rethink things.

 

[robby]

Well, you brought up the subject and made it sound like the parallel Hondas are providing split phase which I found hard to believe, this was for information for others, just so they don't go down that rabbit hole.

As for the sine wave synchronization, it is done with an output filter, nothing fancy about it.

I doubt it's a filter. I would think that the second Inverter samples the first ones output and modifies it into a usable signal that takes over driver control of the firing of the FET's on the second Inverter. Unplug the two wires and it goes back to using it's own internal clock.

 

[kjswiley]

When this 1st came out I contacted several of the influencers on Youtube as I am sure many others did, and the main one we all know well has approved comments since then, none of which include any info on this topic. He is blatantly ignoring it so as not to lose likes/subscriptions/$$. And of course, Youtube will keep pointing people to him and more units will be sold to people completely ignorant to any potential safety issues.

I appreciate Will and Ian bringing these potential issues out into the open, as I am sure we all do. There was not a whole lot of info out on these when I made my purchase. I felt like I had done my research and made decisions based on these influencers and the few other relevant sites I could find, so it is a shame they will just keep pushing these knowingly.

 

[Rustic]

I started to watch Will’s video but it was pulled just as I started watching. Ian has posted a detailed eye opener part 2 video today.

 

[Cornwallav8r]

Has anybody posted pic of the screws internally bonding the neutral to ground inside these SPF5000ES models? 9 pages of thread comments and found nothing, LOL. I really need it presently. I found one screw, poking around before I disassemble, but before I go tearing into these to delete them and insulate the PCB, it'd be nice to know where others have removed them in case there are multiple. Regarding insulating the PCB, wow people get all technical. Just put a good insulator between standoff and pcb and be done.

Despite getting all my AC wiring install and new panels installed correctly without cross-shorting L1 and L2 on any of my (4) new units, I simply didn't think to check for that internal neutral-ground bond. It happens. It's not mentioned anywhere. It didn't dawn on me, mostly because there's no reason this should have been done internally, in the first place by the manufacturer, it should be an external removeable jumper wire to make it US or EU compatible. My fault for not thinking of it, but I ended up with blown up inverters as a result, when parallelled. Why Growatt had no idea what I was experiencing, is beyond me. I was in communication with (3) separate Growatt engineers over this past month, one in UAE, one named Francis in California, and another in China. Then again, the same engineer in California told me my warranty was invalid because the unit's display showed 175% load (overload) after the failure despite it being obvious there was no load connected at all. He said I was overloading it. Mind you, the unit as designed wouldn't even run for a second at that loading, and he somehow didn't know this.

Some thoughts, since I've been through the mill already on this:

Opening your unit won't void the warranty. Why? Growatt engineers ( at least in UAE, China and California) insist the customer open the units themselves, they have no intention apparently to ever replace the whole unit under warranty. Because they're too cheap to replace the whole faulty unit. At least not to my experience. They insisted I open the unit and diagnose which boards were faulty. Then simply told me they've never seen this problem before, it's unusual, and all my fault. Not sure how a non-technical person would have even handled doing this.

I've obtained one for one replacement units for those blown up here due to the situation. With no help from Growatt, mind you. Their engineers had no idea of this thread's ground-neutral situation, and simply blamed me for blowing them up. No warranty-case closed. So be warned, Growatt support is no better than any of the China fly-by-night clone manufacturers, so save your money, these things are basically disposable. They looked for any reasons to deny warranty, then insisted on knowing the serial numbers, I'm imagining to maintain record of the rejected units' warranty in their system. Unless local selling guys like Signature Solar want to replace them out of the goodness of their hearts. China manufacturers are not like those in the US, or Europe or Japan for that matter, they're generally heartless and fraudulent. Back when I put myself through university repairing electronics, we could most always get wiring diagrams/repair manuals/schematics from the Japanese and German manufacturers. In my best Changlish: "China no help". "You steal technorogy".

Those remarking we're mistaken buying these for the US are incorrect. Yes, there are US models with transformers already inside, but when I priced these, I needed (4) and with shipping all that weight and the price difference, it was literally thousands of dollars saved. And I already own a couple good autotransformers. There was a newer US model that looked perfect, but it was unavailable at the time. If not for this silly internal neutral bond issue, all would be well and this thread wouldn't exist in the first place. These work fine with 50Hz or 60Hz, 230V or 240V. No reason not to use them, except maybe for UL-listing, which most China factories fake anyway, it's useless. Just buy from a reputable manufacturer, should be fine without a UL sticker.

Regarding grounding, yes, don't drive another ground rod just for these inverters, separate from the main building ground. Or at least if you do, interconnect them. As I stated before, you'll get current between them possibly causing RF interference, and added lighting risk, because there will always be some tiny ground potential between them. Remember, per NEC we're supposed to have 2 driven ground rods a minimum 6 feet apart (code changed to require this a few years back) , so you could kill 2 birds with one stone here adding the second. But interconnect them with a #6 copper conductor. Or just use one rod(other NEC option) , and get a testing device to ensure below 25 ohms to ground, I believe it is. Most don't bother with the latter.

I hope this info helps others. I'm not interested in arguments over this stuff. If I'm incorrect in any of the above, you're welcome to correct me. If somebody could post the pics of the PCB for others to handle themselves, it would be most appreciated.

Edited: Found the pics, thanks. I removed the screws, looks to be ok, there's no copper or via exposed at the bottom. The first inverter, I disassembled, and removed the standoff, it comes off easily with a vice grip, bend a few times, it then falls out the bottom. Doesn't look necessary now though, at least for the models I received. There is no copper at the bottom of the hole or via present.

One more note. I keep seeing smartass comments about these being inappropriate for our use, and we should be using the split phase version. But the neutral issue that most of these threads discuss, is just the same regardless. Connecting the split phase inverter's supplied neutral to ground without eliminating it at the service, versus our supplied separate autotransformer, it's the same issue. The only difference I see, is some installations will have a neutral relay and some will have temp protection for the transformer. The first isn't needed if you design your setup correctly, nor in my opinion is the second, if you trial test your installation for a week or two, see how much imbalance you can possible have at any one time, and move a load or two to the other phase. Just my thoughts.

Bill