diy solar

diy solar

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

Tony Scott said:
Life was better for before this topic on the forum, too many people throwing wrench’s and don’t even have the product.
Click to expand...

Right up until someone suffers a serious shock.
Your local AHJ and inspector wouldn't have the product either; think they should weigh in on the design?

We, just looking at the information, have seen possible concerns. With a schematic from Signature showing connections including utility feed, inverter, autotransformer, loads panel, I think I spotted a safety issue.

If you present the schematic of what you're having installed for review at your permit office, that is an opportunity for their review. Some designs might require signature of a Professional Engineer licensed in your state, which is another opportunity for problems to be detected, but this isn't necessarily required.

If if an appropriate professional reviews and signs plans, that isn't a guarantee no hazards remain. Stacks of drawings get dumped on a desk, cursory review performed, stamp and signature applied.

You also get to decide how carefully to have it reviewed. Do you want to accept the first "All Clear!" you hear? Or do you want to chase down all reports of potential hazards, even though you yourself don't know whether they are credible, and ask someone who is suitably qualified to consider the specific concern? (How serious is the potential problem; what are the consequences if it turns out to be real?)
 
Quattrohead said:
Just to be 100% crystal clear, the pictures I posted and refer to are a knock off unit from Vevor that has failed, not a Growatt or MPP unit.
However, as my knockoff and most others seem to be a copy of the MPP or Growatt, I suspect the circuitry could be similar in this area.
Click to expand...
It we are not talking about the actual unit, then anything you say doesn’t apply to Growatt. All knock off do not follow the same manufacturing standards
 
Last edited:
So I guess signature solar is refusing to participate in this thread.
 
Tony Scott said:
It we are not talking about the actual unit, then anything you say doesn’t apply to Growatt. All knock off do not follow the same manufacturing standards
Click to expand...
I suggested the Growatt and MPP seem similar because my knockoff may be a copy, and if so I am showing what I see on the circuit board.
Now if SS or Ian want to show us what their circuit boards look like in the same area, we can come to a collective solution.
That solution may be a safe way to use them or a do not use in the USA except if 100% off grid or stop sales totally and ship remaining product back to whence it came.
 
Quattrohead said:
I suggested the Growatt and MPP seem similar because my knockoff may be a copy, and if so I am showing what I see on the circuit board.
Now if SS or Ian want to show us what their circuit boards look like in the same area, we can come to a collective solution.
That solution may be a safe way to use them or a do not use in the USA except if 100% off grid or stop sales totally and ship remaining product back to whence it came.
Click to expand...
So this whole discussion has all been
Bud Martin said:
Did they just remove the screw as shown on the video only? for true safety, they need to install insulator (that has high Voltage break down) between the chassis and the PCB where ground point is since the gap between the PCB and the chassis is very very small as mention on the video, some flexing on the PCB can easily make contact with the chassis.
Click to expand...
Why would a board flex when no one is touching it? It is this kind of false speculation that has created this fear mongering. The guy in the video never shows a unit that was damaged, he states he has received phone calls about units "Blowing up", but nothing about how they were wired or used. If you used a European then the blame lies with you. The schematic from Growatt is correct and you will need to sink a ground rod just like you had to with your panel's.
 
Tony Scott said:
So this whole discussion has all been

Why would a board flex when no one is touching it? It is this kind of false speculation that has created this fear mongering. The guy in the video never shows a unit that was damaged, he states he has received phone calls about units "Blowing up", but nothing about how they were wired or used. If you used a European then the blame lies with you. The schematic from Growatt is correct and you will need to sink a ground rod just like you had to with your panel's.
Click to expand...

This entire thread is about using a European model inverter (single phase 220V 50 Hz) set for 240V 60 Hz, and wiring for use in US 120/240V split-phase operation. In this case, factory removed a screw to let what had been "Neutral" float ungrounded, so the 240V output could be centered around a new grounded neutral.

We're not sure the physical configuration of "Neutral" on the PCB vs. grounded case. It ought to have "Creepage and Clearance" in compliance with UL requirements. If I'm reading table 2K correctly, that needs to have 2.6mm airgap.



The reason for testing the PCB while touching it is to see if the gap is only produced by the PCB not being firmly clamped to case. If a gentle push causes PCB to make contact, that indicates it was maybe 0.1mm away or at least well under 1.0mm. The test is performed with low voltage (and ohm meter), but line voltage or typical transients could arc across. A 12" wide PCB held by screws on the edge, when it expands and contracts at different rate from aluminum casing, will bow, possibly making contact. For "Neutral" and ground to be reliably isolated, they need significant distance between them in X, Y, or Z direction. Because a screw was used to join them, it appears they are less than 2.6mm apart.

These units are guilty until proven innocent. Fearmongering is raising valid concerns, which can be set aside when examination or test shows they are unwarranted. The "blown up" units were ones where this screw was still present. If the PCB can make contact in one with screw removed that is the likely outcome. It is possibly removal of the screw provides decent isolation (perhaps not as large a gap as UL requires), in which case it could function satisfactorily. But due to how PCB holes and vias are usually designed, I don't think that would be the case unless designer planned for being able to isolate it.

Schematic, ground bar, and panels - difference is, although we were required to sink a ground bar, we were also required to connect a ground wire from panels back to SCC/inverter. And wire from there back to ground rod of utility entrance.
 
Quattrohead said:
I suggested the Growatt and MPP seem similar because my knockoff may be a copy, and if so I am showing what I see on the circuit board.
Now if SS or Ian want to show us what their circuit boards look like in the same area, we can come to a collective solution.
That solution may be a safe way to use them or a do not use in the USA except if 100% off grid or stop sales totally and ship remaining product back to whence it came.
Click to expand...
Well we are still waiting for them to supply something that shows a safe way of hooking it up.
I do not think it's possible without doing changes at the factory
 
bgflyguy said:
I think there is too much reliance on regulations without understanding the fundamentals. Regulations aren't physics, sometimes they aren't actually the best choice.
Click to expand...
Most Regulations are created in response to an event that happened in the past. Many of these past events are older than us, so we don't even know why it is in place. They are typically there for a good reason and to prevent history from repeating itself.
 
Hedges said:
This entire thread is about using a European model inverter (single phase 220V 50 Hz) set for 240V 60 Hz, and wiring for use in US 120/240V split-phase operation. In this case, factory removed a screw to let what had been "Neutral" float ungrounded, so the 240V output could be centered around a new grounded neutral.

We're not sure the physical configuration of "Neutral" on the PCB vs. grounded case. It ought to have "Creepage and Clearance" in compliance with UL requirements. If I'm reading table 2K correctly, that needs to have 2.6mm airgap.



The reason for testing the PCB while touching it is to see if the gap is only produced by the PCB not being firmly clamped to case. If a gentle push causes PCB to make contact, that indicates it was maybe 0.1mm away or at least well under 1.0mm. The test is performed with low voltage (and ohm meter), but line voltage or typical transients could arc across. A 12" wide PCB held by screws on the edge, when it expands and contracts at different rate from aluminum casing, will bow, possibly making contact. For "Neutral" and ground to be reliably isolated, they need significant distance between them in X, Y, or Z direction. Because a screw was used to join them, it appears they are less than 2.6mm apart.

These units are guilty until proven innocent. Fearmongering is raising valid concerns, which can be set aside when examination or test shows they are unwarranted. The "blown up" units were ones where this screw was still present. If the PCB can make contact in one with screw removed that is the likely outcome. It is possibly removal of the screw provides decent isolation (perhaps not as large a gap as UL requires), in which case it could function satisfactorily. But due to how PCB holes and vias are usually designed, I don't think that would be the case unless designer planned for being able to isolate it.

Schematic, ground bar, and panels - difference is, although we were required to sink a ground bar, we were also required to connect a ground wire from panels back to SCC/inverter. And wire from there back to ground rod of utility entrance.
Click to expand...
Yep. The chassis, components, heatsink, copper traces, PCB substrate will expand/contract with the change in temperature. The PEMS standoff I spec in has +/-10mill tolerance so the biggest gap will be 20mil between the top of the standoff and the bottom of the PCB.
The schematic has two ground points, so they are relying on the the Earth ground resistance of the ground on the bottom right back to the utility ground to be low enough that the fault current will trip the utility breaker, I would not bet my life on that.
 
Last edited:
Tony Scott said:
So this whole discussion has all been

Why would a board flex when no one is touching it? It is this kind of false speculation that has created this fear mongering. The guy in the video never shows a unit that was damaged, he states he has received phone calls about units "Blowing up", but nothing about how they were wired or used. If you used a European then the blame lies with you. The schematic from Growatt is correct and you will need to sink a ground rod just like you had to with your panel's.
Click to expand...
Nope it is not, and I guess you still believe that the chassis does not to be grounded either according to what SS tells you, right.
Ask SS to trace out fault current path when L1 or L2 from the unity makes contact with second ground at bottom right of the diagram, will it trip the L1 or L2 utility breaker. Right now SS is hoping that the resistance between the two ground points are low enough that the fault current will trip the breaker, as mention by other, the two grounds must be bonded together to insure the fault current will trip the breaker.
When SS tell you that the chassis does need to be grounded, it is obvious to me that safety is not their concern.
BTW, read reply in post #83 from SS:
"see our diagram, this is not our position"
So it looks like SS will give you different answer depends on whom you talk to when it comes to safety, just great! So you got bad information when you talk to SS.
diysolarforum.com

Safe Grid Use of the 5000ES and transformer

Full stop - Big issue here. Adding a second grounding electrode for the separately derived system, not bonded to the buildings ground system and calling it good is not acceptable. Per 2020 NEC 250.50 all grounding electrodes at a building or structure shall be bonded together. This code...
diysolarforum.com diysolarforum.com
 
Last edited:
Bud Martin said:
Yep. The chassis, components, heatsink, copper traces, PCB substrate will expand/contract with the change in temperature. The PEMS standoff I spec in has +/-10mill tolerance so the biggest gap will be 20mil between the top of the standoff and the bottom of the PCB.
The schematic has two ground points, so they are relying on the the Earth ground resistance of the ground on the bottom right back to the utility ground to be low enough that the fault current will trip the utility breaker, I would not bet my life on that.
Click to expand...
Correct, Mike Holt goes into detail about ground rods in his many videos. There are too many variables with ground rods, type of soil, moisture of soil, length of rod.

Electricity will always want to return to the source. In the example of these inverters without a bonding ground at the source (main panel/ meter) and running GEC to inverter units, they are asking for trouble.
 
Last edited:
Bud Martin said:
Nope it is not, and I guess you still believe that the chassis does not to be grounded either according to what SS tells you, right.
Ask SS to trace out fault current path when L1 or L2 from the unity makes contact with second ground at bottom right of the diagram, will it trip the L1 or L2 utility breaker. Right now SS is hoping that the resistance between the two ground points are low enough that the fault current will trip the breaker, as mention by other, the two grounds must be bonded together to insure the fault current will trip the breaker.
When SS tell you that the chassis does need to be grounded, it is obvious to me that safety is not their concern.
BTW, read reply in post #83 from SS:
"see our diagram, this is not our position"
So it looks like SS will give you different answer depends on whom you talk to when it comes to safety, just great! So you got bad information when you talk to SS.
diysolarforum.com

Safe Grid Use of the 5000ES and transformer

Full stop - Big issue here. Adding a second grounding electrode for the separately derived system, not bonded to the buildings ground system and calling it good is not acceptable. Per 2020 NEC 250.50 all grounding electrodes at a building or structure shall be bonded together. This code...
diysolarforum.com diysolarforum.com
Click to expand...
How about this, crack the case open and place a piece of electrical tape or shrink tape between the board and the contact point or use a plastic screw to hold the board in place.

They have sold 5k of these and have not had a problem. Look again at that video and see how thick that board is. No one has measured the heat on that board nor has anyone measured the gap.
 
Bud Martin said:
Nope it is not, and I guess you still believe that the chassis does not to be grounded either according to what SS tells you, right.
Ask SS to trace out fault current path when L1 or L2 from the unity makes contact with second ground at bottom right of the diagram, will it trip the L1 or L2 utility breaker. Right now SS is hoping that the resistance between the two ground points are low enough that the fault current will trip the breaker, as mention by other, the two grounds must be bonded together to insure the fault current will trip the breaker.
When SS tell you that the chassis does need to be grounded, it is obvious to me that safety is not their concern.
BTW, read reply in post #83 from SS:
"see our diagram, this is not our position"
So it looks like SS will give you different answer depends on whom you talk to when it comes to safety, just great! So you got bad information when you talk to SS.
diysolarforum.com

Safe Grid Use of the 5000ES and transformer

Full stop - Big issue here. Adding a second grounding electrode for the separately derived system, not bonded to the buildings ground system and calling it good is not acceptable. Per 2020 NEC 250.50 all grounding electrodes at a building or structure shall be bonded together. This code...
diysolarforum.com diysolarforum.com
Click to expand...
I stand by my position and grounding the chassis would create bonding. This all began with a video and not one ounce of proof and not one of you can produce a problem in the U.S
 
Tony Scott said:
How about this, crack the case open and place a piece of electrical tape or shrink tape between the board and the contact point or use a plastic screw to hold the board in place.

They have sold 5k of these and have not had a problem. Look again at that video and see how thick that board is. No one has measured the heat on that board nor has anyone measured the gap.
Click to expand...
1) Plastic screw will not help since the PCB has pads on both side of board connected by VIA, see pot #66, #75.
2) I already mention about installing insulation sheet on my post #85.
3) So far SS did not show any pictures of their boards, or if the standoff has been removed or not which can easily be verify by trying to install the screw back in, if the standoff is removed then you cannot put screw back in place. So far SS did not tell us what they find out how the unit is modified by the factory, I wonder why.
 
Tony Scott said:
I stand by my position and grounding the chassis would create bonding. This all began with a video and not one ounce of proof and not one of you can produce a problem in the U.S
Click to expand...
Well, it is up to you, if you think it is safe good for you use it.
I design products (Circuit design, PCB design, mechanical design, compliance testing) for living for 25 years and have to deal with many UL, TuV, CSA, ETL intertek compliance testing.
 
Bud Martin said:
Well, it is up to you, if you think it is safe good for you use it.
I design products (Circuit design, PCB design, mechanical design, compliance testing) for living for 25 years and have to deal with many UL, TuV, CSA, ETL intertek compliance testing.
Click to expand...
Then explain why this problem is just now showing up?
 
You must log in or register to reply here.

Similar threads

Offgrid Aussi
Can Growatt SPF 5000ES handle AC coupling?
Replies
5
Views
387
flintofgullies
F
??????
Growatt SPF 5000ES parallel charge controller w/ dynamic scale-up
Replies
0
Views
84
??????
??????
Y
Off Grid shed connected to RV getting shocked
Replies
16
Views
429
yettieu
Y
P
4 nos 200ah 12v in series connected to Growatt spf 5000es.
Replies
0
Views
136
Pope
P
B
Growatt SPF 5000ES + Seplos BMS - earth leakage protection (grid) triggers
Replies
2
Views
296
kirkdis
kirkdis
Back
Top