Terrible Signature Solar / EG4 18Kpv Experience - Returning and Replaced with Sol-Ark 15k

[Watts Happening]

I tried to read the thread but this thing is massive. Can you summarize the actual problems that you have in the evidence of those problems and all of the error codes with the 18K? I just learned about the Net Zero issue 4 days ago and it seems like the new firmware has it resolved. But I have not tested this feature yet.

And I don't think you understand my business model. If the 18K fails then I will sell more sol arks. I still haven't even added the 18K to my site because I've been busy with other things. I'm here to review and criticize every product on the market.

Let me try to reenact what problems you had. I also need the exact error code so I can try to reproduce it myself.

I understand the business model, I merely meant that you've said it's an amazing product and I don't want to be the cause of people no longer purchasing it. Of course you'll move other products, I totally get it.

There has never once been an error code.

Regardless of which person at SS or Gilbert sets it up, it does not peak shave reliably, as a general rule it will pull 150-300w from the battery bank the entire time from 3-9pm that it was programmed to only use batteries. It will then backfeed the grid at random times. It also doesn't follow times in the way you'd expect a computer to, IE "at 3pm" switch to batteries. In Gilberts own words, "it's not perfect and is + or - 8 minutes". Which doesn't sound like much, and agreed isn't much, but it doesn't stop there. I'm seeing it happen hours later.

It also decides to pull from batteries as it wants, IE it's supposed to do 3-9pm but at 10am I see it drawing near 100% from batteries.

It does not have a remote rapid shutdown solution as indicated in the manual, I have asked repeatedly to no avail.

Marcus claims I'm asking for something that doesn't exist, peak shaving from 3-9pm then charge batteries at 10pm. The Sol-Ark does this out of the box.

 

[Will Prowse]

I understand the business model, I merely meant that you've said it's an amazing product and I don't want to be the cause of people no longer purchasing it. Of course you'll move other products, I totally get it.

There has never once been an error code.

Regardless of which person at SS or Gilbert sets it up, it does not peak shave reliably, as a general rule it will pull 150-300w from the battery bank the entire time from 3-9pm that it was programmed to only use batteries. It will then backfeed the grid at random times. It also doesn't follow times in the way you'd expect a computer to, IE "at 3pm" switch to batteries. In Gilberts own words, "it's not perfect and is + or - 8 minutes". Which doesn't sound like much, and agreed isn't much, but it doesn't stop there. I'm seeing it happen hours later.

It also decides to pull from batteries as it wants, IE it's supposed to do 3-9pm but at 10am I see it drawing near 100% from batteries.

It does not have a remote rapid shutdown solution as indicated in the manual, I have asked repeatedly to no avail.

Marcus claims I'm asking for something that doesn't exist, peak shaving from 3-9pm then charge batteries at 10pm. The Sol-Ark does this out of the box.

Can you post the logs of it doing that? I've never seen that happen on mine. That doesn't sound right about 8 minutes. Let's see a picture of that happening.

Or do you have a video of this happening? Can you post it and link it here? Did you verify it with your meter?

Yes you need to buy the remote rapid shutdown separately if it is required for compliance in your area. It's the same with sol ark.

And yes the 18K can do that as well. I don't understand why yours is not able to. We should start with evidence then to figure out what's going on here. Post everything you can. How long did you have the system running for?

 

[Watts Happening]

And how in the world did your inspector laugh at you when solark has the same disconnect? If you need it to be its own box then whether you have a sol ark or 18K will not matter.

When I installed my Wi-Fi dongle it worked instantly. Let me know the steps on how to reproduce your error. I could mention that in a video if I can make it not work. Can you oppose a video of this Wi-Fi dongle issue?

I didn't see you had added to your post when I was writing the last.

As for the remote disconnect, please see what I originally posted. The manual clearly has wires going to the board for a remote disconnect switch. But, there is not a place on the board for this, I have asked them repeatedly where those wires go with no response. I've seen them tell others "there is a white paper coming".

This was literally posted yesterday: https://diysolarforum.com/threads/eg4-18kpv-external-rapid-shutdown-switch.66723/

As for the Wi-Fi dongle please see the image below. The entire process was fine other than the fact that the dongle "wasn't on their server". IE I could change the last digit from a 1 to a 2 and the error would go away, but below I'd get an error for an incorrect pin as you'd expect.

It took multiple SS members and my detailed emails to eventually get somewhere. And guess who fixed it.... Gilbert.

 

[Will Prowse]

Ohh I thought you were talking about a separate disconnect shutdown module for pv conductors. I've never used a small switch one. I did not see that in the manual or try out that feature. I am curious how signature solar will respond when they actually do.

Oh that's unfortunate about the dongle. Has anyone else had this issue? That's not good.

 

[Watts Happening]

Can you post the logs of it doing that? I've never seen that happen on mine. That doesn't sound right about 8 minutes. Let's see a picture of that happening.

Or do you have a video of this happening? Can you post it and link it here? Did you verify it with your meter?

Yes you need to buy the remote rapid shutdown separately if it is required for compliance in your area. It's the same with sol ark.

And yes the 18K can do that as well. I don't understand why yours is not able to. We should start with evidence then to figure out what's going on here. Post everything you can. How long did you have the system running for?

It has been disconnected and removed from service.

Rather than logs and me attempting to prove it, how about this information direct from Gilbert:



I'm not worried about buying a remote switch, I have that switch and had no expectation of getting one with the inverter. I'm saying it had no instructions as to how to connect it whatsoever. My inspector looked at me quite puzzled when I said, here's the instructions from the manufacturer but I'm hearing nothing back as to where these wires actually connect:




System ran for maybe 10 days, continually with issues.

 

[Watts Happening]

Ohh I thought you were talking about a separate disconnect shutdown module for pv conductors. I've never used a small switch one. I did not see that in the manual or try out that feature. I am curious how signature solar will respond when they actually do.

Oh that's unfortunate about the dongle. Has anyone else had this issue? That's not good.

Gotcha, yeah I never got to the point of testing the rapid shutdown modules, I purely meant the switch itself. My inspector had a hard no for putting panels up without a button a firefighter could push to shut the system down. As they clearly don't have direct access to the garage, that had to be at the meter and there was no way to do so.

 

[G00SE]

You don't have any of that data at all? Can signature solar pull that data? You didn't take any screenshots or save it or share what you were seeing with signature Solar?

On page one you can see screenshots of his PMs where data was shared with signature solar. Signature solar also remote connected to his inverter so wouldn’t they also see that data then?

 

[Will Prowse]

You don't have any of that data at all? Can signature solar pull that data? You didn't take any screenshots or save it? This unit graphs the data over time so we can see exactly what's going on.

 

[Will Prowse]

On page one you can see screenshots of his PMs where data was shared with signature solar. Signature solar also remote connected to his inverter so wouldn’t they also see that data then?

Yes absolutely. I just want to have access to it here so we can see what's going on.

Went through the screenshots right now and I can't see any of the numbers or what's going on. Has anyone else had this issue that can show us this data?

 

[G00SE]

Yes absolutely. I just want to have access to it here so we can see what's going on.

Went through the screenshots right now and I can't see any of the numbers or what's going on. Has anyone else had this issue that can show us this data?

My apologies I didn’t realize you meant for your own use/testing. I read it as “did signature get the data” ?‍

 

[Will Prowse]

There is a possibility that you have a manufacturing defect unit. I don't know why it would do that. I've not heard of anyone having that issue, and I've never seen that issue on my unit. That is very frustrating to say the least.

I think that's why the support team had a tough time helping you, because what would actually cause that? I think that's why they wanted to do a return as well.

 

[Will Prowse]

I'm not sure what would even cause that. Did you verify the current readings of the units shunt with a meter?

 

[Watts Happening]

You don't have any of that data at all? Can signature solar pull that data? You didn't take any screenshots or save it? This unit graphs the data over time so we can see exactly what's going on.

I will connect the inverter to power and do what I can to download logs. I certainly have screenshots of issues I was having that were sent to them in communications, but they wouldn't really "prove" much in the way you're asking. My last message to Marcus relating to this unit was after he promised he himself would fix it, and that the "engineer that designed the unit himself" would set it up. Screenshot of that conversation and the attachment below. At this time we were many hours away from when it should be discharging, this was the nail in the coffin.




Attachment from messages above:

 

[Watts Happening]

I'm not sure what would even cause that. Did you verify the current readings of the units shunt with a meter?

Absolutely, at no point did anyone in the organization make ANY suggestion that the unit was faulty at all. The only suggestions were that it was programmed wrong (by their employees) or had too old of firmware (which is what Gilbert put on it because it was the only firmware that didn't backfeed).

Please see screenshots below, without me having to prove current measurements (which I'm happy to do), you'll see in their own words they know it was happening. They verified it themselves.

This is from Ty, after I gave him detailed instructions (all available in the original post):


After he did that the first thing it did was export to grid, I messaged both him an Marcus, this is the email I received:

 

[Will Prowse]

Did you verify those readings with a meter? So you still have the unit? Can you hook it back up again temporarily so that we can see what's going on? Just connect it to the grid with a cable to a meter somewhere.

 

[Watts Happening]

The last communication I had from Ty was the following (that same day, this Monday)


I have not heard a word since then. They CLEARLY agree and know it's not working as expected.

 

[Will Prowse]

Absolutely, at no point did anyone in the organization make ANY suggestion that the unit was faulty at all. The only suggestions were that it was programmed wrong (by their employees) or had too old of firmware (which is what Gilbert put on it because it was the only firmware that didn't backfeed).

Please see screenshots below, without me having to prove current measurements (which I'm happy to do), you'll see in their own words they know it was happening. They verified it themselves.

This is from Ty, after I gave him detailed instructions (all available in the original post):
View attachment 161101

After he did that the first thing it did was export to grid, I messaged both him an Marcus, this is the email I received:

View attachment 161103

Ok nice, what happened after this?

 

[Watts Happening]

Did you verify those readings with a meter? So you still have the unit? Can you hook it back up again temporarily so that we can see what's going on? Just connect it to the grid with a cable to a meter somewhere.

I'll get that done this weekend, part of the downside there is that I also have to connect it to a battery to really show anything via a meter. The logs should show it previously back feeding no doubt, but it obviously can't backfeed the grid if it has no other source connected to it like batteries or panels.

 

[Will Prowse]

The last communication I had from Ty was the following (that same day, this Monday)
View attachment 161105

I have not heard a word since then. They CLEARLY agree and know it's not working as expected.

Yeah, absolutely. They did offer to return it right? If they cannot figure it out remotely, having it on Markus's desk might be the only solution. I want to know what's going on with that unit

 

[Watts Happening]

Yeah, absolutely. They did offer to return it right? If they cannot figure it out remotely, having it on Markus's desk might be the only solution. I want to know what's going on with that unit

Well, yes. He has stated multiple times I could return it for a full refund which is great, just sucks the amount of time and energy put into it and cutting holes in a wire way etc. But hey, that just comes with the territory so I'm not faulting them there. He "promised" someone would reach out Tuesday, they never did. Then they didn't Wednesday, then they didn't Thursday. Then I messaged him last night and he had the gall to bullshit me saying they had reached out and he had been CC'ed. To which I immediately called him on and asked for a simple screenshot and what do you know, no response whatsoever.

Then magically this morning I get a message. It's just completely unprofessional stuff. See below: