Sol-Ark 15K FAILED! - My experience with Sol-Ark ; Their response was underwhelming
- Thread starter tomlu_2000
- Start date
robby
Electronic Engineer
- Joined
- May 1, 2021
- Messages
- 5,290
I talked to him today and he said that he was very aware of the problem and that they are working to rectify it.
As I suspected, he said a lot of the senior CS people have been reassigned to work on the Industrial side and that has left a serious Gap in the amount of people on the Tier 3 residential CS side. He said they are in the process of hiring and training more people but that it would take time to get enough Tier 1 & 2 CS staff to the level of the Tier 3 people who got transferred.
Maybe they should have a moratorium on new sales until they can get the support staff up to speed?
That relay is rated for over 30k make/break cycles at a full 200a at 830VAC. You wont be able to kill it in your lifetime at 240VAC even with extreme usage.
Last edited:
Not according to the data sheet.
timselectric
If I can do it, you can do it.
- Joined
- Feb 5, 2022
- Messages
- 25,303
And Sol-Ark confirmed that it is the relay that failed in the OP's unit.
That begs the question;
"what load was switched?"
It's something I don't worry about. I've looked at my daily history for an entire year. My maximum load was 9KW. My worst case would be clothes dryer + heat pump + kitchen appliance = 12KW max. I'm not doing laundry when I use my 40A range.
timselectric
If I can do it, you can do it.
- Joined
- Feb 5, 2022
- Messages
- 25,303
No idea, but I guess that it could have been as much as 200a. Since it's supposed to have 200a passthrough.
Guess something really had it clicking! Assuming it was not under full load--but such a load as could cause it to fail after 60,000 operations, and assuming it might click 10 times per second if something or other was oscillating in the system, it would only require an hour and forty minutes to run out its life expectancy. Granted, those are big assumptions given my ignorance of the relay in question, but it would fit the timeline of the OP reasonably well. Now, the OP claims that by the time he got home after his dinner, the inverter was clicking only once or twice a minute. If the relay was still operational at that point is questionable, because he disengaged the inverter at that point and apparently SolArk claims the relay is bad.
The most logical alternative option is to accept that the relay failed far short of its claimed life expectancy. If so, that does not portend well.
timselectric
If I can do it, you can do it.
- Joined
- Feb 5, 2022
- Messages
- 25,303
I can only assume that it had already failed, and that's why it was cycling.
Cycle
On > check for correct voltage > not correct > off ....... repeat
A couple actuations of make/break over 50A would be a good way to fry it, from looking at that data sheet. Then, yeah, starts the click click click trying to operate it and it failing to work properly anymore.
That relay is under-specced. If you say it has 200A passthrough, well, OK technically it does, however, how do you make absolutely sure it is under 50A when it cycles? OK the inverter can only handle that 50A. Makes sense there, because the inverter would trip out trying to run more than the make/break rating of that relay. Doesn't mean it won't try to do it. Heck, could be running say 60A through from the grid, the inverter goes CLICK to switch to inverter mode and damages the relay contacts in the process, inverter overloads instantly, CLICK it cycles back to grid because it can't handle that overload, another bit of arc damage to the contacts. Do that a handful of times and now it is fried? Unless the inverter knows there's more than 50A flowing in the pass-through and it is smart enough to not try to break that current and switch to inverter mode until the load drops low enough.
That relay is under-specced. If you say it has 200A passthrough, well, OK technically it does, however, how do you make absolutely sure it is under 50A when it cycles? OK the inverter can only handle that 50A. Makes sense there, because the inverter would trip out trying to run more than the make/break rating of that relay. Doesn't mean it won't try to do it. Heck, could be running say 60A through from the grid, the inverter goes CLICK to switch to inverter mode and damages the relay contacts in the process, inverter overloads instantly, CLICK it cycles back to grid because it can't handle that overload, another bit of arc damage to the contacts. Do that a handful of times and now it is fried? Unless the inverter knows there's more than 50A flowing in the pass-through and it is smart enough to not try to break that current and switch to inverter mode until the load drops low enough.
It should be smart enough to sense 60a passing through and not try and operate, otherwise it’ll just like its guts each time.
You'd think so. I'd hope so. But, I'm just raising the question, does it know how much is going through the relay when in grid AC pass-through mode? If not, that could explain the failure.
Brucey
Velcro Man
"World-class support......"
Literally abandoning the poor residential owners while mgmt focus shifts to commercial. It's OK come back in a year or two things will be better, please understand....
Brucey
Velcro Man
I guess that's why they strapped a fan on there try and keep it from cooking. This is the big innovation solark brought to town?
Bound to have CTs on the load and source side, then it’s just a matter of applying some logic.
Brucey
Velcro Man
Based on the datasheet it's rated for 30k make breaks at 50A. Could just be a handful if it's breaking 200A which is four times it's rating?
Isn't there one for each leg? so it would be only 100A.
robby
Electronic Engineer
- Joined
- May 1, 2021
- Messages
- 5,290
Okay what am I not seeing because the Graph data indicates 30K cycles at 200A at 830Vac
Seems to be either some discrepancy in their data sheet, or confusion in how to properly read it 
The table seems to indicate only 50A make and break, but also says max switching current, and the graph appears to indicate operation cycles at various amperages...
The table seems to indicate only 50A make and break, but also says max switching current, and the graph appears to indicate operation cycles at various amperages...
robby
Electronic Engineer
- Joined
- May 1, 2021
- Messages
- 5,290
Interesting. I get what they are saying but either way that 50A at 830V.Seems to be either some discrepancy in their data sheet, or confusion in how to properly read it
The table seems to indicate only 50A make and break, but also says max switching current, and the graph appears to indicate operation cycles at various amperages...
This relay is not going to die unless it's being clicked like a clock.
Last edited:
Has anyone seen an AIO around a 12K rating that has better rated relays? Hopefully won't even be switching them monthly but definitely running some current through them since we have been averaging 110 to 175 KWH a day through the Churod 830 Volt 200 amp units in the 18kPV.
Have seen On grid use surges to 27 KW to load a few times. Most days over 100KWH through the load relays, jmo but this is where I see the 10 year warranty having real value, because the 15k SolArk and 18kPV are advertised with the Full 200 amp service pass through capability.
From what is posted about the markup on these imported AIO's companies should have plenty of income to keep replacement parts and units in stock. Take care of the customer if you want repeat business and more word of mouth growing sales from satisfied customers.
But the contacts won't be switched at 830VAC , they could be switching the full 200 amps current @ 120V - 24000 watts to match a grid single pole disconnect.
It would seem the mode the Inverter is run in would be a big factor in how the lifecycle ends up on the relays. run in Back up mode very little switching involved if the Grid is reliable. "Off Grid mode" with loads exceeding the Inverters capability it will/could be cycling those contacts under load multiple times daily to meet power requirements for zero backfeed.
Scroll down a bit further to page 2 for the graph with more detailed cycle life data:
Similar threads
