diy solar

diy solar

Sol-Ark 15k - Surge Amps

So this is a thought i just had (ie no vetting whatsoever) but if you had inverters configurable for 220 or 240v and none of your loads really cared, would you gain any slight headroom for surge if you ran it at 220? I am prepared for bashing lol.
 
220V would reduce surge current 10%, also starting torque.

They're already wired for 240V

Here's a 3HP, 230V motor


14 FLA, 3220 VA. LRA not quoted, will assume 5x which would be 16,100 VA.
Sol-Ark rated 24 kVA surge for 10 seconds should have been able to start that. (would have been too much on 120V)

Capacitors in good shape? I have a compressor that was having trouble starting (on the grid). I replaced capacitors, but that didn't help. I think I had damaged windings with brown-out. When it wouldn't start any more I replaced the motor, will get original rewound.

Check battery voltage on input of Sol-Ark during this event. If voltage from battery or through cables sags, that would limit its output.
Maybe specs are only achieved at top battery voltage? But would expect power output to derate linearly, not enough to drop from 24 kVA to below 16 kVA.
 
This is hopefully a quick easy question.

I’m reading the manual for the 15 Sol-Ark trying to make sure the AC units I wanna use will not trip the inverter.


So if I’m reading this right, the Sol-Ark 15 can surge it’s output to 24,000VA at 240V, essentially 100 amps for 10 seconds.

So if I had an AC unit turn on that used 80 at startup, then idled down to say 40 amps, on battery and or Solar, the Sol-Ark should have enough surge output to startup the AC load without tripping?

Is that correct?
you need a clamp meter with inrush function to measure the inrush. only then you determine if you can run this AC.
 
14 FLA, 3220 VA. LRA not quoted, will assume 5x which would be 16,100 VA.
Sol-Ark rated 24 kVA surge for 10 seconds should have been able to start that. (would have been too much on 120V)

"4 x 200Ah MeritSun LiFePO4"


Rated 120A continuous, 300A (1 second) per battery, should be able to supply 500A for 24000 VA surge from 48V battery.
But look into how your batteries are bussed together - do you connect positive and negative to opposite ends, so current draw is reasonably balanced?
How much resistance from batteries to inverter, and what voltage drop expected?
 
Rated 120A continuous, 300A (1 second) per battery, should be able to supply 500A for 24000 VA surge from 48V battery.
But look into how your batteries are bussed together - do you connect positive and negative to opposite ends, so current draw is reasonably balanced?
How much resistance from batteries to inverter, and what voltage drop expected?
The batteries are not daisy-chained, each is connected to a bus bar with about 5' of 6 AWG. The bus bar is connected to the inverter with 5' of 0000 copper cable. The total resistance is about 2 milli-ohms, so the voltage drop is a fraction of a volt.

The 12K has separate faults for DC voltage low and DC over-current, but the manual says any of these can occur for the same underlying fault.
 
14 FLA, 3220 VA. LRA not quoted, will assume 5x which would be 16,100 VA.
Sol-Ark rated 24 kVA surge for 10 seconds should have been able to start that. (would have been too much on 120V)
maybe that motor should start, but the empirical evidence is that mine don't.
Capacitors in good shape? I have a compressor that was having trouble starting (on the grid). I replaced capacitors, but that didn't help. I think I had damaged windings with brown-out.
New SawStop with 3HP, new 3HP motor on bandsaw. They start just fine on my generator (no grid available).
When it wouldn't start any more I replaced the motor, will get original rewound.
Is it really worth getting the motor rewound? Around here (Albq), rewinding starts around $500. A new motor is around $300.
Check battery voltage on input of Sol-Ark during this event. If voltage from battery or through cables sags, that would limit its output.
Maybe specs are only achieved at top battery voltage? But would expect power output to derate linearly, not enough to drop from 24 kVA to below 16 kVA.
Problem occurs even with full battery plus 10kW of solar input available and generator running.
 
The 12K has separate faults for DC voltage low and DC over-current, but the manual says any of these can occur for the same underlying fault.

This is the only clue we've got left. It is complaining about DC, not AC current.
Have any way to test AC current peak? DC current peak? Can you measure the momentary DC voltage dip?

See if SolArk has a firmware update, or can help you with the issue.
I think we heard they started out with conservative limits, to protect their equipment, and were able to loosen them.
 
This is the only clue we've got left. It is complaining about DC, not AC current.
why do you say this? The fault is an F18 AC overcurrent fault, not theF20 DC current fault.
Have any way to test AC current peak?
48A. Voltage sags to 160-170V (Peak readings on Fluke 325)
DC current peak? Can you measure the momentary DC voltage dip?
Not easily - the tools are in the shop, the inverter/batteries are in the house.
And I've gotten tired of blowing the house offline and having to reset all the clocks.

See if SolArk has a firmware update, or can help you with the issue.
I think we heard they started out with conservative limits, to protect their equipment, and were able to loosen them.
The firmware is only a couple of months old. I can check, but I doubt there is anything relevant available.
 
Have you tried just sticking some leads in the bottom of the fluke and seeing what the dc voltage does with min-max?

Honestly, 4x 6ga does not sound all that great to me. You probably need 12 or more 6ga to have an equivalent amount of copper to a 4/0 (online calc probably maks that math easy smewhere). So i am suspicious of the connections from your batteries to your busbar. Not saying that’s the issue or even ‘an’ issue, but id be doing some min-max to at least try and find out.
 
I have 4/0 from my batteries to my bus. My 15K starts everything in our house….1 1/2 hp well pump, 4ton Geothermal, 2hp pool pump, sewage ejection pump etc. One thing you might do is to move some breakers around so your load is more balanced. Look at L1 & L2 on your Sol Ark see if they are balanced to begin with. The closer you can get them the better obviously. Move some breakers around to even them out.
 

Attachments

  • 89B07C03-C169-4853-8D47-9D18D767B923.jpeg
    89B07C03-C169-4853-8D47-9D18D767B923.jpeg
    1,005.6 KB · Views: 27
Ok. I thought 6ga x4 was believable if you originally never intended to run large loads but 2ga makes me feel a lot better. I would still set it up to do voltage drop measurement with minmax and just run whatever the biggest thing is that doesn’t crash the system and see what the number is.
 
why do you say this? The fault is an F18 AC overcurrent fault, not theF20 DC current fault.

"The 12K has separate faults for DC voltage low and DC over-current, but the manual says any of these can occur for the same underlying fault."

Do you mean the 12k manual lists DC faults, but your Sol Ark didn't give those errors?

48A. Voltage sags to 160-170V (Peak readings on Fluke 325)

48A x 240V = 11520 W, should not be a problem. (12k? 15k? Which model?) But either way, should be within surge rating.

Not easily - the tools are in the shop, the inverter/batteries are in the house.
And I've gotten tired of blowing the house offline and having to reset all the clocks.


The firmware is only a couple of months old. I can check, but I doubt there is anything relevant available.

Might be the latest firmware, but the deal could be they were conservative, had it go into shutdown to protect itself. But that would seem to mean it isn't delivering to published specs. Someone mentioned updates for earlier models to deliver more power.
 
"The 12K has separate faults for DC voltage low and DC over-current, but the manual says any of these can occur for the same underlying fault."

Do you mean the 12k manual lists DC faults, but your Sol Ark didn't give those errors?
I said my 12K reported an F18 AC over-current fault. That was the only fault it reported in multiple shut-down events when starting any of these motors.
The manual contains this wording that I find bizarre:
F18 Tz_Ac_OverCurr_Fault Overloaded the Load Output, reduce loads. Wiring Short on the AC Side can also cause this error. Overloads
can result in F15, F18, F20, or F26.

I believe the last sentence is telling you that an overload can result in a variety of errors, but it is not saying that the reported overload isn't the actual source of the error. For example, with too little battery capacity the overload will result in a DC Current overload whereas on a system with lots of battery you will get an AC current overload, which is what I'm seeing.

48A x 240V = 11520 W, should not be a problem. (12k? 15k? Which model?) But either way, should be within surge rating.
W != VA for reactive loads
The 12K (and the 15K) are rated in terms of kVA, which takes into account the load factor for reactive loads. This is not measured by a readily available meter. As far as I can find, meters for measuring reactive loads are on the order of $10K.
 
I use a $25k scope that I bought for $2000. Much slower scope would be sufficient for 60 Hz, but very convenient to have integral and X vs. Y.
Voltage probe and current probe, then do math.


48A x 240V would be 11520 VA, possibly less than 11520 W, doesn't seem to be a reason why inverter should quit.
But I do wonder how well it handles inductive loads, making it swallow current out of phase.

I think my Sunny Island is OK with that, being 4-quadrant.
At work, we're using a 2U rack size UPS for some tests. When I put a transformer on the output its power consumption goes up, can trip a breaker on the branch circuit feeding it and other loads.
 
I said my 12K reported an F18 AC over-current fault. That was the only fault it reported in multiple shut-down events when starting any of these motors.
The manual contains this wording that I find bizarre:


I believe the last sentence is telling you that an overload can result in a variety of errors, but it is not saying that the reported overload isn't the actual source of the error. For example, with too little battery capacity the overload will result in a DC Current overload whereas on a system with lots of battery you will get an AC current overload, which is what I'm seeing.


W != VA for reactive loads
The 12K (and the 15K) are rated in terms of kVA, which takes into account the load factor for reactive loads. This is not measured by a readily available meter. As far as I can find, meters for measuring reactive loads are on the order of $10K.
Llaves, have you called Solark? I had f18 faults that didn't make sense and they updated my firmware several times and each time they updated, my f18 errors happened less. I think they are still fine tuning things and loosening the parameters as Hedges said.
 
Did anyone even look at the specification sheet for this guys AC unit? Or is this just a typical completely derailed thread.
The OP linked to LG heat pump inverter systems therefore there is no startup surge whatsoever.
So there should be no problem starting those things up with any inverter...... in theory lol
 
So my very limited experience running inverter-driven compressor air conditioners off of whole house inverter is just one Midea window unit.

It has no hope in hell of ever exceeding my inverter limits. BUT it does seem to confuse the inverter! When the Midea is running the ‘load’ reading on the inverter, whether viewed by watts, % or whatever, fluctuates rapidly and constantly. However, i have plugged the ac unit into two different power meters and neither one fluctuated.

I do not know if the Midea has PF correction circuitry or whether that is 100% the issue, but i do know it is possible for an inverter driven compressor device to make the house inverter THINK the load is higher than what it really is, at least on an instantaneuous basis. So if you scaled that problem up a whole lot it may be relevant to having a large inverter ‘talk itself out of’ powering something because it thinks something is happening that really isn’t, because of some vagary of sampling and averaging or whatever by which it assesses ‘load’.
 
Did anyone even look at the specification sheet for this guys AC unit? Or is this just a typical completely derailed thread.
The OP linked to LG heat pump inverter systems therefore there is no startup surge whatsoever.
So there should be no problem starting those things up with any inverter...... in theory lol
I got my verification @Quattrohead, even starting both of these up, since they have internal inverters and DC motors they slowly ramp up the load and should be able to run fine on a signle Solark 15K.


Afterward, the other guy was mentioning problems with his SolArk 12K starting certain pumps etc which I think his advise was to use a soft start.

I appreciate the followup, I'm feeling really confident about my proposed design, which I've shared below for anyone interested.

 
Back
Top