First 3kw DIY solar with PowMr split phase inverter10kw

[Mattb4]

I have 18 Trina and 4 Longi.

They are all cracked panels but I verified all the diodes and they are putting out rated voltage but a about 20% less amps than the rated amount.

The Trinas are about 4-5 amps short circuit current each the the Longi are 8.5 amps each.

Broken glass panels are really iffy if you can get much out of them. I tried out 3 last year that voltage looked OK but 2 of them were worse than not having them in circuit. They literally drew down other panels output. 1 (that I am still using I figure I get ~100w out of its 200w rating.) works OK in parallel but it seems to suffer when in series.

BTW. if you don't seal the front from moisture you also will get odd things going on with production.

 

[Rconce01]

Broken glass panels are really iffy if you can get much out of them. I tried out 3 last year that voltage looked OK but 2 of them were worse than not having them in circuit. They literally drew down other panels output. 1 (that I am still using I figure I get ~100w out of its 200w rating.) works OK in parallel but it seems to suffer when in series.

BTW. if you don't seal the front from moisture you also will get odd things going on with production.

I sealed the big holes with clear silicone. Do you refer to sealing all the cracking?

 

[Mattb4]

I sealed the big holes with clear silicone. Do you refer to sealing all the cracking?

Yes. Cracks allow water to get in.

 

[Rconce01]

Happy Easter.
Do have any shading at all?
Is each 9s string aimed the same way?
What mode are you running in if it's pulling from utility and using solar?

Yes I just laid them all flat and I checked there was no shade. I was running in "Sol" mode which is solar priority with utility backup.

I just don't see this inverter pulling anywhere beyond 50% of the short circuit current the PC array makes.

 

[42OhmsPA]

Yes I just laid them all flat and I checked there was no shade. I was running in "Sol" mode which is solar priority with utility backup.

I just don't see this inverter pulling anywhere beyond 50% of the short circuit current the PC array makes.

I'd recommend switching to SbU. With panels laid flat it's not surprising your seeing nowhere close to rated output, combine that with suspect panels and you're asking for problems.

 

[Rconce01]

You didnt' answer my question. You did the "poor man's" panel test. Once in operation, shit can go to hell even when Voc and Isc values check out.



Again, Voc and Isc are not confirmation of fully functioning panels.

To put it another way, when the array is operating and producing power, did you measure the current from each string with a clamp DC ammeter?

My apologies I meant to answer it. Yes each of the series strings were measured. Each string puts out close to 300v and 6.2 amps. If I'm checking total Voc of the series strings and Isc of the entire string what other test can i do?

I am kinda poor but if I need to buy a special device to confirm the watt production let me know whatever be.

I'll post pics now

 

[Rconce01]

My apologies I meant to answer it. Yes each of the series strings were measured. Each string puts out close to 300v and 6.2 amps. If I'm checking total Voc of the series strings and Isc of the entire string what other test can i do?

I am kinda poor but if I need to buy a special device to confirm the watt production let me know whatever be.

I'll post pics now

 

[sunshine_eggo]

My apologies I meant to answer it. Yes each of the series strings were measured. Each string puts out close to 300v and 6.2 amps. If I'm checking total Voc of the series strings and Isc of the entire string what other test can i do?

I am kinda poor but if I need to buy a special device to confirm the watt production let me know whatever be.

I'll post pics now

It's been noted that having panels oriented in different directions will pull down ALL panels in that string down to the WORST performing panel. It looks like you've addressed that.

It's been noted that ANY shade may completely destroy your array performance. I can't say conclusively, but it looks like there is potential for shading. I can't emphasize how critical it is that shade be avoided at all costs.

Debris on panels also affects them.

I did not see it noted (might have missed it), but if your battery is full, the array will only produce what the battery (edit: and loads) needs. To ensure that the array is performing at its maximum, you could apply a 3000W load at high noon to ensure the power has somewhere to go.

Have you tried one string at a time? The two unshaded 9S arrays should perform nearly identically. Best to confirm them individually then together.

 

[Rconce01]

It's been noted that having panels oriented in different directions will pull down ALL panels in that string down to the WORST performing panel. It looks like you've addressed that.

It's been noted that ANY shade may completely destroy your array performance. I can't say conclusively, but it looks like there is potential for shading. I can't emphasize how critical it is that shade be avoided at all costs.

Debris on panels also affects them.

I did not see it noted (might have missed it), but if your battery is full, the array will only produce what the battery needs. To ensure that the array is performing at its maximum, you could apply a 3000W load at high noon to ensure the power has somewhere to go.

Have you tried one string at a time? The two unshaded 9S arrays should perform nearly identically. Best to confirm them individually then together.

Yes my space is limited so i laid them flat in an effort to keep consistency. They are free of debris. As far as shading potential I've minimized it within my abilities as you can see full sun exposure on all panels.

I have tried each string individually and yes even with a fully charged battery. The inverter registers each string putting in only .4 to .5kw each. Both connected in parallel are registering 1kw.

 

[G00SE]

If your batteries are full the energy from the panels has nowhere to go, does this same happen when you don’t have a full charge and are pulling loads?

 

[Rconce01]

If your batteries are full the energy from the panels has nowhere to go, does this same happen when you don’t have a full charge and are pulling loads?

It always pulls from the battery. The power from the PV array drops when the battery starts to hit the low voltage threshold and then it pulls from the grid.

I need to look at batteryless options than can pull the watts I'm generating

 

[sunshine_eggo]

I have tried each string individually and yes even with a fully charged battery. The inverter registers each string putting in only .4 to .5kw each. Both connected in parallel are registering 1kw.

Again, if your battery is full, the array will only produce what is needed to charge the battery or power your loads. The array power has to go somewhere. If there's nowhere for it to go, it won't produce it.

In order for you to assess your array performance, your battery charger AND loads must be greater than the array power.

It always pulls from the battery. The power from the PV array drops when the battery starts to hit the low voltage threshold and then it pulls from the grid.

I need to look at batteryless options than can pull the watts I'm generating

I'm sorry, but I really don't follow you on any of this logic, especially the last statement.

 

[Rconce01]

Again, if your battery is full, the array will only produce what is needed to charge the battery or power your loads. The array power has to go somewhere. If there's nowhere for it to go, it won't produce it.

In order for you to assess your array performance, your battery charger AND loads must be greater than the array power.



I'm sorry, but I really don't follow you on any of this logic, especially the last statement.

I understand the battery load and the loads connected to the inverter need to be greater than the PV array. What I'm trying to communicate is that the inverter will not pull the full capacity the panels can produce even when the load exceeds the PV array production.

Here is the logic under full load the amp meter clamped to the PV input shows 5-6amps being pulled from the array that I showed produced approx 12 amps. This is taking place while the inverter is pulling current from the battery on a load that is less than the PV array production capacity.

What I tried to say in my last statement is I want an inverter that can utilize the PV power I am actually producing which is closer to 12 amps instead of this thing that barely pulls 40% from the array capacity and needs a battery in order to compensate for what it lacks in pulling from the PV array.

 

[Mattb4]

Power out (<or=) to power in. Your unit is not capable of taking in more power than it is consuming unless there is a problem like a short.

Amps in from panels at one voltage will never equal amps out to battery and load at another but the watts have to balance out.

You do not produce any PV power without load. It is also true that without a battery your system just functioning on PV will be erratic because at any moment that load exceeds PV supply things stop working.

You will never find an inverter* that will output what your panels could produce. Your panels must output what your inverter demands based on loads.

* Exception of grid tie. That is because the grid is an infinite load.

 

[42OhmsPA]

I understand the battery load and the loads connected to the inverter need to be greater than the PV array. What I'm trying to communicate is that the inverter will not pull the full capacity the panels can produce even when the load exceeds the PV array production.

Here is the logic under full load the amp meter clamped to the PV input shows 5-6amps being pulled from the array that I showed produced approx 12 amps. This is taking place while the inverter is pulling current from the battery on a load that is less than the PV array production capacity.

What I tried to say in my last statement is I want an inverter that can utilize the PV power I am actually producing which is closer to 12 amps instead of this thing that barely pulls 40% from the array capacity and needs a battery in order to compensate for what it lacks in pulling from the PV array.

Your inverter can definitely do what you are asking, your panels or wiring must be the issue.
Mine does it every time the sun is out.


Have you changed to SbU mode?

 

[sunshine_eggo]

What I tried to say in my last statement is I want an inverter that can utilize the PV power I am actually producing which is closer to 12 amps instead of this thing that barely pulls 40% from the array capacity and needs a battery in order to compensate for what it lacks in pulling from the PV array.

One of four things is happening:

1) Your used panels are not performing anywhere near specification. Confirmation of Voc and Isc are NOT guarantees of proper panel function.

2) your MPPT is insufficient for the size of your array.

This seems unlikely as the subject implies you have a 10kW PowMR with 2X MPPT capable of 5500W each.

3) Your charging + loads are insufficient to draw max power from the array.

You seem to understand this.

4) Your AiO MPPTs are defective.

#1 is the most likely by far and I'm assuming you understand that your maximum power output is going to be at high noon with all other times being lower and that panels RARELY output rated power. Additionally, panels flat on the ground can't produce max power for two reasons, 1) they're going to get hotter than normal and 2) they're not optimally tilted.

If you attach those panels to a different MPPT under the same conditions, they will perform no differently (unless your AiO is defective).

1. Are these strings in parallel on 1 MPPT, or are they each on their own?
2. Your panels seem to have white sand on them, but in the pictures where they are vertical, it is still visible. What is this? If this isn't debris, I would question if the panels have been damaged. Further support that #1 is the issue.

IMHO, you need to find your bad panels.

I recommend you pick five of the most perfect looking panels and put them 5S on a single MPPT and confirm output at high noon. If you get 70-80% of rated power and the array voltage is close to the calculated Vmp, you have 5 good panels. If not, you have one or more bad panels. Go through the rest of them one at a time adding a 6th panel and confirm you get a 20% boost to power and Vmp. If you fail to get that level of increase, the 6th is bad.

When you buy used panels, you're rolling the dice.

 

[Rconce01]

Power out (<or=) to power in. Your unit is not capable of taking in more power than it is consuming unless there is a problem like a short.

Amps in from panels at one voltage will never equal amps out to battery and load at another but the watts have to balance out.

You do not produce any PV power without load. It is also true that without a battery your system just functioning on PV will be erratic because at any moment that load exceeds PV supply things stop working.

You will never find an inverter* that will output what your panels could produce. Your panels must output what your inverter demands based on loads.

* Exception of grid tie. That is because the grid is an infinite load.

Understood the point I was trying to make with the amps is that the inverter doesn't pull the amps I register as short circuit current at the PV input cables even under full load.

For example If I take both cables that I would feed from the PV into the inverter and check the shortcircuit current I'm at more than 50% over what the inverter pulls while connected to the PV panels With a full load on the AC output.

I would like an inverter that can use the grid to supplement the load if needed and have the battery as optional.

This one goes crazy if you don't put a battery on it. It throws all sorts of error codes.

I'm thinking if my PV array produces the 3kw it should I can always get a smaller inverter closer to that capacity and see if under the same loading conditions and with this PV array it will pull more KW.


Is there no logical test or device other than the Inverter that can load the PV circuit to assess PV power output?

 

[Rconce01]

One of four things is happening:

1) Your used panels are not performing anywhere near specification. Confirmation of Voc and Isc are NOT guarantees of proper panel function.

2) your MPPT is insufficient for the size of your array.

This seems unlikely as the subject implies you have a 10kW PowMR with 2X MPPT capable of 5500W each.

3) Your charging + loads are insufficient to draw max power from the array.

You seem to understand this.

4) Your AiO MPPTs are defective.

#1 is the most likely by far and I'm assuming you understand that your maximum power output is going to be at high noon with all other times being lower and that panels RARELY output rated power. Additionally, panels flat on the ground can't produce max power for two reasons, 1) they're going to get hotter than normal and 2) they're not optimally tilted.

If you attach those panels to a different MPPT under the same conditions, they will perform no differently (unless your AiO is defective).

1. Are these strings in parallel on 1 MPPT, or are they each on their own?
2. Your panels seem to have white sand on them, but in the pictures where they are vertical, it is still visible. What is this? If this isn't debris, I would question if the panels have been damaged. Further support that #1 is the issue.

IMHO, you need to find your bad panels.

I recommend you pick five of the most perfect looking panels and put them 5S on a single MPPT and confirm output at high noon. If you get 70-80% of rated power and the array voltage is close to the calculated Vmp, you have 5 good panels. If not, you have one or more bad panels. Go through the rest of them one at a time adding a 6th panel and confirm you get a 20% boost to power and Vmp. If you fail to get that level of increase, the 6th is bad.

When you buy used panels, you're rolling the dice.

Thanks, it's not white sand on the panels is the cracked glass. It may appear as sand but is the cracks.

I will give your troubleshooting tips a shot and I'll start looking for a new set of panels. For now I'm going to slow things down and try to get my $1800 back on this inverter since definitely doesn't give me the outcome desired.

 

[sunshine_eggo]

Thanks, it's not white sand on the panels is the cracked glass. It may appear as sand but is the cracks.

Okay, so you have known damaged panels. Obviously, they are the most likely issue if you don't get the output you expect.

I will give your troubleshooting tips a shot and I'll start looking for a new set of panels. For now I'm going to slow things down and try to get my $1800 back on this inverter since definitely doesn't give me the outcome desired.

Again, the inverter is almost certainly NOT your issue. In case you've missed it the two other times I've said it, Voc and Isc measurements on a panel or string in no way confirm they will perform properly.

Your desired outcome is almost certainly associated with the panels you have.

 

[Rconce01]

Okay, so you have known damaged panels. Obviously, they are the most likely issue if you don't get the output you expect.



Again, the inverter is almost certainly NOT your issue. In case you've missed it the two other times I've said it, Voc and Isc measurements on a panel or string in no way confirm they will perform properly.

Your desired outcome is almost certainly associated with the panels you have.

Thanks for your patience! I'll start looking for some new panels and hopefully I'll have better luck. Any guides on here on the best value on inverter for Approx 240v 3000watt ?

 

[42OhmsPA]

Thanks for your patience! I'll start looking for some new panels and hopefully I'll have better luck. Any guides on here on the best value on inverter for Approx 240v 3000watt ?

It's not the inverter.
Are the longi panels cracked or damaged at all?
Try your longi panels in 4s configuration with no other panels hooked up and see what you get, angle them for your latitude.

 

[Mattb4]

Understood the point I was trying to make with the amps is that the inverter doesn't pull the amps I register as short circuit current at the PV input cables even under full load.

...

The current your panels can deliver when fully loaded is the Imp rating not the Isc. Short circuit amperage is a bit of a fictional amount. Yes tehre are testers for panels. Image below for one on Amazon for $90

It sounds like you want a full grid tie hybrid AIO if you are going to use grid and PV only with a small backup battery. Although some hybrid AIO's will run batteryless this way they tend to be a bit flaky.

 

[gotbeans]

The max pv input voltage is 500v on the inverter. I figured the higher the better. Should I string them up as 7s 2p? I figured 14s is more effective use of the power.

14 in a string? so their VOC is 34volts?
Are you sure they're not 37.6 VOC because those look identical to the 250 watt ones santan solar sells

480Voc is VERY high unless you have a 600Voc MPPT. The MPPT voltage is a NEVER EXCEED value.

All panels in the same string must face the same direction. All panels will perform no better than the worst orientation panel.

Shading may also challenge the MPPT's ability to find the max power point.

I think you are guessing at the 480 being VOC. I'm pretty sure the 480 is running voltage. I'm guessing the VOC is actually 526.4

The current your panels can deliver when fully loaded is the Imp rating not the Isc. Short circuit amperage is a bit of a fictional amount. Yes tehre are testers for panels. Image below for one on Amazon for $90

It sounds like you want a full grid tie hybrid AIO if you are going to use grid and PV only with a small backup battery. Although some hybrid AIO's will run batteryless this way they tend to be a bit flaky.

I have one of these, they're pretty cool. Powered by the panels themselves too
$50-60 on aliexpress

 

[Rconce01]

It's not the inverter.
Are the longi panels cracked or damaged at all?
Try your longi panels in 4s configuration with no other panels hooked up and see what you get, angle them for your latitude.

Yes the deal i got was on cracked panels. My dumb mistake I guess to think that voltage and the current confirmed watt output.

I am still going to return the inverter because it was nearly $1800 and I will never need 10k output.

 

[Rconce01]

14 in a string? so their VOC is 34volts?
Are you sure they're not 37.6 VOC because those look identical to the 250 watt ones santan solar sells


I think you are guessing at the 480 being VOC. I'm pretty sure the 480 is running voltage. I'm guessing the VOC is actually 526.4



I have one of these, they're pretty cool. Powered by the panels themselves too
$50-60 on aliexpress

So those testers have more technology than the clamp on amp meter and voltage tester? It can actually confirm the true power out of each panel?