abysmal poor sun performance

[mike90045]

I now have the task of installing a 3rd array, and am choosing a controller for it.

I currently have 2 evergreen arrays, one on a tristar MPPT60, other on a classic 200.   The Tristar kicks the Classic into the mud whenever there is anything less than pure sun.   A little haze, 15% performance.  Clouds, it sleeps.  a break in the clouds, it wakes up, till the sun fades out.  sunrise, sunset, it's a half hour later/earlier than the Tristar.

Is there a solution ?   any troubleshooting I can try ?  Since this week is going to be cloudy, when I'm not tending the generator, I'll try playing with the legacy modes, fixed MPPT voltage, anything to get some harvest that I told my wife we'd have.
The classic array is 2Kw, tristar is 3Kw, so I'd expect to see the same ratio of output.  In full sun they track pretty well, as I'd expect.
Firmware:     
- Classic Rev: 1849
- Network Rev: 1839

[ChrisOlson]

When I used to use my Classics for solar, and first bought our XW-MPPT60's to integrate with our new (at the time) XW Power System, I had the same problem.  The XW's pretty much stomped on the Classic during cloudy conditions.  I switched the Classic to Dynamic Mode and got a significant improvement, as well as more hours online during the day.  The only problem is that the Classic's Dynamic Mode is a little slow on tracking when it's sunny with scattered clouds that block out the sun intermittent.  The regular Solar Mode works better on sunny days (or at least did for me).

I'd try Dynamic Tracking and see what you get.  The Classic doesn't seem to unload the array on Dynamic like it does on Solar Mode every few minutes.  I remember that it used to slowly vary the Vmp to find where it got maximum power and then sort of "lock in" on that.  It worked pretty good here.

[boB]

Mike, what  is your battery voltage supposed to be ?   I am seeing something like 62.8 volts battery V on
the Local App screen.  What is the target voltage of the Classic ?  You will need to go to the MNGP screen
under CHARGE---T-Comp---View  and see what that is supposed to be.

If the Classic is seeing battery voltage at or above the temperature compensated target voltage,
it won't try to charge at that time.  It may also be that the MS is bringing that battery voltage
up to that point ?

AND, what is your Voc of the PV array into the Morningstar ?  They may be different.
Also check the LIMITS screen to make sure the MAX temp-comp'd battery voltage is
not limiting that charge voltage.  It's pretty high as I see on the screen.

If all is good in that case, then you may want to try either dynamic mode or Legacy P&O mode.
I would try Legacy P&O mode before Dynamic tracking but if the PV input voltage and Voc
is too low, SOLAR may not be the best for this condition.

Please report back as to what you find.

boB

[ChrisOlson]

If all is good in that case, then you may want to try either dynamic mode or Legacy P&O mode.

boB, as long as we're on the subject, what is supposed to be the difference between the Legacy P&O and Dynamic?  I tried both and couldn't really see that they did much different from one another.  I did try a wind turbine once with both too, and the Classic did a pretty good job of tracking the turbine on Dynamic.  Not so good on Legacy P&O.  But with solar panels I couldn't see much difference between them.

[Vic]

From earlier Threads,  Mike's battery is NiFe,  which needs HIGH   Absorb voltages,   believe that the TS runs about 90-ish Vmp string and the Classic 200 runs string Vmp of about 130 V (Mike specs String Voc at 110 V for the TS and 160 for the Cl 200).

Believe that the PV array for the Classic looks SE,  ...
Some info here:
http://midniteforum.com/index.php?topic=2014.0

To me,  the configurations are not very similar and makes direct comparison difficult and perhaps not fair.

EDIT:  What was the time if day when the snapshot of each CC's performance was taken?
And guess that there was no shading,  on either array,    other than clouds (?).  What is the azimuth bearing of each array?


FWIW,   Vic

[mike90045]

.....http://midniteforum.com/index.php?topic=2014.0
To me,  the configurations are not very similar and makes direct comparison difficult and perhaps not fair...

The cells are evergreen, just the arrangements in the panels and array sizes differ.  So, the power-light ratio should be very close.
If MPPT works on one array. it should work on the other.  Or even if one controller wastes 50w in heat, that still doesn't cover the wide variation of output powers.   The strings in each array all match each other closely, so I've discounted that as an issue.

My settings
absorb  66.7V 45 minutes.

I have tweaked the meter on the classic, to closely match the Tristar readings
(Battery Voltage offset  -0.1V)  and only when many amps are flowing, do the meters drift apart.

MNGP screen  under CHARGE---T-Comp---View  and see what that is supposed to be.

I'll look at that tomorrow. but for now,   Local app shows this at  Compensation: 66.7 V   

[mike90045]

.....http://midniteforum.com/index.php?topic=2014.0
To me,  the configurations are not very similar and makes direct comparison difficult and perhaps not fair...

The cells are evergreen, just the arrangements in the panels and array sizes differ.  So, the power-light ratio should be very close.
If MPPT works on one array. it should work on the other.  Or even if one controller wastes 50w in heat, that still doesn't cover the wide variation of output powers.   The strings in each array all match each other closely, so I've discounted that as an issue.

My settings
absorb  66.7V 45 minutes.

I have tweaked the meter on the classic, to closely match the Tristar readings
(Battery Voltage offset  -0.1V)  and only when many amps are flowing, do the meters drift apart.

MNGP screen  under CHARGE---T-Comp---View  and see what that is supposed to be.

I'll look at that tomorrow. but for now,   Local app shows this at  Compensation: 66.7 V

Attached - sunny AM,
Tristar 73% of PV nameplate
Classic 71% of PV nameplate  So with great sun, these track pretty well, but it's less than perfect conditions that the classic looses it.

But this 2nd screen shot, shows the Classic with 145V input  0w output, while the tristar array is still in shadow and not lit up yet, but it's charging the batteries, a bit anyway.

[Vic]

Hi Mike,

The pics are a bit small  ...

On the second screen shot,  is the TS in Absorb??

Thanks,   Vic

[TomW]

Not being boots on the ground, I would suggest  swapping the arrays to the other controller to ensure the wiring is not being flaky. It would possibly throw some light on the problem and should be fairly simple to accomplish.

If the arrays are not exactly the same orientation and type you will not be doing a fair comparison as pointed out by another poster. Swapping the controllers may give a better idea what is going on.

Just a thought from the P-nut gallery.

Tom

[mike90045]

On the second screen shot,  is the TS in Absorb??

No, that's early AM,  it's in MPPT which is too short of a label to appear behind the local app window.

As to swapping the arrays, I can put the 130VOC array on to the classic 200, but I can/won't put the 160VOC array on the 150V tristar.    Since the melted buss bar in the ePanel incident, I've been keeping an eye and IR thermometer on all the connections,  and none seem to be flakey.

Changing modes this AM.  Solar = 0 watts  Legacy P/O gives me about 35w, and trying to play with UsetVOC%, I can't figure on where to set the % on the local app?   Going back to Solar, still gives me 0w
And 15 minutes later, it's up to 45W, while the tristar is at 133w  (Solar is still 0)

[mike90045]

some more screen shots.   
P&O  vs Solar vs Tristar & arrays (left array is 2kw/Classic, right is 3kw/Tristar)
  I'm still working on getting the images small enough to be uploaded, but still readable.

[phxmark]

I also noticed that Legacy P&O and Dynamic modes give better low light performance than Solar mode does.  Only drawback is that Legacy and Dynamic have slower tracking.  I used Legacy P&O with a setting of 12% and had great performance on cloudy days, but poorer performance on fully sunny days.

The MPPT algorithm in the Classics solar mode is excellent for full sunny days, but not cloudy days.

[dgd]

As to swapping the arrays, I can put the 130VOC array on to the classic 200, but I can/won't put the 160VOC array on the 150V tristar.     

Ok, just changing the 3Kw array over to the Classic what does the power output look like? 

dgd

[dgd]

I also noticed that Legacy P&O and Dynamic modes give better low light performance than Solar mode does.   I
The MPPT algorithm in the Classics solar mode is excellent for full sunny days, but not cloudy days.

A few posts have said something similar but I have never seen this on my C150 running in solar mode. Low light/cloudy performance is excellent and when I tried P&O/Dynamic there was no noticeable improvement.
My C250 is different, it suffers the dropping input voltage to just above battery voltage and staying there bug that Vic first pointed out some time ago.
Solar mode will not work on it but Legacy P&O is good.

dgd

[mike90045]

changing the 3Kw array over to the Classic what does the power output look like?

That will happen next week when we hope to have some sun.