Stop and sweep MPPT

[stephendv]

I'm the happy new owner of a Classic, and firstly want to say what a solid and full-featured piece of kit it is!  Did a firmware update on day one and currently have the local app running on my mac desktop, no other charge controller would let me do that.

The classic has replaced a Morningstar MPPT, and one thing I noticed is that the classic seems to do a stop-and-sweep where the controller stops charging for about 18 seconds while presumably it does it's MPPT sweep.  The morningstar didn't appear to do this (I say "appear" because I'm just going by what they put on the display not measured output) - it seemed to do a dynamic MPPT that didn't interrupt charging.

Is this stop-and-sweep behaviour something that is here to stay or can it be changed to a dynamic MPPT without charging interruption?

[boB]

What mode are you in? No it should not be stopping for 18 seconds at all.  Maybe half a second but certainly not 18 seconds In regular solar mode.

Maybe it went to resting?  If so, in the main status screen, hold down the left arrow button and tap the enter button and report back the top middle number.  That number is the reason for resting.  But 18 seconds is still a long time for it to come back on.  Maybe there is some reason that it's going to resting if that is what this problem is related to.

[boB]

Could you try something for me please?  Update the code again (just the classic) but this time wait until the 100 percent progress black DOS screen goes away on its own in about one minute.
I have heard reports and have to verify that sometimes the code is not finish updating until you wait for that screen to go away on its own....  Or, just wait 1 minute.  I have to look closer at that and see what's going on.

I am assuming that you stopped the update after you saw the 100 percent progress.

boB

[stephendv]

It's in "Solar" mode.

When I did the update I waited for the message that comes after the 100% complete message.  It took a while to appear after the 100% message, and when it did I closed the window.  So I'll do it again and wait for the window to close on it's own accord.

Thanks!

[boB]

It's in "Solar" mode.

When I did the update I waited for the message that comes after the 100% complete message.  It took a while to appear after the 100% message, and when it did I closed the window.  So I'll do it again and wait for the window to close on it's own accord.

Thanks!

Great.  Thank you!   But if it still stops and goes to resting, check that reason for resting number so we can see better what is going on.

boB

[stephendv]

Performed the update again and waited for the window to close on it's own.
I'm still seeing the same behaviour where it stops charging for a few seconds, then starts up again.  The resting reason was "5 Low light", even though Volts input read 101V.  Screenshot attached.

[boB]

Performed the update again and waited for the window to close on it's own.
I'm still seeing the same behaviour where it stops charging for a few seconds, then starts up again.  The resting reason was "5 Low light", even though Volts input read 101V.  Screenshot attached.

OK...  NOW the question is...  What is the open circuit voltage of the PV input ??  (Voc)   I am going to guess that the 101 volts
you are seeing is 101 volts and the Classic is not (for some reason) bringing that input voltage down to V max.

Try this.   Make sure the mode is OFF so it is resting.  Turning the mode to off or powering the battery breaker off will make sure
the Classic doesn't try to come on.  Then, with the top cover off, check and/or push the control PCB down so that it is
definitely latched down and solidly connected.   You can look to make sure the nylon standoffs (white) are holding the
control board to the power board.

We found one case where they were not solidly pushed together and evidently the slight movement when the USB cable
was plugged in moved things just enough so that the control board did not tell the power board to change the
input voltage.  If this is the problem, it will most likely be on the top end of the Classic PCBs.  (opposite end of
the USB and main power terminal block)

boB

[stephendv]

Yep, Voc is around 101V and Vmp around 80V.  I tried what you suggested, turned it off at the battery disconnect then pushed the board down, checked the standoffs and they seemed fine.  Turned the controller on again and it's still doing the same thing, it happens with regularity, about every 8-9 minutes. 
I hear a click, the mode goes to "Resting: 5 low light", the V goes to Voc and the amps go to 0.  Then another click a few seconds later and the V goes to Vmp and it starts charging again.
I've got a data logger attached to it, so you can see what's going on, available in real time here: https://cosm.com/feeds/75889

[WizBandit]

I see your data @ around 8:30 and I'm assuming in the morning. There is not a lot of power available then and I found my Classic 200 would do this if I had "LOWMAX" enabled. Even my 3Kw east facing trackers don't make much power this time of the year at 8:30AM.  Just a thought.

[boB]

OK....    I think I know what it is now that I know the Classic is controlling the input voltage OK....

When you updated the firmware, I don't suppose you did a "Vulcan Mind Meld" to re-initialize
variables ??   This is where you hold the LEFT and RIGHT arrow keys down as the MNGP (remote)
is re-booting (power up) until you see the selection screen  SOLAR--WIND--HYDRO come up.

Then pick your mode (SOLAR in your case), nominal battery voltage (it should guess the proper
voltage), adjust the Absorb, Float, EQ voltage, then set the clock.  The MNGP will then send
a bunch of modbus data to the Classic which will initialize some newer modbus registers
that may very well not have been in the old firmware.

One of those registers is the low watts that the Classic will start Resting at if the Classic isn't
producing at least that value.  I have a feeling you just need to do this re-initialization.

TO re-boot, just power down and power back up again.  This will reset any custom
battery voltage set points to factory default but it isn't working correctly now anyway.

Please give this a try.   The new registers should be documented in the latest and
greatest modbus document coming up soon.   (I hope soon, anyway).

We need to stress this VMM process to be done after certain (most ?) updates for
this reason.

Sorry for all the trouble.  I think this might just be it though.
boB

[boB]

I see your data @ around 8:30 and I'm assuming in the morning. There is not a lot of power available then and I found my Classic 200 would do this if I had "LOWMAX" enabled. Even my 3Kw east facing trackers don't make much power this time of the year at 8:30AM.  Just a thought.

Yes, that could also be it but I think that Stephen has been having this problem during the day as well.

Voc of 100 and Mpp of 80V is just about perfect...   80% of Voc is very typical for MPP voltage.

boB

[stephendv]

When you updated the firmware, I don't suppose you did a "Vulcan Mind Meld" to re-initialize
variables ??   This is where you hold the LEFT and RIGHT arrow keys down as the MNGP (remote)
is re-booting (power up) until you see the selection screen  SOLAR--WIND--HYDRO come up.

Woooooot, we have a winner 
20 minutes since the VMM and no dips.

[stephendv]

Many thanks for you help boB!!!  Glad this beauty is working nicely now.

[Vic]

...  Well  I also have been seeing some unexpected Resting during the day.  When I did the updates here,  did NOT do the VMM procedure,  which I will try.   One controller has been behaving a bit differently than the other,  each with "identical" PV arrays on a single batt bank,  and each with 6-29 FW.

From the traffic here recently,  had thought that I should re-update each cc with its current 6-29 FW,  but make certain to wait after the U/D for all processes to complete.  Now appears that the VMM should be done as well.  Am going to that site today,  so that should not be a big deal.

I did see an early Absorption termination in the Log of one CC which had a periodic Reating situation.  I had wondered if this Rest might have caused my EA criteria to have been met ( 00 A resting current met 15 A EA criterion).

Will do the updates and report any change.  Thanks for the sleuthing.  Vic 

[boB]

Will do the updates and report any change.  Thanks for the sleuthing.  Vic

Excellent !!!

That's why I let these discussions continue on until we resolve the issue or give up.
It may help others.  I especially like the thought process of elimination and dialog
that goes along.  I notice that other companies a lot of times will just have the
customer call customer support.

Of course the Classic should just be smart enough to take care of itself but as I
remember, in this particular case, the numbers in those uninitialized registers are
somehow close enough to values someone might want to use that it doesn't
always automatically catch it.  Still fixable by being just a bit smarter in the code.
We'll get there.

Let us know how it goes, Vic.

boB