Local App background color

[kitestrings]

Happy Holidays.

We just had a pretty windy spell.  I was sitting here with the local app pulled up so I could watch it for a bit.  On one particularly strong gust I noticed the background screen color briefly changed (from the default green) to red.  Nothing else seemed out of the ordinary.  Aux 1 had triggered the water heater diversion, but otherwise I don't believe the input voltage or loading was out of the margins.  I will check the peak input (Vin) and high output, but my guess is maybe 85-100V, and ~1200-1400 watts.  There were no alerts showing.

I searched the board and saw one post where the red indicates an error, but there doesn't appear to be anything in the app manual explaining this (only the warning icons).

Any ideas?  Thanks, ~ks

[boB]

Not sure what's up with the red.   Maybe Andrew will chime in with an idea.

Could the Classic have gone into  current limit ?

[Halfcrazy]

Red indicates an error yes. Orangish indicates a warning like over current.

The big gauge labeled power would have an icon come up on it at the same time to show the error.

Unfortunately like the Classic the Local app does not log errors so it makes it very hard to trouble shoot.

[kitestrings]

Could the Classic have gone into  current limit ?

This was my first thought.  It's quite possible I've just never been watching it on the local app.  IIRC correctly at the Classic there is a yellow LED that lights when the current limit is reached.  This one logged a high Vin of 135.1V and a high output of 1,695w.  I didn't think to check the second Classic (there are 2-250s), but at 135V I would have expected Classic A to be fully loaded (45A), and Classic B to be contributing another +/-20A.

I've also on occasion noticed a red LED on the Classic, and briefly a "got comm?" message that appears - maybe this is unrelated.

What are the other types of errors that would explain this?

~ks

[kitestrings]

The big gauge labeled power would have an icon come up on it at the same time to show the error.

If it did, I didn't catch it.

[kitestrings]

Happy New Year.

Back home now. I checked the second controller (B). It logged a high of 832w, so yes, current limiting on (A) was likely.  High temp was only 35-36 degC. Everything else looks normal.

[kitestrings]

I've got another question.  When a Classic goes into current limiting mode, what does the display indicate for wattage?  Is there a relay action?

I'm still trying to understand a puzzling condition.  This time I was watching the Classic MNGP's during another windy period tonight.  Classic A was loading up normally.  There's a really big gust (one of several tonight).  The second Classic (B) is now set to wake up at 90V.  It clicked on.  The Aux1 light goes on, so I know we're now over 100V.  Then just as  Classic B starts to contribute, I hear the relay on A click, and the wattage display goes to zero.  Normally the only time there is the audible click is when the unit awakes, or the voltage has dropped and remained below cut-in for a spell.  I'm not sure why, and if it is, as it appears, dropping to -0- watts.  If so, the total output is shifting to B albeit it a brief event.

Any ideas?  ~ks

[atop8918]

On the original problem, Ryan nailed it -- The background color of the monitor window turns red or yellow depending on error or warning conditions, respectively. Unfortunately we don't log/latch these events though that would be very useful information. I will add this to my list of things to do and see if I can't squeeze another info page in the menus to show events like this if management approves.

[kitestrings]

Thanks.  There are not many folks using wind with the Classics, let alone multiple units on a homebuilt turbine.  One of the problems is these conditions are typically really fast, and it can be hard to capture all that's happening.

I gathered a bit more information from the data logs this morning.  It's dead calm this morning, -12deg F, clear head.  Classic A logged a high output of 2,672w and input V of 169.7.  Classic B logged 1,300w & 168.7V.  The voltage is significant, as I currently have my homemade clipper - a resistor bank on the 3-phase input from the turbine triggered off Aux2 - set to limit at 170V (PV V on high) with a 5V PWM width.  Here's what I'm thinking could be happening:

Fluctuating winds, within capabilities of Classic A, everything is normal.  This unit cuts in about 59V, the curve is pretty linear to step 15, 45A & 115V.  Step 16 is 46A & 116V.  The output current limit is set at 45A.

Big gust hits.  Classic A loads up.  The voltage climbs.

At 90V Classic B "clicks"; it's ready (Step 1: -0-A & 90V).

At 100V Aux1 actuates.  This just diverts a bit of power to water heating.  About 1,200-1,500w on the DC side of things (48VDC elements).

The voltage climbs.  At step 2 on Classic B (3A, 115V) the second controller starts to contribute power.  The curve is pretty linear from there to step 15 (46A, 163V).  Step 16 is 52A & 163V.

At 165V, however brief, Aux 2 is going to start pulsing the SS relay to the load bank.

At 170V Aux 2 would completely change states (12v+) and the effect is the equivalent of adding about 4.2 kW of load to the alternator.

All this is happening very fast.  The peak condition is very short duration.  I'm speculating on this part, but I wonder if the resulting load across the 3-ph line is dropping the input voltage very abruptly - about the time the relay gets busy, the rpms may already be dropping - and Classic A may be "clicking" out on low V input (>59V).

Otherwise, I can't explain why Classic A would be "clicking", or going to -0- watts, unless it happens during current limiting?

Thanks again for any thoughts on this.  ~ks

[boB]

Kitestrings,  if your  2 Classics are next to each other, I would take a short video of the 2 dancing together in this fashion.
It is much easier to see what is going on by playing this back and being able to step through and watch both displays
at the same time.

Your system is complicated by the Aux output telling things to divert power so it may be even more useful to
do a little video.  I'd love to look at it and help diagnose.  There is more than likely nothing wrong,
just that those gusts can make things change rather quickly.  It may also be that one of your classics
are seeing the battery voltage go up to or above the battery absorb/float set-point and at that point, it
must raise its input voltage to keep that battery voltage from rising any further by raising its input
voltage (unloading the turbine).

boB

[kitestrings]

if your  2 Classics are next to each other, I would take a short video of the 2 dancing together

They are; side-by-side.  Oddly, I was thinking this the other night.  I'll experiment with it...just need to get the lighting so it will be legible.  I agree seeing it all together, and having the ability to stop - or rewind would be a big plus.

It may also be that one of your classics are seeing the battery voltage go up to or above the battery absorb/float set-point

My absorb set-point is 59.2V; Float is 53.8V; EQ is 62.0V.  On the 17th (01/17 the high Batt reading recorded was 55.8V, but on the prior big blow (01/05) it was 59.0V.  The only thing is I don't recall if it was sunny mid-day, which could have been when that reading occurred.  On the original post, where the screen color changed, the high was only 52.4V.

There is more than likely nothing wrong

That's my hope.  I'm thinking there may be a tweak or two to either the power curves at the hand-off, or a change to the Aux settings that could make the transition smoother.

I'll try the capture the next event.  Thank you for your assistance.  It's much appreciated.

Kindly, ~ks

[kitestrings]

I still haven't been able to film this as I've been hoping to, but I stumbled on something important for wind mode users.  It was light bulb for me anyway.

Awhile back I'd had been having some occasional problems with the mymidnite data not updating.  I would get a message that says the Classic has not been connected to in over __-weeks, check network communications & consider a firmware update.  The local app had been fine, so I'd assumed that I needed the firmware update.  When asked about it someone suggested that enabling the auto reset (A-RST in tweaks menu) might solve this.  I tried it; and it did.

What I hadn't noticed was that the A-RST basically does a little reboot - perhaps boB, atop, Ryan can explain this sequence - where it temporarily disconnects from the input, and then "resets".  Watching it reminds me of a PV sweeping routine.  Anyhow, I'd been noticing the fan doing a little system check at midnight (I hadn't tied the two things being related).  Last night I watched it in modest winds as we struck midnight.  For about a half a minute or so the Classic disconnects, the input voltage continues to register, but the output goes to -0- watts.  This is not a good situation for wind.  I'm pretty sure this was the situation on at least occasion where, in higher winds, I couldn't understand CC dropping out.

I've disabled the A-RST, but thought others might want to note.  As Ryan suggested the communication issue can probably be resolved with a firmware upgrade if it persists, or I can on (windless) occasion reset.

Brutal cold here in VT.  Kind regards, ~ks

[kitestrings]

I had taken this one off-line to sort through with Ryan & Roy at MS.  I'd managed to film it happening on the one MNGP.  It was a big help to be able to break this down frame-by-frame.

Anyhow, the just of the problem I believe was due to an over-looked bottleneck in my wiring.  Everything is was properly sized between the Classics and the DC panel, and I have 3/0 CU to the batteries.  There is, however, a short (~10") wire that connects the DC+ bus to the hot side of the inverter breaker lug.  This had been sized for our original PV system, or 60A (48V).  When we added the wind, I overlooked increasing the size of that wire.  In high winds we are pushing 80+ amps through that bus.

I've change it over to a piece of 2/0 CU.  I also moved the water heater diversion to the Aux2 (of our second unit) as planned.  So now the water heater load can be eased into a bit, rather than all- or nothing with Aux1.

I want to again thank Ryan, Roy & boB for their help with this one.  You guys rock.

~kitestrings