random Classic time resets

[kitestrings]

Hi,

I occasionally find that the time (date actually) has reset on one of my Classic's.  Typically the time will be correct, but the date will be something wild, like "2000".  It is easy to reset, but is there something that can help avoid this?  Maybe a firmware update?  We're currently Classic Rev: 1401, Network Rev 1190, S/N# 0415 & 0418.

At one point someone suggested enabling the auto reset, which worked, but with wind it is not a good option as the reset can occur during a high wind and leave the turbine unloaded.

~ks

[boB]

KS, you''re having reset problems ?  Still ?

Well, I ~think~  this new upcoming software rev IS GOING to fix this !  That is, IF it is actually from
the software which apparently is the problem mostly.

Also, you don't have a Clipper ?  What happens when you batteries get full ?
You have battery side diversion ?

boB

[kitestrings]

you''re having reset problems ?  Still ?

Yes, but I have not done any firmware updates, in part due to fear of the unknown I guess, but maybe I should wait a bit for this next version?

We don't have a MS Clipper.  We do have a home-made clipper/load bank similar to the one Ryan did at his neighbors I believe.  We have six 6.8 ohm, 800w resistors; pairs in parallel wired in wye and controlled off aux2 on Vin -Hi via SS relay.  This is on the 3-ph source and intend to keep voltage limits safe to the CC's.

We also have a diversion load on the DC side.  This is a water heater fitted with 48V elements and controlled using "waste-not" off the second controller, Aux2, via SS relay.  This works great, especially in the summer when the PV surplus can be squirreled away.

Still having the "resting" followed by "OCP" issue.  We're pulling the thing down next month, and crazy as this will sound, I've got a pair of pulleys fitted so we can motor the alternator (once blades are off) from our EV.  That's right folks, "an EV powered wind turbine".  Well, really I was trying to just find a means of variable controlled speed so we can try to resolve the OCP/resting thing without smoking something.  I've been working with Ryan & Roy, but the current thinking is to disable OCP and try it under controller conditions.

Our bank is admittedly now a bit undersized.  It is 48V, 840 AH Trojan L16s.  And, we have a 15' axial now added to 3 kW of PV.  We plan to increase capacity next time batteries are replaced to get somewhere close to C/20 again.  The resting/OCP issue, however has always been at night in moderately high/variable winds.  Our thinking is that OCP is reacting too quickly (i.e. before the clipper can react enough to prevent this).

Open to suggestion if we're off track.  Sorry for the length response.

~ks

[boB]

OK, I'm sure we've talked about this before but some particulars again...

Classic genre ?  150, 200, 250 ?
As for OCP and resting, (not resetting), how is the Classic's battery wires wired to the battery
and how are your loads (inverter, etc) wired to the battery.  Usually, OCP will happen because
the Classic is electrically wired too close to the inverters' battery input and the batteries are
wired many many feet away from both the Classic and the other battery loads.

As for the resetting, yes, hang on.

One more thing, on your INPUT side dump load (I think you meant that it was on the input), is that
load connected DIRECTLY to the Classic's input ?  Or is it isolated by a diode or even better,
on the turbine side of the turbine rectifier diodes ?

boB

[kitestrings]

boB,

Classic genre ?  150, 200, 250 ?

It is 2- Classic 250s.  They are set up much as Ryan had done with his neighbors.  In our case the first one takes the first ~40A, then current limits.  The second one starts to load above that point.

Usually, OCP will happen because the Classic is electrically wired too close to the inverters' battery input and the batteries are wired many many feet away from both the Classic and the other battery loads.

Well the Classics are very close to the inverter input; right off the +pos bus bar in the DC panel.  This is #2/0 CU about 12" to the inverter +pos terminal.  The inverter is an OB VFX 3648 (2007).  It is close-coupled to a PS2DC breaker panel, and connected via 175A breaker to the batteries with #3/0 CU about 12-15'.  There hasn't ever been much inverter load when this happened, maybe 500 watts.

The rectifier outputs are on 50A 300V breakers; the Classic outputs are on 80A 150V breakers. Rectifiers & Classics are within about 3-5' of the inverter & DC panel.  At Roy's suggestion we moved the DC diversion load so it comes directly off the +Bat terminal on the bank; instead of the +pos bus bar.

The two Classic are on isolated full-wave stud mount diode bridge rectifiers.

It is resting that occurs.  In moderate/gusty winds when the first Classic is nearing full load, the unit will sometimes go to resting, -0- watts, in some cases reconnects, or tries two a second time.  The voltage shoots up, our clipper/resistors may see some load, and the second unit comes in (usually as the turbine rpm increases).  After it reconnects the OCP error appears.  I've sent a video of it to Ryan & Roy.  Can cc: you if it might help.  I've also sent schematics & load curves.

A frame by frame description is here:
http://midnitesolar.com/Forum/index.php?topic=2430.msg23313#msg23313

The dump load is on the 3-phase turbine side of the circuit.

Thanks for any suggestions.

~ks

[boB]

OK, KS, I see...   So the "OCP" that is occurring may or may not actually be responsible for the "Resting" actually...

When you see that Classic go to Resting, please go into the "secret" status screen from main status and let's
catch a couple of Reason For Resting (RFR) numbers at the top, in the middle.  I have a feeling that you may
also be seeing a RFR #9 which would be reverse current.  Or, something else....  But this might be
fixable by an adjustment or something.  I have seen the Resting on another double-Classic wind turbine
installation.

Pasting in how to find the RFR number...


From the main status screen hold the left arrow and tap Enter. The top middle number is the "Last" RFR

Remember this is over written the next time it goes to resting

boB

[kitestrings]

So the "OCP" that is occurring may or may not actually be responsible for the "Resting" actually...

I think Roy was thinking it was related to, or even in response to, Aux2 triggering the load bank.  I could be wrong, but I don't think so.  The last time I recorded it the voltage (Vin) was only at about 97V (1,717 watts), and then there was an audible “click”.  Classic A dropped out and went to “resting” mode.  The voltage then climbs very rapidly and aux2 engages within the next two frames of our video.

At 130V the load bank should be looking like a 2.5 kW balanced load (R_tot is 3.4 ohms), but I don’t think we’d have ever gotten to that voltage without the controller first (reacting to something, and) disconnecting.  If I intentionally lower the aux2 setting and tickle the load bank at lower voltages is seems fine.

I also tried a simple test where in light winds I moved the I limit way down to like 5A, but then also moved my Vin set-point (aux2) down to ~70 volts.  It did what I would have expected, which is hit the limit, the amber current limit light came on, and continued to hold current at that level.  This of course pushed the voltage up until the load bank came in.  So functionally the controller is fine, just have to figure out why it dislikes this at 100V.

We recently (02/26/15) increased the size of the wire between the inverter lug and the +pos bus bar (to now #2/0) thinking that was the bottleneck, but unfortunately it did not solve this situation.  And, we've seen with and without the load diversion in play.

let's catch a couple of Reason For Resting (RFR) numbers

I’d hoped to catch the last reason for resting earlier, but when I have to shut it down the last reason becomes low voltage (after I’ve furled the turbine).

Simulating the conditions are problematic right now because of the lack of wind, and fact that the turbine is only up 25' (which has been good for monitoring behavior).  As I mention earlier, however, we'll be taking it down soon and should have the ability to spin it to a (relatively) controlled rpm.

Are the adjustments you're thinking, a change to the system configuration, or something that involves internal, or software mods?

~ks

[kitestrings]

boB,

Usually, OCP will happen because the Classic is electrically wired too close to the inverters' battery input... I have a feeling that you may
also be seeing a RFR #9 which would be reverse current.

I wonder if would it be worth trying to move the Classic output leads, after the breaker, so that the termination is at the battery bank as opposed to the bus bars in the DC panel?  As in this illustration:

http://www.midnitesolar.com/pdfs/Ryans_stacking.pdf

Electrically it would seem to be adding more resistance - a pair of what ~#4 CU wires to the bank, vs. #3/0 - but if the Classic is somehow responding to the direction of flow maybe...

If the wind is meeting or exceeding the inverter input at any given time, I assume those electrons flow directly from the Classic(s) output; the path of least resistance?

[boB]

KS, could you please post a simple (or complicated)  drawing of how the Classics (mainly the inputs)
are connected together and how your dump load and rectifiers and turbine is connected ?

I am guessing here.  That drawing would be very helpful !

Thanks,

boB

[kitestrings]

This is my schematic (architectural), schematic (electrical) drawing.  I had this description from a year or so ago:

In normal operating mode, the charge controllers are attempting to first bulk, absorb and then float the batteries.  Once they are fully charged, or whenever we bump against the upper charge limits set, the voltage starts to climb if there is no place else for power to go.  Initially the load diversion comes into play.  Through the Aux 2 â€" Classic B (PWM relay signal) the Classic diverts, or burns off, excess available power into the water heater.  This is a Marathon storage tank with 48V DC elements (via SS relay).

When the batteries are full and the water heater is satisfied, the Aux 2 (PWM relay signal) starts to engage our homemade 'clipper' or resistor dump via the 'dump relay' in the rectifier box.  This one, unlike the diversion load, is on the 3-phase side of the rectifier.  It's mission is just to keep the voltage input into the CC's below safe limits (250VDC, 180VAC).

Okay, now worst case - the Classic or one of these components fails - we still have the Omron looking to limit voltage independent of the CC's via our 'hi-limit relay' (also in the rectifier box).  Otherwise, if the Classic(s) shuts down, we’ve now moved the problem to the tower.

The relay is in parallel with the Classic Aux 2, but if the Omron actuates it will both disconnect the turbine from the rectifier(s) and close in the load bank.  Rob Becker has done this with a DP relay, but the result is the same.  The diode shown allows the signal to operate from either 12V (from the Classic), or 24V from the Omron without a fight.

What's not clear, or not accurate, on the attachment is that the input for the Classics is from the +pos and neg bus bars in the DC panel.  #6 CU between the bus bars and the breakers, and also between the breakers and the Classics.

[kitestrings]

Attached are:
1) the DC panel,
2) physical layout, Classic A, B, inverter above (DC panel just out of sight to the right), rectifier box below.  The input breakers are in the rectifier box.  The batteries are directly below the floor in a 4'+ crawlspace.
3) the rectifier schematic

Don't laugh.  ~ks

[kitestrings]

Here's the dump load

[dgd]

Ks,

AN interesting setup. The images you posted are rather tiny so its almost impossible to see any detail. Can I suggest some over 100k images to give a more detailed/bigger view.
The use of the Omron contacter is different, I assume its one of those relatively expensive voltage set and detect types. I note you say its the backstop if other components (the Classics?) fail. As a mechanical device I would have thought the Omron was the most likely component to fail     (ok my experiences with Omron contacters is not that great)
But surely the only time the voltage would rise too high is when the Classics have unloaded the turbine because the batteries are fully charged and the water is hot enough. So why bother keeping the turbine spinning and just diverting its power to the dump load resistors?
Why not just turn off the turbine by using a 3phase SSR to short out the turbines 3 AC outputs?  rather than wearing out the tubine and burning up the resistors?
That diode separating the 12v output from the Classic AUX and the 24V control output of the Omron better be a decent type otherwise some smoke might get out of that Classic. Could you not get a 12v Omron?

dgd

[kitestrings]

Hi dgd,

Yeah, sorry about that - I hadn't posted pics here and didn't have a good sense of what it took to compress to a 200 KB file.  I will try again when I can.  Hopefully it may help boB at least see what were doing.

The Omron was not an original design.  Rob Beckers who frequents the OP board has used them in conjunction with his Aurora grid tie inverters for some time.  He claims this unit is pretty robust.  They are a user adjustable, voltage sensing relay good to 600V.  Unfortunately, They did not offer a 12V input; that would have been my first choice and I could see using 120VAC just for that.  I'll check the diode, but IIRC it is 3A 600 PIV.  I have both control inputs fused at ~.5A.  I could be wrong, but aren't the aux relays also protected for overload?

I think this resting/OCP issue is a good example of why the secondary protection is good.  We tried a number of things to resolve this, but when it occurs the turbine is unloaded until the load bank slows things down.  Fundamentally I dislike shorting the windings of an axial.  It is very harsh compared to the load bank.  It also puts heat where it is unwelcome IMO.  I have, however, considered power furling.

Turning off/shutting down the turbine is pretty easy when someone is here, but wind is far from predictable.  Our goal has been to have useful diversion loads where we can, and have the 3-ph load bank as the means of keeping the voltage safe to the CC's; much the way the MS Clipper does.  It can be set to contribute at a pretty conservative voltage.  To your point though, we routinely manually furl the turbine during the sunny days when we're not here.

Thanks for the comments.   Everything has worked well except for this resting/OCP thing.  Hoping to sort this out before the winds return in the fall.

[boB]

KS, your schematic looks just fine.   Should work great.

OK, so the "resting"...  Is that mainly the secondary Classic that is resting or both of them or the primary Classic ?

Also, I think we need to get some of those RFR (Reason For Resting) numbers after the Classic has actually
gone to Resting to see if it is the OCPs or maybe #9 (negative current)...

The one install I know of in Spokane Washington was pretty much the same as your but he was using
our Clippers with the two Classics.  It was his secondary classic that was having the problem.

What happens if you disconnect the secondary Classic while you are getting the OCPs and Resting ?

I wonder if your Clipper just Clips a bunch (because the system is over-powered) and no OCPs or
Resting happens then ?

Or, disconnect the Primary Classic (flip the input switch to disable either Classic)

boB