Arc fault at dawn and when power cycled.

[Mainewoods]

Hello,

I just started my new system yesterday and I am working through a few bugs and fine tuning.  I have two classics running two separate arrays.  One of the classic's gets an arc fault at dawn and if the input breaker is power cycled.  The other classic does not seem to have this issue and is also charging at significantly higher (2-3 times) amperage - both the arrays are the same size, although they are at different angles.  I have checked the cables and not found a loose connection.  Maybe the breaker?  Any ideas?  I am hesitant to adjust the sensitivity until I confirm that I don't have a problem.

Thanks,
T

[Westbranch]

With the arrays not being aligned in the same direction there WILL always be a difference until each array gets the same amount of solar insolation....  it should happen once each day and then the leader becomes the follower and visa versa..

hth

[Mainewoods]

Thanks for your reply Westbranch.

Sorry, I should have mentioned that both the arrays face exact true south.  The only difference between them is the mounting angle and my planning calcs show no more then a 20% difference between the two from winter to summer.

Any idea on the arc fault?  This has happened two more times today (both seemingly without reason as far as I can tell).  I did adjust the sensitivity from default 10 to 11 on both classics and have not had a repeat yet.

[Vic]

Hello Mainwoods,  welcome to the Forum.

Congratulations on getting the new system on-line.  This is always a fun to do.

Regarding the Arc Fault,  when a breaker is switched,  on  a circuit that is powered,  there will be some arcing.   Why is it that you are switching the PV input breaker near sunrise?

It is always a very good idea to check and re-torque all cable connections after about 24 hours,  as  there is some relaxation of the torque at wire connections that use screws (this is often called Cold Flow).  Even some connections with lugs might relax a bit  --  particularly on all lug connections on the batteries,  as lead is a fairly soft metal.

Arrays with differing azimuth and elevation angles  will produce differing amounts of power at various times of the day.  Maximum solar production requires the sun to be at right angles to the front surface of the PV module.   This is probably most of the difference in PV production that you are seeing from your system,  with identical PV arrays.  BUT,  it is very important to re-check ALL connections in the system,  for proper torque.

Vic

[Mainewoods]

Thanks for the welcome and reply Vic.

I was pretty excited to put it online after months of work (running on gen) and overall I am happy with everything except the batteries.  I bought 24 used batteries and weeded out a few bad ones until I was down to 16 of the best, did a lot of testing charging and equalizing but in the end the capacity is still too low.  New batteries are in order - once I figure out how I am going to pay for them.

As for the arc fault, I did not switch the breaker at sunrise, I switched it off later in the day to check all of the combiner fuses and power from each set of panels - all was well from the array and fuses.
I did adjust the sensitivity again after a few more faults, now at 12 and it seems to be okay, no faults so far.

I will be sure to re-torque all of the cables tonight.  I used to be a project manager for a yacht outfitter and this was a common problem on yachts as well.

The charge controller outputs seem to be relatively arbitrary, I have noticed it switch from one array to the next without a sun angle correlation as far as I can tell.  I am thinking that we have not put more then a 30% load on the system yet and so the classics have not been working very hard and are switching the work between themselves.  At full array output each classic should be around 4-4.5kW and the highest I have seen was only 3.5kW - the other classic was around .6kW at the time.  I don't think I need to spend much time thinking on this, I will see how it goes once our workshop is up and running and the system is under heavy load.

Again, thanks for the welcome and reply,  I am sure I will be around quite a bit.

T

[Halfcrazy]

So the fact it is arc faulting and lower on production than you expect makes me think there is a bad connection some where. If I where trouble shooting I would swap the classics but that can be a pain. The other thing you can do is look for heat rise on breakers, mc4 connectors and connections.

Curious where in Maine you are? and has it stopped snowing yet today?

[Mainewoods]

Halfcrazy,  Thanks for your reply. 

I just found and got my infrared thermometer working yesterday - perfect timing.  I should be able to use it to check the MC4's and all the other connections for that matter.  I will also try an amp clamp on each panel set when I get a good sunny day. 

We are near Farmington (Central Maine).  We did not get much snow today but had 2" yesterday and 3" the day before, it has not melted yet.  It was great to see the snow slid off the winter angle array (60deg) yesterday morning about 1 hour after sunrise - just as hoped.  The summer array (30deg) has not cleared yet, I think the warm temp (20-30F) allowed the snow to freeze to the array, time will tell.  It is beautiful here.

[mike90045]

Nice shedding array,  but what is happening with the left panel that did not clear ?

[TomW]

Nice shedding array,  but what is happening with the left panel that did not clear ?

Good question.

From here my guess is it is not connected or not loaded somehow.

I would check on it to see why it is not pushing any current when the others clearly are. My experience is a panel will show signs of melting when it has current flow. That one appears to be at the same temperature as the roof behind it.

I would definitely advise MaineWoods to investigate this apparent anomaly. Every watt helps, especially in Winter.

That is a nice looking array. Too nice to have unused modules for sure.

Can't see the picture now but it seems there were unused modules on the hip roof above those with snow on them? Maybe  one of that array and may be unused for some reason?

Tom

[Mainewoods]

Mike and Tom,  Thanks for your replies.

I have been trying to figure it out too.  It's both the far East and far West panels on the lower array.  I would also suspect that they were not connected, except that these panels are wired in pairs, side to side (13 strings per array).  So the adjacent panels to both the mystery panels are connected and producing power.  Making this even more puzzling.  We just finished insulating the attic today, the picture is post insulation and the far West and East panels are above the gable eves by about 1'.  So, I am guessing (and hoping) that the heat from our woodstove helped clear the other panels first and that the snow stuck on those two because they were over the end walls.  I am keeping an eye on it and will let you guys know after the next snow. 

[Mainewoods]

Just wanted to update in case anyone was interested. 

After insulating the attic the panels in question now shed snow about the same time as the rest of the array.

The arc fault has gone away after the adjustments listed in these posts. I never found a loose connection but it could be a MC4 that is out of reach.

I have seen production as high as 4.8kW on each array when I have sufficient loads to pull it - the arrays are rated 5kW, so I am very happy with that too.

My new Rolls S-550 batteries should be here this week and then I should be home-free:)

Thanks everyone for all of your help.