An unlikely issue - current from solar is out of control

[harryn]

Hi,

I realize that the official specs for the classic are 96 amps output, but this makes some assumptions as well.

They design that unit assuming stationary use, regularly checked highly torqued down screws, and perhaps not particularly high cabinet temperatures. It isn't actually all that easy to achieve and maintain the spec torque.

If you then go back and look at the actual screw terminal and see how it can just barely accept 4 awg wire (and not all 4 awg wires, just solid and low strand count), in my mind, it starts to paint a picture of thinking 80 amps max output vs 96.

I attempted to get them to make some versions of their products with actual bolt down ring lugs but didn't succeed.

If it were me, with the solar power levels that you have, I would use two controllers and two breakers rather than one.

Magnum makes a slightly higher power single unit, (PT-100) but it is still marginal for your application. The wire connections are better.

Victron makes great inverters, but I am not nearly as impressed by the charge controllers. It seems like they use their inverter based brand strength to sell the companion items.

[CountryFit]

i forwarded a few photos and followed up with a call to midnite tech support. i recommended that

(1) enlarge the terminals to accept 2ga or 1ga wires. (your idea of flat lugs even better).
(2) replace the small/ineffective cooling fans at lower part of the control boards behind the terminals.

my feeling was they are not going to do anything - as replied 'we have been doing this for 60 years...' (well, i certainly understood his/their proudness of 60 years in business, but i found hard to corrolate the years and the room to improvement).

what i have done and am doing -
(1) rimmed the terminals with 2ga wires in (mentioned in previous posts)
(2) got 4 longer cover retaining screws and put 1/2" spacers between the cover and the casing. this opened up the interior to ambient air. (my battery bay is sealed and no gassing issues so no worry about the dirts getting inside the controller).
(3) upon arrival of the thermo switch ordered, i am going to install two more cooling fans at the very bottom of the box; connected to "aux 1". see fans below.
(4) will move over one of the paralelled panels to another array (need to get a 200v controller).

when all of these are done i'm pretty confident my system will perform fine.

[CountryFit]

after done the additional fans, it has been running at 96.1a for the past hour. the temp appeared having been subdued.

but the current keeps hitting the ceiling. i still need to do more to fully utilize the power.

[harryn]

In case it is useful, on the midnight web site there is a "end users supported forum". Mostly it is visited by contractors doing home and business installations.

There are some discussions there on how to slave multiple classics together if that is ever needed.

Officially MS personnel don't support the forum but you see some posts there occasionally.

If you are looking for more ways to utilize the excess generated power, I seem to remember that there is a way to divert some to hot water heating but not 100% certain.

Interestingly, your challenge of too much production might go away as the outside air temperatures rise. Panels produce less power as they get hotter (typically Vmp drops).

Places like S CA this time of year can be relatively cool but also have a high UV index - so lots of power.

I guess in theory one way to solve the problem short term would be to disconnect one panel from the string and tape off the terminals. Sometimes easier said than done.

[harryn]

BTW - I like your improvements.

I made some copper pins that fit into the terminals and will crimp into 2 awg high strand count wire for one project.

It is tough to make the 90 degree bend you have in there though.

[CountryFit]

Thanks Harry!
Good idea on the Midnite enthusiasts' forum. I'll dig a little there.

[CountryFit]

somehow i over-read this question. my eyeglasses really need to be replaced .

Originally Posted by Smitty77

So both pair of solar panels comes together on the inside of the breaker?

my combiner box for 8 panels is on roof, there is only one pair of 6ga wires for this array coming down to the battery bay.

another pair is for 3 panels in series, which does not go through a combiner box and gets in battery bay directly.

each array has its own controller and breakers.
-----------------

an update -
i had a chance to communicate with an engineer who participated in the design for midnite classic controllers. he confirmed that it won't work with multiple controllers in between a single array of panels and the battery.

[harryn]

Originally Posted by CountryFit

somehow i over-read this question. my eyeglasses really need to be replaced .


my combiner box for 8 panels is on roof, there is only one pair of 6ga wires for this array coming down to the battery bay.

another pair is for 3 panels in series, which does not go through a combiner box and gets in battery bay directly.

each array has its own controller and breakers.
-----------------

an update -
i had a chance to communicate with an engineer who participated in the design for midnite classic controllers. he confirmed that it won't work with multiple controllers in between a single array of panels and the battery.

That is correct - you would have to break the larger array into two independent arrays to run two controllers.

[CountryFit]

update -

remember i had two arrays - one had 8 panels all in parallel at 42v; another had 3 panels in series at 126v. now, i moved one panel from the group of 8, to the group of 3 to form a 2 x 2 at 84v.

the result is better, but not without shortcomings - the large one got 88.3a, still has a room of 7.7a to the ceiling of 96a; the smaller one reached at 818w, almost hitting the ceiling of 60a (846w equals to 60a at 14.1v).

so i will need a bigger controller for the smaller array, but no hurry...

(notice for 4 panels i got 818w, that's 81.8% efficiency. but 7 panels i only got 1234w, that's 70.5%... big difference! why? those are identical panels, no shades, sampling at the same time... the only difference - one has higher input voltage, another has lower...)

[kenn_chan]

take a soldering torch and tin the wires together solid...now you can take a file and work it down to the size needed to fit into the receptacle without drilling on an expensive piece of equipment... an as far as the former nuke....what you expect? a bunch of snark and "I'm a nuke" are about typical fair.

[CountryFit]

Originally Posted by kenn_chan

take a soldering torch and tin the wires together solid...now you can take a file and work it down to the size needed to fit into the receptacle without drilling on an expensive piece of equipment... an as far as the former nuke....what you expect? a bunch of snark and "I'm a nuke" are about typical fair.

i like the idea of tinning the wire ends, that'll effectively increase the contact areas, thus reduce the heat generation.

[harryn]

Originally Posted by CountryFit

update -

remember i had two arrays - one had 8 panels all in parallel at 42v; another had 3 panels in series at 126v. now, i moved one panel from the group of 8, to the group of 3 to form a 2 x 2 at 84v.

the result is better, but not without shortcomings - the large one got 88.3a, still has a room of 7.7a to the ceiling of 96a; the smaller one reached at 818w, almost hitting the ceiling of 60a (846w equals to 60a at 14.1v).

so i will need a bigger controller for the smaller array, but no hurry...

(notice for 4 panels i got 818w, that's 81.8% efficiency. but 7 panels i only got 1234w, that's 70.5%... big difference! why? those are identical panels, no shades, sampling at the same time... the only difference - one has higher input voltage, another has lower...)

If you look carefully through the midnight solar manuals, while the solar charge controllers "can operate" over quite an input voltage range, they are more efficient at some input voltages than others.

The simplistic explanation is - Ratios.

For reasons that I don't pretend to understand, if the voltage ratio (input vs output or vice versa) is less than 10:1, things tend to be more efficient than if they are more than 10:1.

With an input voltage of 130 ish volts / output of 13 volts you are at roughly 10:1.

With the reduced input voltage, it is now more like 7:1 and the result is - better efficiency.

It is possible to make them equivalent - but at about double the components cost.

Inverters do the reverse - trying to hit a peak voltage (at the top of the wave) of nearly 170 volts. This is much easier to accomplish with 24 or 48 volts feeding in vs 12, but the designer have gotten pretty good at it due to market demand.

[CountryFit]

Harry,
Sorry for the late response. I just saw your post. I am learning new things everyday.
Big thanks!!

[baphenatem]

Originally Posted by CountryFit

i like the idea of tinning the wire ends, that'll effectively increase the contact areas, thus reduce the heat generation.

This is not a good strategy in most cases as quite a number of terminals are not designed for solid conductors. There have been a lot of failures in industrial power cabinets of late due to installers using ferrules on stranded conductors. Victron even goes so far as to specify the minimum number of strands in the cables to be connected to their terminals.