Disconnecting panels under load

[rha9]

I’ve read that panels should not be disconnected under load, and therefore should be covered with a towel before disconnecting in daylight , e.g. to store or reposition. I would appreciate advice from more experienced users on this point. Is it a problem, and if so is there a better approach than covering with a towel?

Thanks, R. Andrew

[boatpoker]

Originally Posted by rha9

I’ve read that panels should not be disconnected under load, and therefore should be covered with a towel before disconnecting in daylight , e.g. to store or reposition. I would appreciate advice from more experienced users on this point. Is it a problem, and if so is there a better approach than covering with a towel?

Thanks, R. Andrew

Disconnect between panel and controller - no problem
Disconnect at battery - PROBLEM

[goboatingnow]

Why is it a problem disconnecting at the battery. ?

[HeywoodJ]

Panels get disconnected under load all the time. It is the primary protection feature in power plants. The code requires that the disconnecting not cause damage. The risk is more to the disconnecting means, it must be designed to disconnect DC under load and quench the ensuing arc. Disconnecting by unplugging an MC4 connector under load will damage the connector. Disconnecting with a properly rated DC load-break switch not a problem.

To Boatpoker’s comment, the risk of disconnecting at the battery side is similar to the risk to alternators - the collapsing field in the controller power electronics might cause damage when their output is disconnected under load.

[goboatingnow]

Again why. An alternator is a very specific case ( And one that alternator manufacturers could easily solve if they bothered )

In my case there is no such stipulation , I can disconnect anywhere I like

An mppt controller is essentially a switched mode power supply , there is no issue disconnecting its output.

[skipmac]

Originally Posted by goboatingnow

Again why. An alternator is a very specific case ( And one that alternator manufacturers could easily solve if they bothered )

In my case there is no such stipulation , I can disconnect anywhere I like

An mppt controller is essentially a switched mode power supply , there is no issue disconnecting its output.

Well before I added my panels and MPPT controller, neither the manual nor the internet told me I couldn't disconnect at any point so I did a few times at the battery and so far nothing blew up or caught on fire.

[team karst]

Arcing contacts doesn't do them any good. Especially with DC circuits. DC doesn't have that nice waveform that goes thru zero Amps many times a second. Its that zero crossing that helps AC switches blow out the arc.
So, its best practice to switch off loads at the lower current branches, rather than at the source of much higher currents. So, if there is no big current, then it doesn't matter where you shut it down. If there is a large current, even the larger Batt switches will eventually complain about it.

This is why 125V DC rated circuit breakers are so expensive.

Here is a question. How does China Grid and Hydro Quebec shut off their 1MV
and 735kV DC transmission lines?

[goboatingnow]

Ah here turning off a 50A or 100A solar feed ( assuming you have 1200W ) is no biggie. Arcing might occur but the switch is built to withstand it. You’re not looking for 100,000 operations in 50 secs

[team karst]

Oh. Solar panels, not power panels. In that case, i’ll have start over. Thinking here about a towel over an electrical panel to fix an arc flash issue. 🤯

I disconnect my SOLAR controllers from the house batts sometimes. No ill effects. Even with current flowing.

[donradcliffe]

Most controllers instruct you to connect the battery first. Some will withstand ignoring that instruction, so will not. I looked for the controller I cooked when the battery got unplugged, but I have binned it. It showed serious signs of overheating.

[HeywoodJ]

Originally Posted by goboatingnow

Again why. An alternator is a very specific case ( And one that alternator manufacturers could easily solve if they bothered )

In my case there is no such stipulation , I can disconnect anywhere I like

An mppt controller is essentially a switched mode power supply , there is no issue disconnecting its output.

Yes, and most SMPS have a flyback system to prevent the damage that might occur when disconnecting under load. Serving inductive loads this is much more a problem than straight resistive loads. The mechanism is nearly identical to what happens in alternators, however, any decent power supply can deal with it. As could any reasonably designed alternator with a couple of extra dollars worth of components. The problem comes in with cheap/poorly designed systems where the flyback isn’t present or isn’t adequate. Hence “might”.

In all cases, much better to just tell the controller to shut down than to disconnect on either side, that way you don’t have to disconnect under load.

[goboatingnow]

Originally Posted by HeywoodJ

Yes, and most SMPS have a flyback system to prevent the damage that might occur when disconnecting under load. Serving inductive loads this is much more a problem than straight resistive loads. The mechanism is nearly identical to what happens in alternators, however, any decent power supply can deal with it. As could any reasonably designed alternator with a couple of extra dollars worth of components. The problem comes in with cheap/poorly designed systems where the flyback isn’t present or isn’t adequate. Hence “might”.

In all cases, much better to just tell the controller to shut down than to disconnect on either side, that way you don’t have to disconnect under load.

It’s simpler then that , an alternator is a big inductor , smps isnt, especially modern high frequency.

But I agree with what you say. A few bucks and alternators could easily design out the disconnect issue

[skipmac]

Originally Posted by team karst

Here is a question. How does China Grid and Hydro Quebec shut off their 1MV and 735kV DC transmission lines?

Don't know if they still do it this way but years ago a friend that worked with HV systems told me they used switches that were armed with some kind of cartridge, like a shotgun shell (without the shot of course) and the switches were blown open. The shell blew the switch open so fast that it eliminated or at least reduced the arcing.

[Steve_C]

Most Distribution and Transmission Lines use OCB's (Oil Circuit Breakers) to open and quench the Arc. In some cases the switches open inside of gas. It all depends on if the switch is design to open under load (load-Break) or just open circuit voltage.

[team karst]

At around 100 kV ac and above, SF6 breakers have dominated modern installs.
I’ve not worked with the EHV dc systems directly. But, I suspect faults are managed on the ac/ pre-rectifier areas. Breaking a million volts dc at many tens of thousands of amps is really difficult.