How to prevent DC arcing ?

Thread Starter

Externet

Joined Nov 29, 2005
1,846
Hi all.
A DC supply of 400V conducting 10 Amperes; when unplugged/disconnected/switched off like at place ---->>---- below, can spark a considerable gap.

400VDC(+)------------------------------> >-------------------------------4KWload-------------------------------(-)

If a light bulb or a low-current of few mA resistive load is placed across the disconnection point --->>--- ; would it prevent arcing ? The bulb would turn on and there would be no spark? Any thoughts ? (Yes, the large load would stay somewhat energized)
 

Thread Starter

Externet

Joined Nov 29, 2005
1,846
4KW load is an inverter, 240VAC output with all sorts of household loads. PSU is a series of 10 x 40VDC solar panels like
1645647816653.png
 

MaxHeadRoom

Joined Jul 18, 2013
25,198
The big problem when using contact switching, especially on high voltage DC is the sustained arc when the contact opens and plasma is created which is hard to extinguish and can often sustain itself if allowed when the contacts are fully open!
The contact separation has to be very fast and as distant as possible.
Arc prevention, chutes, magnetic blow out etc. are generally used.
 

Thread Starter

Externet

Joined Nov 29, 2005
1,846
If a light bulb or a low-current of few mA resistive load is placed across the disconnection point --->>--- ; would it prevent arcing ? The bulb would turn on and there would be no spark? Any thoughts ? (Yes, the large load would stay somewhat energized)
What about a resistor across the point of circuit opening ?
 

Ya’akov

Joined Jan 27, 2019
5,657
The big problem when using contact switching, especially on high voltage DC is the sustained arc when the contact opens and plasma is created which is hard to extinguish and can often sustain itself if allowed when the contacts are fully open!
The contact separation has to be very fast and as distant as possible.
Arc prevention, chutes, magnetic blow out etc. are generally used.
Yes, but that switch is rated for 1000V at 25A, it's designed to break a high voltage DC circuit.
 

crutschow

Joined Mar 14, 2008
29,777
Since the arc is likely generated by stray wire and load inductance, why not use a 500-600V, 10A diode from the switch output (cathode) to ground.
 

Thread Starter

Externet

Joined Nov 29, 2005
1,846
Thanks, gentlemen.
The question is about avoiding the spark at breaking a contact/connection; perhaps I expressed wrongly :(
I do have the DC switch at post #6 in the circuit, rated for 1000 or more DC volts / 40 or more Amperes for when I want to turn off the load.

Please allow me to rephrase: all panels are in series. Each panel P is wired to the next. There is connectors C that joins their wiring.

Output (+)DC-------------P1-------C--------P2-------C-------P3--------C-------P4--------C-------P5--------C--------P6------------------(-)output

The series circuit can fail by open circuit at
- any connector C poor contact, pulled, breakage, corrosion.
- inside any panel P, a wire melts or a solder joint fails, or corrosion, or stone thrown, or a tree branch falls....

If across every P and across every C a resistor is added to ensure always some continuity at a open circuit failure event; what happens to the arcing if a C or a P fails open but the resistor survives fine in-circuit ?

1645669478462.png

(image borrowed from the web)
 

MisterBill2

Joined Jan 23, 2018
11,877
Since the arc is likely generated by stray wire and load inductance, why not use a 500-600V, 10A diode from the switch output (cathode) to ground.
NO!!! The arc IS NOT "generated by inductance", an arc is what happens as a DC circuit is interrupted. Itstrikes as the conductors separate. That is how DC current works. There are contactor relays that do utilize a magnetic blow-out with the magnetic field created by the current. Those devices are expensive and subject to contact welding. The solution is heavy contacts that open rapidly to a distance the arc will not jump. The heavy contacts are to reduce the amount of metal vapor that tends to stretch the arc. There are also systems that use a blast of compressed air to blow out the arc. THAT DOES work, but it is noisy.
 

Ya’akov

Joined Jan 27, 2019
5,657
Thanks, gentlemen.
The question is about avoiding the spark at breaking a contact/connection; perhaps I expressed wrongly :(
I do have the DC switch at post #6 in the circuit, rated for 1000 or more DC volts / 40 or more Amperes for when I want to turn off the load.
This is oldish, but seems like an excellent starting point. It seems that PV Arc Fault Circuit Interruptors (AFCI) are not a standalone item, though at one point they were. Instead, the AFCI is built in to the inverter. But there are some PV circuit breaker with AFDD (Arc Fault Detection Device) inputs, and Siemens makes an AFDD that is quite expensive.

My takeaway from the research is that if your proposed simple method worked, it would be mentioned somewhere and possibly adopted along with a rapid shutdown component. It's interesting though. It makes me wonder if the Tesla solar roof thingies have AFCI protection, though it appears that the NEC actually requires it.

 

DNA Robotics

Joined Jun 13, 2014
625
Electric railroads use DC contacts in something like a hydraulic cylinder. They use dielectric oil pressure to separate the contacts and quench the arc.
 

MisterBill2

Joined Jan 23, 2018
11,877
OK, NOW I see that the intent is to prevent any arcing at any random opening of the current path caused by any means. So the goal is to interrupt the circuit if any disturbance in the current that carries the signature of an arc is detected. If that is a correct understanding then a single arc-fault detection device at any point in the loop will protect the entire loop, as opening at any point will cease a series-arc. The effect of a shorted-segment arc within the loop is not anything that I can predict. And in addition, is not a failure that opening the loop would alter the effect of. And it was not until post #18 that the actual requirement was made clear. Up until that point we all presumed that it was an arc caused by intentional actions.

In addition, the extreme damage shown in the photo is due to a very poor installation. No PV array should ever be installed in a position where a catastrophic failure can ignite roof material. That installation is horribly unsafe!!
 
Top