Standard accessories used with DC

[sparkticus]

Yes, I understand that. What I was 'struggling' with is the high degree of 'voltage de-rating' for DC. Once ionisation has occurred, is the arcing not primarily current-, rather than voltage-dependent - or have I got that wrong?

Yes, I understand that. What I was 'struggling' with is the high degree of 'voltage de-rating' for DC. Once ionisation has occurred, is the arcing not primarily current-, rather than voltage-dependent - or have I got that wrong?

That is correct for part of the process. Ionisation essentially presents a low resistance path since there are plenty of liberated electrons and ions which act as charge carriers.

Now I don't have any numbers easily available at the moment so I'm going to make them up (at the risk of being attacked for doing so ) but if you had a 5mm switch contact gap already open then you might require let's say 2000 volts to jump the gap (to breakdown the gas between the contacts) Once ionised then you may only require 200 volts to keep the arc going (again, these are not real values but they are not unrealistic either)

That very same principle is used to start florescent lamps.

Now, if you have a closed switch contact (let's say carrying 100 volts DC with a load current of 50 amps. When the switch is opened the following process tends to follow

1) The amount of contact area is reduced slowly (slow in relative terms)
2) The reduced contact area will increase contact resistance and therefore volt drop across the contacts.
3) As the contact area is almost open the residual contacting material will start to get very hot (and ultimately flash off- look up wetting current) This part is a function of the load current.
4) As that residual hot metal finally disconnects, an arc is struck utilising a combination of electrons from the thermionic emission process and electrons from the heated gas at the contact site.
5) Now the ionisation has a higher resistance than that of the metal-metal contacts so the circuit resistance is going to increase making the low load resistance less of a factor. At this point the potential difference and the final contact gap will be the major factors for how long the arc will persist.
Well the contact material will factor significantly too.


It has been a very long time since I looked at such events so I am sure I have missed something but in general that is roughly the process.

There is also a problem of stored energy in inductive parts of the circuits but that's for another day.

Finally if the voltage applied were AC then the arc would almost certainly extinguish at the first zero crossing.

So it is the voltage which is used quite often to derate an AC switch for DC applications. However, in my personal experience AC switches do a good job in many DC applications.

 

[sparkticus]

PS, on reflection I recognise that I have meandered a bit in my attempt to explain above. However I think I eventually attempt to answer your questions

 

[JohnW2]

.... At this point the potential difference and the final contact gap will be the major factors for how long the arc will persist.

I guess that's the bit which most directly answers my question; thanks.

So it is the voltage which is used quite often to derate an AC switch for DC applications. However, in my personal experience AC switches do a good job in many DC applications.

I must say that I have rarely used 'AC switches' to switch high DC currents. Over the years, I've often used standard ('AC') 'electrical switches' to switch DC, but it's generally only been low current, and most commonly ELV - and certainly had no problems.

It would obviouly be nice if manufacturers did more testing and then produced more helpful ratings. It's all very well to say that those octal relay contacts are rated at 10A 240V AC or 10A 28V DC, but I presume that, per your argument, the usable DC voltage for, say, currents of 0.5A or 1A would be appreciably higher than 28V.

Kind Regards, John.

 

[JohnW2]

PS, on reflection I recognise that I have meandered a bit in my attempt to explain above. However I think I eventually attempt to answer your questions

Yes, it did help - and your edited version was an improvement. The bit in the original version I liked was that which read something like:
"the following two things occur : (1)... (2)... (3)... (4)... (5)..."

Kind Regards, John.

 

[sparkticus]

PS, on reflection I recognise that I have meandered a bit in my attempt to explain above. However I think I eventually attempt to answer your questions

Yes, it did help - and your edited version was an improvement. The bit in the original version I liked was that which read something like:
"the following two things occur : (1)... (2)... (3)... (4)... (5)..."

Kind Regards, John.

Yes, mathematics was never my strong point

 

[SimonH2]

One thing no-one seems to have picked up on ...

As I read the description, there is the AC supply and dimmer plus DC supply which goes to a contactor acting as automatic changeover. After this, there is then a grid switch for the individual circuits.

So if someone switches the lights off, the DC won't do anything and so you've no emergency lighting ? It would seem to me that this aspect of the system needs some reconsideration, although other than that it does seem a neat arrangement and avoids having a pile of luminaires with separate batteries to be maintained.

Actually, I do have an idea for dealing with that. Use changeover (two way) switches, and arrange the wiring so that when the lights are switched off, they are actually connected to a third pole on the changeover contactor (ie wire the lights to Common, 1 to the normal supply, 2 to the third pole). When mains is present, this third pole has no supply and so the lights can be turned off, but on the DC side it's connected to the battery - thus when on battery, the lights are on whichever way the switches are set.

As an aside, my brother spent several years on the Anastasis with mercy Ships. Some of that time was spent as an electrician.
I did a couple of short spells (couple of weeks only), and it was an interesting experience with 50 year old Italian electrics - which incidentally were a work of art. Originally all DC, and open switchgear with a grab handle to stop yourself falling into the open sparky stuff if the sea's a bit rough At some point, some of it had been converted to AC - luckily all the original CBs and switches were AC/DC so mostly they just swapped out a DC generator for AC and connected most of the hotel loads to a new AC distribution board.

Anyway, the emergency lighting was 240V DC - just a bank of batteries, and a subset of the lighting connected to it. Some had local switching, but mostly all the corridor lighting (or "lucio corridio" on the CB tallies) was just never switched off.

 

[JohnW2]

Actually, I do have an idea for dealing with that. Use changeover (two way) switches, and arrange the wiring so that when the lights are switched off, they are actually connected to a third pole on the changeover contactor (ie wire the lights to Common, 1 to the normal supply, 2 to the third pole). When mains is present, this third pole has no supply and so the lights can be turned off, but on the DC side it's connected to the battery - thus when on battery, the lights are on whichever way the switches are set.

That sort-of reminds me of my very first house. It had only only one lighting circuit and all the one-way light switching was actually done with two-way switches. The third terminal (say L2) of all the switches in the house were joined together (lamp going to common and power to L1). Switches by the front door and main bed, and also a relay operated by an intruider alarm, connected that 'house-wide' line to L. In that way, if one entered the house, or woke up, and 'heard noises', or if the alarm was activated, all the lights in the house would come (or stay) on, regardless of the position of their switches.

Kind Regards, John.

 

[RF Lighting]

The system that is installed in the theatre will not operate if the circuit breaker has tripped, or if the grid switch is turned off.

In normal usage the system is operated by skilled persons, so the grid switch will be turned on, but the light turned off via the dimmer.

The whole system is non compliant, as none of the wiring is fire proof or even fire resistant.

There is a proper system of etap self contained fittings throughout the building, and the

system is actually redundant, but it's pretty cool so has been left in for posterity

 

[SimonH2]

That sort-of reminds me of my very first house. It had only only one lighting circuit and all the one-way light switching was actually done with two-way switches. The third terminal (say L2) of all the switches in the house were joined together (lamp going to common and power to L1). Switches by the front door and main bed, and also a relay operated by an intruider alarm, connected that 'house-wide' line to L. In that way, if one entered the house, or woke up, and 'heard noises', or if the alarm was activated, all the lights in the house would come (or stay) on, regardless of the position of their switches.

Neat idea.