I have a 220VAC, 20A rated switch and want to use it on a 12VDC, 3A circuit.
Is there a rule of thumb on what is the maximum equivalent DC voltage of an AC rated switch?
There isn’t always an equivalent, especially with high-power (>10A) switches and contacts. Different materials, contact coatings, lubricants, switch speeds and gaps, etc. are used for DC, AC, and low and high voltages.
In very general a voltage and current rating applies to all forms of electricity, but switching 50A of DC on a 220VAC, 50A switch might cause it to wear out or fuse eventually because of the non-compatibility of the contacts. I would never switch AC with a DC-rated switch, though.
I would convert the RMS value of 220 to its peak value of 340 volts. That’s the greatest potential that could appear across the switch contacts without exceeding its rating (as long as this switch isn’t going to be used in a rectifier or multiplier).
One of the critical issues is the notion of fault current. Whilst you are happy to consider the current under normal operating conditions, what is the worst case current that will flow if the switched device shorts out? If you were running it off a car battery the answer could be hundreds of amps. This would cheerfully weld your 20A switch closed and lead to real problems. This is one reason everything gets fuses in circuit. Even then the fuse might only blow after the switch is welded shut. Switches rated for use in AC settings (ie mains power) are designed and rated assuming they are installed in a system where there is proper (and typically solidly standardised) upstream protection. Thus they are designed with fault current capabilities assuming these standards. It isn’t easy to translate this into another setting.
The main problem when switching DC circuits is that there is a lot more arcing between the switch contacts. This isn’t a problem with AC because the arc self-extinguishes at the zero-crossing.
Here’s a video showing the difference: http://www.youtube.com/watch?feature=player_embedded&v=Zez2r1RPpWY
Correct.
The best answer is that you should contact the switch manufacturer and inquire on the switch’s DC specs. Or use a switch that is spec’d for both AC and DC operation.
If you’re looking for a rule of thumb, there isn’t one. That’s because it depends on a *lot *of things, such as the magnitude of the DC voltage, contact alloy, thermal mass of contacts, contact spreading velocity, and the gap distance between the switch contacts when they’re open.
This paper lists AC and DC current ratings for some toggle switches made by Microswitch. The data are in a table on Page 2. Interestingly, the current ratings at 28 VDC are slightly better than the current ratings at 115 VAC for each switch. So perhaps you’re O.K. But just keep in mind that these data are just for those switches manufactured by Microswitch, not your switch.
Oh, and regardless of what switch you use, and I would definitely include a snubber circuit to help reduce the amount of arcing across the contacts. A 1N4007 flyback diode, Trasorb, or Zeners across the load should do the trick if the load is inductive. An RC across the switch contacts can also be used in addition to the snubber across the load. Do some googling on DC snubber circuits for more info.
Apart from the general question, for this specific case is there any way that wouldn’t be OK? That’s a factor of almost 7 in current, and much more in voltage.
Couple things:
-
There is no “rule of thumb.” See my post above.
-
Just because the peak voltage is less, and the current less, does *not *automatically mean it is O.K. to use the switch. As an example, it would probably *not *be O.K. for him to use his switch at 110 VDC and 2 ADC, even though the peak voltage is much less (110 V vs. 308 V) and the current is less (2 A vs. 3 A).
This is one of those cases where I would look at how many cycles I expected the relay to last. If it was only a few thousand, I would just use it and not give it a second thought.
Agree, but for completeness sake it should be mentioned that in other applications it’s not just about longevity. Depending on the DC voltage, current magnitude, and gap width, it is possible for a DC arc to sustain itself after the contacts are open. Which is… obviously a bad thing. This won’t happen at 12 VDC, though, as long as the distance between the open contacts in the switch is greater than 4 mm or 5 mm. For a 5 mm gap, the minimum DC voltage to sustain an arc is around 27 VDC, and that occurs in the 10 A to 100 A range.[sup]1[/sup] At 5 mm / 3 ADC, the minimum DC voltage to sustain an arc is around 46 V. Not sure about gap widths less than 5 mm; I don’t have data for that. The minimum DC voltage to sustain an arc increases as the gap width increases, all else being equal.
[sup]1[/sup][sub]Electric Arcs in Open Air. A D. Stokes and W.T. Oppenlander. 1990[/sub]
He doesn’t have a 3A switch. He has a 20 A switch. That’s why I’m wondering. It’s not just “less voltage and less current” It’s “A whole lot less current and a whole lot less voltage”.
I’m suspicious of the DC voltage being 220 volts not 220 VAC x 1.414 (311 VDC), since the resistors heat up much faster and get much hotter on DC, which wouldn’t happen if they were equivalent. In any case, I know from experience that even voltages well below 100 volts DC can cause sustained arcs, with the drawn-out length dependent more on current than voltage.
On a related note, circuit breakers may not work properly on DC either - even ignoring potential arcing:
The breaker also emits smoke when it trips on DC, indicative of arcing burning away the contacts; standard size fuses (e.g. 5x20 mm) should be OK on common DC voltages; I often see them used in older TVs in the main power supply of around 120 volts, in addition to an AC fuse.
Also, depending on what you are switching and whether isolation is needed, I’d use a MOSFET to switch 12 volts at 3 amps, which can be both smaller and more reliable and use less power (effectively no drive power at low frequencies).
ETA: As for using switches on DC, if it is pulsating DC, like from a rectifier, it should be OK to use it up to its rated voltage since current flow isn’t continuous; I often see this in appliances like hairdryers which use a diode to reduce power on the low setting.
I am not qualified to speak at length of the differences between AC and DC.
For the general public: House (AC) wiring uses solid conductor. Automotive (DC) uses stranded
Stranded is more expensive and car manufacturers are noted for saving money whenever possible. AC runs cooler because of the peak/low cycle, so it doesn’t need large surface area to cool. DC has no such self-limited heating and requires the much greater surface area in stranded to keep cool.
Please do not think a device rated for AC iis suitable for DC - switches and relays especially.
10 amps AC will cause the same resistive losses as 10 amps DC; stranded wire is only used because it is more flexible, although for the same gauge stranded wire is thicker because the individual strands have space between them, so it can dissipate more heat for the same temperature rise. In fact, AC will cause higher losses if the frequency is high enough so that the skin depth becomes less than the radius of the wire, which is about 1/3 of an inch at 60 Hz for copper wire; this is why overhead power lines often use a multi-layered structure, usually aluminum (which is better than copper with regard to skin depth, and thus may actually have lower losses) over steel (a poor conductor but a cheap and strong substrate) and SMPS transformers use multiple strands of thin wire (at 100 kHz, anything larger than #26 is a waste of copper).
Another interesting tidbit is that, even at DC, stranded wire has slightly higher resistance (Ω/m) than solid wire for the same gauge.
Yes, I understand that.
Let me use another example. Let’s say you use his switch (rated for up to 220 VAC and up to 20 AAC) on a 30 VDC circuit that can produce up to 10 ADC. Used in this DC circuit, the peak current is 2.8 times lower than the switch’s peak current rating, and the peak voltage is 10.4 times lower than the switch’s peak voltage rating. So it should be O.K., right? Not necessarily. If the gap between the contacts is only 4 to 5 mm when the switch is open, then it is *possible *the arc won’t extinguish when you open the switch.
Again, there is no “rule of thumb.” If you want to use a switch or relay on a DC circuit, you should ignore the AC specs and *only *look at the DC specs. If they’re not available, contact the manufacturer or use a switch that has the DC specs you’re looking for.
Why use another example? Why not just use the OP’s case: 12 V and 3 A?
I’m pretty sure it would be safe to use the switch on DC in this case; just looking around, I have a switch rated at 16 amps at 250 VAC* and 6 amps at 30 VDC, which would translate to 7.5 amps for the OP’s switch (although I know you can’t just assume the DC rating is 25% higher if the AC rating is). Based on previous posts; e.g. Crafter_Man’s post (#10), I assume that almost any switch with the proper current rating will be safe on 12 VDC at 3 amps.
*Some switches will also have a rating for 125 VAC, which can be up to 2x the 250V rating, presumably also for arcing reasons, more for reliability in this case (AC can also sustain arcs if the voltage/current is high enough, as in this example, around 2 kV at 1 amp).
That’s interesting. But for welding, they seem to use HF (doesn’t that imply AC?) to ignite arcs. Isn’t that contradictory?
AC current and voltage goes to ZERO 100 times every second (at 50Hz) or 120 times (at 60Hz). When the voltage or current is ZERO, it is much more difficult to start up across an open switch. This means that ratings for AC are completely different and unrelated to DC ratings.
So there is your rule of thumb: There is no relation between AC ratings and DC ratings.