Hello,
I work with, design, build & fix small electric vehicles. (36VDC, 48VDC & 60VDC e-bikes, electric scooters, go karts etc.)

1.) I am interested to know, for ELV (extra low voltage) systems (75VDC max), what is the minimum amperage that is needed to create an arc?
...enough of an arc to damage the contacts in a switch.

I ask because the On/Off circuit for many of the speed controllers (ex: Yuyun YK31 48V 1,000W), for little PM brushed motors, operates at pack voltage.
(the amp draw for this circuit is ~15mA)
…& the brake signal circuit for these speed controllers also operates at pack voltage too.

2.) So, to switch these small current (ELV) circuits, do I need to use switches that are specifically DC rated?
Most of the DC rated switches I have found are rated 250V AC/DC ~15A which, to me, seems "overbuilt" just to control a 48VDC ~15mA circuit.
...or could/should your average/standard (250VAC 10A/125VAC 16A) AC rated switches be able to handle up to 54VDC & up to 25mA?

Thanks,

 

An arc is created across switch/relay contacts are often due to switching inductive devices where collapsing BEMF occurs, causing a 'plasma' arc.
In some cases this causes destruction or welding of the contacts.
Switch and relay contacts are often rated at resistive loads.
Relays that switch high voltage DC or high current/inductive loads often have an arc blow-out method such as a P.magnet adjacent to the contacts.
Max.

 

1.) I am interested to know, for ELV (extra low voltage) systems (75VDC max), what is the minimum amperage that is needed to create an arc?
...enough of an arc to damage the contacts in a switch.

It's quite possible you should be worried about the opposite scenario. Standard silver contacts will oxidize over time and they actually rely on a small, reasonable amount of arcing to break through the insulating oxide layer. When you run very low voltages and currents, there may not be enough arcing potential to keep the contacts clean. Gold, or gold plated contacts are one common solution to this problem. Sealed reed relays and mercury switches are other possibilities.

 

As Max said, you are switching motors, which are highly inductive .It's very likely that this is the cause of your premature switch failure. Old style electric window circuits in vehicles were similar and switches were always a problem.
The underlying theory is: when a switch is turned off the magnetic field in the (Solenoid, Armature, Stator etc. etc ) winding collapses as the source sustaining that field is removed. The voltage generated in a circuit is related to the length of the conductor ( determined by number of turns and diameter ,length etc.) x the strength of the magnetic field x the speed that that field moves through the windings. - See: Faradays laws of electomagnetic induction & Lenz's Law. Now the field will collapse a lot faster than the normal change due to the rotational movement of the armature of your motor, thus a much much higher reverse voltage is generated than the standard system voltage. This is known as back EMF. which can approach 90-100V in the kind of motor you describe. This voltage is sufficient to ionise the air in the switch contacts as they open and may sustain a spark up to 1.0 mm or so. This ionisation will quickly burn contacts. In addition this will also tend to transfer ions across the switch contacts leading to pitting and deposition - much like a mini electric arcwelder.
The solution is to provide an alternative path for the induced voltage by either using a suitably rated diode reverse biased (See "Freewheel Diode" ) or a capacitor across the contacts, if a diode is not possible e.g.due to directional switching in the circuit.
I hope this goes some way to explain the phenomenon. You could try try doing a search for "Back EMF" & "Electromagnetic Induction" in the articles on this site.

 

As Max said, you are switching motors, which are highly inductive .It's very likely that this is the cause of your premature switch failure. Old style electric window circuits in vehicles were similar and switches were always a problem.
The underlying theory is: when a switch is turned off the magnetic field in the (Solenoid, Armature, Stator etc. etc ) winding collapses as the source sustaining that field is removed. The voltage generated in a circuit is related to the length of the conductor ( determined by number of turns and diameter ,length etc.) x the strength of the magnetic field x the speed that that field moves through the windings. - See: Faradays laws of electomagnetic induction & Lenz's Law. Now the field will collapse a lot faster than the normal change due to the rotational movement of the armature of your motor, thus a much much higher reverse voltage is generated than the standard system voltage. This is known as back EMF. which can approach 90-100V in the kind of motor you describe. This voltage is sufficient to ionise the air in the switch contacts as they open and may sustain a spark up to 1.0 mm or so. This ionisation will quickly burn contacts. In addition this will also tend to transfer ions across the switch contacts leading to pitting and deposition - much like a mini electric arcwelder.
The solution is to provide an alternative path for the induced voltage by either using a suitably rated diode reverse biased (See "Freewheel Diode" ) or a capacitor across the contacts, if a diode is not possible e.g.due to directional switching in the circuit.
I hope this goes some way to explain the phenomenon. You could try try doing a search for "Back EMF" & "Electromagnetic Induction" in the articles on this site.

That would all be true for switching the motors themselves, but the thread starter claims to be wanting a switch which only controls the driver circuitry, with a current draw of only 15mA. A switch in that situation shouldn't be dealing with back emf from the motors.

I think we might need a schematic of exactly where the switch is meant to go so we can make sure we're all on the same page.

 

Yes, we are talking about a control circuit.
It just to turn the style speed controller On or Off. (aka key switch or power lock)
Below, is an old (rough) diagram I drew up a while back & also a wiring definition.

It's NOT that I ever had any switches fail or want to use the "wrong" components.

Many, no most "Chinese" electric scooter/e-bike parts do not have any actual ratings.
…& it's really hard to find any or specs anywhere for a specific part.
Even the Wuxing (I've seen this name/brand many many times) brake levers that come on factory made scooters don't have any visible ratings.
…& the brake lever is another control circuit that is pack voltage but, also very low draw & uses a switch.

That's why I was asking, Mainly wondering if this was something I should be concerned about.
Hence: What is the minimum amperage needed to create an arc @ 60VDC?

 

Still interested in this topic
So to test, "What's the minimum amperage needed to create an arc @ 60VDC?"
...one would simply need an adjustable power supply capable of 60V?

Then, set the voltage at 60V & the amperage at 0.01A (~7.81mA) & start testing?
...maybe using a couple of pieces of ~20g. wire attached, as expendable electrodes. (as to not damage the actual electrodes)
…& simply touch them together & observe for an arc? (in the light & also in the dark, just to be sure)

Then, do a few tests at 0.02A (~15.63mA)
...& then at 0.03A & then at 0.04A (~31.25mA) etc.
...until a strong visible arc is created?

Then, keep testing to find out how much of an arc is needed to damage/pit or melt the end of one of the 20g. "expendable" conductors?
…& this should indicate what the minimum amperage that is needed to create a damaging arc @ 60VDC?,
…

 

I know that some of the DIY, home made EDM's use 60VDC at a low amperage to cut metal with. Amperage's in the 5 amp and up range.

 

I thought your carts were for personal use: https://forum.allaboutcircuits.com/threads/iec-320-c-14-receptacle-questions.155692/#post-1342794

In any event, can you define the difference between sparking and arching? DC switches are particularly prone to sparking/arching at quite low amperage.

 

Switches have ratings. If they're rated for 60 volts it means that at voltages up to 60 will not likely cross arc (electricity jump from one part of the circuit to another). As long as you don't exceed the rated voltage then the switch is not likely going to present a danger to the electronics. Imagine if you apply 100 volts to a 60 volt rated switch - and say it's a DPDT (Double Pole Double Throw) switch. On one of the poles is control power and on the other is a signal wire. IF power can jump from one circuit to the other it can present a potentially hazardous voltage to your sensitive electronics. So voltage rating is important.

Switches also have amperage ratings, meaning how many amps they can conduct. You can take a switch rated for 250 volts and 10 amps and use it for any voltage that doesn't exceed the voltage and doesn't exceed the amperage rating of the contacts. If you exceed the amperage rating then you burn up the switch.

In order for us to help you it's important to understand a few things: You ask about how much current (amperage) is needed to make 60 volts arc. If the switch can handle 10 amps then you can assume any arcing at or less than 10 amps should be no problem. Well, ALL switches arc. You might not be able to see it - but it happens. Especially with inductive loads (as others have said). So there's no magical answer that I know of to answer your question about arcing.

 

You can take a switch rated for 250 volts and 10 amps and use it for any voltage that doesn't exceed the voltage and doesn't exceed the amperage rating of the contacts.

Not really. A 250V AC switch won't last long on DC of a lower, say 100VDC circuit. AC/DC switches always have a much lower DC rating than the AC rating. Or at least any I've ever looked at were that way.

 

A 250V AC switch won't last long on DC of a lower, say 100VDC circuit.

Agreed.

Without going into ALL the possibilities, the gist of my statement is to find a switch rated for what you plan on doing with it. Yes, DC is harder on a switch than AC. Hence, the reason for sticking with the ratings.

 

I thought your carts were for personal use: https://forum.allaboutcircuits.com/threads/iec-320-c-14-receptacle-questions.155692/#post-1342794

In any event, can you define the difference between sparking and arching? DC switches are particularly prone to sparking/arching at quite low amperage.

Yes, my karts are/have been for personal/family use.
So, I want to be sure their safe for me & my kids to be around &/or use.
...but, also if I ever sell any of them (I've already designed & built over 10 small EV's & am running out of space)
...I want them to be safe for others to use too.

*For the record, I have no formal electrical education
...but, I do lots of research, am self taught at most things & have been working with & testing these small EV's for many years now.

A spark (or spark breakdown) is a brief event where an insulating medium (often air) is electrically stressed sufficiently to cause it to break down. Once this occurs, a conducting filamentary region of the insulating medium changes into a low impedance conductor. This event is usually accompanied by a sharp click or bang, since the mechanisms associated with the formation of the spark also creates a shockwave in the surrounding medium.

Once spark breakdown occurs, the voltage across the gap plummets, and current surges through the spark discharge. If the surrounding circuitry has a limited amount of stored energy (such as a charged capacitor), the spark is quickly extinguished afterwards. However, if the power supply can continuously supply sufficient current, the spark rapidly evolves into an arc and current continues to flow. The diameter of the arc channel is a complex function of the available current and maximum arc length scales roughly with available supply voltage. A spark is a transient event. An arc is a longer duration event that can often be treated as being in local thermal equilibrium (LTE). Note that you can also create an arc using a lower voltage source (typically 10's of volts) by briefly touching one electrode to the other and then separating them to form an arc, as in arc welding. http://www.gozuk.com/forum/what-is-the-difference-between-a-spark-and-an-arc-499476.html

Interesting note:
"The first continuous arc was discovered independently in 1802 and described in 1803 as a "special fluid with electrical properties", by Vasily V. Petrov, a Russian scientist experimenting with a copper-zinc battery consisting of 4200 discs."

Switches have ratings. If they're rated for 60 volts it means that at voltages up to 60 will not likely cross arc (electricity jump from one part of the circuit to another). As long as you don't exceed the rated voltage then the switch is not likely going to present a danger to the electronics. Imagine if you apply 100 volts to a 60 volt rated switch - and say it's a DPDT (Double Pole Double Throw) switch. On one of the poles is control power and on the other is a signal wire. IF power can jump from one circuit to the other it can present a potentially hazardous voltage to your sensitive electronics. So voltage rating is important.

Switches also have amperage ratings, meaning how many amps they can conduct. You can take a switch rated for 250 volts and 10 amps and use it for any voltage that doesn't exceed the voltage and doesn't exceed the amperage rating of the contacts. If you exceed the amperage rating then you burn up the switch.

In order for us to help you it's important to understand a few things: You ask about how much current (amperage) is needed to make 60 volts arc. If the switch can handle 10 amps then you can assume any arcing at or less than 10 amps should be no problem. Well, ALL switches arc. You might not be able to see it - but it happens. Especially with inductive loads (as others have said). So there's no magical answer that I know of to answer your question about arcing.

The way I understand it, there must be sufficient current (amps) available to even create a "spark"
...or (by extension) if more/constant current is available an "arc.
…& "if" a spark or an arc is created, "it's" size (damage potential) would be dependent on the available voltage & current.

So, my theory is:
If an arch cannot be created at 60VDC until there is at least "hypothetically, lets say" 1A, then
...there should not be any "potential for damage" to (pretty much) any switch, when switching circuits of up to 1A.

So, my question is to find out/establish a minimum current level needed to even establish an arc & to maybe help confirm my theory.

This is Damien, one of my latest creations.
It's a "work in progress" & when done, it will be powered by a 48V 1,800W brushless motor.

 

So, my theory is:

My theory is, why are you fussing so much over this? Use a switch rated for the job and be done with it.

 

No fussin', just want to know more
...so I can do, what I do, better.

Yes, to use properly rated switches is plan A.
...but, there are a couple of reasons I am interested in this.

1.) As I mentioned in post #6
"Most "Chinese" electric scooter/e-bike parts do not have any ratings. (actually listed on the components)
…& it's really hard to find any specs anywhere for a specific part.
Even the Wuxing (I've seen this name/brand many many times) brake levers that come on factory made scooters don't have any visible ratings.
…& the brake lever is another control circuit that is pack voltage but, also very low draw & uses a switch."

2.) Most DC rated switches I have found (Digikey & Mouser) are rated for 72VDC ~15A (~$10.00)
https://www.mouser.com/ProductDetail/633-JWL21RA1A
…& to me that just seems like "overkill" for switching/controlling a 60VDC circuit that's only carrying ~14.5mA.
(If my math is correct 14.5mA is equivalent to .02A or 1/64 of an amp)

 

I said it once before, but I'll reiterate. At the very low levels of current that a logic IC draws, you're more likely to have a problem with not arcing enough than problems with too much arcing.

I'm sure my explanation won't be quite right, so you should research this yourself, but it has something to do with the contact material. I believe most switch contacts rated for higher currents (maybe 1A or higher) are silver. The silver oxidizes readily unless it's hermetically sealed in a very dry environment, and the oxidation prevents good conduction across the closed switch, so you end up having high contact resistance, in some cases even appearing as an open circuit.

In most applications, the small amount of arcing that occurs in switches *which is perfectly normal and OK* cuts through the oxidized layer and allows for a good, low resistance contact. When switching very low currents at low voltages, there isn't enough arcing to cut through the oxidation, so you end up with bad contacts over time.

For this reason, special switches are designed for low power applications with gold (or gold plated) contacts which aren't prone to oxidation, and therefore don't rely on arcing to ensure proper operation.

I may not have the exact details above correct, but I think the general concepts are right. The part I'm really unsure of is whether or not 60V is high enough to mean you don't need gold contacts, despite the low current you'll be dealing with. I don't know if it's just current, or current and voltage that determine the material choice.

The key points here are that arcing isn't always bad, and over-spec'ing a switch for higher voltage and current than you need can actually backfire!

 

When designing for a 60 volt circuit, engineer a safety margin of 1.3 to 1.5 times the minimum requirements. 60 volts at 1.3 times is 78 volts. 72 volts is close enough and you shouldn't encounter any problems.

When designing for a 10 amp system, engineer in a safety margin of 1.3 to 1.5 times the minimum requirements. 10 amps at 1.3 times is 13 amps. At 1.5 times that's 15 amps. Your 72 volt 15 amp switch would be just fine.

When designing a system, engineer in a safety margin of 1.5 to 2 times the minimum wattage. If your system draws 600 watts you want to plan on 900 to 1200 watts.

Follow those practices and you'll never have an issue. Even if you don't know the ratings on Chinese switches, the math is not that hard. Since you're talking about a control circuit it's very likely you'll have very low currents; I believe someone mentioned 20 mA. Have you determined that 60 volts is being supplied to your control switches? I would not expect to see that much voltage. It's probably regulated down to or below 12 volts. My thoughts is that it is likely going to be 5 volts. A switch that handles 20 volts at 2 amps might be good to go. On the other hand it might be too big. If it IS 5 volts, those common tactile switches (tiny push button type) are likely going to be fully sufficient for the job.

Bottom line, you need to know what voltages are present at your switches. Your controller handles the high current for the motor. Switches and potentiometers are likely low voltage low amperage.

 

Like the others ahead of me have said. The only switch that should ever see the full power in this application is the main switch. And that should not really see any load if the control is not active when it is being switched. The controller is made to vary the speed and reverse if included in the controller. All of that is done by semi conductors not analog switches. So that means your throttle and other switches going to the main controller are only seeing a very low voltage and next to no amperage, the controller is doing the heavy work.

 

When designing for a 60 volt circuit, engineer a safety margin of 1.3 to 1.5 times the minimum requirements. 60 volts at 1.3 times is 78 volts. 72 volts is close enough and you shouldn't encounter any problems.

When designing for a 10 amp system, engineer in a safety margin of 1.3 to 1.5 times the minimum requirements. 10 amps at 1.3 times is 13 amps. At 1.5 times that's 15 amps. Your 72 volt 15 amp switch would be just fine.

When designing a system, engineer in a safety margin of 1.5 to 2 times the minimum wattage. If your system draws 600 watts you want to plan on 900 to 1200 watts.

Follow those practices and you'll never have an issue. Even if you don't know the ratings on Chinese switches, the math is not that hard. Since you're talking about a control circuit it's very likely you'll have very low currents; I believe someone mentioned 20 mA. Have you determined that 60 volts is being supplied to your control switches? I would not expect to see that much voltage. It's probably regulated down to or below 12 volts. My thoughts is that it is likely going to be 5 volts. A switch that handles 20 volts at 2 amps might be good to go. On the other hand it might be too big. If it IS 5 volts, those common tactile switches (tiny push button type) are likely going to be fully sufficient for the job.

Bottom line, you need to know what voltages are present at your switches. Your controller handles the high current for the motor. Switches and potentiometers are likely low voltage low amperage.

Yup, I agree.

Definitive info & a logical path/practice to follow, I love it! Thank you.

I figured for a 60VDC system the top charge voltage of (5) SLA (sealed lead acid) batteries would be ~66.5VDC
(5 x 13.3VDC = 66.5VDC) plus a 10% (6.5VDC) = 73VDC

So, yes everything I work with is in the ELV (extra low voltage) up to 75VDC, range.

I just used 60VDC in the question because, I hoped someone on this forum would have a 60V adjustable power supply (which I don't have)
...& could easily perform a simple test like this.

Yes, I have tested the circuits that I am working with.
As for voltage, on most of these systems, when switched (Off) all plugs show 0 VDC, except the on/off plug (labeled "electric lock") & the Charge Port plug (they both show pack voltage)

Then, with the controller switched (On) the Battery Indicator plug showed pack voltage (of course), the Indicator Light plug, the Brake Light (lever) plug & the Brake Light plug, all showed pack voltage (~64.5VDC)

The Throttle plug, Reverse plug & the 3-speed plug are the ones that showed 5VDC.

The On/Off "Electric Lock" circuit is the only one I have done an Amp test on. (so far)
...just to see what was flowing thru it.
At first nothing was showing on the meter (must be below .2A) so, I changed the setting to 200mA & then, got a reading of 14.5mA.

Then, just to see, I did a "tap/strike" test on the, "splayed" (spread out into individual strands) bared ends, of the (2) 20g. wires
...but could not see any spark or arc.
...even turned the lights off & still nothing (this is kinda what led me to the no arc no damage theory)