Why most Dig Switches Will Fail

[PunkRC]

Hopefully this will help people understand why may Dig Switches are failing over time.

In designing the PunkRC Dig Switch, I noticed that many people are using switches rated at 12A and more on each motor. This are first seemed like major overkill until I learned that the failures were being caused by contact failure due to switching an inductive load (motor).

Here is how to compute the required switch rating:

Battery Capacity (mAh) / 1000(mAh/Ah conversion factor) / Run Time (hrs) / Number of Motors = Current per Motor (Amps)

So a 4000mAh battery on a Dual Motor berg that will run for 30 minutes full throttle is:

4000 / 1000 / 0.5 / 2 = 4 Amps per Motor

So why does anyone need a higher rated relay or switch? The property of inductors is that they resist any change in current because they actually STORE energy in the form of a magnetic field. So you cannot instantly turn off an inductive load. The current flow will not stop instantly, it can only decay.

Here is the WILD part. If you disconnect an inductive load, it will develop ANY voltage required (infinite in theory) to maintain the current. (Think of this a Water-Hammer for electronics)

Here is how the math works.

Voltage (V across motor terminals) = Inductance (Henries, property of the motor) x Change in Current (Amps) / Time (sec).

What happens if I instantly turn off (not possible) a 4A motor? So assume all motors are rated at 1 Henries.

Voltage = 1H x 4Amps / 0sec = Infinity Volts

What about more realistic 1 ms Switch on the same motor?

Voltage = 1H x 4Amps / 0.001sec = 4000 Volts!!!

Here is the problem, it is not any current that is melting the contacts, it is the thousands of volts that is burning the contact material. This is why turning off a stopped motor causes no damage.

So what is the solution? Not bigger switches or relays. A 10A switch is more than adequate. You need to suppress this voltage spike that will destroy your contacts.

Capacitors will not fix this, they will actually add to the problem if used alone. What is needed is a surge protection circuit. These are designed to SHORT any voltage spike above a certain threshold. The protection I’ve used on the PunkRC Dig Switch (www.punkrc.com) are will short any spike above 32 Volts. So no voltage above 32 Volts will flow through my relays which are rated at 12A @ 36VDC.

So, unless you run surge protective circuit you will still be doing damage to any switch, even a 100A switch. It just becomes a matter of WHEN not IF an unprotected dig switch will fail.

EDIT BELOW: Added more info from below!

Originally Posted by Greatscott
Having the dig switch shunt the motor to a ground when the motor is disconnected should fix this problem, you'd want a diode in there to ensure electron flow was one-way and you'd want a couple of resistors in line with the diode to control electron flow.

No insult, but I don't think what you are suggesting, as I understand it, would work. I am hope this will provide more info. (No trying to hurt feelings, just provide info)

In the fraction of a second after the motor has been turned off, but before it was shorted, there is a huge voltage that will arc dua to the inductance of a motor. The switch cannot solve this it has to be another component.

Using a diode would work, except the motor has a reverseable polarity, so shorting the current through a forward current in a diode will not work. It would be ideal, but since the load is both polarities, something else has to be done.

The circuit that I use on my Dig Switches do ALMOST the same thing. Only they act like a SHORT to any voltage above 30Volts any polarity. That way, the contact voltage is always legal as less than 35VDC which the switch is rated for.

Since these relay switches are operated within spec, they are rated at "20,000,000 operations at 18,000 operations/hr", they will last alot longer. In english that means "Good luck wearing it out"

If any switch operated with arcing voltages, depending on the contact material, I'd give them a MUCH shorter life (alot less the 20million).

The arc supression circuit i used is also rated at 150A and 1 Joule. So the use of resistors on this one is unneeded. But, if one was to use capacitors to supress arcing, they will need resistors to bring the capaticance into (closer to) phase. A capacitor alone as a noise filter is not only worthless, but also more damaging.

[ohmyCLOD!]

So, will it not make a difference whether you switch the positive or negative lead as far as limiting the arcing effect?
I only ask because i have read several posts stating that switching the neg will cause less damage to the switch.

[vavaroutsos]

No, I can't see how switching the negative lead would make a difference. You still have the energy in the magnetic field of the motor that is going to develop a large voltage when the switch is opened causing the contacts to arc and burn. You need some type of snubber circuit across the switch to limit this voltage and dissipate the energy from the magnetic field. PunkRC is correct that a capacitor across the switch is not a complete solution. The voltage/current through inductors and capacitors is 90 degrees out of phase. Current leads voltage in a capacitor and lags in an inductor. Without some resistance in the circuit to dissipate the energy as heat, it will oscillate back and forth between the cpacitor and inductor. There are other types of snubber circuits in addition to the RC snubber previously described.

~petev

[EGRESSor]

http://www.crouzet-usa.com/downloads...ch_catalog.pdf
there are some ac/dc inductive/resistive ratings
and some tips to prevent arcing (page 41)

[70duncan]

Very nice post PUNKRC . This should help people understand what happens when you switch on and off current when it is under a load. Also, I have been looking at your DIG switch for a while now. You did a great job on the design.

[Greatscott]

Great write-up!!!


Having the dig switch shunt the motor to a ground when the motor is disconnected should fix this problem, you'd want a diode in there to ensure electron flow was one-way and you'd want a couple of resistors in line with the diode to control electron flow.

I think this is just about as complicated as electric RCs get, but I still think it is easier than nitro .

[PunkRC]

Originally Posted by Greatscott

Having the dig switch shunt the motor to a ground when the motor is disconnected should fix this problem, you'd want a diode in there to ensure electron flow was one-way and you'd want a couple of resistors in line with the diode to control electron flow.

No insult, but I don't think what you are suggesting, as I understand it, would work. I am hope this will provide more info. (No trying to hurt feelings, just provide info)

In the fraction of a second after the motor has been turned off, but before it was shorted, there is a huge voltage that will arc dua to the inductance of a motor. The switch cannot solve this it has to be another component.

Using a diode would work, except the motor has a reverseable polarity, so shorting the current through a forward current in a diode will not work. It would be ideal, but since the load is both polarities, something else has to be done.

The circuit that I use on my Dig Switches do ALMOST the same thing. Only they act like a SHORT to any voltage above 30Volts any polarity. That way, the contact voltage is always legal as less than 35VDC which the switch is rated for.

Since these relay switches are operated within spec, they are rated at "20,000,000 operations at 18,000 operations/hr", they will last alot longer. In english that means "Good luck wearing it out"

If any switch operated with arcing voltages, depending on the contact material, I'd give them a MUCH shorter life (alot less the 20million).

The arc supression circuit i used is also rated at 150A and 1 Joule. So the use of resistors on this one is unneeded. But, if one was to use capacitors to supress arcing, they will need resistors to bring the capaticance into (closer to) phase. A capacitor alone as a noise filter is not only worthless, but also more damaging.

[tim the toolman]

Very interesting info! I now feel better about investing in a punk dig switch

[Greatscott]

Originally Posted by PunkRC

No insult, but I don't think what you are suggesting, as I understand it, would work. I am hope this will provide more info. (No trying to hurt feelings, just provide info)

In the fraction of a second after the motor has been turned off, but before it was shorted, there is a huge voltage that will arc dua to the inductance of a motor. The switch cannot solve this it has to be another component.

Using a diode would work, except the motor has a reverseable polarity, so shorting the current through a forward current in a diode will not work. It would be ideal, but since the load is both polarities, something else has to be done.

The circuit that I use on my Dig Switches do ALMOST the same thing. Only they act like a SHORT to any voltage above 30Volts any polarity. That way, the contact voltage is always legal as less than 35VDC which the switch is rated for.

Since these relay switches are operated within spec, they are rated at "20,000,000 operations at 18,000 operations/hr", they will last alot longer. In english that means "Good luck wearing it out"

If any switch operated with arcing voltages, depending on the contact material, I'd give them a MUCH shorter life (alot less the 20million).

The arc supression circuit i used is also rated at 150A and 1 Joule. So the use of resistors on this one is unneeded. But, if one was to use capacitors to supress arcing, they will need resistors to bring the capaticance into (closer to) phase. A capacitor alone as a noise filter is not only worthless, but also more damaging.

Dunno, I was trouble shooting off the top of my head (it was posted after 14 hours on the road, I was checking the forums to get away from the in-laws). I'd have to draw out the ciruit and look at the switch options, but why go through all of that trouble if you have already got it figured out (no need to reinvent the wheel)? When I go cheater **cough**, I mean MOA I will see about getting a Punk RC dig switch.