Eliminate Huge Spark when Connect ESC to battery

[blueangel]

May I suggest a good high amp switch ?

I think Mosfet circuits are a bit on the BMW car building way - engineering overkill

With the switch open, connecting battery will not see a spark since it is open circuit.

Once ready, flick switch to close, there you have it.

To surge protect, a Capacitor in parallel to the switch can do the job too.

[Super-Hornet]

With switch, the huge spark still exist...but it is in the switch instead. So, instead of damage the plug, we might end up damaging the switch instead.

What I trying to create is something that can remove the spark completely and easily to connect/disconnect battery.

EDIT: I don't mind a bit of overkill anyway. If it is working, then it is another alternative we can use to eliminate spark.

SH

[Adrianli]

Just make sure u make one can take 14S lipo and 300A.

My normal running of 10S lipo 180A after 3min, the 10AWG wires are hot and the plugs already can give u a nasty burnt mark on the bare fingers. I cant imagine wat it will do to ur circuit.

[mave]

Hi,

is what you suggested similar to a relay circuit?

I am not an electronic guy but would like to know how some circuit makers remove that sparking when connecting the esc to the battery.
I suspect that the way to do it is not to connect the battery source directly to the esc, but to have some form of "switch" as a mechanism between the esc and the battery.

Nevertheless, I experience less of this problem when I am using the plug(as per attached). Not sure why.

For those who has R/C that required high voltage and current, do u always having "uncomfortable" feeling each time u plug in the battery and it give u a big spark and loud "Piak!" sound? Do that too often and u notice your battery connectors starting to wear out from spark (like spot welding).

I know there is a commercial version of a device to remove the big spark but I was thinking maybe we can actually develop something like that easily using the following items:

MOSFET => IRFZ44 (which is 50A, 60V)
( 2 MOSFET in parallel in order to increase the current to 100A)
A capacitor
A Resistor

What I thinking is that, the MOSFET is connect in series between ESC connector (dean, bullet) and ESC power input (in other word, separate ESC to the dean/bullet connector)
Then a high resistance resistor is in series with a capacitor. The +ve side of the capacitor is connected to the "Gate" of the MOSFET. The resistor and capacitor is connected to the dean/bullet connector.

With that, when u plug in the connector to the battery, the capacitor will slowly charge up to the voltage of the battery. It is slowly charge up due to high resistance. Once the capacitor reaches certain voltage, the MOSFET is turn ON and your ESC will have full power from battery.

I haven't test this out yet. Maybe it will work

I see whether I can test it out and let u all know the result.

SH

[Super-Hornet]

If we are to use any physical contact either a connector, switch or relay, there sure being some sort of spark. Some may have lesser spark then the other but we still have spark.

The spark is come from charging up our ESC's capacitor(s) (See EDIT below)

During initial state (no battery to ESC), our ESC capacitor do not hold any charges inside. Because of the behavior of capacitor, it is like a conductor (that is, current can flow from +ve point of capacitor to -ve point easily). When current flow through from +ve to -ve, it charge up the capacitor. As the capacitor beginning to charge up, it become less conductivity till the capacitor is fully charge (See note), the capacitor become non conductive to D.C. voltage.

Now, because of the fully conductive when capacitor is empty on charge, when we connect battery to ESC, in terms of battery point of view, it sees as a short circuit from its +ve terminal to the -ve terminal. That is why we see sparks!

The size of the spark is depend on the voltage, current and the size of the capacitor (in terms of farad). The higher the farad, the longer it take to charge up. Because of the longer charging, the battery actually see longer short circuit than smaller farad capacitor.


BTW, I just finish doing a quick test on a prototype board (will post it after I take some photos of it), and it works perfectly!! NO MORE SPARK!
Due to mid-night now, I can't spin up my 500 size heli with blades to see the loading effect.

Note:
well, in real life, it never 100% charge. But we do not want to explain in detail on it here


EDIT: When we close a physical contact, since the capacitor is in full conductive at initial stage, the point between the 2 physical contact point will act like a spot/electrical welding. That is where the spark comes from.
Using MOSFET to turn on after the physical contact is make, the battery still see as a short circuit at initial stage (or maybe lesser due to MOSFET is not instantly turn on..or there is a ramping up) but there is no longer a physical contact point for spark to appear.

[Super-Hornet]

Here is the 1st prototype board. Only the MOSFET board.
I intend to create 2 boards and stack on top of another. Notice the black row of connectors? That is where I will connect the controller board. Haven't develop the controller board yet.

The purpose of the controller board is to delay the turn on time.
=> It can be just a simple tiny switch that connect the MOSFET "gate" to high to turn on or "low" to turn off.
=> Or it can be using a resistor with capacitor (slow charge up the capacitor till the voltage reaches the turn on voltage on Vgs of the MOSFET).
=> Or it can be a microcontroller that delay 1 or 2 seconds (with some LED indicator also) before sending "high" signal to the MOSFET gate.

This simple, testing board is just to test whether can the MOSFET be use to turn on and off and whether driving a heli (lets say 500size heli) with load and whether will the MOSFET heat up using just 3 pieces of MOSFET.

So far I tested spinning my 500 size HDX500SAS heli without blades (due to mid-night time), it can spin up without any problem and the MOSFET is cold as if it is not turn on at all. But I really need to test it with load (that is, high current flow) to see how.

As u can see in the photo, I just connect a wire that short the MOSFET gate to ground to turn off the MOSFET or short to +ve battery to turn on. I using IRFZ44N because it is easily available in Sim Lim Tower. Because of N-Channel MOSFET, the MOSFET is connected at the low end. That is, current flow from battery +ve to ESC and ESC -ve is connected to the MOSFET's Drain pin and the Source pin of the MOSFET is connected to ground or battery -ve.

SH

[Super-Hornet]

Due to I do not have access to PCB development and etching, I use normal soldering to create the PCB track instead. For high current pathway, I use solder wick as wide conductor surface in order to allow high current to flow through.

SH

[aeromech]

bro, SH actually that is also my problem before when i was a car audio installer before specially when i was working with 1000 RMS watts of power. .every time i have to connect the audio system to the battery it would tend to spark real big. .lucky the size of my wires are 4AWG and 2AWG hehehe. .normally i would try to drain the 1.5A capacitors by connecting a 12v bulb to the system then after the stored power from the capacitor is finished then i connect the system to the battery and there is very minimal spark

my experience just want to share

[Super-Hornet]

Here is an updated working prototype.
I going to try using only 2 IRFZ44N MOSFETs..which should give me in calculation of 98Amps (each IRFZ44N can go up to 49Amps, 55V). That should be more than enough for my TRex500 (clone) for my bo-105 (4 bladed head).

1st picture shows the comparison to the 1st prototype.
2nd picture shows the controller (the thing that turn on the MOSFET after some delay).
3rd picture shows the Driver/Solid State Switch (or MOSFETs)
4th picture shows both Driver and Controller stack together.

I'm using PIC12F629 for the controller. When powered up, the sequence is as followed:
Connect battery,
Delay for 800ms then Red LED turn on.
Yellow LED blink 3 times (each on and off takes 600ms)
Green LED turn on and at the same time, turn on MOSFET.
Then Yellow LED goes blink every 1seconds to indicate that the controller still running.

SH

[Super-Hornet]

Here is the video showing without "Spark Eliminator" and with "Spark Eliminator".

{ Note: Video is encode using XviD. But due to the limitation on this forum that it cannot accept ".avi" file extension, I rename it to ".mp4". If you cannot play it, try rename the extension to .avi }

[Adrianli]

Looks very big but no spark.

U making sort of "Bob Finless how to do" videos.

[Super-Hornet]

Hehe!

Well, I do not have PCB etching device/equipment. All are manually routed by hands...therefore, size is big. If I can produce my own PCB, the size can go even smaller size. Oh! other things like heat sinks, connectors, if I were to use smaller size type, the overall size can reduce further.

SH

[LittleGiant]

May I suggest a good high amp switch ?

I think Mosfet circuits are a bit on the BMW car building way - engineering overkill

With the switch open, connecting battery will not see a spark since it is open circuit.

Once ready, flick switch to close, there you have it.

To surge protect, a Capacitor in parallel to the switch can do the job too.

I don't see how a capacitor in parallel to a switch to have surge protection.

When u on the switch the capacitor is shorted across its terminals. Then u open the switch, the capacitor acts like a short circuit until it is charged. What is the capacitor doing exactly?

[LittleGiant]

Here is an updated working prototype.
I going to try using only 2 IRFZ44N MOSFETs..which should give me in calculation of 98Amps (each IRFZ44N can go up to 49Amps, 55V). That should be more than enough for my TRex500 (clone) for my bo-105 (4 bladed head).

1st picture shows the comparison to the 1st prototype.
2nd picture shows the controller (the thing that turn on the MOSFET after some delay).
3rd picture shows the Driver/Solid State Switch (or MOSFETs)
4th picture shows both Driver and Controller stack together.

I'm using PIC12F629 for the controller. When powered up, the sequence is as followed:
Connect battery,
Delay for 800ms then Red LED turn on.
Yellow LED blink 3 times (each on and off takes 600ms)
Green LED turn on and at the same time, turn on MOSFET.
Then Yellow LED goes blink every 1seconds to indicate that the controller still running.

SH

Nice work. The mosfets look very small. But for constant voltage without switching the efficiency should be quite high. May I suggest u use protection diodes across the drain source terminals of the mosfets so that back emf can bypass the mosfets straight to the battery if you are using regenerative braking. You can also try solid state relays (basically high powered mosfets) if you need more juice.

[Super-Hornet]

Thanks

OK...thanks for the tips on using flyback diode protector. Will try to add it next time I visit Sim Lim Tower because I do not have any diode component at home


EDIT: The IRFZ44N is called "HEXFET". I think it already has a schottky diode build into it. I can't confirm yet. Need to dig up more info.

SH