BroMikey's Capacitor Dump Circuit

[SeaMonkey]

Aye, BroMikey still has a ways to go before he acquires full
and complete technical understanding of what he's up
against with his circuit.

Good education always seems to come with some sort of
pricetag - be it in dollars or burned out MOSFETs.  But the
rewards can be enormous if the "student" has the gumption
to stick with it until he is finally able to make it do what it is
supposed to do reliably and without fear of premature failure.

And know the reasons why...

[Farmhand]

Tinsel he is charging the capacitor bank from a series set of batteries to get the higher voltage "through a resistor" to limit the current when he discharges the capacitor, I explained how much power the resistor is dissipating when he does that, then he is dumping the huge cap bank through the paralleled mosfets to the batteries. I think he is using a large capacitance 10,000's of uf and charging the dump caps to more than double the battery voltage, maybe triple.

When we dump caps the caps end up not discharging to zero volts they end up at battery voltage, when the cap is first connected it charges to just under charge battery voltage.

I suggested he use a mosfet to isolate the supply from the load when dumping or use a series inductance coil to partially isolate the supply for at least part of the dump.

A better way would be to use a transformer to charge the caps. And prime the caps if the bank is high uF.

Or if using a series batt set to charge the bank he really needs to isolate the supply series string of batts from the batt he is dumping into.

What would be the peak current of 36 volts in a 200,000 uf cap dumped into a paralleled set of four 12 volt batteries in good condition ?

I haven't taken the time to verify what circuit he is using right now, I'm just going by his descriptions.

I haven't destroyed a mosfet for a long time now, I mustn't be trying hard enough.

..

Fairly sure a NC relay could be used to rout the transformer output through a resistor to initially "prime" the large cap bank, then when the caps are charged the relay would turn on to give a low resistance charge path from Xformer to caps..


..

[TinselKoala]

What would be the peak current of 36 volts in a 200,000 uf cap dumped into a paralleled set of four 12 volt batteries in good condition ?

The peak current depends on the total resistance of the circuit plus load. The capacitance _does not enter into it_.

The peak current into a _direct short_, when switched by the mosfet, is found by applying Ohm's Law: I = V / R. 
Since v = 36 volts and R = 0.27 ohms, the mosfet's minimum on-state resistance, the maximum current is just over 133 amps. Into a dead short, from whatever you are using that provides 36 volts, whether it be a capacitor of infinite capacity or a battery or a hydroelectric power plant.

For _how long_ can that peak current, or any current, be sustained? NOW the capacitance matters. Higher capacitance, more energy, more time to discharge it. As soon as the discharge from the cap begins, the voltage will begin to go down and the current will drop from its peak value, and as the voltage decays exponentially so does the current. More capacitance, longer time constant. More _voltage_ more peak current.

Now let's imagine a more realistic load impedance. Say it is one single ohm. Now solve I = V / R, and find that you can only push a bit over 28 amps through the circuit, no matter what the source is, no matter what the capacitance you are dumping is. That solution to Ohm's Law is the _maximum_ current that 36 volts will push through your circuit.

Now I don't know what other elements he has between the mosfet switch and the load, or what the equivalent internal resistance is of his battery banks, but I will bet that the total circuit resistance, even with all six mosfets properly paralleled, is more than 1.27 ohms. Which means that the _maximum pulse current_ from a 36 volt source will be less than 28 amps.

However in his post he says he's using 80 volts and is attaining 6 amps per mosfet, or 36 amps total. Again, solving Ohm's Law, we find that using those numbers at face value, R = V / I gives us a total circuit resistance of a little over 2.2 ohms.

So going back to the original question, if your capacitor of a million billion Farads is charged to 36 volts, what will be the peak current when it is discharged thru the mosfet (or thru a simple mechanical switch closure) into BroMikey's 2.2 ohm circuit? I = V / R = 36/2.2 = less than 17 amps. A single mosfet will work safely for a 36 volt supply, IF the backspike is handled properly.

[Farmhand]

Yes I understand that the peak current is limited by the voltage / resistance. basic stuff I'm not sure exactly why I wrote the uF value  except to say it is a high uF bank of several paralleled capacitors, won't that affect the resistance of the current path ? And if the batteries were paralleled then less resistance as well. I understand that no matter what the capacity of the supply it is voltage and resistance that determine current in DC.

Anyway it seems the problem is improper switching like most cases and since I don't have the time to help him, Matt over at EF will have to continue to do it. See no one helps anyone over there and he has been invited over here so, it's up to him.

With so many mosfets in parallel they probably are not turning on properly, I had to explain to him that there is a high peak current in a coil discharge to a battery, he was led to believe there was none, and also that to drive a mosfet requires current when some nong posted that mosfets do not require current for switching. Meaning I had to try to explain displacement current to him.

With the mosfets not turning on properly he has no chance.

..

To put it very simply if he cannot do it with one mosfet he is doing it wrong. Not much point trying to help people that want to parallel up hand fulls of mosfets.  And he is charging the caps from a battery series set through a resistor. Wow.

..

The other thing I meant to mention was he may be exceeding the Gate to Source voltage limit. Gate abuse maybe.

..

[SeaMonkey]

BroMikey is struggling with some internalized misconceptions and still
lacks full understanding of some critical details.  He's still in the embryonic
stages of his technical development.  Ways to go yet.

And yes, BroMikey you're reaching some of us.

The good news is that we've all been there; it too will pass. Then the
real understanding will congeal with eye-opening results.

Don't stop the quest!  True understanding is built upon numerous
failures which are temporarily disappointing but serve to fuel the
fires of desire.  The desire to know with certainty.