Cap Dump Circuit

[BroMikey]

Back up and running

Hello Group members

I wanted to give you a report from my last video where I had thought the 555 timing circuit went pop. It did not and I guess I am bugging out every-time things pops so i think it is a cap or a driver but instead it was none of the above.

Yes one of my Fets weaken and would not switch after 30vdc but that was probably due to my bad brush on my variac that was blowing breakers.

You think?

The brush needed to be adjusted and so it would arc till the breaker popped every time I reset it. Opened it up and everything was fine on all other components as they are over rated.

Also will be showing a new video with the timer circuit still running along plus all of the new amp meters bouncing.

Each time I add another Mosfet in parallel with the rest I go up in amps for each pulse.

So far I can really ring that bell so to speak when it comes to putting the hammer down.I have been puttering all day adding fan temp switch and overload protection.

Now the dump fan comes on by itself and so if I run the power up by increasing the pulses per second it will stay on so I need to finish the Fet's for a total of 9 more.

I bought these Fets and the legs break off ever so easy if I move them much. Has anyone every had that happen? They are IRFP460's.

These Fet can handle high power at the voltages I require.

Even those big batteries don't stand a chance against this big dump.

I can get some high powered current pulses out of this box that forced me to go to a less sensitive amp meter so i wasn't pegging the nettle.

Before I used a mini 5 amp meter but now with the same power level it only shows 2 amp pulses so now i have approx 5X that energy to move this shunted amp meter.

It shows as high as 15 amp pulses when I know it is in the 50's and 100's and this is all at 12vdc. When i move up to the 24vdc charging I fill my caps to 85 voltage so as the repetitive discharges come there is always 75 volts available between RC charge ups.

I will then need to use a bigger amp meter that shows a 30amp pulse.

Mike

[BroMikey]

Blown Mosfet (ONE)

Howdy Howdy Group Members

I am coming to you today to give you this video

Pulse Charging BroMikey Cap Dump circuit - YouTube

It is about how my progress is coming using only 1/4 of the Fet's to pulse the huge cap bank.

Before when I used IRFP250 Mosfet's using 6 devices or six Mosfet's they got hot and blew out at 50vdc. Really I am struggling to find all of the right words.


Now with only 3 Mosfet's I was able to pulse charge the entire bank of caps for days in 95 degree heat using a fan. I ran the supply up to 65vdc to pulse a 12volt battery big no no for any small battery, mine are as heavy as some cars.

I am running the upper limits of these three Fet's using a 10amp @65vdc 3 pulses per second. They warm up and the fan does not stop the sinks for staying mildly warm.

These devices are a 16amp average part so running them at 6-8amps will extend the life of the component dramatically.

I really don't know how to relate to you the difference that cap sizes make. For the same rating the physical size may vary.

My cap bank holds more energy than most setups I have seen. 80 caps in parallel each one with a 14awg wire, caps 2.5 inch high and 1.5 inch dia for 1000uf.

A single pulse from these caps has burned out 1 or two Fet's on the spot, especially the IRFP250's.

I am exploring the upper limits of these Fet's as I said and now that I know where they over heat and know what to expect, I am going to drive these same IRFP460's with a new driver to see if wasted heat generated from sloppy switching times will lower heat.

Pulsing 20-100 watts is easy and you need not be concerned much about much of anything as just about anything from an automotive relay to a reed switch works.

Yet when going up in power around 1000 watts sloppy driver circuits will show problems with heating. I think many guys tell me that I need to stay with 1 or 2 Fet's as this is hard enough to balance so one is not doing all of the work.

I am going with 12 Mosfets and thinking about how to get them to automatically balance the load between themselves.

I would like to learn how to set a Fet at a specific amp handling amp pulse using some form of feedback loop. If I could adjust one Fet and one driver for a single Fet to a given output, I could do all 12 Mosfets the same way so they could all pull together.

I can't be the first man to think this up.

Mike

[BroMikey]

Genmode

Are you wanting to pulse the bats with voltage or amperage.
boost converter will give voltage - pos pulses

generic circuit-google boost converter.

buck converter will give amperage - neg pulses

google buck converter

Or you could incorporate both into the same circuit

Thanks Dave

I will stick with the hybrid converter Oscillator Bedini patented using big coils of magnet wire. Using long wires WITH the converter Oscillator topology is far superior to anything i have seen.

Most booster converters today employ tiny coils that are far less forgiving than the larger coils. I have noticed that these small coils can be no larger than a golfball in most cases.

My coil weights 7LBS while those tiny booster circuits have say 3 oz coils at best and depend on high freq's to achieve operation using 3 feet of winding.

My coils are 130 feet long 14awg wire and I am about to pulse up to 6-8 amps per wire at 9khz using 4 wires. This machine will input 24dc at 25- 40amps sending the output to my caps for dumping.

This oscillator will briefly fail when the dump comes and reestablish filling caps for the next dump. This is a great way of filling caps because this circuit can tune itself.

All other circuits are made for one set very narrow very ridged design criteria that would burn out under these dumping operations.

Mike

[Dave45]

Are you wanting to pulse the bats with voltage or amperage.
boost converter will give voltage - pos pulses

generic circuit-google boost converter.

buck converter will give amperage - neg pulses

google buck converter

Or you could incorporate both into the same circuit

[BroMikey]

Joule count?

But if you load your caps correctly with reactive or radiant power and not with real power than those things become less of an issue as the total joule count in the cap is greater than the joule count taken from the source power.

Matt

Hi Matt

I looked back at your earlier posts. Are you saying that the joule count charging caps from a battery charger is less than when using a radiant one?? Same volts in the cap yet more joules?

Mike

[BroMikey]

Batteries as heavy as a Car

For 12 volt battery my cap dumper charges to 28 volt. But the cap I use is very small 20k uf. Even on the big batteries. I hit the battery 2 times a second with a 9-12 amp pulse and the cap discharges to 14.2 volt.
If I were to use a bigger cap say .1f I would charge only to 20 volt.
For a 36 volt system I would charge to 75 volt with 20k uf cap and 50 volt with .1f cap.

I am not following what you are saying so that should give you an example.

To high of voltage and you can damage the plate material on the battery. Like wise if you create to much current you are waisting power into heat.

Matt

Yes true Matt

To much power running into a battery with high surges can not only desulfate but damage as well. Sometimes over the last 1 year I will run my caps up to 40-50vdc and dump down to 20vdc FOR A HUGE HUGE HUGE BATTERY.

I have one 110ah AGM battery that I do this with. This battery is capable of giving back 70 amps. This battery single handedly can run my 120vac 7 Amp "Saws All" out in the woods tree trimming half the day.

If any damage is going to occur it won't be from zapping it with 20 watts of surge energy. I am giving you the details of my experiment now. In the beginning the battery was abandoned as junk.

I add ALUM.

I use high voltage to start using the spikes from a Bedini forced oscillator as high as a 75watt input. Then from there I use a small dump, not my biggest dump to high current surge this 110ah Square AGM monstrosity.

I can start my 40hp air-cooled Wisconsin engine with this battery that requires 50-80amps, so damage to this battery will take some doing.

I have found many old batteries that will never come back to life and these batteries fooled me. Let me say it this way, if a battery charges up in two minutes and discharges fast that battery is already ruined.

I have used 5-10 amp current surges at 1 per second by winding up the voltage as high as 45vdc before letting it go.

Yes this is a 12vdc battery and my battery works better now than it even did. However if you use to much energy you will reach a point that no increase in charging will occur.

Every battery should be approached in this way. The operator should keep experimenting with input and output surges to see where the battery begins to start to charge. Then continue exploring the higher pulse rates and current surge values.

For a 1ah battery use very little and so on.

If my battery is as big as a car and my charging system is very small plus my 100ah wire to the bank is 50feet remember the voltage drop.

I will be showing off the new setup soon. Voltage drop is a consideration plus size.

Every battery bank setup requires a different capacitor discharge voltage and sometimes to get the amps high enough and to over come the distance the operator may need 40vdc.

Also when I lower the pulse rate at night time I can run 45vdc but only one of those per 3 seconds.

I understand what you are saying Matt and thank you for your input.

If you have a specific battery size in mind and charger please share your setup so others can gain a perspective.

These are my practices that I use and my batteries are all high density. My 110ah battery use to only be good for 50ah but now it produces 170ah under the same amp draw rate.

I have damaged batteries and you are so right Matt. I won't go into that but burning up a battery is due to beginners over charging and over powering it and everyone has to go through this bump in the road.

Some say only use a max of 10 watts and 20 watts with no battery size given.

The battery bank size must be known and the distance away form the energizer to deal with loses. It may not be a great lose but still significant.

I have used 50 watts input where the battery just hangs there in limbo not charging much and when I went to 60watts the battery would charge 10x as fast. This was done by raising the cap bank voltage discharge.

Mike

[BroMikey]

Wow Thanks A million

I agree with Matt, a fast turn off is only important to creating higher voltage
from coil discharges and both turn off and on quickly to allow and disallow
current is mainly for reducing the power dissipation in the switch. And too
fast of a turn on or off can cause unwanted ringing ect. Sometimes we may
use a gate resistor to slow down the turn on and off or suppress ringing at
the gate.

With the cap dumping, I look at it two ways, the way I used in my posted
setup was to dump a relatively small capacitance at higher frequency and
not fully discharging the capacitor, this is mainly to hasten the desulfating
process with low input power, pinging the battery with short high current
spikes.

Then there is dumping a large capacitance relatively slowly and totally
discharging the capacitor to get high surge currents through the battery.

In the later case the turn off time is unimportant in any way except maybe
to save time so the cap can be recharged sooner, maybe.

This video shows some wave forms and stuff from a boost converter
powering a cap dumper/desulfator.

Note the groups of step discharges of the dump cap towards the end of the video..

https://www.youtube.com/watch?v=7ZBE4i5UR58

..

Thanks so much Farmhand

Yes I see clearly now and the truth is I have always known that my coil rings but my cap doesn't. So cool of you to show me this video

I have those forms on my scope with coils.

And the Sawtooth form is a further verification of how the cap looks on the power side like Matt said.

@MATT

Yes you are right and getting me back on focus because I was going from gate wave forms to the power wave forms and back and forth in my mind.

Just like a toddler nearly falling over, I told you guys I needed help to walk here

Both of you gentlemen have helped be today a great deal at a very critical time for me to stay on track.

Yes I am spending a good deal of time building my forced Oscillator to fill caps proper. Very important I agree.

So if i have anymore bright ideas I know who to ask

Thanks so much you don't know how important this is to me.

Mike PS I thought we wanted ringing? And that ringing is part of radiant? That was my only puzzling thought.

[Farmhand]

I agree with Matt, a fast turn off is only important to creating higher voltage
from coil discharges and both turn off and on quickly to allow and disallow
current is mainly for reducing the power dissipation in the switch. And too
fast of a turn on or off can cause unwanted ringing ect. Sometimes we may
use a gate resistor to slow down the turn on and off or suppress ringing at
the gate.

With the cap dumping, I look at it two ways, the way I used in my posted
setup was to dump a relatively small capacitance at higher frequency and
not fully discharging the capacitor, this is mainly to hasten the desulfating
process with low input power, pinging the battery with short high current
spikes.

Then there is dumping a large capacitance relatively slowly and totally
discharging the capacitor to get high surge currents through the battery.

In the later case the turn off time is unimportant in any way except maybe
to save time so the cap can be recharged sooner, maybe.

This video shows some wave forms and stuff from a boost converter
powering a cap dumper/desulfator.

Note the groups of step discharges of the dump cap towards the end of the video..

https://www.youtube.com/watch?v=7ZBE4i5UR58

..

[BroMikey]

Opto's

Again let me stress the significance of a circuit that produces a sawtooth wave shape. I this simple wave a transistor or Mosfet is turned off with a resistor. This is a crude way to turn off a Fet and produces this wave form.

Notice the cap dumps in a very short time and all of the wasted time after it fully discharges.



Very poor management by comparison to a state of the art drivers ability to turm off the Fet or transistor.

It is my understanding that a sawtooth wave shape is a starting point for beginners.

Opto's can turn on faster than 1 ms say a normal speed of 100uS or less and this is fast.

It is the turn off time that we are concerned with here in order to come out of the beginners box. Turning off a fet or transistor is the point.

Does anyone see all of the wasted time in the diagram and do you realize that the after the cap is done discharging that the Fet does not shut off for three times a long as it is open?????????????

Anybody?

To get past a sawtooth wave shape will take some doing.

Yes it makes a difference. When ole Bruce Lee struck his boards his unidirectional impulse was fast in and fast out. This is where the extra energy is realized.

Everything matters.

Talk to me guys.

Mike

[Dupe]

Cap discharging

@ Mathew Jones,

Your replies are spot on Mat, and as you say working on the basis of discharging at around 3 sec intervals really makes little difference to the stored energy when dumped into the load at that rate, The CR network denotes the rate of discharge as you have already pointed out. I have experimented myself with cap discharge circuits in the passed, some batteries work better than others which depends on what the circuit output see's (impedance)
I currently use inductive pulse circuits but would have to use some rather large inductors to charge lead acid 12v deep cycle batteries quickly. I charge my 14v NiCad drill battery using this type of circuit which lasts much longer between charges compared with a standard charger, not to mention it charges quicker.
Regards Dupe

[BroMikey]

Trying To Learn More

You cannot get a square wave with caps. Or pretty impossible anyway. The minute you turn on the dump power falls out of the cap down the battery level.

To get a square wave you would have to switch from a source that has enough current to maintain a voltage level. This would be alot of current as a battery has a negative resistance. Meaning the more you feed it the more it takes to the point it will destroy itself. And thats what will happen. Thats why normal battery chargers use Ferro Resonant transformers, so they can regulate the amount of current passed.

If you are concerned solely about the gate signal you can use true Gate driver like a FOD3120 with an isolated power supply.
My favorite is the FOD3184 with dual polarity isolated power supply to power it.

You can use that power supply with the gate driver by tying the positive polarity of the PS to the positive (+15 vdc) of the driver and tying the negative (-15 vdc) side of it to the negative side of the driver. This gives you 15 volt below ground to empty the gate. The shut of times are unreal.
On times can be as fast as the driver allows i think 50 microseconds or something.
But you might also be looking at a complete rework of the whole circuit if you do that.

Matt

You cannot get a square wave with caps?

[shylo]

Hi Matt, so are you saying , that if I charge a cap to 24 volts ,let it discharge to, say 12.6 (on a battery reading 12 volts) , then let the cap recharge?
That is don't let the cap discharge to battery level.
Just had to ask.
artv

[BroMikey]

circuit diagram

Here is that diagram for a sawtooth waveform



I removed the plus sign.

Mike

[BroMikey]

See what I mean?

@ BroMikey, check the partial circuit polarity labeling !! Dupe

Hi Dupe

I found this on the web. See what I mean about the garbage you find on line sometimes.

Anyway the polarity was not my focus. I am wanting experimenters to look at the voltage divider starting with the 10k ohm resistor.

The claim al around the web is that this circuit produces a square wave. That is bare faced lie. This circuit produces a sawtooth wave form.

That is what I am saying.

They may have goofed on plus and minus?

Mike

[Dupe]

Cap dump charging

@ BroMikey, check the partial circuit polarity labeling !! Dupe