Capturing high frequency pulses

[Reynard]

Hi

my aim is to capture high frequency pulses (duration of about 0.1 - 0.2sec) into a capacitor and able to store these charges over a period of reasonable time. In order to store these charges over a period of time, i will need to use a 1000uF or above capacitor. But these large cap will not be able to charge up by such fast impulses.

I have tried using charging 3pcs of 220uF cap (in parallel) with these impulses which in turn will charge up the larger cap. But the results are not very favourable. Are the values of the 220uF too large as well? Or should i use more 220uF?

Can any1 help me. I do welcome any other ideas. Thxs!!

Cheers

 

[Nigel Goodwin]

You need to provide more details - where are the pulses coming from, and what current can they supply? (or what is their source impedance).

 

[Analog]

voltage, frequency of pulses?

 

[JimB]

You asked this same half baked question here just recently:

https://www.electro-tech-online.com/threads/ideas-to-operate-applicances-from-waveform-acquired.22741/?highlight=capacitor+charge

The idea appeared silly then, and seems no better now.

JimB

 

[audioguru]

You said that the current and power of the pulses are small. You showed a pic of the pulses that show they have a very small duration. Therefore you want to store nearly nothing in a capacitor. Why?

 

[Reynard]

Hi all.. im sorry if my idea seems stupid but i still got to get this project working hehe.

Ok just to give a more detailed idea... i got a series of pulse ranging from 2V up to 50V and the duration of each pulse is about 0.3sec. Btw this pulses are obtained from electric eels ....really. However the current from the eels are very small (in mA).

My main idea is to store them into a energy storage device which i have narrowed down to supercapacitors. However over a duration of about 30hrs, i can only charge up a supercapacitor of 1F by 0.1V. I'm still looking for supercapacitors of smaller capacitance such as 0.5F to try out.

I have tried using a pulse waveform generator to simulate the pulses from the eels. And im able to charge a supercapacitor of 1F by 0.5V in about 2mins. So i guess these supercapacitors are able to charge up with pulses of such duration (ie. 0.3sec)

Attached is a circuit of my charging circuit. The 16.4Mohm is to raise the load impedance so as to maximise the power transfer to the capacitor.

Pls advise me on anything if i did some silly implementations. Thank You!!

 

[stevez]

Reynard - have you done the math, so to speak, to get a sense of just how much power you might be starting with? You have some sense of the voltages and current - and you've estimated the time so there's enough information to perform a calculation as a place to begin. I would expect the results of the calculation to help you understand what might be possible in terms of the end result (I am not sure what that is).

You might then consider the losses, over time, of the capacitor, to get a sense of just how much net power you'll have to work with.

I am not sure of this but I do seem to recall reading about eels and other creatures with similar characteristics - and I think I've seen data expressed in terms of the power. This might only be another way of stepping back from the problem to see if whatever it is that you are doing makes sense. Your thought of capturing pulsed energy certainly does make sense at some level.

 

[ikalogic]

Ok, i don't really see a clear question.. but however, i was wondering,

1-why add the 16Mohm R?
2- why use the 5.1 zenner regulator? why not letting the maximum voltage across the capacitors, this could decrease the charge time..(i think)

 

[audioguru]

I thought electric eels give shocks. Only 20V doesn't give a shock.

If all the pulses are 20V at a current of 2mA for a duration of 0.33 seconds for each second, then a 1F capacitor will be charged to 5V in about 66 hrs.

The actual time will be much longer because many pulses are less than 20V, the average current is probably less than 2mA and the duration of the pulses don't appear to be present for 1/3 of the time.

Your test circuit has a current-limiting resistor of only 220 ohms which limits the current to a whopping 80mA. About 8.2k will limit the current to 2mA.

 

[lord loh.]

Is it feseable for you to use an opto coupler for isolation and amplify the voltage pulse by a high gain op amp and chgarge the capacitor?

I was wondering why no one suggested this.... Or have I missed something?

 

[audioguru]

Is it feseable for you to use an opto coupler for isolation and amplify the voltage pulse by a high gain op amp and charge the capacitor?

I was wondering why no one suggested this.... Or have I missed something?

The circuit doesn't need isolation.
The voltage pulses are already 20V or higher so why use high gain?

What is going to provide a power supply for an opamp?
The power to charge the capacitor must come from somewhere, and it looks like there isn't enough power from the eels.

 

[ikalogic]

dear Lord loh

I don't think an op-amp will solve the problem, at all, because, the op-amp will need a powersupply to work, right? and i think this is what we are missing here..

and even if we had a power supply, don't confuse between op-amps and DC-DC "boost" converters, op-amps wont give you voltages higher than the supply rails, while BOOST converts can..

So your idea - i think - is feasable..

Then, why do you want to isolate??

 

[ikalogic]

haha!!

me and audioguru posted barely the same message at the same time!!

 

[audioguru]

Hi Ibrahim,
The max voltage for some super capacitors is only 2.5V, so he has 2 in series and is using a 5V zener diode to protect them.

The power from the eels is so low that it will take a million years for the capacitors to charge to 5V. Maybe the capacitor's value should be 1uF.

 

[Reynard]

Hi all

thxs for giving me advice.. just to clarify some doubts being raised.. the 5.1V zener diode is to protect the supercapacitors that i used (which has a rated voltage of 5V)

the 16.4Mohm is to raise the load impedance so as to maximise the voltage being transferred to the capacitors.

reason for not using any op-amp is because i dont want to use any external power supply to power the op-amp.

in literature, the eels may emit voltage up to 600V but due to attentuation, the highest voltage i have acquired so far is 60V.

meanwhile im still trying to get hold of some smaller value supercapacitors so that they can be charged up in a shorter period.

i have also tried MAX1682 voltage doubler which is capable of working up to 12kHz. But the results are not very encouraging as well...

with the components that i used....are any of them being misued which caused power loss of the components?

Sob sob....

 

[audioguru]

in literature, the eels may emit voltage up to 600V but due to attentuation, the highest voltage i have acquired so far is 60V.Sob sob....

Maybe that is why the power from the eels is so low. Maybe your circuit needs a step-down transformer.

Even if you do manage to charge a small-value capacitor, what are you going to use the small charge for?

 

[Reynard]

Maybe that is why the power from the eels is so low. Maybe your circuit needs a step-down transformer.

Even if you do manage to charge a small-value capacitor, what are you going to use the small charge for?

Hi

yup i agree that my source power is very low. Can i know how does a step-down transformer help in my circuit? Sorry for being such a noob hehe.

ok if i managed to charge a supercapacitor of 2F, i can use it to light up a LED for quite some time and i guess it will be a major step for my project.

Cheers.

 

[audioguru]

Your eels can't charge 2F, you tried it. They might charge 0.2uF then it can blink an LED for a few milli-seconds, one time per day.

 

[Reynard]

Your eels can't charge 2F, you tried it. They might charge 0.2uF then it can blink an LED for a few milli-seconds, one time per day.

Hmm....yup i have tried it before but such a small capacitor does not prove any usefulness in the practical world....

Cheers

 

[audioguru]

Such a small amount of power from eels does not prove any usefulness in the practical world.