Desulfating circuit Which switch to use

[hatamata]

So first, the usual white flag, I'm not a professional as most of you guys, which means that I would understand more if you use simple terms OK?

I'm building a power desulfator. I need to pulse batteries with high Amps at different frequencies and pulse widths. The part of generating the pulse is already covered in another thread somewhere in this forum, my problem is with how to deliver the high Amp pulse to the battery.

Basically, there are two power sources at the two sides of the switch, one is the transformer and a large capacitors bank, my pulsing circuit should control the suitable switch that will deliver the square pulses to the battery, and on the other side of the switch there will be the battery, and the batteries I'm targeting are as large as 200 Ah/ 24 V.

My problem is that although my pulsing circuit can produce almost square pulses when not connected to load (the battery in our case) but when connected to the battery, distortion appears at the generator side, ringing or echo diminishing sine wave also appears on the side of the battery (I place a 0.01 ohm R at the output of the mosfets I use leading to the battery - and it some times measure up to 4.5V on the scope = 450A).

I pulse at two ranges: one generator produces variable square pulses, but only up to 200 Hz max with 1% - 15% duty cycle, the other is a fast one, it is also a variable one, it generates from 200 Hz and up to 50 KHz and with pulse width as narrow as 50 ns and up. In both cases, the switch needs to deliver a pulse with up to 500 A/ pulse (yes, I'm using a huge power supply and capacitor bank) and V not more than 40 V

So which switches should I use? are there any relays that can handle the low frequency circuit? relays although has their mechanical problems but I think that at less than 200 Hz, a relay will allow for a sharp square pulse and will eliminate any effect of the battery on the circuit at the off time.

To share the project's idea with you, pulsing a battery with high amp and square pulse do improve the condition of the battery as long as it does not have a mechanical damage, I wasnt sure why it is taking so long to desulfate a battery until I used a scope to see what is going on, this is when I discovered the distortion that occurs when the pulsar is connected to the battery.

Thanks

 

[FvM]

Apparently you have used MOSFET pulse generators before (which sounds reasonable) and intend to use mechanical switches (contactors) instead. It's a bad idea and won't work reliably.

You're also discussing some observations about unexpected or unwanted waveforms ("distortion"). It's actually not clear, if you are talking about voltage or current waveforms. In my opinion, a "desulfator" would primarly feed a pulsed current to the battery, so the voltage waveform won't matter at first sight. If you are preceeding to ns pulsewidths, you should be aware of parasitic wiring inductances and transmission line effects. You'll need more then casual knowledge of electrical circuits to handle it.

 

[dick_freebird]

Well, batteries do resemble large electrolytic capacitors.
So some resonant behavior is to be expected. One question
to answer is, what does the useful work? Current (if you are
cracking and then replating to the electrodes)? If so, the
on time should be enough to accommodate the slower
drift rate of the ions. Joule energy (if you are trying to
simply destroy shorting whiskers)? If so, do you fare better
with a narrower, higher pulse than a lower, fatter one?
How well do you understand the physics at play, the
thermal ideal for getting what you want and less of what
you don't (bulk heating vs adiabatic heating within the
matter-to-be-obliterated)? Anyway, do voltage waveform
artifacts or ugliness matter at all? Is it the current you
should really optimize? The pulsed power? Does ringing
mean anything to effectiveness?

I have seem some people discuss more mechanical methods,
removal of the sulphate detritus at the bottom of the cell
(a final resting place, and where electrical desulphating
will have little effect - infinite supply of it, relatively
speaking). Have you looked into this, and EDTA chemical
methods? At least for wet cells, this seems to enjoy some
success.

For gelled cells, you have the problem that once a sulphate
crystal is even marginally dislodged from the plate your
electrical access to it is gone, and so would be your
ability to return the lead to a useful position even if you
were able to (c r a c k - stupid censor) it.

 

[hatamata]

Thank you for your help
FvM, if I use relays, it will be only for the slow frequency, up to 200 Hz, but I dont know if there are any that can handle a current of 500A /pulse (that will be as wide 50 us wide). That will provide a nice real square pulse which is a must in my opinion for this thing to work.

"In my opinion, a "desulfator" would primarly feed a pulsed current to the battery"
Yes, that is exactly what I need, but it must be square or at least with as sharp rise as possible.

"If you are preceeding to ns pulsewidths, you should be aware of parasitic wiring inductances and transmission line effects. You'll need more then casual knowledge of electrical circuits to handle it."
Fortunately, I dont have a problem with this side, I bought this part of the circuit already built, it works fine with almost perfect pulse shape, but the problem and distortion only appears when I try to deliver the pulse to the battery, I think the reason is that the battery as it has a huge power stored in it, it bypasses the mosfets and affects the generator circuit, that is why I thought of relays, reed switches, even tubes if any will do the job. obviously, semiconductors can not hold back the effect of the battery.

Dick-freebrid: wide pulses is useful in case the battery can not be charged with straight DC. One must be careful in my case, 30+ v and average (non pulsed) current of over 20 Amps may cause the battery to dry in a short time due to excess heat, That is why shorter duty cycle is preferred. if you look at the commercial desulfators, the ones working with inductors and use the own battery power to pulse it back, you will find that they use 1000 Hz and the pulse width is about 50 ns, the duty factor is about 0.00005, now this works but it takes ages. What I'm trying to do is to pulse high amps with different frequencies and pulse widths till I find the range (and it will not be only one frequency but a range of frequencies and pulse widths) that will give the fastest desulfating.

"Is it the current you
should really optimize? The pulsed power? Does ringing
mean anything to effectiveness?"
Yes, it is current that should be optimized, preventing the ringing (the reflex of the pulse on the battery) from bypassing the switch for two reasons:
1- it affects the generator and hens, the frequency is affected as seen on the oscilloscope
2- the ringing is in the form of a sine wave that fades (tail) and that means lots of -ve electricity that I think it minimizes the effect of the +ve pulse.

"I have seem some people discuss more mechanical methods,
removal of the sulphate detritus at the bottom of the cell
(a final resting place, and where electrical desulphating
will have little effect - infinite supply of it, relatively
speaking). Have you looked into this, and EDTA chemical
methods? At least for wet cells, this seems to enjoy some
success."
EDTA is not a good option, it works in SOME cases but on the long run, it causes internal shorts, and to remove sulphate detritus from the bottom is correct, however, the only way to do it right is to drill holes at the bottom. sulphate detritus will not pass the thin space between the plates if you try to turn the battery over.
IMHO, leave sulphate detritus where they are and clean the plates, the sulfate on the plates is what is preventing the battery from accepting a full charge, also, if it is allowed to grow, it will cause an internal short.

 

[FvM]

200 Hz aren't slow. Contactors can't work with higher pulse rates than a few Hz. With continuous switching, their lifetime will be rather short.

I don't agree with your explanation of the observed "distortion", but without knowing the pulse generator circuit, I can't but guessing.

 

[Johansen]

This "distortion" appears to be 90% transmission line inductance between your pulse generator and the battery.
can you provide more info on what the waveform looks like?
some of this you can't get around, the battery itself will form a loop on the order of tenths to single digit uH, and 450 amps is a lot of energy stored in even a tenth of a uH.