Genius! Please tell me the theory which apply on this

[ranatungawk]

I found this circuit in a cheep rechargeable LED touch. However, as far as I could understand, I saw an issue of this circuit (other than the safety issue). So genius! Please tell me what I think on this, is correct!!

In order to drop the voltage, a Mylar Capacitor is used. Further, the Ni-Cad battery (3.6/700mA) will be charged with 4V/70mA source. However as we know, when the battery-charge go up (after 1-2 hours), the drain current from the source go down (<70mA). However, at the start, the charging current and the voltage was calculated according to “V=IR” and then a capacitor was selected in order to fulfill the above charging-values. So according to this argument, after 1-2 hours, charging voltage on the battery must be high than starting 4V (Ex; if I=30mA then voltage on the battery should go up to 134V). can this be happened ?? Can it make any harm to the battery?
View attachment 69025
Please see the attach diagram

 

[audioguru]

Why is this garbage on TWO (or more) websites?
On the other website we told you that it will blow up.

 

[dr pepper]

The answer to your question is within the schematic, the capacitor acts as a resistor dropping the mains volatge to the battery volatge.
This circuit is dangerous and shouldnt be built, audio has a point on this one.

 

[Timescope]

Replace this circuit with a small transformer and rectifier circuit then use a resistor to set the charging current. Electricity is a good slave but a bad master.

Timescope.

 

[MrAl]

Hi,

Off line power supplies are a lot more common than most people seem to think. However, when there is human intervention with parts of the circuit you have to be a LOT more careful. Since one side of the battery is connected to one side of the line, there is always a shock hazard. If you unplug the circuit before you swap the batteries to be charged then there's no problem, but what if you forget and touch one of the live terminals and happen to be grounded? You better have a ground fault interrupter and you better hope it works. That's why everyone is suggesting that you dont use this circuit but instead use a transformer type circuit which is safer to handle.

Off line power supplies are used when there's no chance that there can be any human contact with any part of the circuit. Steps are always taken to make sure this cant happen in a professional design. These days wall warts are very common for circuits that might have a little human contact and those things usually incorporate double isolation from the mains line, so you can see that this is something that is taken quite seriously by the industry.

 

[4pyros]

These days wall warts are very common for circuits that might have a little human contact

They are also cheep and most of them are certified to be safe.

 

[ranatungawk]

ust needed to get clarified some basic theory-parts

Dear friends;

Thanks for the information and advices ! i really know this is very unsafe and very primary design. as well as i'm not going to build this. To be frank i just needed to get clarified some basic theory-parts from experts like u guys, by using this. Other than a few members, most of them were talking on the un-safety side of the diagram than useful theory matters.

However i have made some changes. Please see the attachment.
1. i put a full-wave bridge rectifier for pure DC
2. put a 1uF 450V Cap for extra safety
3. put a fuse at the main AC input

i have seen these type of re-chargeable torch are being widely used and sold in online markets like E-Bay!! And they are not too bad as it seen.

So please help me to understand following matters.

1. one of my friend has suggested to install a 100 ohms in series with the AC input to limit inrush current! - Doesn't the bleeder resister (R1) do this ? or is it necessary to put another one ?

2.what do you think on my main question: "Further, the Ni-Cad battery (3.6/700mA) will be charged with 4V/70mA source. However as we know, when the battery-charge go up (after 1-2 hours), the drain current from the source go down (<70mA). However, at the start, the charging current and the voltage was calculated according to “V=IR” and then a capacitor was selected in order to fulfill the above charging-values. So according to this argument, after 1-2 hours, charging voltage on the battery must be high than starting 4V (Ex; if I=30mA then voltage on the battery should go up to 134V). can this be happened ?? Can it make any harm to the battery? "

thanks for your time to write me.
best regards!
Kushan

 

[4pyros]

1. one of my friend has suggested to install a 100 ohms in series with the AC input to limit inrush current! - Doesn't the bleeder resister (R1) do this ?

No R1 just takes the charge off the cap.

Can it make any harm to the battery? "

Yes it can over charge your battery.

 

[ranatungawk]

No R1 just takes the charge off the cap.

Yes it can over charge your battery.

Thanks! pls tell me to where should i put a resister to limit inrush current ? how can i decide the value of the resister ? (mathematical formula ?)

 

[audioguru]

A "3.6V" Ni-Cad battery is fully charged at about 4.2V. The voltage must be limited to no more than 4.2V (then the charging current decreases) or the battery will continue over-charging at 70mA (it might reach 4.8V) which makes it hot. A cheap battery might burst and release deadly Cadmium all over the place and ON YOU.

Energizer shows that their obsolete Ni-Cad AA cells (they replaced them with Ni-MH cells) still work with a small capacity loss after two years of over-charging at 60mA. They say not to over-charge Ni-MH cells. If you over-charge a lithium battery then you will have a very hot fire!

 

[alec_t]

However as we know, when the battery-charge go up (after 1-2 hours), the drain current from the source go down (<70mA).

No it won't! There will be a negligible change in the current because 230V is very much greater than 4V. The battery, when fully charged, will still have ~70mA forced through it and will probably overheat. That is one reason (apart from shock hazard) why this circuit is dangerous and shouldn't be built.

 

[MrAl]

Dear friends;

Thanks for the information and advices ! i really know this is very unsafe and very primary design. as well as i'm not going to build this. To be frank i just needed to get clarified some basic theory-parts from experts like u guys, by using this. Other than a few members, most of them were talking on the un-safety side of the diagram than useful theory matters.

However i have made some changes. Please see the attachment.
1. i put a full-wave bridge rectifier for pure DC
2. put a 1uF 450V Cap for extra safety
3. put a fuse at the main AC input

i have seen these type of re-chargeable torch are being widely used and sold in online markets like E-Bay!! And they are not too bad as it seen.

So please help me to understand following matters.

1. one of my friend has suggested to install a 100 ohms in series with the AC input to limit inrush current! - Doesn't the bleeder resister (R1) do this ? or is it necessary to put another one ?

2.what do you think on my main question: "Further, the Ni-Cad battery (3.6/700mA) will be charged with 4V/70mA source. However as we know, when the battery-charge go up (after 1-2 hours), the drain current from the source go down (<70mA). However, at the start, the charging current and the voltage was calculated according to “V=IR” and then a capacitor was selected in order to fulfill the above charging-values. So according to this argument, after 1-2 hours, charging voltage on the battery must be high than starting 4V (Ex; if I=30mA then voltage on the battery should go up to 134V). can this be happened ?? Can it make any harm to the battery? "

thanks for your time to write me.
best regards!
Kushan

Hello again,

First, yes you need a surge resistor of maybe 100 ohms but it should probably be higher than that. This resistor limits the surge current when the circuit is turned on at the peak of the sine wave after sitting for a while with no power. The current peak is limited to:
Isurge=Vpeak/R
so if you have a 220vac line your peak is 311v so the surge current is:
Isurge=311/100=3.11 amps which is still quite high. Going to 220 ohms might be better, with the appropriate power rating.

What the bridge rectifier does is makes the cap value more effective in delivering current to the circuit. With half wave you get less current, with full wave you get more current, with the same value cap. To get the most out of the cap then you need to use a full wave rectifier.

The other issue is the charge current over time. As alec pointed out, the input voltage is much higher than the battery voltage so the only thing that limits current is the parallel LED and 1k resistor, which doesnt say much because the 1k resistor only passes at most 4ma (even without the LED). So you need to add current limiting.
The simplest current limiting comes from adding a zener diode of the right voltage rating in parallel with the battery, but also with a resistor between the zener and battery to actually limit the current at any time, not just near the end of charge.
The zener limits the voltage getting to the resistor and battery, and the resistor limits the current. The current then decreases over time as the battery charges up.
The drawback here is that the battery voltage cant be exactly specified, and the zener voltage cant be exactly specified either (although better than the battery). So it's hard to make a perfect circuit like this.

The best bet is probably to use a charge regimen that includes charging the cell for a limited amount of time, knowing something about the nominal current flow during charge. To do this you have to measure the current with a given set of batteries and calculate how long to charge them given their initial state is zero charge. That's about the only way to do this and not ruin the batteries. It's not that good of an idea unless you are willing to take the time to measure the current and do the calculation, and it must be done over for a different set of batteries and also again as they age.
Then again if you are going to time it (which you need to do anyway) then you probably dont need the zener and resistor addition. The current will be roughly 100ma.

There will still be some amount of overcharge, but with this new zener the overcharge will be reduced if you do it right. So the next thing is to figure out which batteries can stand this kind of charging, even with the timer.
NiCd's can take a bit of overcharging without a lot of damage. Older NiMH cells cant take much overcharging at all, but the newer ones can take a lot more abuse. I've read that they can stand 200ma overcharge for a significant amount of time now, but i havent actually did any tests to confirm this. They said the same thing about NiCd's in the past, and that lead to designs that CONSTANTLY overcharge the batteries and of course early failure. So the first rule then is dont overcharge for too long, if at all.

We've talked about NiCd's and NiMH cells, but what about Lithium-Ion? Sadly the answer is a resounding, "DEFINITELY NOT". In other words, if you use this charger or anything like it for Li-ion cells you will create a dangerous situation where the cell could really blow up and start burning right before you. So the main point here is that this charger can never be used with Li-ion cells.
Granted, it would charge them, but since there is no exacting voltage control they would seriously overcharge and blow up if left on their own. You'd have to sit there with a volt meter and monitor the voltage at all times and maybe halt the charging when they reach a safer 4.150 volts. I dont recommend this procedure however because if you leave the room and they overcharge, you'll return to a fire.

So just to recap:
1. NiCd's and NiMH's might work ok with the charger at low enough currents (less than 200ma) with the zener and resistor addition, and a time limit on the charge and starting from zero charge.
2. Absolutely not to be used with Li-ion cells.
3. Shock hazard doesnt go away until the charger is unplugged from the wall so extreme caution is recommended.

 

[Mr RB]

The circuit as shown is a half-wave capacitive reactance supply, and will supply approx 35mA constant current NOT 70mA. As the battery is 700mAh the charge rate is 1/20th C which is reasonably safe for continuous charging (ie 24 hours). The LED and resistor might even drop that 35mA to 30mA or so.

I have one of these type torches and it has worked fine (same battery) for about 10 years, the only fault is that the cap failed open circuit about 3 years back and I replaced it with a slightly better spec cap. The reson for cap fail (like in the circuit above) is that there is no series resistor with the cap, this is generally considered bad practice in cap reactive supplies as it exposes the cap like a short circuit to any high power mains noise. Generally a resistor dropping about 5% to 10% of the mains voltage is used in series with the cap for a better (more reliable) design.

 

[ranatungawk]

Hello again,


So just to recap:
1. NiCd's and NiMH's might work ok with the charger at low enough currents (less than 200ma) with the zener and resistor addition, and a time limit on the charge and starting from zero charge.
2. Absolutely not to be used with Li-ion cells.
3. Shock hazard doesnt go away until the charger is unplugged from the wall so extreme caution is recommended.

Dear Mr. Al;

Thanks for the very clear and simple answer! further, spending your precious time to write such a detail answer is highly appreciated.

i have a few little doubts on some matters, if it's possible, please answer them when you can.

1. when the resister value is calculated to reduce the " surge current" effects, what need to be considered ? you have said "Isurge=311/100=3.11 amps which is still quite high. Going to 220 ohms" 3.11A is quite high >> against what value ? is it against our charging value 70mA ?

to be in the safe side, is there a limit or ratio: such as the MAX " surge current" should be up to this limit?

according to the circuit, the main component to drop voltage is 1uF cap, so, i put a 220Ω resister, should i need to consider the resister value when calculating the CAP value too ?

2. i assumed that by putting a 1uF cap (Cx=3200Ω) i would be able to get 70mA/ 4V to my battery , is it wrong ? (specially when we don't know the charging current and voltage of the battery )

3. how about using Lead–acid battery (4V/700mA) instead of others types ? is it safe than others?

4. i hoped to build a rechargeable camping-lamp (with LEDs) using removed Li-iron batteries from laptops. if it's charged only one hour when i see dim LED light, is the common risk still there ?

Thanks and regards!
Kushan

 

[MrAl]

Hi again,


Before i start, i have to ask the question that's on everyone's mind here i think...
That is, why dont you want to use a wall wart type transformer?

1. Since 100 ohms leads to about 3 amps surge, then 200 ohms leads to about 1.5 amps surge. I like to get as close to 1
amp as possible so that we dont tax the other components including the fuse itself. Fusing current goes by time as well as
current, and the higher the current the faster the blow out time. You may get away with 100 ohms but 220 is better.
The 220 ohm will require a 10 watt rated resistor, but You need a power resistor there anyway even with 100 ohms.

2. You will probably get more close to 65ma average charge current (multiply your result by 0.9). With a 2uf cap you can get up to 130ma.

3. Lead acid battery will charge too, but again you have to keep an eye on the voltage like Li-ion or it could cause
problems.

4. You can not specify just a time for charging an Li-ion battery at a known current unless you know the capacity of the
cell. The capacity of the cell goes down with age so it may be hard to judge this parameter.
The best i can do here is tell you that in order to avoid problems the voltage has to be monitored and charging has to stop
when the cell reaches 4.200 volts, but a better safer voltage is 4.150 volts which takes into account the inaccuracy of some
meters. The very high efficiency Li-ion cells can go up to 4.35, but that's a different kind of cell really and you better
be sure you have that type before you allow the voltage to go as high as that.
Now theoretically if a cell has capacity of 1AHr and it is totally discharged, then we can charge it with 100ma for 10 hours
and it still wont be charged to full capacity so we're safe. But if we misjudge some measurement or something else happens
that we didnt think of it could mean disaster. So avoid any charging of Li-ion cells unless there is some way to monitor
the voltage.
I've actually done this in the past, but i was willing to sit there and watch the volt meter waiting and waiting for the
voltage to get close to 4.150v, but that's because i was doing a test and i didnt mind sitting there all that time. That's
what it takes though, unless we use a monitoring circuit of some type. These days the circuit of choice is usually the
microcontroller chip, one type or another.


So just to put this into perspective for you...
You can get away with overcharging several types of batteries, but not the Li-ion type. That also goes for discharging
because you can not discharge the Li-ion too low either without risk of damage in the form of electroplating of the
electrodes which makes the two terminals of the battery physically closer together and thereby increases the risk of
internal short (an external protection circuit does nothing here to stop this either). An internal short causes fire and
possibly explosion.

Lead acid is a little more forgiving, but to get long life you have to charge that type correctly too.

NiMH and NiCd are usually charged with a controller that monitors the voltage, and even possibly the temperature. If you
want to charge it with a known current for a known period of time, then you should make sure the battery is discharged all
the way before starting. That way you'll get long life out of the battery.

I take my batteries quite seriously as some of them cost quite a bit of money to purchase if they start to fail and have to
be replaced. This reminds me of the time i bought a "Jump Pack" for the car and that was 10 years ago, and to purchase an equivalent battery now would cost twice that.

 

[dr pepper]

I love it when someone mentiones a off line capacitive dropper, theres allways a good response.

 

[unclejed613]

er..... um...... easy for you to say..... i do remember somewhere in the rules for this board that we post in English.....


yeah, direct line powered capacitive dropping supplies are DANGEROUS. just because XYZ company uses them in their designs, does not mean they're safe. some countries have very lax safety requirements, others do not. people seem to forget there are TWO common failure modes for capacitors, open circuit and SHORT. open circuit is not a big problem, but a shorted cap will set the battery on fire, as well as the rest of the charger. all it takes is a good size voltage spike to do this to most capacitors. so the question remains, why would you want to build this? the only reason the manufacturer made this is because it's cheap, less than $0.50 US in components, where a wall wart would have cost $1.00US.

if you want to continue on this course, i'm sure i could scrounge up a Wile E. Coyote Super Genius Award for you......

 

[Ian Rogers]

Sorry Guy's... I'm on my own.... I keep missing one or two... Gone now..

 

[MrAl]

Hi again,


Ian, did you post to the right thread or did i miss something somewhere

Unclejed, the shorted cap failure mode should be covered by the fuse. We could look at this in more depth too i guess: what else could go wrong...
A long time ago it used to be "ok" to tie one side of an audio amplifier to the line neutral, but people started getting killed when the plug was put in the line socket flipped because that connected the metal of their microphone right to the live side of the line. Needless to say, that was dangerous. I compare these circuits to those a little because the human can easily come into contact with one of the battery terminals of the charger when inserting or removing cells. So same thing can happen.

Some ideas:

1. We'd have to come up with a way to make it physically impossible to come into contact with the charger cell terminals if the unit was plugged in. Is this even possible without huge cost. This means when the unit is plugged in if some sort of battery 'cover' is open the line is disconnected and doesnt connect unless the cover is closed (like microwave oven door) , or some other idea.
2. It would be interesting to look on the web and make a total cost of parts list for both a wall wart version and a simple offline version, then compare the prices. I bet they are not hugely different if we go by the average cost of parts. Wall wart prices vary quite a bit from place to place, but 9v low current models are usually pretty cheap (say 100ma output). Anyone here up to the task? Would be very interesting.

 

[Ian Rogers]

No... I'm on spam alert ( uncleged613 saw it... and it was reported) When I remove the post... The continuity is a bit peculiar...