a switchless battery charger. Is this circuit OK or is the author unreal?

[mik3ca]

I'm doing some research because I looked at commercial chargers for my batteries and I thought maybe I can save $20+ by making one that could even be better than the commercial ones already out there.

Here is the URL:

**broken link removed**

It claims "This circuit may be used to replace the single current limiting resistor often found in dirt cheap battery chargers. The alternative shown here will eventually pay off because you no longer have to throw away your NiCds after three months or so of maltreatment in the original charger."

And here is the circuit the website shows:

**broken link removed**

If I use this circuit, then of course I'd eliminate the second and thirs sets of transistors, diodes, resistors and batteries. and keep only the first set. (meaning I'd remove D5,D6,T2,T3,R3,R2,D2,D3).

Question is, would this circuit reliably charge a large capacity battery (about 4000-6800mAH) in a few hours? or is the author of the article on the site nuts?

 

[dknguyen]

Reliable? Questionable due to all the heat generated. Maybe reliably if it was heatsinked enough. Efficiently? No.

That circuit just turns all the extra voltage into heat and can never charge batteries with a higher voltage than the input voltage.

Modern large capacity chargers are switching. Much more efficient, much less heat, and therefore smaller, and don't requires the input voltage to always be higher than the output voltage.

If your battery is expensive, don't skimp on the charger. That circuit is better than a resistor but is by no means good. I wouldn't waste my time on it but maybe you're really cash strapped for a charger.

 

[mik3ca]

I might have found a starting point for one. I'm looking at documentation from https://bristolwatch.com/ccs/LM317.htm and the website states you can use that regulator and a resistor to produce a constant current output which they say can trickle-charge batteries if the input voltage is high enough. They show this circuit:

But then again someone says modern regulators are switching. So would I be able to make the above circuit switching if I cut wire somewhere in this circuit and replaced it with an oscillator of some sorts?

 

[dknguyen]

I might have found a starting point for one. I'm looking at documentation from https://bristolwatch.com/ccs/LM317.htm and the website states you can use that regulator and a resistor to produce a constant current output which they say can trickle-charge batteries if the input voltage is high enough. They show this circuit:

But then again someone says modern regulators are switching. So would I be able to make the above circuit switching if I cut wire somewhere in this circuit and replaced it with an oscillator of some sorts?

Switching circuits are an entirely different animal. The simplest switching charging circuit would a buck regulator working in constant current mode. Do you know how a buck regulator works?

They get more complicated from there (with smart monitoring and charge algorithms so they can fast charge the battery without damage, or accept lower input voltages to charge higher voltage batteries, etc).

That said, switching chargers are pretty big projects on their own.

If you have an bench supply with adjustable current limiting, it will do the same thing as your constant-current charging circuits (just don't set it wrong or you'll blow up the battery). It should be as efficient as the circuit if it's a linear supply. More efficient if it's a switching supply.

 

[mik3ca]

Sounds like I may need a new set of IC's to get something rolling. I'm surprised the charger circuits shown all over on the internet are mostly the linear type from commonly available parts. My goal is to completely charge a 6800mAH battery from dead to full

 

[dknguyen]

Sounds like I may need a new set of IC's to get something rolling. I'm surprised the charger circuits shown all over on the internet are mostly the linear type from commonly available parts. My goal is to completely charge a 6800mAH battery from dead to full

That's because switchers are complicated, let alone switching chargers. Switching battery chargers are enough effort that they almost always processor controlled because for just a little more effort than what was already required you get a lot more flexibility. You can get switcher ICs specifically for battery charging instead of getting a regular SMPS switching IC and building it to be a charger.

How much voltage is this battery and what chemistry?

 

[mik3ca]

7.2V NiMH. I'm tempted to go for a MAX713 IC and follow the schematic included in the datasheet but I'm gonna read it more first. Yes the chip is a bit expensive but its ok since I won't make that many chargers.

 

[dknguyen]

7.2V NiMH. I'm tempted to go for a MAX713 IC and follow the schematic included in the datasheet but I'm gonna read it more first. Yes the chip is a bit expensive but its ok since I won't make that many chargers.

That's not a switching IC. But if you're going to make your own linear charging circuit you might as well go with a nice one like that since it will be more functional (fast charge algorithm!) and relatively simpler than any of the other ones you listed. A switching IC will require an inductor or have one internally and they will advertise this fact if it is the case.

 

[mik3ca]

hmm.. What IC should I go for then? I mean (as you should know), I want to charge a 6800mAH battery within a few hours from dead status. If there's an IC specifically for a task like this that isn't ultra expensive I'd order it from an online electronics store

 

[dknguyen]

You're fine with that one. I'm just letting you know it's not a switcher. 'm not certain (I haven't looked) if there are switching ICs for NiMH. NiMH is getting replaced by Lithium Ion so there are plenty of ICs for those. And a few hours isn't really considered a fast charge either so you're probably fine.

The MAX713 can potentially charge your battery in 15 minutes if you let it. The problem you may run into trying to get that is the heat to be dissipated since it is a linear charger. With a 6800mAh battery that's 27A of charging current at that rate which is too much to be dissipated linearly in most cases.

 

[mik3ca]

So after all this I should do the simple regulator circuit I showed? and one document showed C/10 as an OK charge rate for overcharging, but what about C/5? hmm..

 

[dknguyen]

So after all this I should do the simple regulator circuit I showed? and one document showed C/10 as an OK charge rate for overcharging, but what about C/5? hmm..

I guess, haha. Survival at C/5 is dependent on how much trickle charge the battery can take when full. You don't want it to trickle charge for hours after it's already full. That's not good for the battery.

But C/5 at 6.8Ah is still more than 1A through a linear regulator dissipating the excess as heat. Calculate if your LM317 can actually handle that for your desired input voltage and charge current?

Temp Rise above ambient = (Input voltage - output voltage)*Current*Junction to Ambient Thermal Resistance

 

[mik3ca]

well I could pick a different regulator and I'll likely check the progress every 30 mins.

 

[dknguyen]

well I could pick a different regulator and I'll likely check the progress every 30 mins.

As long as it's a linear regulator it won't matter. The wattage dissipated is always the same. The only thing you can do is pick a package that dissipates heat better (rather difficult) or heatsink it better (might still not be enough since at 1A you could easily be talking about multiple watts and a temperature rise of 30C+ per watt).

 

[ChrisP58]

An important part of properly charging batteries is terminating the charge when the battery is full. A good battery charge control IC, such as the Maxim or Texas Instrument parts, will do that. The simple circuit in your first post won't. It would just keep on dumping current into the battery long past the optimum full point.

 

[mik3ca]

An important part of properly charging batteries is terminating the charge when the battery is full.

Thats why I was inquiring on the very first circuit in my post. Maybe its more efficient? but then I could throw in some zeners and cause a cut-off when voltage is over a limit?

What about the charger (schematic below) from here: https://powerfromearth.blogspot.ca/2015/05/diy-12v-solar-panel.html

**broken link removed**

except I could replace the darlington with a TIP41 and limit current maybe? But does that zener diode approach seem to work (ZD1) I mean I want a method that will work wonderful before I attempt to build this myself. and I'll replace the solar panel with a 9-12V wall wart and capacitor across it. and I may remove ZD2 or maybe replace it with a 7.2V zener to match the voltage of the battery I'm trying to charge and maybe ditch the LED and resistor in series with it because to me those two things are accessories.

What do you guys think? Is this a good circuit for nimh batteries for charging at C/5 or faster?

 

[MikeMl]

NiCd and NiMh require a constant-current charger for several hours, and then they need to be disconnected. NiCd and NiMh do not have a reliable, detectable voltage rise when they reach full charge like lead-acid chemistry; rather prior to reaching full charge, most of the charging power goes into the electrochemistry, and when they reach full charge, the power is dissipated as heat, causing a very noticable temperature rise. For this reason, most NiCd and NiMh battery packs have an embedded thermistor, and the matching charger watches the temperature of the battery back, and terminates charging when the temperature increases by X degrees.

Read the treatise on charging NiMh at the Battery University....

Most of the charger circuits you show here are constant-voltage chargers and are totally inappropriate for NiCd or NiMh.

 

[mik3ca]

...and the matching charger watches the temperature of the battery back, and terminates charging when the temperature increases by X degrees.

Matching? This is strange because I bought a 6800mAH and a few 4000mAH from two different sellers on ebay and each order came with a charger but the charger has 200mA label on it which means 20+ hours to charge the battery. I'm gonna need something better than that. I've seen some that do 2A for charge but that seems to violate the C/10 rule for charging batteries?

NiCd and NiMh require a constant-current charger for several hours, and then they need to be disconnected.

Thats why I was suggesting the last circuit but I guess something with it is not quite right?

Rather prior to reaching full charge, most of the charging power goes into the electrochemistry, and when they reach full charge, the power is dissipated as heat, causing a very noticable temperature rise.

So this means I should only use chargers with built-in thermistors to sense when any nimh battery is hot? (I haven't seen a charger in my life with a built-in thermistor but if thats the main issue, I'd be tempted to add a thermistor to the last circuit and attach it to the NPN base so when the surroundings are hot enough, the charger turns off and I could add a loud squealing circuit too so I hear when the charging is complete)

For this reason, most NiCd and NiMh battery packs have an embedded thermistor

Since mine come from ebay, I could assume mine doesn't come with one, but how will I know for sure without literally ripping open the battery pack if a thermistor is included? Do I change the temperature of the battery pack and look for a special measurement on a multimeter?

and the matching charger watches the temperature of the battery back, and terminates charging when the temperature increases by X degrees.

So these thermistors must then be inside the charger itself? or?

Most of the charger circuits you show here are constant-voltage chargers and are totally inappropriate for NiCd or NiMh.

Ok. so my best start so far is that mini current regulation circuit but I just have to somehow expand on it.

 

[ChrisP58]

Thats why I was inquiring on the very first circuit in my post. Maybe its more efficient? but then I could throw in some zeners and cause a cut-off when voltage is over a limit?

What about the charger (schematic below) from here: https://powerfromearth.blogspot.ca/2015/05/diy-12v-solar-panel.html

except I could replace the darlington with a TIP41 and limit current maybe? But does that zener diode approach seem to work (ZD1) I mean I want a method that will work wonderful before I attempt to build this myself. and I'll replace the solar panel with a 9-12V wall wart and capacitor across it. and I may remove ZD2 or maybe replace it with a 7.2V zener to match the voltage of the battery I'm trying to charge and maybe ditch the LED and resistor in series with it because to me those two things are accessories.

What do you guys think? Is this a good circuit for nimh batteries for charging at C/5 or faster?

You do want a constant current source. There are many choices for that. Since you are going linear, one will be no more 'efficient' than another. They will all dissipate heat based on the same equation, (Vin - Vout) x charge current. You want enough headroom to take the battery to 1.4 Volts per cell.

There is no fixed voltage point that tells you that the battery is full. Yes, many chargers get away with doing so, but they risk either undercharging the battery, or damaging the cells by overcharging. Read the sections of the datasheet for the MAX713 titled "Voltage-Slope Cutoff" and "Temperature Charge Cutoff."

 

[dknguyen]

C/10 is not a rule as much as a rule of thumb when you know nothing about the battery and have no smart algorthm since it is unlikely to overcurrent the battery. You can charge NiMH at 4C if you know what you're doing.