Building Batteries with 1 Port Play and Charge

[martinwels]

Will lipo batteries work with the 2s cmb?

 

[Stitches]

Will lipo batteries work with the 2s cmb?

They'll work, yes. They're just not recommended for first time builds due to safety concerns.

 

[Space Puppy]

Can't you do the same thing with a circuit like this? Just a switching DC Jack and a Diode. The forward voltage of the diode must be taken into account, but should be a simple solution. Apologies for the badly drawn schematic.

 

[Aurelio]

Can't you do the same thing with a circuit like this? Just a switching DC Jack and a Diode. The forward voltage of the diode must be taken into account, but should be a simple solution. Apologies for the badly drawn schematic.

This is not enough. Lithium batteries can’t be charged with a constant voltage supply

 

[Space Puppy]

Yes you are right. The diagram I drew assumes the protection module has a circuit which stops charging the batteries once they reach 4.2V. I assume this is what you are talking about.

 

[Miceeno]

Can't you do the same thing with a circuit like this? Just a switching DC Jack and a Diode. The forward voltage of the diode must be taken into account, but should be a simple solution. Apologies for the badly drawn schematic.

What Aurelio said and also a mechanical switch is far too slow even for a CPU switching at 93.75mhz (93,750,000 times a second); otherwise a relay would be a viable solution as well. Think of a transistor like a variable resistor, like a potentiometer that switches extremely quickly. However it never truly completely disconnects the battery, it just puts a massive amount of resistance between it and the load, like a volume pot. The voltage from the battery drops out as the input voltage from the external power switches the diode back around and the capacitor keeps the voltage from dropping. This switching is more analog with the parts working in beautiful harmony to maintain voltage.

What you are trying to illustrate is how a guitar pedal works (and other simple, non CPU based electronics). This switching is very clever but not viable because a CPU has to have perfect, stable voltage. In these applications you generally use a center negative PSU. EEVblog has a great video on the topic.

 

[Aurelio]

Yes you are right. The diagram I drew assumes the protection module has a circuit which stops charging the batteries once they reach 4.2V. I assume this is what you are talking about.

Not really. Lithium batteries need first a constant current charging, followed by a constant voltage one. This is done because the internal resistance of lithium cells is very low, therefore applying a 4.2V/cell would push too much current in the batteries, destroying them and causing fire hazard

 

[Miceeno]

Yes you are right. The diagram I drew assumes the protection module has a circuit which stops charging the batteries once they reach 4.2V. I assume this is what you are talking about.

What Aurelio said and by abusing the over voltage protection of a battery protection board you can't get a full charge anyway. A smart charger switches from CC to CV while tapering the current down when it reaches 4.1-4.2 volts (see the graph in the original post). Without this switch the protection board will disconnect the battery at 70% charge.

 

[Space Puppy]

I believe I understand now, thank you for clarifying. I was under the impression that the circuitry inside the protection module will provide the CC and switch to the CV at the correct time. I would think this depends on the protection module? For example does this protection module work the way you two describe, or will it do what a smart charger does? I would think it would still need a smart charger from what you are both saying. https://www.amazon.com/gp/product/B00N48RCS8/ref=oh_aui_detailpage_o01_s00?ie=UTF8&psc=1

Additionally, Miceeno are you talking about issues with switching speeds in the device being powered? Such as a CPU on an N64 freezing because of an insufficient switching time for when the battery power and the wall power switch? I would think a capacitor would solve this.

I design guitar pedals for a living so I am familiar with the negative tip configuration, this is why I was hoping to use it.

 

[Miceeno]

I believe I understand now, thank you for clarifying. I was under the impression that the circuitry inside the protection module will provide the CC and switch to the CV at the correct time. I would think this depends on the protection module? For example does this protection module work the way you two describe, or will it do what a smart charger does? I would think it would still need a smart charger from what you are both saying. https://www.amazon.com/gp/product/B00N48RCS8/ref=oh_aui_detailpage_o01_s00?ie=UTF8&psc=1

Additionally, Miceeno are you talking about issues with switching speeds in the device being powered? Such as a CPU on an N64 freezing because of an insufficient switching time for when the battery power and the wall power switch?

I design guitar pedals for a living so I am familiar with the negative tip configuration, this is why I was hoping to use it.

That board doesn't provide smart charging. The dead give way is generally a smart charger has a coil on board. If you are planning on a 2s battery just use the Red Board described above.

As for the switching speed basically yes. If the CPU undervolts on any one of it's cycles the system crashes. It wouldn't be an issue if you could physically flip a switch at the clock speed of the CPU (and in sync). I haven't measured but I'm pretty confident I can keep up with a 8hz CPU. 8hz is 1/16th notes at 120bpm. If you rig the switch to a midi pad and give me drum sticks I could probably do 12hz (1/16th notes at 180bpm) more if I buzz rolled, lol.

 

[Space Puppy]

I was planning on using a 2S2P configuration with some 3.7V 3400mAh (7.4V 6800mAh) and using that protection board I linked since it has a high enough current rating. Was curious if I could get a charge and play with this configuration with that diagram I made.

I understand the switching speed of the mechanical jack is extremely slow. But the clock speed for a CPU is determined by the 3.3V regulator in most cases, lets assume, so if an RC time constant is used you should be able to slow the discharge of a cap enough to maintain the 3.3V line. Or one would think. Either way the lack of a smart charging circuit complicates the diagram I had in mind, as the smart charger wont know what to do when its trying to switch from CC to CV. I can either get a different protection module or use a simple FET circuit as you showed. Please correct me if I am wrong.

I like your drum stick stats bro, I think I got an easy 220BPM with my fingaz on a tap tempo, lol.

 

[Miceeno]

The MOSFET circuit is a diagram of how the Red Board works. It needs the right components to work efficiently (battery life). I haven't dug into this because of other hobbies robbing me of my time lately. The Red Board handles everything the only downside is the charge current is limited to 1 amp. The only advantage to building your own switching circuit is you can get a higher rated charger module. The only module readily available that I know of is the TP5100 which is limited to 2 amps of charge current.

 

[Space Puppy]

This is a PFET used for power switching in portable applications.
https://www.mouser.com/ProductDetai...DBzk1/Wi90KS4xNEreAqpw%2bazpwlG3n20a/hTm1HA==

I would imagine the schottky diode just needs a low forward voltage, I would imagine the 1N5819 would work for the circuit.
The pull down resistance is likely 100k or 1Meg. The cap value just needs to be big enough to hold a charge when the FET turns on or off.

 

[Miceeno]

This is a PFET used for power switching in portable applications.
https://www.mouser.com/ProductDetail/Micro-Commercial-Components-MCC/SI2305-TP/?qs=sGAEpiMZZMshyDBzk1/Wi90KS4xNEreAqpw%2bazpwlG3n20a/hTm1HA==

I would imagine the schottky diode just needs a low forward voltage, I would imagine the 1N5819 would work for the circuit.
The pull down resistance is likely 100k or 1Meg. The cap value just needs to be big enough to hold a charge when the FET turns on or off.

Yeah sounds about right. Let us all know how it turns out. In my original post I ordered a handful of parts so I could build a demo circuit as fast as possible with the intention of fine tuning it later but I got busy with another project. It turns out my "let's hurry and body work and paint a truck before it gets too cold" project turned into a two vehicle, two month ordeal because of rust repair and weather.

 

[Space Puppy]

Later today I will try to post a schematic of what I plan to implement. However, I will need a smart charging circuit. Not sure if anyone is familiar with stand alone smart charging circuits, but I know it is needed in addition to the other parts. Im going to assume the battery in the diagram Miceeno posted has a protection circuit on it separate than that of the smart charging circuit.

 

[Miceeno]

The TP5100 is a smart charger circuit. It's less than $4 shipped and it works.

The Red Board is $8 shipped and it does everything.

What are you trying to accomplish by building the whole thing from scratch? There are IC's that handle smart charging but you'll need to dive into the labyrinth that is Mouser to find them and all the additional components to make it work.

Aurelio is building some beautiful boards that handle the entire power side of a portable, including voltage regulation.

 

[Space Puppy]

I would use the Red Board, but the charging current of 1A is annoying if my batteries are 6.8A. Its going to take a long time to charge. The TP5100 is a better option if I want to use a 2s2p configuration with my cells.

I would prefer not to build the whole smart charger from scratch if that is what your asking. If Aurelio is selling/willing to share CAM files of said beautiful boards I would be inclined to go with that alternative to avoid any extra work . However, I would like a long battery life and a small footprint for my battery. Using four 18650 3.7 3600mAh cells in a 2s2p configuration should give me about 4-5 hours of play time on a N64 portable and a small enough footprint.

I was considering however moving to a 3s2p configuration as I have the space in my current design and would love the extra play time, but the 3s charge and play board is rather large. Having different modules, such as the protection board, smart charging board, and charge and play circuit is ideal as I can individually mount them where it is most convenient. This also allows for everything to be fairly cheap and swappable if I decide that I want to put more cells in parallel.

I will be honest in saying that this is based on my understanding of how I have seen most portable N64s made. If there is a different battery configuration that would be a better suited for what I am looking for I would love any suggestions.

 

[MasterNate]

I would use the Red Board, but the charging current of 1A is annoying if my batteries are 6.8A. Its going to take a long time to charge. The TP5100 is a better option if I want to use a 2s2p configuration with my cells.

You have to be careful with charging lithium ion cells. The faster you charge them, the more likely something bad could happen. Also, charging lithium ion cells fast can shorten the life of the battery.

 

[Space Puppy]

Of course, you dont want to exceed 1C of the individual battery cells. I plan to use a 2s2p config with 3.7V 3600mAh cells, 2A is just under 1C of 3.6A (in reality it will likely be more like 2.9A), so the TP5100 should work to charge the whole battery in about 3-4 hours. Much better than 1A which would take 6-8 hours. I could try to look for something which could charge at 4A since the batteries are in parallel, but I dont need my portable to charge that fast, and as you said it decreases the battery life.

 

[Booshman]

Thanks for all this useful info! I'm in the process of planning and buying all the parts for an N64p and sharing of info like this is really appreciated.