The question:
The battery has been charging for over 6 hours, and I feel like that's kind of excessive. Even after charging this long it seems to struggle with the lights. I see flickering, and if I increase the brightness the whole the stops respond and I have to reset it
My voltage meter comes in today so I'll be able to see actual numbers, but I feel like this batter shouldn't 1) take this long to charge and 2) struggle to light 42 lights at 10% brightness
The charger works properly, and will shut off at full charge, only if the battery is disconnected from any load. The TP4506 module does not support load sharing.
struggle to light 42 lights at 10% brightness
The LEDs draw about the same peak current, which could be as high as 42x60 mA = 2.5 Amperes, regardless of the brightness setting. It is only the average current draw that is regulated.
Normally it would take about 4 hours to recharge a drained 3500mA battery.
With something connected it will take longer. If the ESP is on and transmitteing it can draw a lot of current
LED brightness is regulated by turning them on and off too quickly for the eye to notice the flicker.
The longer the OFF time, the dimmer the LED appears to be, and the lower the average current. The peak current during the ON time is the same, regardless of the OFF time.
Huh, I didn't realize it was just an on/off state, I really thought it was adjusting the actual brightness level (which I guess it is doing)
The LEDs need 5V. Do you have a boost converter somewhere?
No I don't, but the datasheet lists +3.5~+5.3, so I figured the 3.7 from the battery would be sufficient. And I know this is anecdotal, but I've seen a bunch of projects others have done that power twice the amount of LEDs that I am with a single 18650 battery
Can you provide a link to your TP4056 module? And can you confirm that your Samsung 18650 is not protected? A schematic or diagram of your circuit would also be helpful.
If your TP4056 module is the version with protection built in (the DW01 chip), then the current the protection will permit may be limited to about two amps. So even if it's an 8A battery, any draw above 2A would trigger the overcurrent protection. Electronoobs has a recent video on this problem. Apparently the voltage drop across the twin mosfets triggers the protection, and that drop is a function of the RDSon of the mosfets. His solution I believe was using better mosfets with lower RDSon. Normally 2A would be plenty for a microcontroller project, but maybe not with all the LEDs.
I'll watch the video when I get home from work, thank you for that info
This is the module, ordered it awhile ago so didn't remember the specifics. It says that maximum charge current is 1A, does that mean an 8A battery will take 8 hours to charge?
The seller's page lists the battery as unprotected
Is there a tool to use to create a diagram? The way I have it wired is - TP4056 module: B+ and B- to the battery, the OUT+ and OUT- both both have 2 wires - 1 to the lights, and 1 to the ESP ESP: GND and 5V from the TP, D3 data pin to the lights, D4 data pin to momentary switch, and another GND to the other momentary switch pin
The setup works, switch does its thing and the lights turn on/off, change modes, etc
That is copied from the listing you linked. You could try connecting your circuit directly to the Samsung cell and test that way. If that works you may have to run your project in t.hat manner. You can do your own overdischarge protection using an analog input to monitor the battery voltage.
If that doesn’t work you will have to provide a schematic diagram of your circuit so we have a better understanding.
There are schematic drawing functions in the various board layout apps, but everything I've looked at so far is too complicated for just drawing a readable schematic. Also, most of those don't produce readable schematics. So I usually draw simple ones by hand, or with Paint.
No, those are both instantaneous rates. The capacity of the battery is what determines the charge time. So nominally a 3.5AHr battery charging at 1A would take 3.5 hours to charge. But that's not really the case. First, the battery may not be fully discharged, so would require less time to get it back to fully charged. Also, the charging process ends in the constant voltage stage where the charge current gradually declines. So the average charging current is less than 1A. Generally, it takes "a few hours" to charge.
But that not likely to be 6 hours as in your original question. The TP4056 terminates charging when the charge current at 4.2V drops below 10% of the full charge current. So if your load current during charging is greater than 100mA, charging will never terminate. Opinions differ on this, but I don't think that's something you want to let happen. Remember that there are really only two things in hobby electronics that can ruin your day. The first is anything connected to mains. But the second is rechargeable lithium batteries. So you want to take some care with them.
Anyway, if failure to terminate is a problem, then you'll have to look at adding a load sharing circuit, which lets your 5V USB supply power the load directly when present. That's just adding three additional parts, but I don't think you will find a module with load sharing built in.
Hey man, I just wanted to thank you for all of the info. Super helpful and interesting!
I ended up doing the following:
Cleaning up my soldering on the battery
Used a different ESP device (same board, but a second one I had laying around)
One or both of these seemed to have fix my issue. I don't see the option to close the thread, but I'll mark your earlier response as a solution because it's what got me to reevaluate my setup
Giant thanks to everyone that commented on this, I really appreciate the insight and knowledge dumps
I made sure it was depleted, charged it for about 4.5 - 5 hours and gave up waiting. Plugged in my controller and lights and went to bed. 7 hours later they were still going
A few months ago I tested just with the battery charging via TP module, and the LED indicator eventually went from red (charging) to blue (fully charged). I forgot about it and it sat over night, so no idea how long it took. I also later found out it's bad to let it charge over night, so I'll watch for that
TL;DR: ~5 hours it never reached a full charge, I'm wondering if it's due to the power drain of the ESP, but had enough juice for more than 7 hours
The problem would be if it actually did reach full charge, but charging didn't terminate because of the load current. Then you would be continuing to apply 4.2V to a fully-charged lithium battery. You shouldn't do that.
Just a thought.
I made a few solar LEDs for the wife just before Christmas.
Cheapo 5V LED strings. Cheapo 6V solar panels. Lipo cells harvested from disposable e-cigs. TP4056 module and a little boost converter.
The schematic was stolen from one of Big Clive's solar projects. Very simply just a transistor, diode and a couple of resistors.
They have worked without failure, every day.
The last one I made about six months ago involved an ATtiny85 so that I could "twinkle" the LED string. Again a 6V solar panel and TP4056. This failed once when the cat from next door got angry and attacked the solar panel enclosure, but apart from that, it has worked perfectly.
They don't draw huge amounts of power so the cells will last all night and then get fully charged during the day.
