TP4056 myth busting
Вставка
- Опубліковано 2 жов 2024
- Following another video where I showed that you can make a handy emergency backup light from salvaged lithium cells, I got a few messages implying that with a load connected to a cell the charging may not stop and risk overcharging the cell.
To be fair it does look like that is happening, but in reality its down to the way the TP4056 and its many clones and variants indicate the end of charge state.
I set up an experimental rig to monitor current and voltage, and confirmed that the reason the charge never seems to end is because of the programmed current threshold the TP4056 chips use to detect end of charge.
When you set the desired charge current on these chips with the programming resistor it doesn't just set the charge current, but also a threshold of one tenth of that current that is used for both initial trickle charging to get an over-discharged call gently back up to 3V, and also the point at the end of charge when the current gradually drops to that lower current level.
If the load is above that level then the current will gradually tail off at the end of charge, but because it ends up powering the load directly it never reaches the end of charge threshold.
It will only be supplying enough current to drive the load and hold the cell at its nearly fully charged state, but will just not switch the LEDs to display charge completion.
With a load current lower than the end of charge threshold it will terminate the charge, but then kick back in again when the voltage drops to a slightly lower voltage, and top the cell up again. That results in the charge status LEDs slowly toggling back and forth.
The cell will not be charged above its voltage limit in either instance.
If you enjoy these videos you can help support the channel with a dollar for coffee, cookies and random gadgets for disassembly at:- www.bigclive.co...
This also keeps the channel independent of UA-cam's algorithm quirks, allowing it to be a bit more dangerous and naughty.
#ElectronicsCreators
Thank you so much for shrinking yourself down to show the PCBs in greater detail. It’s very appreciated and I know it’s difficult on your physiology
All without reducing the tone of his voice to a squeak - Amazing!
@@Derek_Garnham he probably fixed his voice in post.
Now we can sleep peacefully ever after :~}
First used in 50's films then modified to Wonka Vision.
He's breathing SF6 with oxygen in order to lower it back to normal.
I love the combined charge/boost circuits that are available now.
Input 5V, charges your battery and output can be set between 4.2 and 27V.
Nice for 9V and 12V applications.
Do you have any suggested part number / search terms for those? Sounds super handy.
@@KeithJewell CD42 probably the smallest and best ones. Stay away from the powerbank "usb out-microusb in" modules. They are getting REALLY hot.
@@KeithJewell I think your best search term would be "buck boost converter". I don't know if that's what op talked about or if it's even the correct term, but you'll find it that way.
Looks like MH-CD42 is the best search term, for anyone interested.
Just be careful with what you want to power with these chips. Most claim a minimum current of 30-50 mA. I ordered two different boards with two different CD42 clones, and both disable the output already below 80-100 mA. No fun for low current applications.
I picked up some of these modules after watching your previous video. Ridiculously useful and ridiculously cheap 😀👍🏻
Good for modding lights that dont require more than 4V to be powered by the module under USB power as well instead of using cells.
Me too, but am still waiting on the shite that is evri to deliver them been in this country since 9th November, last updat was 14th November... Did I say I think evri are SHITE... TFS, GB :)
I like those too. My last project with the TP4056 was modding my wife's epilator, together with a buck converter when the original NiCd battery died, and it is working quite good without overheating and a smother motor than in the original circuit.
Me too.
I bought five.
Me too I bought 15 for $10. I powered a marble machine and some custom lights so far but have more plans. Very easy to work with.
I have several 4056 modules, very useful
I find the best way to explain it is to compare to a water storage tank with a float, if you don't use water, refilling stops, if you start using water the input starts, but will never overflow ;)
I love your analogy Sir! Kinda like a standard diode (say IN 4007) is akin to a one-way valve with a small step into it; Mr Voltage has to rise over the step before Mr Current can get through, he can then go forward but he can't go back?
Of course you could explain the theory (should explain, depending on the audience) but if you impart just that much knowledge, then the recipient can understand it in a basic functional way. The only flaw to my logic (and yours also Sir, no offence intended) is that the spotty-faced oik that is nodding his head wisely hasn't got a 'bulls notion' of a one way valven or a water storage tank either!
God, I miss good 'old fashioned' hardware electronics, in industry, it is surely a dying art.
Pax
@@conmcgrath7174 I understand your point, it's an impossible task to make an anology for someone who can't even figure out how to sit the right way around on a toilet. It's a sad example of our societies willful ignorance and so-called education-system. I despair sometimes, I really do.
As a relative amateur in newbie in the electronics hobby. Your videos are very inspiring. I feel like I'm sitting in a classroom learning about something I actually want to learn about. And have actually used these modules in a couple of projects. Thank you so much for posting these videos
I respectfully disagree. You are right in that keeping a load on does not 'overcharge' the battery, but keeping the cell at 4.2V does stress it out, and that is why charge termination exists: after the cell is done charging, it can 'relax' and the voltage will fall back to a lower one. If you never terminate the charge cycle, the battery is kept at the stressful 4.2V, and that'll cut into its life. I've destroyed LiIon cells that way, and now I add the diode+mosfet required to run my devices from 5VUSB whenever the LiIon is charging.
I know, this is a bit late, but... a similar chemistry, LiFePO4, does not care if you keep the voltage floating, which is why many modern garden solar lights now use them. They're also showing up as drop-in replacements for vehicle start batteries. A disadvantage is that the charge voltage is 3.6V instead of 4.2V, so your circuit needs to take that into consideration.
Add a diode in series at output terminal + can work for charge Lifepo4 ?
Could you please share the circuit diagram for the diode and mosfet connection
A usual point and it sounds valid. Can you suggest a modification that would correct or improve the situation?
No matter the day I've had, I can always come home and feel like I've learnt something new when I watch your videos, so thank you for that!
Incredible timing, Clive. I've been looking for lithium charger modules for some solar powered ESP8266 sensors I'm building... I actually went searching for your older video just yesterday to check which chip it was people had said you couldn't use for a constantly connected load.... And with incredible timing, you're back to help me out again 👍
ESP8266 works best on LiFePo4-batteries. Voltage range of those is 2.8V - 3.5V, which is exactly the same as on ESP8266. LiFePo4 is also more resilient, similar to NiMH. You can tricklecharge and overcharge it. No special charging circuitry needed.
@@TimoNoko you're right, but I guess lipo or liion is what most people have on hand.
@@TimoNoko
I've got about 100 18650 cells sitting here. For around $1 each, this is the cheapest option.
Normal 18650 Lithium with ESP8266 is HELL. First you have to lower the voltage and then you have to prevent it getting TOO LOW. It is known fact that if you start ESP below 2.5V something breaks inside and it is permanently bricked. Andreas Spiess explained the exact reason quite recently. LiFePo4 with protection circuit solves all these problems.
@@tin2001 I've got a collection of about 500 now, schools just toss them out like butter every time they get new laptops. 6 cells per laptop and they usually end up tossing 20-30 computers at a time. Honestly really wasteful and they're not the crappy $1 chinese sand batteries either, some are used pretty heavily but most actually keep pretty close to factory specs...
I actually have one of these ICs in an upcoming design and I've never worked with them before, so this video was useful in better understanding the practical operating behaviour. I was considering adding a fairly expensive PMIC to the BoM because I couldn't quite figure out the shutoff behaviour (the perils of auto-translated Chinese datasheets) so this has saved me a bit of cash!
I would really like to see you use the reclaimed cells and make a 12v battery pack with charge and discharge protection as well as balance charging.
Fantastic vid as always
Yes, this will be super. Also, different ways of doing it. I'm currently using a boost converter for low current scenarios. But apparently one could daisy-chain these modules.
I'm currently using what your describing in my cordless drill. I used a 3S 40A BMS board. The board is around $1 in aliexpress
I have used these modules while drawing current without fail for some time now. It is like a mini uninterruptible power supply for small devices that need to work 24/7. Thanks for thanks for reinforcing my cognitive bias with the truth!
Except 4.2v is not the most healthy state for lithium battery. If you want a mini-ups that lasts long time without issues it would be much better for the battery to keep it at 3.9-4v, not 4.2. I'm at the same boat, I'd love to find a piece of hardware like this one but that would stop charging at 4v not 4.2v
@@pawelszpyt1640 What if you connected the battery to the board with two reverse parallel diodes (on one of its leads) in order to introduce a bi-directional forward voltage drop? 🤔
@@pawelszpyt1640 I have been doing this for years with success and never encountered any problems, To many people over engineer things and never get anything done.
@@ozonesama I guess the official response is:
1) Nothing, it just behaves as it always did.
2) It would detect something was wrong and would never actually charge the cell.
3) It would work as intended.
My guess would be both 1 & 3: it would cut off at a diode drop below 4.2V, but then when the current had cut off, it would probably then detect that the cell was not charged (no, or very little, current through a diode can produce very low forward voltage drops), and it would then kick back in and try to recharge, repeating the cycle ad infinitum.
@@ozonesama it will work but you loose the power in the diodes. Actually no ability to set the end of charge voltage and fact that the chip is linear (no pwm) puts it into crap bin, that is why it is so cheap.
Another problem of those chips is they need nice and stable 5V - while in most cases one has variable input voltage, like coming from solar cell. this adds another regulator to the circuit..
Actually it is far better to just use attiny instead - more options, more control over simultaneous charge and discharge (two pwm channels) more easy to set all desired parameters and spare inputs for f.e. temperature control.
So glad you made a video explaining what exactly this does and how it does it.
I'm using it in a project and I was unsure if the step-up converter was actually needed or not. Now I know.
Thanks a lot!
You should generally pick up the version of this board which not only has the TP4056 but also has the DW01 under-voltage battery discharge protector built onto the same board. The boards are close enough in price that any cost savings are not worth the risk.
Undervoltage is such a risk factor with lithium cells, and aside from uninformed users even the best of us make the mistake of zeroing a cell once in a while, so it's a very important feature indeed. I took a look in my stash and was very happy to see my 4056 boards have the DW01 on them :)
Not worth buying the non Dw01 boards
Cents of difference in price.
In fact I’m finding it nearly cheaper to buy them in lots of 20.
Those non-DW01 boards are used for cells that already have protection circuits on them, like cell-phone batteries. i keep both version in my drawer to match what i need at the time.
Worth noting that DW01 is not great for long life because its cutoff voltage is too low. You can replace it with FS312F-G which is pin compatible chip but has 2.9V cutoff instead of 2.4V so it's better for your battery.
@@martincerny3294 I didn't know it was so low. I'd not feel safe with 2.4, I like to baby my cells.
Cannot wait for the channel to hit a million always providing excellent videos
All looks fine to me no problems with video. Closer zooming in on the schematic is good for those viewing on mbl phone screens. Keep up the good work!
But you could just remove the screen from the phone and not worry about it being too small.
This is a year old but I just instantly subscribed having scoured the internet for hours trying to figure out this exact thing! THANK YOU
It was nice getting these in USB-C. I still have a bunch in USB mini and micro, but I don't much care for either of those connectors. And I do agree on the charge limit resistor. Mine came with the ultra small 0402 resistors(I think, it's the uber teeny one without numbers on it.). I ended up wiping it off with my iron, then soldering one end of resistor to the outer pad and the other directly to the controller pin. Fortunately, I do have soldering tips that come to a needle-like point, otherwise that would have been a real pain. But these little modules are a huge boon to anyone who tinkers with lithium cells. The fact you can get an absolute boatload of them for just a few bucks is quite nice as well.
Oh, wait till you see 0201, 01005, and 008004 size components. (Mercifully, they’re rarely used outside of high tech mobile devices, since they require extra-precise machines to assemble the PCBs, increasing assembly cost.)
0402 meaning 4mm x 2mm? And that was 2mm x 1mm, 1mm x 0.5mm and 0.8mm x 0.4mm?
@@K-o-R Fucking UA-cam. Something about explaining SMD sizes triggers the comment auto-delete. I’ve tried three times.
@@K-o-R So let’s try it little by little.
@@K-o-R No. Standard (“imperial”) SMD sizes are based on inches.
I love those little boards, so handy to have in the drawer of stuff when you're mucking around with lipo cells.
That is why I use an USB Voltage & Current tester (Keweisi) when charging, especially when use an unknown condition cell. With this you can better understand what is the state of charge. When it is at the end of the charge cycle, finally the current will drop to zero (= battery is full). It is a good way to monitoring the charge process and progress. Using such tester in between gives you more information to understand what is going on.
Fantastic video and explanation Clive! I bought 15 of the chips with the overload protection for $10.00 and that included shipping. I have a motion sensor nightlight in my bathroom that took 4 "AA" batteries so I replaced it with a cell from one of those "devices" that get tossed out and now I can just charge the battery back up when it runs low, I use a solar charger as well. The nightlight had six LED's which was too bright so I removed a resister and that lights four LED's which gives plenty of light. Winter has arrived in the Upper Peninsula of Michigan and we are getting slammed with a snow storm today, maybe 10 inches, so finding those "devices" will become harder, I will stock up come spring when the snow melts. Thank's Clive!
I use-while-charging all the time...kind of like having built in battery backup...though I make sure the charger/battery combo is able to handle the loads properly without damage.
Yep! Some people don't know how chargers work, they also often get confused in thinking if they plug their 5v 1A phone into a 5v 80A power supply 80A is going to flow into their phone. They need to learn about Ohm's Law. The phone limits the current all by itself.
Side note, the modules with the DW01 are nice however you didn't mention that most Li-po packs that you might salvage out of something will have that same or better protection already attached. So no point in giving it "double protection".
Amazon sells these in packs for less than one dollar apiece with free shipping. Very useful to have in the kit.
my favourite module is the HW 357 which i found on aliexpress, I bought like 20+ and I put them in anything I can, the fact it has built in boost and supports 2A makes it an amazing little power supply module
I just got one of those modules.
Hi Clive, thank you so much for doing this video, it's very reassuring as I've wondered this about the TP4056 for a long time! I built an Arduino project a while ago with a PIR sensor that checked to see if post came through my porch letterbox and started flashing an LED through the window, however was always wary of it as it had a TP4056 hooked up to a mini-solar panel to charge an 18650 cell. Always worried about whether it might over/under-charge! Glad to know it's not a problem!
Can't believe I hadn't noticed the absence of charring on your work surface until you said you were on location ;-)
Nice!
For clarification it may be helpful to briefly explicitly explain (or link if you already have) the charging phases/characteristics of a Li-ion: trickle charge, CC, CV, termination.
It is just using a “trickle charge”, then constant current, then constant voltage that cuts off at a certain current, by the looks of things, that is a pretty common and recommended way to charge lithium batteries.
Used these on my uni project a few years ago. Was essentially a solar powered weather station using an Arduino so the overcharge protection these give (while still allowing the circuitry to work) was core to the build as you cant just turn off the sun when it's charged your batteries :)
I'm making a small wind turbine battery charger for a flashlight battery with a dump load and am hoping it diverts excess current to the load and doesn't drain the battery when the wind stops! Am I on the right track here?
Thank you for setting my mind at ease. There is so much conflicting information out there about the TP4056.
Great info and clearly presented facts.
I did actually buy bunch of these at some point, but I've been avoiding using them since I felt they're so cheap they're just going to end up igniting my battery or destroying whatever circuit I attach to them. Good to know these are actually usable, thanks!
you could always wire a fuse inbetween the cell an the charge board, just in case the charge chip decides to short internall one way or another, rating to suit charge and discharge currents, whichever is the largest
Your voice is a dulcet tone through my Sennheiser headphones.
My Sony's make him sound very baritone with all the low end
I can't use my bose Bluetooth speaker with Clive 🤣 Too much Bass
Try it through a dbx processor and a CKKIII headphone amp :)
I use IEMs tho
@@SvrM_ I only got a dbx compressor with srs
Fantastic video, dude! Thanks! 😃
I played a LOT with those boards! My only complaint about them is the 1a limit of discharge the battery on the protected board. If you put a bigger load you can even burn it. 😬
Anyway, I like to have opinions!
Anyway, stay safe there with your family! 🖖😊
Updated an ESP8266 based outdoors sensor project with one of these (with the DW01 chip), a small 5V solar panel and a 1500mAh lithium cell I found at the side of the road. The only change I made was to cut off the charge-in-progress LED so that it doesn't waste the tiny changing current available on winter days. Seems to be working well. The previous power for this was four AA alkaline cells which would usually last about 4 months.
I've been using these chips and modules for years now, and I've never had one overcharge a Lipo while drawing power from the battery while charging.
Like you stated in the video, the charge in current just gets shunted to the load if the cell fully tops up while charging and drawing power at the same time. I'm not sure why people think it's harmful, and don't understand that it works this way. 🤷🏻
It is harmful to hold an 18650 cells at 4.2 volts for a long time. You clearly do not understand that BOTH the cell and the load will draw milliamps at 4.2 volts.
A while back, Great Scott had a look around Ali and found an IC called the ip2312. It doesn't appear to dissipate nearly as much heat as the 4056 and is about as cheap.
Bringing the cold hard facts to the people as usual. Btw I'm seeing this as posted 2 minutes ago, but comments are 3 weeks old. I redact my initial statement about the facts... something fishy is going on here. Possibly a big TP4056 conspiracy :p
Patreon supporters get access early.
Over >10yrs using these I've seen some variation in these "TP4056" modules. Some charge to 4.1, other batches as high as 4.22. Some seem to lock-out(or lock-up?) in certain states or charge times , requiring re-power to function again. Some will charge from zero others will blink the red LED and not charge. Some have various LED blinking schemes like solid blue with fast blinking red, not sure wuts that about.. I've also had a lot of the protected ones with the green (not blue) circuit board version, connected properly , let the magic smoke out for some reason. I don't use the protected version at all now, as even the blue ones seem to have some odd behaviors that I don't understand and don't experience by using standalone charger + protection modules.
One thing they all share is instant smoke when the battery polarity is reversed, making them poor choice for use with 18650 cell holders or alligator battery clips!
I'm thinking of moving over to the buck converter versions for UPS applications, hoping for better efficiency and an adjustable termination voltage. .
They should market the lack of polarity protection as Reverse Polarity Pyrotechnic Indicator.
Interesting look at a handy module. Great use of CliveCAD there: No curve tracer required!
Might get a couple and run some tests, some modules I use in my projects get multiple extreme tests of the control IC's inputs/outputs to look for potential problems down the road.
EG: Very high/low resistance at various points to make sure a bad connection or failed component won't under/overcharge a cell etc.
Or throw many amps at a cell without warning, had a component failure that caused that once. Due to a current sense resistor (shunt) going wildly out of spec and a lack of inline resistance to the pack. I temporarily replaced it with a calibrated length of wire in the interim. No harm done however, it was a large rugged lead acid fortunately.
I admit it wasn't a charge module but I was in a hell of a pinch and it was monitored.
I've been collecting discarded cells and packs for years.. must make use of them before they are completely knackered like some of my stash of emergency lighting lead acids and SLI lead acid batteries. I don't mind them being weak though; the price of free was enticement enough and the transportation provided entertainment.
I interpret the data sheet as it being: after the bulk CC
, Go to CV@4.2V until the draw is 10% of CC.
That would mean that a load of 10% to 99% of the set current would mean the battery will be indefinitely floated at 4.2 volts, and that is what would hurt the battery, as I understand it
Agreed. I wrote a blog post about addressing this, with a dead-simple circuit - using a p-channel mosfet. Just search for "A circuit to charge a LiPo and drive a load at the same time".
These modules are super, super handy. I keep a bunch on hand and have used them in tons of things which I've converted to rechargable lithium cells. They're about .79 cents each if you buy 10 or more from most Amazon sellers.
Wow, thank you so much! I'm learning electronics as a hobbyist since May and I was overwhelmingly confused with all the warnings about it over the Internet. It always looked like I was missing some fundamentals on how lithium batteries work and it scared me everytime I saw those board LEDs alternating when a full charge was expected.
It's basic physics applied to electronic properties of materials, chuck in the physical chemistry and your sorted. ;-)
"I'm in a different location at the moment"
I figured based on the lack of workbench skidmarks :)
Thanks for this Clive! It explains something I've always wondered about my mobile phone as well: It comes to 100%, then takes absolutely a.g.e.s. before saying 'charge complete' - it must be reducing the charge current, as you explained, until the battery is completely topped off.
I'll get myself a couple of modules as I've had some cells just lying around waiting for re-use, never actually knew what to do with them!
Correct. It's also why public EV fast chargers generally are configured to only charge to 80% - beyond 80% the charge rate will slow right down, effectively wasting a charging bay that someone else might want to use.
@@tin2001 Not charging to 100% has a few other benefits as well: For example regenerative braking will work right away. When the battery is at 100% there's nowhere to put that energy and you'll have to use the normal brakes losing some of that energy.
@@M4RC90 interesting. I thought most EVs reserved a fair whack of battery capacity to extend the amount of recharge cycles.
@@M4RC90 - Just shows how far behind the automotive world is. That regen braking issue with a full battery was solved in material handling industry years ago.
@@leepower2717 How was it solved?
I have some Bluetooth earbuds that have amazing sound quality but less than awesome battery life... One of these modules with a reclaimed cell and a 3d printed enclosure: now I have days of battery life, I can charge while using them, and I added a flashlight to the opposite side of the battery to balance it out. I love them.
It really helped me to understand how chargers work when I found out they limit current by providing the voltage necessary to draw the set current. The required voltage keeps increasing until it reaches the max voltage and it maintains that voltage with less and less current being drawn until the battery is fully charged. That is why fast charge phones brag about how fast they can charge from 0% to 70-80%, not 100% because that's where they reach max voltage and max current becomes irrelevant.
Thank you so very much for these videos!! you are an excellent source and always explain everything in the simplest terms without omitting anything important.
Did you actually see the charging current go to *ZERO* with a load connected? No, you did not ...
I use these for a lot of projects.
It would be good if the board had another setting (Resistor perhaps) to allow a resting charge level to say 60% where it's being used as a UPS.
Holding a Li-Ion cell at its max charge for long periods is not good for it. Since in UPS usage, the battery is effectively being stored, it would be good to allow the voltage to come back down to 3.7, it's nominal storage charge level, after a full charge.
Sure that means less usable capacity when needed, but you can always oversize the cell for the requirements, but know that the cell is being looked after for the longer period, sometimes years.
- Great video dispelling the myths. Generally the charge profile is Bulk (Max current, rising voltage), Absorb (Max voltage, reducing current) then float (maintenance current).
Having a version that terminates charge at something like 3.9V is definitely desirable, but currently requires extra circuitry.
You can use the modules intended for LiFePO4 to accomplish this. I use them to charge NMC cells to approximately storage level voltages.
It would be nice to be able to purchase 3.8 , 3.9 , 4.0v and 4.1v modules ..
Question I would ask is "are Li-Ion cells used in UPSs?" All the ones I've serviced/repaired have lead acid cells... But I haven't seen any really new/recent models, so might be talking out of my whatsit...
@@stepheneyles2198 All the bigger units seem to use SLA batteries. But there must be versions with lithium cells. I suppose the Tesla power wall is a UPS in a sense.
@@stepheneyles2198 You are correct re: traditional computer UPS.'s. The term is just being used generically here. For example I use a single cell bank which is boosted to 5v thru a dc dc converter to power a raspberry pi that's functioning as a server. The mains adapter runs the unit but if power is lost the Lion cell takes over. It is maintained at fully charged state all the time the mains is active.
A lot of equipment runs on 5 to 12v DC and home spun UPS like this work far better and longer than traditional mains inverter based tech..
There are actually 2 issues I can think of are as follows. If the voltage of battery is low 2.9 voltage, the charger will only supply the trickle charge rate. If the device draws more than this current, the battery will continue to discharge until the low voltage cut-out occurs. On the other end ~ 4.2v... you are never sure when the charger has completed because the charge complete light never comes on. A power pass circuit would solve these problems.
The problem is that Lithium batteries don't like to be "float charged". ie: Maintaining a voltage of 4.2V across it's terminals for longer periods of time will damage the cell even though the battery has stopped accepting current.
So leaving the LEDs ON drawing 170mA while charging will degrade the battery over time because the TP4056 will hold the voltage at 4.2V indefinitely. The solution is to have extra circuitry that runs the LEDs directly from the 5V supply during charging rather than getting their current via the TP4056. Or simply disable the LEDs during charging.
I was wondering about that. Continuously cycling between 95% and 100% doesn't seem like a good way to treat a battery.
@@strayling1 Well, lead batteries absolutely *love* if. But not Nickel or Lithium cells.
I do wish there was a standard chip that only charged to around 3.9V instead of the full 4.2V.
@@bigclivedotcom A crude hack could be to put a germanium or schottky diode in series with the TP4056 output terminal... Plus a ~100 ohm resistor in parallel with the diode so the TP4056 can still sense the battery voltage and doesn't "think" it's open circuit. I haven't tried this myself since I haven't had a need to float a Lithium batt yet.
Just ordered a few of these modules to play with. I am looking forward to getting them.
I prefer those higher current switching chargers though. They are also dirt cheap and much more efficient, ideal for multiple cells in parallel. But you have to check the C-Rating for charging of your cell array and calculate it to check if 2 or 3A are okay for the number of cells you use.
Been using the standard 1 amp on 4x650 ma packs and charging from when tp5046 cuts output and I have not noticed any heating.
As big Clive said in the video if the chip starts to heat it cuts back the milliamperes it supplies.
On 4x650 ma packs using the DW01 version of the tp5046. From where it cut supply to load to fully charged takes I think a little over 2 hours.
@@patomahony9747 1 amp is too slow in my personal applications. Nice that the TP chips work for you. I use them as well for low power stuff.
Thanks Clive for very helpful and useful tips.
Hi Clive, I think that it won't work well if your load is higher, like an amp. Consider a scenario of a UPS. Most of the time you just want to power the load form the supply and not from the output of the charging IC. The output of the charging is assumed to go entirely to the battery. There are more complex boards that have "load-sharing" where they can power the load directly from the supply if there is a supply connected. See the "PowerBoost 1000 Charger" and the MCP73871 IC.
If you wanted to power the load from the 5 volt USB supply, you wouldn't need one of these modules at all. But if you look at the zoomed in picture of the circuit board in this video, you should see that the two outputs from the module (marked B+ and OUT+) are common, so both are coming from the output of the TP4056 charge controller IC. Only the ground outputs (marked B- and OUT-) are separate, with the B- going through the twin-MOSFET chip to provide protection for the lithium-ion cell.
In any case, if whatever you want to power can be run directly from a single lithium-ion cell (nominal 3.7 volts, fully charged 4.2 volts), then you can use one of these charge control boards plus a lithium-ion cell as a mini-UPS: Connect the "OUT+/OUT-" pads to the device you want to supply power to, and the "B+/B-" pads to a lithum-ion cell - either a single cell or several wired in parallel. To start with, let it fully charge the cell(s) before it does anything else. That way you guarantee that none of the output current from the charging IC is going into the cell(s), so all of it is available to power your device. As long as your device doesn't draw more than 1 amp at 4.2 volts, it will not use any power from the cell(s) until the 5 volt USB input to the module is cut off. Which is precisely the behaviour you want from a mini-UPS like this.
Thanks Clive! I'd wondered why my 4056s weren't ending charge, but the cells never went over 4.2.
Great video Clive! There is a version of the TP4056 that has power output pads aswell as the battery charging pads, the output maxes out at around 400ma, useful for low power projects. I have used many of these in projects and I can verify that you can charge and draw power at the same time. the higher the power consumption the longer it will take to charge but it is possible..
Add big mosfets or igbjts, with a little tweak, and the tp4056 can control really big currents.
They are as ubiquitus/useful as the 555 timer.
Anyone know if there is a similar board I can use to charge two cells that are in series? Battery pack is rated at 7.4. Some type of a decent BMS with USB.
I've always wanted these to have an adjustable top-off voltage as well. In many applications, the cells would last a lot longer if charging could be cut off at 4.15V or less, but some of these chips will charge up to 4.25V :|.
I built a charger around the TI BQ25895 charger IC, which lets you configure all the charging parameters using I2C. If your application has a microcontroller, that chip (or one of its brethren, or one of the very similar devices from Monolothic Power, whose dev boards and tools are way, way cheaper than TI’s) might be a decent option.
Totally agree. My last batch of ten had an average 4.25V topp so cells all got toasty (not good)
I’ve just bought a pack of 5v solar cells to add to my reclaimed eZig powerbank..
This info came just in time..👍
They come even in USB mini and Type C without extra cost. You can even cut the PCB and use it as a cheap USB C breakout board.
I also use the big brother, the TP4333. It's in principle a power bank chip (with flashlight) and it does not go to sleep with low powered devices (like most powerbanks do).
I've built two projects using the version with the DW01 chip and they work very well. One of the projects is a rechargeable emergency light and it has worked perfectly for about eight months.
Great video as usual! But have you experienced the situation where TP4056 starts to charge the battery after LEDs drain it a bit? I've just discovered that if you first apply USB power, and then connect battery, the module will go crazy (both LEDs dimly lit), while if the same battery is connected before USB, it starts to charge properly. I think the first situation looks the same as with LEDs discharging a battery for some time, and the IC just doesn't kick in properly.
Seen one of those chips fail so bad the chip inside was glowing hot in a black Friday special tablet. 😂
Would be nice if people would actually name their projects where they use these modules, instead of just mentioning "in lot of my projects".
I really wish there was a version that showed fully charged over 4v or 90% of charge.
Oh no, it was a "OwO" chip
Clive, are you sure all those TP4056 modules with the W01 chip will toggle on/off charging a cell that's on load?
I have found some TP4056 modules with the W01 chip will not go back to the charge state until the power to the module is cycled.
Thanks for another informative video!
K Watt.
The ones I tested did toggle. I wonder if there are brands that don't.
ive found significant differences in 4056 versions -if your cell is under voltage some tc4056a trickle charge at 60ma some at 200ma (with a 2.2k mode resistor) but these chips really need reverse battery connection protection if using loose 18650,s and cheap battery boxes that have the same terminals both sides of the battery its so easy to blow the chip
The slope at almost fully charged battery is not feature of chip
It is actually physics of battery itself
You are not likely to keep throwing same size boxes into almost full storage at exactly same rate as when it was empty
Same goes to electrons, they actually have to find space to rest
Probably worth noting that in their standard set up the TP4056 gets hot at first, especially with 2 cells in parallel, so a good heatsink candidate. Also note the overcurrent circuit shuts the MosFets down if the current draw is too high... LEDs = good... Motors = bad
A risk with motors is voltage spikes damaging the MOSFETs. I'd recommend a flyback diode or small capacitor across inductive loads.
useful information that i didn't know i needed to know, thanks Clive
If you look at the lithium batteries that have small circuit boards in them, you'll see that they already have the DW01 and the mosfet.
So cool. I'm literally holding one of these circuts.! Gonna have to search other video. Thank you for helping people to learn to "recycle" and "reuse" these batteries. / circuts. The past 3 or 4 years I've been picking them up off the ground and when the box got full I'd take them down to Erecycle . Wasnt untill last year when the batt died on my bone conduction headphones I started on this journey of can I reuse the pouch cell from a vape?. It wasnt untill this year the quality and size of the cells are quite excellent and i think i can !so lets keep recycling repairing and reusing stuff!
Merry Christmas and Happy New Year Clive! 🎄🎉🎊🎆
(I apologize if I am out of line.) 😁😍🥰❤🌈‼
Brought to you, Straight from the United States! 🙂🎄
Clive - As YT blocks my comments - See your - Odysee Community Page - for my Aka @Super-Grand-Ad Comment Pictures and Links on this video.
Very clear presentation. Thank you.
Ah, I liked the Monolithic LTH7 a bit more than the TP4056. Although the Top Power is fine depending on who the manufacturer is some of them weren't correctly reading the program resistor (for low amperage).
I still have a few rolls of different Top Power charging ICs from LCSC that I've been meaning to try out. There's some neat integrated features in some of the variations.
Very nice and detailed explanation. Somewhere I read that you cannot be charging with these while running a load at the same time. I have some applications that need these. I am experimenting with reducing the peak charge voltage for longer battery life. 4.2 volts is not very gentle on a battery life cycle. I am also working on a board with various resistors to change the current for various sized Lithium batteries.
Totally. And what is the open terminal voltage of a full cell?
It does take some voltage to drive the reaction so having it extended time on charge instead of idle voltage likely does nothing good to the cell.
@@erlendse I am aiming for more like 3.9 volts per cell myself for off grid solar power use. I am also aiming for the highest cycle life of the cells. They do settle down to a lower voltage after you remove them from charge. I have thousands of batteries from Battery Hookup and am seeing a wide variation in voltages between different types of cells.
I feel like I am one of the ones that has inadvertently pushed the "myth". I put a load sharing circuit on a TP4056 because I wasn't sure the charge would cut off properly when using external power and having the device switched on, as the circuit is enclosed in a nice die-cast aluminium enclosure I want to make sure I am not overcharging that 18650 in there. I have heard of no instances where a fire or anything bad happening due to a battery being on charge whilst under load using a TP4056, but its seen not as a good idea to keep holding an 18650 at 4.2v forever under load also its good practise to switch charging off on a battery when its reached it peak.
So I do use these modules to charge up 18650s (all of my galliums and A4 super computers are powered this way) without a load share and I sleep soundly at night, however these are one offs and for my use only.
I know that a lithium cell degrades faster when stored at full or low charge due to lithium ion density bias on the electrodes, but I wonder if that really impacts modern cells as much as earlier electrode chemistry.
Should have mentioned the chips does support adding a ntc thermistor for the battery with a minor mod to the circuit you can solder thermistor to pin 1 after isolating pin 1 from ground.
To elaborate; melt solder on pin 1 with iron and lift the pin with a small screwdriver about 1mm. Wick away the solder and slide some kapton tape between the pin and the pad. Connect NTC to Battery- go through a voltage divider and back to Temp/pin 1. I used 10k NTC, 4k7 and 68k to roughly stay within +5C to +45C.
Those CC/CV Modules based on the XL4015, LM358, 78L05. Operate in much the same way. But I prefer the modules Clive has demonstrated here.
Thanks Clive.
;)
I love those updated boards with DW01 and that 8205A, I use them in my DIY solar chargers for light/phones and just lights, especially chrismas ones that normally use AAA batteries that die after few days, and when you have bunch of old 18650 cells from old laptop you slap some resistors for those lights some box and you have lights that last weeks without charge.
Thank you, Clive.
Have been using tp4056 from aliexpress since 2016, had 0 problems. Even tried using two in parallel with different resistors to get 3 A , which still works fine.
@4:37 Isn't there a chance that a cell can fail catastrophically, once it was discharged below a certain voltage? Reading up on the topic, in the best case, you lose some capacity. In the worst case, you get "dissolution of the copper anodic current collector, which can deposit on internal cell components and induce an internal short."
Going down to 0 V sounds dangerous...
There is research on the 'net that showed that reverse charging cells posed a higher risk than discharging to zero. I think the main issue is the density of lithium ions in the electrode material.
A real hazard is charging cells near zero degrees C as it can cause lithium plating on the electrode surfaces.
I wish the pads were on a 0.1 inch pitch. Difficult to solder to strip board or use on breadboard.
when you're high and your hands arn't big but the pcb is
A simple comparable circuit... they make 18650 chargers for vape pens. Super cheap. To give it an output, maybe just wire up the battery pins? Would be interesting to see what's in an e-cig which will charge by USB charger, too.
For an Alec chicken, you sure know your electronicals!
The DW01 also does overcurrent and short circuit protection, using the MOSFET Rds(on) as a shunt resistor to roughly measure the current.
I've used to use a lot of those TP4056 but as you mentioned it dissipates quite a bit of heat regulating the charge to the litium cell therefore I've switched to the IP2312.
I've had to research older Li-Ion cells on viability for re-use as part of a college internship. Testing hundreds of them and it became apparant that old/degraded cells CAN actually start to keep consuming current even if at 4.2V in varying currents from 10mA to 200mA. These cells would keep charging during CV-mode and slowly overheat. It led to me terminating the project out of safety concerns.
In practice most cells should never exhibit this behavior, especially those with protection circuitry, but a bad egg can do that. Which i believe is why nearly all chargers all got a current termination threshold at about 100mA (notably when a ntc function is missing). So while normally putting a load is fine. You are diminishing/losing a safety functionality in the process.
A full-fledged ORing/Switching circuit is still the 'proper' way to go. Can be as simple as a P-Mosfet + Diode that disconnects the cell the moment a different supply is hooked up.
Cells that have developed an internal shunt resistance should be discharged and recycled. It sometimes happens in old laptop packs when one cell has been reverse charged by the others.
@@bigclivedotcom Correct, but the shunts i've seen are not large and able to trick hardware and user into believing they are fine and functionally act fine so long as the cell doesn't degrade further. And they can also develop when a li-ion reaches critically low charge as sometimes seen with cheap/crummy stuff turning volatile after extensive storage.
Without proper termination of the charging circuit once full-load is achieved. You are risking increased likelihood of a user not noticing when the device suddenly stops terminating charge properly and starts to swell up like a balloon.
The charging into a load panic is often overblown and of no consequence to like a hobby project, but not to be outright dismissed and taken in account if going more serious.
Clive I've tried reading the documentation for those TP4056 and I seem to recall that Pin 1 was listed as taking (some sort of) NTC thermistor for battery thermal protection but I couldn't make heads or tails of the circuit diagram and quite frankly the the font the documentation was in made my eyes water.
The thermistor input can be used with the thermistor as part of a voltage divider to set the desired point that it will stop charging a hot cell. But at the typical 1A charge current a 1Ah or higher cell shouldn't get hot.
These things are way too easy to kill, I just keep poping them without any overload.
Back when dinosaurs roamed the earth, the Li-ion charger that I used was the LTC4055 sourced by Linear Technology, now a part of Analog Devices. It was quite sophisticated, protecting the battery at all costs, important if you are concerned about selling into consumer markets and want to avoid lawsuits over fires created by your product! It had one feature that limited the scenario you point out here: a programmable timer. We set our timer to be 3 hours. If the battery was not charged within 3 hours, the charger cut off, letting the load drain the battery. If the battery then dropped 100mV below 4.200V, another charge cycle was initiated.
It was a pretty pricey charger chip at the time, but we were willing to pay the price for it to assure the utmost in safety. We were not your average eBay seller selling junk! When you use a much cheaper chip like the TP4056, you sacrifice some of the safety features that you get when you buy a better charger chip. Indeed, the added safety of the DW01 safety chip is well worth the extra pennies of expense. In our case, our batteries had a small PCB in it that held a similar chip + dual NFETs and a thermistor. (You don't want to charge these batteries above or below a temperature range.) The TP4056 has the capability of supporting a thermistor, but the pin (pin 1) is disabled by grounding it.
So am i right in thinking it is like your car . You are driving along and the alternator is charging the battery but being regulated and then it gets dark and you turn on the lights . The light work but the battery is still being trickle charged
I burnt out some modules like that by applying power (to measure the charging voltage) before connecting the battery. I bought cheap ones though, hopefully more expensive ones don't have that problem.
I fixed charlie the chameleon with this, a reclaimed 18650, a TPL5110 timer module and a case I 3d printed. My Nephew was happy. I did use the updated design with under voltage protection because kids
I know a little about electronics but you make it very interesting and learnable for us idiots. Thank you so much for the education and confidence to try some of these little projects.