 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Microcontroller and PC projects : Capacitor smoothing question
Page 1 of 2   |
|||||
| Author | Message | ||||
| lizby Guru  Joined: 17/05/2016 Location: United StatesPosts: 3252 |
I'm trying to monitor the capacitor of various 18560 batteries--nominal capacities of 3400, 5000, and 9900!! mWh. The batteries are in an 18650 holder/charger with 5V USB output. My display shows voltage, current, and total mWh. My load is a little Harborfreight 3-AAA LED with 3 on positions--red LEDs, white LED, and red LEDs blinking every second with about 50% duty cycle. I removed the AAAs and directly wired 5V and 0V from the USB output. The white LEDs draw too much and the battery shuts off (even though it nominally has a 5V 3A rating). The red LEDs on solid draw between 360 and 390mW. This is nominally close to C/10 for the "3400mAh" battery. The battery shuts down at about 1080mAh. I'd like to be able to get closer to C/20, which I think is a typical test current draw for the ratings for these batteries. WIth the blinking red LED, the reading fluctuates every second between about 120mA and 300mA. I tried putting a 220uF capacitor across 5V and 0V, but that made no significant difference. My question is, what size capacitor am I likely to need to get the display to approximately half of the 360mW reading--to around 180mW? PicoMite, Armmite F4, SensorKits, MMBasic Hardware, Games, etc. on fruitoftheshed |
||||
| twofingers Guru Joined: 02/06/2014 Location: GermanyPosts: 1456 |
I'm confused. Just by the way: I think 5000 and 9900 are fakes. I think 3500mAh is the current maximum. Regards Michael The required capacitor can be calculated if you know the internal resistance of the battery and the PWM parameters (duty cycle, frequency). IMHO. Edited 2025-03-10 07:31 by twofingers causality ≠ correlation ≠ coincidence |
||||
| DaveJacko Regular Member  Joined: 25/07/2019 Location: United KingdomPosts: 80 |
not fakes.. just that Chinese Amps are smaller than the western ones. I've just completed a similar project, I bought a bag of 1 Ohm 1 watt resistors with which one can easily make any load. I used 5 in series, but be aware they get warm and the resistance will increase. so I used the INA 219 current sensor module to measure the current in real time. I did find a new 4500 mAh cell that was more like 450 (Chinese people are very nice, and I buy lots of stuff from them, including food) Try swapping 2 and 3 over |
||||
| lizby Guru Joined: 17/05/2016 Location: United StatesPosts: 3252 |
I think they're all fakes from Amazon. 3400mAh is a reasonable number to hope for, but when discharging at around C/10, I'm seeing about 1080mAh max. The number would be expected to be less than at C/20--but how much less? I would assume that the red LEDs when blinking are ON/OFF with a duty cycle of 50% and a frequency of 1Hz. Current is about 360mA. PicoMite, Armmite F4, SensorKits, MMBasic Hardware, Games, etc. on fruitoftheshed |
||||
| lizby Guru Joined: 17/05/2016 Location: United StatesPosts: 3252 |
I have a USB resistor module which has a switch for 1A or 2A, but the 18650 batteries I tried immediately shut down with 1A at 5V (with the modules I was using). I've ordered a USB current tester with fan and adjustable current--I'll see how that does when I get it this coming week. I was just wondering if I could cobble something together--but I don't have any capacitors bigger than 440uF. PicoMite, Armmite F4, SensorKits, MMBasic Hardware, Games, etc. on fruitoftheshed |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 9800 |
The simple way to roughly know a battery capacity is to weigh it. Take a known good battery of a known accurate capacity and get the weight. Scale this by the weight of a unknown battery allowing say a fixed 50% (could be more) of the good one for packaging and that will give you a good approximation of the unknown. e.g. known good 3000mAH @ 3500 weight unknown 2750 weight likely capacity (2750-1750)/(3500/2)*3000maH = 1700mah You will find the 9900mah battery is surprising light  |
||||
| twofingers Guru Joined: 02/06/2014 Location: GermanyPosts: 1456 |
Actually, you only need a Picomite, a logic-level N-MOSFET (like IRL540), a load resistor approx. 18 ohm/1W, 2 * 470k as a voltage divider, 1x 100k and 1x 100nF. causality ≠ correlation ≠ coincidence |
||||
Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 7339 |
To measure the capacity of a cell you really need a constant current load. You can cobble something together. This might be worth a try. +---------- 5V ! +--------+ --- ! ! 330R ----- \ --- === \! ! ! NPN !----+ ! /! ! ! e/ ! ! ! ! ! ! ! ! --- --- \ Red ! R \ / -/ LED ! --- --- ! ! ! +--------+------+--------- GND R = 1.2/I Now the current won't depend on the cell voltage as much. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| twofingers Guru Joined: 02/06/2014 Location: GermanyPosts: 1456 |
That can work. But I think all you need is a suitable resistor and measurements every second and a shutdown at the threshold voltage. I built a device for measuring capacitance with MicroMites 9 years ago. Regards Michael causality ≠ correlation ≠ coincidence |
||||
| lizby Guru Joined: 17/05/2016 Location: United StatesPosts: 3252 |
However, people have opened 18650s which did not meet the rated capacity and found that part of it was filled with sand. PicoMite, Armmite F4, SensorKits, MMBasic Hardware, Games, etc. on fruitoftheshed |
||||
| Volhout Guru Joined: 05/03/2018 Location: NetherlandsPosts: 4721 |
A 18650 is a huge capacitor itself. There is no-way you can add a smoothing capacitor parallel to it so a pulsed current load (pulsing LED) changes into a constant current load. A big inductor in series may do it. But since the frequency of pulsing is so low, this is not practical (the inductor may weight several kilo's). But in general: capacitors smooth voltage, inductors smooth current. Simplest solution (already suggested here) is to use a resistor to discharge (or accept the pulsing LED as a reference load), and count the time to get from full charge to 80% battery voltage). Volhout @Peter, aha..that is how they do it. A small cell inside a huge shell. PicomiteVGA PETSCII ROBOTS |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 7339 |
For lithium cells you won't be far off with a resistor, Michael. You are only working over about a 1V range from fully charged to the 3V cut-off point. No good for primary cells. The above circuit is easy to control though, just short the base to GND to stop the charge. You'd need a transistor or mosfet in series with the resistor anyway so it only adds a resistor and LED. You could use a 3V3 supply for the LED but increasing the voltage and series resistor causes it to behave more like a constant current source for the LED and increases accuracy a bit. 12V would be even better but you would then need a transistor to short out the LED. Measuring the voltage needs a mosfet to isolate the potential divider if you want best accuracy otherwise the PD forms a non-linear load in parallel with the test current. Easy enough to do if you can be bothered. :) It's an interesting little circuit. The temperature coefficient of a red LED running in its "sweet spot" (usually 5-10mA but it varies) is similar to that of a silicon P-N junction but opposite so the current through R has at least a degree of temperature compensation. :) If you don't get 1V2 across R then just change the 1V2 "fudge factor" and re-calculate R. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| twofingers Guru Joined: 02/06/2014 Location: GermanyPosts: 1456 |
Hi Mick, I'm sure I misunderstood you or missed something. Firstly, a 18650 is not a primary cell (https://en.wikipedia.org/wiki/Primary_battery) and secondly, billions of devices (smartphones, etc.) work with these power sources and the 1 V difference. As I wrote, I (and Geoff too) published a circuit based on this principle almost ten years ago. They work excellently.  The small 2 EUR zb2l3 modules also work in a similar way. I have been measuring my 36V e-bike batteries according to the same principle for many years. You should check your statements. As I said, I suspect a misunderstanding. Regards Michael I wrote above: Edited 2025-03-10 21:49 by twofingers causality ≠ correlation ≠ coincidence |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 7339 |
No, you misunderstood. :) I meant that using a simple resistor as a load is inaccurate for primary cells. Rechargeable cells usually have a much shallower discharge curve, although some lead-acids can be a bit iffy as they age. With a shallow curve a resistor has much better accuracy as the start and end currents don't vary by all that much. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| twofingers Guru Joined: 02/06/2014 Location: GermanyPosts: 1456 |
Hi Mick, are we talking about "primary cells"? Is a 18650 a "primary cell"? Does the inaccuracy play a role when we make hundreds of voltage measurements across a certain resistance? I can only confirm that my measurements always agreed very well with the expected values. A constant current sink is not required. I don't know where we are talking past each other? For me the topic is closed. Kind regards Michael causality ≠ correlation ≠ coincidence |
||||
| Volhout Guru Joined: 05/03/2018 Location: NetherlandsPosts: 4721 |
Hi Mick, As long as the resistor is a constant value (we are not measuring the resistance) it is absolutely usable as an accurate sink. We know the voltage at any time (that is what we measure), so we also know the discharge current (ohms law). So with every measurement we discharge the battery with a known quantity (not the same all the time, but known value), so we can calculate accurately what the cells capacity is. It may take 7 hours, or 8, or 10... doesnt matter. We know exactly the capacity... twofingers project Volhout Edited 2025-03-10 23:41 by Volhout PicomiteVGA PETSCII ROBOTS |
||||
| lizby Guru Joined: 17/05/2016 Location: United StatesPosts: 3252 |
Ah--thank you. This is where my ignorance of fundamentals shows through. I'll await the arrival tomorrow of the adjustable load USB module. It should be far more accurate than what I was attempting with a capacitor. Since the 18650s I have are all vastly over-claimed with respect to capacity, I guess there's no way other than trial and error to find out what the C/20 discharge rate is--run the setup repeatedly with different resistance settings until it is detected that the battery module has shut off after 20 hours. I'm not looking for constant current--I'm looking for the discharge rate at which the battery/module lasts for 20 hours. Then the reading on the current monitor of mWh will give me what the battery should be rated at. PicoMite, Armmite F4, SensorKits, MMBasic Hardware, Games, etc. on fruitoftheshed |
||||
| lizby Guru Joined: 17/05/2016 Location: United StatesPosts: 3252 |
Thank you for providing that link. I think that's what I need. ~ Edited 2025-03-10 23:59 by lizby PicoMite, Armmite F4, SensorKits, MMBasic Hardware, Games, etc. on fruitoftheshed |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 7339 |
A "primary cell" in this context is where the energy is provided by a non-reversible chemical reaction - a non-rechargeable source. An 18650 is a "secondary cell", in other words the chemical reaction can be reversed. Using a fixed resistor, the current through it falls as the battery terminal voltage falls (ohms law). A constant current sink provides a fixed, known current throughout the discharge curve. Unless the battery is completely flat (and probably damaged) we can be pretty sure of it's off-load terminal voltage and we know the current so the internal resistance is the only variable. That's what ultimately decides the capacity. You can also estimate the capacity by measuring the terminal voltage while you connect and disconnect a known load. This gives a guide to the internal resistance and hence the state of the battery. This method is often used for automatic "battery health" indicators in switchgear. It isn't a good guide to the capacity, but does give a pretty good indication of whether the battery could accept a short term emergency lighting or breaker tripping load. It's all a guide anyway as the amount of charge entering the cell and the discharge capacity both vary quite a lot with temperature. If you want to measure the C/20 capacity then it's not accurate to load it with a fixed resistor. e.g. C=1500Ah It will discharge to end point voltage in 1 hour at 1.5A C/20 It will provide 75mA for 20h If start voltage is 4V3 and the load resistor is 56R then current is 76.8mA At end point (3V) the current has fallen to 53.6mA At a voltage of 3V7 the current would have have fallen to 66mA so the run time of a 1500mAh cell would be about 22.5 hrs. A constant current of 75mA would have given the correct value of 1500/.075 = 20 hours. As the voltage has fallen then so has the load so the time to end point has slowly increased. . . Edited 2025-03-11 01:30 by Mixtel90 Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| Bleep Guru Joined: 09/01/2022 Location: United KingdomPosts: 548 |
I use one of these less than £10 on Ebay it'll give you A, W, Ahr, Whr, voltage, if you are measuring a low voltage, like a single lithium, it has a 3pin connection on the side to power it, otherwise it will power itself from whatever you are measuring, you still need a load and to keep an eye on it to prevent over discharge, but otherwise it works great. I just use a relatively powerful LED torch as a load. Remember the faster you discharge the lower the capacity, the usual standard is to arrange a discharge over about 10 or 20 hrs 10c or 20c, but otherwise as long as you don't discharge too fast say a few hours, it'll give you a good idea. I primarily use it to test old laptop cells. Regards Kevin Edited 2025-03-11 01:18 by Bleep |
||||
| Page 1 of 2 |
|||||
 Print this page |
