the ultrafire flashlight and batteries I got have protection ic's (batteries) and yes the flashlight (9000lum) can get over 100c under max brightness but the cheap ass batteries have an internal resistance that limits the light from going thermonuclear and even mid quality batteries do exactly the same fire may never happen but I was so worried it would I added arctic mx-5 between the led pcb and the case edge and it smoked a bit before it dried a little and I do like it as it is a great short term spotlight/worklight and seems to pull a stable 3.8a from a pack of 4 18650's
Like 99% of those are fake with up to 9000mah ratings. That's how you can get free batteries from ebay if you tell the seller that the capacitance doesn't even close to stated one (and risk to burn your house with those).
The highest capacity 18650 cell ever made (afaik) is 3600 mah from Panasonic. But you cannot find these anymore. So currently the highest capacity is 3400 mah again from Panasonic. With the current chemistry its not possible to squeeze more energy into a 18650 cell. Anything above that is automatically a lie, unless it's from one of the top manufacturers, its genuine and a new chemistry has been used. I believe you probably knew all these and I understand this video is for educational purpose. My comment too.
Tomas Schertel You are welcome. My comment is a little bit old and now there can easily be found new 3500mah cells, available from reputable manufacturers.
+TransfixedBunnies There is literally dozen of them popping up everyday on eBay, nothing will happen if you report them, in fact I've seen some of the better sellers will say it's not genuine, but what do you expect for $1 dollar? Most people will buy the battery pop it in the flash light and it will run for an hour and they will be satisfied. Your other option is to spend $11 bucks and buy something like Keeppower with Japanese protected cell so you get what you pay for.
Dima Prok My problem is not that you get what you pay for, that part is a property of the market economy. My problem is that you do not get what you pay for, or to be more precise, what you are promised. There is no gray area or matter of opinion in scientific matters. 9000mAh is 9000mAh, not 1000mAh + imaginary saved cash flow. Sadly I have to at least partially share your pessimism about the end result of reporting fakery in ebay. That is just ebay for ya.
What do you expect of Chinese sellers? I once read a question on Quora, I don't remember exact wording but it was something about why do Chinese copy (steal) and sell cloned technology. And the answer was by Chinese guy is that for them it's a way of life and they don't consider this wrong. So I think this exact same mentality applies to grossly exaggerating specs, it's just part of marketing, how they do business. But you know this is not the worst, I was shopping for Power Supply just few days ago, and I see a big seller with 98% feedback and I am concerned about this seller's reputation, so I started reading complains and there was literally over a hundred complains in the last 6 months. #1 was seller was listing that he was located in CA while the items were shipped from China, then there were up to 10 day delays in shipping stuff out, there were people who never received stuff and didn't get refund, there were people who got wrong /broken / used items and it made me wonder how is this guy still in business? Well I guess if you sell 20 little junk items that people won't complain and then screw someone over a bigger item the feedback looks fine and ebay doesn't care because they get their piece of pie.
+Dima Prok Forgot to mention, I try to look for 99.5% or better feedback with big sellers. With small sellers I look how many negative feedback there was in 6-12 months and what kind it is. If the item arrived damaged, then most likely it's shipping company fault but most people will blame seller.
+Dima Prok you make good points. Having read entrepreneur blog and talked to people who have worked in China I gather contracts are much more abstract and flexible (as in not trustworthy in same degree) than in Western countries.
Bob Brown: "China-amp/hours"!!!! Reminds me of the Hp ratings of Chinese motors. There are those little bitty Chinese horses and then there are those big, strong American horses.............
DD_Ranch Texas most US HP claims are BS. In Japan they under-rated their engines for a while (they started focusing on handling instead of an HP battle, which gave them much faster cars in real life, so they put an upper limit rating of like 275 HP or something even if the engines were 300-400 HP). In the US, we were rating older cars with stupid high HP which is easily disproven, we claimed crap like 1980’s cars with 300 HP that took 18 seconds or more to hit 60 mph... and they were cars, not literal battleships... and the new ratings are also BS, my dads Dodge Dart claims a higher HP rating than his old 1987 Mercedes (which is not even running right), and yet the Mercedes outruns it (and that car has like 200 pound electric seats, I felt like I was breaking my back trying to remove them from another car at the junkyard...). We still claim BS on most of the engines, and of course it’s BS rated on top of not using wheel power (somehow every car has the worlds most inefficient transmission ever produced right? That must be why 240 HP dropping to 170 seems reasonable...) There are a few US cars that are actually rated correctly (typically dyno tested crate engines), and even a few that are under-rated. But then you also have BS like Dodges Hellcat engine bragging 777 HP or whatever, but then require a “race key” (in red obviously), to “unlock” that power (which is what you paid for already...), basically DLC from video games in real life... What a scam to advertise it as such and then lock it to 500 HP! So that’s beyond stupid. As an engineer, I’d call it a faulty reject if it doesn’t hit the spec, if I buy a car and it’s out of tolerance, it should go in the scrap bin. The US marketing teams are about like the Chinese (which is super frustrating, I don’t pick a parts spec out of thin air, if that electronics piece smokes itself or melts at half the power, then why put that rating sticker on it?!?). Anyways, US drag racing is also flawed, it doesn’t measure time until the front wheel has already passed the laser, which means it takes off like 1/10 of a second or more (and yet posting crap in the thousandths of a second like it’s accurate or something...), so you could by their standard hit 60 mph in 0.000 seconds! Unfortunately you can’t get a negative time to 60 mph, but obviously if another car hit it 80 mph it wouldn’t have a lower 0-60 time). I guess it makes tanks with really long tracks look super good or something, it’s pointless, they are purposely measuring it wrong, just have a second laser, but oh no that would screw up marketing!
Fantastic! This is exactly what I was looking for. I tried making a constant current load with an LM317 to test battery capacity, but the voltage was too low resulting in the whole thing falling out of regulation. I think I have most of these parts lying around. Thank you so much, I can't wait to try it out!
Why don't you use a tp4056 charging and discharging module. It will automatically cuttoff you battery connection with your lm317 tester when it discharge alot.... I know I'm little bit late here,😅
When I tested my li-on battery for power tool, I just measured voltage over the resistor load at different times. For example I measured (18v at t=0, 17v at t=30min and 16.5v at t=1h) then I just do polynomial interpolate u(t) in matlab and integrate the function (p = u(t)^2/R) over the test time. This should give you stupendously accurate results since u(t) is changing slowly monothonically over time :) Btw it is easier to measure voltage accurately than current
Good video GS..! Just a few small points to add for all...... 1. The constant load circuit is simple and neat but there are alternatives which will eliminate the ripple. This is relevant if we wish to to test capacity precisely. 2. The relationship between discharge current and capacity is non-linear. The lower the current, the greater the measured capacity. 3. I suspect you will find the Ultrafire is rated for a discharge current less than your 0.5A, which means you will generate a pessimistic result. 4. Low voltage cut-off at 2.5V will not damage the battery. All one can say is that life measured in discharge cycles will be lower compared to 3V cut-off. Keep up the good work..!
Thank You, Greatscott, for all your constant output vids!Robert Murray-Smith has a large vid library devoted to DIY paper/carbon megacapacitors for use in motive apps. The voltage drop on a megacap is like a rock slide. I dunno if homebrew (raw materials start-point) development is your idea of fun, but I propose that megacaps constructed of low-ESR kilocaps would be the solution to this competent alternative to li-Ion, if efficient solid-state power-switching can be done to sequentially boost volts at the expense of amps as discharge progresses. RMS ( a phD chemist) is trying to compete with his professor-brother for the kiddie-crowd and sticking to paper bacitors. I further propose that nickel foil and PET film starting materials would guarantee low ESR and high cycle efficiencies. If you know anybody who wants to wow 'em with high-power motive demos (electric motorbikes, etc) this approach looks ripe-- but I dunno electronics, so maybe you could comment on the utility of the above switching scheme. Yours is the most awesome DIY electronics channel I've found in many years of YT searches. Thanks again, for being so cool!
Fantastic Experiment. I also made the same circuit to test a couple of fake lithium ion batteries that I had. The actual capacity of those batteries proves out to be only 418mAh. And not 2500mAh as mentioned on them. Thank you very much for letting me know this capacity test experiment.
900mah Capacity at 2.5V discharged is even worse than the average 3.0V/2.9V most smart dischargers use. But those batteries are known for this. 900mah is actually higher than you see when other test the same batteries. I actually purchased 6 once and the capacity said 3000mah, but they actually tested 2200mah at 3.0V(I was floored). But the basic wear and tear(charge and discharging) cause it to lose Mah quickly. After 3 or 4, the MAH dropped to around 1100. I just put them in my recycled 18650 bin and called it a day! Great Video Great Scott!!
I am following your channel from Iraq and I found the videos very useful, educational and practical for hobbies and professional engineers, keep it up, many thanks.
When it comes to 18650 Li-Ion battery, the valid ranges are from 2,000 to 3,500 mAh depends on its drain characteristics. Avoid the batteries claiming anything below or above this range. Btw, higher the drain current(>30A), the smaller the capacity (close to 2,000 mAh), and of course, lower the drain current(
The capacity of a battery shall be measured at 1/10C current (hence, you must have something near10 hours of discharge); if the drained current is too high the measured capacity of the battery become lower than expected (and, on the contrary, if the drained current is too low the measured capacity become higher than expected) I also suggest the use of some smoothing & decoupling capacitors to remove the oscillations.
teslacoiler But 1/10C is far away from reality. I will do some further test with other big brand batteries to see whether they can supply their rated capacity with 500mA.
Li-ion batteries are pretty good at supplying most of their rated capacity a higher discharge rates than C/10, so the error is unlikely to be great. I have a "Multi Smart 3000" battery charger/tester from 7-Day Shop and it can measure battery capacity through a full charge-discharge-charge cycle at a choice of several set currents. I haven't seen much difference when testing the same 18650 batteries at any of the test currents, so I'm fairly confident that the capacity I've measured is close to its value for C/10. I do strongly agree with the need to properly decouple the supply to the opamp to remove those oscillations. I'd suggest a 100nF ceramic capacitor between pins 4 and 8 of the opamp, as close as possible to it, along with a 10μF - 100μF electrolytic capacitor across the incoming 5V supply.
Love your work.. I wouldn't mind seeing the same test being done on a highly regarded battery brand, such as Duracell or for those people in the RC hobbies maybe turnigy.. Thanks for your informative video keep them rolling.
Just a simple and dumb question, but only for being sure, in the constant curent load diagram, you use the logic level z44n, but i have the normal irfz44n, if i supply it with anyting between 10 to 12 v i suppose it will work as well, the only question is if the potentiometer to regulate the current pat has to change a lot or it can be the same 500k. Thanks.
jotasgasco Potentiometer value stays the same. But you may need to increase the supply voltage for the LM358 and potentiometer. From 5V to 12V should do the trick.
GreatScott! BIG Thanks! that did the trick for me, here in Taiwan i only can get the IRFZ44V and the C358C, with 5V i was getting only 200mA max load. With 12V i can increase the load to 1A or even higher, but 0.5A is enough for testing LiIo's... i finally have a constant current load i always wanted, now i just need to dig out the arduino + gnuplot thingy i made a while ago to make real accurate testings. Besten Dank und.... Ohren steif halten und weiter machen! tt
@@gamingSlasher It has 100% negative feedback already, so you can't lower the closed loop gain any further. It's a non-inverting buffer with the mosfet simply adding extra current capability (which it happens to draw from a separate supply - the battery under test).
To eliminate the oscillation, I'm pretty sure the solution is just to decouple the supply properly. Add a 100nF ceramic capacitor between pins 4 and 8 of the opamp, as close as possible to it. It wouldn't hurt to put a 10μF electrolytic capacitor across the incoming 5V supply, especially if you have long supply leads.
(I'm not a natural english speaker so if theres ugly errors about right writing , please do not account . Im still learning) So than , to the real comment. I just spent like a hole day watching your videos. Excellent channel and videos !! If you can , please make some video about using ir reciever with raspberry pi ( using php , or the languages that you use as well .. they are beeing very usefull to me , to learn some things and , get some ideas for my own project) . Thanks , and again awesome channel !
Hey Great Scott! I would like to know, why You disregard the transistor consumption power? You said that it's really getting hot... Hot means energy... I think that You don't need constant current, but constant power. I would choose buck-boost for constant voltage on the resistor. Constant voltage on resistor means constant current means constant power consuming minus buck boost efficiency.
Yes a very good video from someone who is far more technical than me. I am not anywhere near as technical as this guy but he certainly knows what he is talking about. I know a better way though. Don't buy cheap Chinese batteries. I buy only the Panasonic enelop rechargeables or the Optimum Power 18650's from Japan. These have never let me down and I have had them for years and still going strong.
2:34 With a negative feedback, you do not have a comprarator. I may be wrong, but here's what I think: Through the feedback path, the OP is trying to get a 0V difference in between its inputs. (Comparators do not have feedback, just two inputs. ) So, it is taking the voltage on the potentiometer as its reference, and regulates the mosfet in its linear region (that's why it gets hot), in order to let so much current pass, that I_R1•R1 = U_ref at potentiometer. In orher words; The voltage drop at R1 were the OP is measuring, is equal to the voltage at its non inverting input (+), due to the current which it allows to pass through the transistor and through R1. Meaning, you have an amplifier with the resistance of the Mosfet dictating the -let's say- degree of amplification. Said degree of amplification changes with the voltage of the battery, because it takes a smaller resistor to allow equally as much current to pass, as the voltage of the battery drops.
The other way of looking at it is that the circuit is an opamp with a mosfet in source-follower mode to buffer the output current to a much higher value than the opamp is capable of supplying. Then, as you say, the opamp action under negative feedback will ensure that the voltage at its inputs is the same, leading to the mosfet sourcing a constant current (Vref / R1) to the resistor R1. Since that current must come via the drain of the mosfet, it will draw a constant current from its supply, i.e. the battery under test. It will do so regardless of that battery's voltage until the battery no longer has enough voltage to keep the mosfet's source at Vref. There is no inherent oscillation or ripple in the circuit - that's just a misunderstanding of how a feedback loop works, along with poor layout.
Great video! Could you modify this circuit to use the current for charging an empty battery? Basically, run a battery discharge test while charging an empty battery. Might be a great idea for future video!
I've made a couple of discharge tests on a small 12v 1.3Ah battery. I took a rough average of the discharge current. I was annoyed that I only got 60% of the rated capacity. Does anything from China do what it supposed to do or do they just write anything they want on the label?
Great :-) . I´ve checked an 9800 mAh LiIon Cell for a friend, it has had exactly the same: 900 mAh. And thanks for the circuit idea. Ich finde es übrigens auch gut dass die Videos - für alle - in englisch sind, und nicht nur in deutsch. So haben mehr Leute etwas davon :-) .
I did the same test a while ago, but i used a diy usb powerbank where you can insert your own 18650 (1€ from ebay), an adjustable load (3€ from ebay) and a usb charger tester (same price) and tested some vander batteries with it. The promised 6.000 mAh cells only took about 1.2 Ah..
I know that this video is more than 4 years old, but I just bought 9800mah UltraFire Li-Ion cells. They were really cheap when purchased in a lot of 50. I have a bicycle headlight that the 8.4v 3000mah battery pack was not powering the headlight more than 20 minutes when 3 years ago lasted as much as 1 hour. So I assembled 2 rows of 6 cells of these UltraFire with a BMS expecting this battery pack to last more than 2 hours. I charged it and it lasted about 35 minutes on the headlight. I charged one of the 9800mah cells using a Li-Ion battery charger and then discharged it using the same charger/discharger and only got 1300mah. You can bet I was disappointed. But when purchased in a lot of 50, they were much cheaper than other 3000mah cells. I guess you get what you pay for.
What's worse is when a Chinese seller sells an item more quickly, they have an algorithm which automatically increases the price. So the ones who bought later are scammed even more than earlier buyers.
Not sure if it true, but I've had a couple people tell me that the "cheep" batteries need to be discharged at a very low rate. For example a good battery can 35amps at 500mah. While a "cheep" 6amps at 100mah. And that's how you see 9800mah batteries.
Hello GreatScott! I have made this dummy load several years ago, all by your instructions. It really works great and I have found it to be more than suited for my use for many DIY projects. Still, I would have 1 question please. What is the max voltage that I can apply to the dummy load input? So far it worked for me with 25V, but I dont know if I can go any higher. Also, what is the max current that you would suggest - I try to avoid anything above 1A (I do have a large heat sink, so heat is not an issue). Many thanks for all your videos and for a possible reply. Dont forget to make more of those easy projects for us dummies ;) Kind regards from Slovenia
The max current would be the continuous current limit of the Mosfet. The max voltage will be the op amps limit, if it is lower than the Mosfet's drain to source voltage limit. Also, consider the resistor and mosfet's power limit or you'll have some interesting smoke. You can use a beefy heatsink with fan for the mosfet if it does get toasty.
I used an IRFZ44N instead of IRLZ44N, my circuit not works. Why? Please can you help me??? ---UPDATE--- I solve the problem. It was the mosfet that have some problems 😂. I changed it with an IRF3205 and it seems to work perfectly. Now i can test my batteries! Thanks!
Best channel i ever seen about electronics... Pls sir can you make a video about IR proximity sensor using LM358. And can you show how to make it??????
A better way about this would be buying an lipo charger/balancer. Imax B6 can be had for 15-20$ and using cheali-firmware it can do capacity check with a single click. You can also set constant current or constant current and variable current with fixed voltage, when it drops to cutoff like 3V. So you can get a decent idea what the ideal capacity is.
Clever. But wouldnt be easier to use one transistor instead opAmp and mosfet. By establishing base current you will obtain practicaly constant current Jc in CE part, even though the load is changing. No current ripple, no heated resistors. I have checked it with TIP35C. It works.
Another possibility was to use a microcontroller circuit that monitor and saves the current in a SD card. Then, use some computer program to plot the current x time and do the integral.
Attention. FET IRLZ44N is critical. It has Threshold Gate voltage (VGS) 1.0 to 2.0V. This is important.....Usually, Power FET have VSG from 3.0 to 4.0. Those common FET will not work. 🙂
In this schema this mosfet is not working in impulse mode. The opamp keeps it in semi-open state, so his DS resistance is keeping high enough to provide needed voltage drop. So its not impulse mode, and not a key mode.
A clamp meter or clamp tester is a DMM with an in built current transformer. The CT can measure an ac current without having to connect the terminals of the meter or without having to cut open the cable whose current flow needs to be measured.
>be me >theese cost 1 euro where i live >go to ebay >buy a empty power bank that fits 6 of theese for like 2 euro >buy batteries for 6 euro > assemble in my homemade workshop >got an great power bank for 8 euro > my friends ask where can they get that beast that can charge your phone like 4 times >tell them i made >you sell them? >idea.jpeg >tell them its 15 euro >buy shitload of batteries > buy some cases > sell them >everybody in school finds out >everyone wants to buy > sell like 15 more > get cash >there is an empty room in my house >really small like some clostet or something > use cash to buy a nice workbech and tools >making projects like this guy >make something cool and go to some science competition >win >everyone knows >my face when i just sold some shitty batteries and now im here
Nice and simple circuit. I made it work, but mine is quite inaccurate. I suspect the voltage drop in the traces between the mosfet and the shunt resistor is the reason.
The mosfet is being used as a source-follower inside the feedback loop of the opamp which is arranged as a non-inverting buffer with a gain of 1. That results in the voltage at the source of the mosfet being forced to be equal to the voltage set by the potentiometer. Because we want that voltage to appear across a resistor that is referenced to ground, we need the source-follower to be operating in its linear region with the source at the negative potential. That dictates the use of an N-channel mosfet. We can't "flip the circuit" and have the resistor connected to the battery positive because the LM358 can't accept inputs any close to the positive supply rail. The LM358 can however include ground in its input range.
Resistance is not entirely constant during discharge. Internal resistance is in series with the load resistance. Internal resistance is related to both temperature and state of charge (STOC). You can easily compare the capacity of batteries without a constant current source. Just charge the battery to its maximum voltage and discharge it over a fixed resistor (4ohm). The time it takes to reach the minimum voltage of, let's say 2.5 VDC, is relative to the battery's capacity. Obviously, a battery with the highest charge will last longer under equal conditions of which temperature is important.
The internal resistance of the battery doesn't affect this circuit's measurements because the load is a constant current. We want to know the numerical value of the battery's actual capacity in mAh, and that is most conveniently done by keeping the discharge current constant and measuring the time it takes for the battery's voltage to fall to a predetermined value. Your method will tell you that the battery that lasts longest has the highest capacity, but we still won't know what that capacity is.
UltraFire batteries are interesting. I bought a set of "4000mAh" red UltraFire's a few years back, and have been getting a fairly consistent capacity of around 2000mAh out of them when charging from around 3.2v to 4.2v which is very surprising considering. Despite my fairly good results with a set of UltraFire, I still wouldn't recommend them for anything of importance as I doubt they will handle much of a load and despite being labelled as protected mine do not have protection circuits in them. Also I think I was extremely lucky with the set that I bought as 2000mAh is very high compared to what others seem to get. UltraFire is comparable to playing the lottery :D
Hi, my mi power bank take 8 hours to charge by 2amp charger but discharge only charge my phone around half , the capacity of phone battery is 3300mah , now my question is when the capacity of cell of power bank is low then why it take longer to charge full almost 8 hours via 2amp charger... reply soon... Your videos are awesome keep it up...
i got 6 of these ultrafire 4800 mah in parallel on a diy power bank. as soon as i charge my phone the digital indicator in my powerbank drops from 100% -1% every seconds until it gets empty under heavy load of 1.2 amp lol. but noticed it last a little longer when drawing around 0.3 amp max like charging bluetooth headsets. these batteries are extremly light weight, i dont have a scale so i dont have figures. i charge each of them from 3.4v to 4.2v and my usb voltmeter calculates around 400 to 500 mah each
Hi, I'm not clear where did you get 1.8 h of charging if you stopped your stopwatch at around 1.48 h? Taken that into equation you'd multiply the discharge current of 500 mA * 1.48 h = 740 mAh (a bit more) of that battery's actual capacity. Am I wrong? And one more thing: if you had the digital balance charger like Turnigy Accucell 6 that I use, you could just set it to Li-ion and discharge mode, set the desired dicharge current and after it finishes, just read the mA value on the right bottom of the display and multiply that with the time (in h). Cheers!
these usually are filled with sand and inside is a tiny nimh cell (knowing from experience) you should take a worn out cell from a good manufacturer and short circuit test it. also do the same with a china cell thats not sand filled.
Nice video GreatScott! If something looks too good to be true it probably is. They seem to do similar things (but not so bad) with big leisure batteries. Either re- badge car batteries &/or quote inflated AH figures at the 100 hour (or even longer) rate. Ordinary people can get confused. I've tested several that don't meet their stated capacity.(Anybody watching, never run a lead acid anywhere near flat. 10.8 V with a C/20 load (5 a Have a nice day.
Sorry error in 1st reply, should read- 10.8 v for a 12 volt battery should be considered flat with a load of 5 A drawn from a 100 AH battery. Don't even go this low, especially with a car battery. Best to stop when down to 11.5 v or so, if you want reasonable life out of a 12v leisure battery.
GreatScott. I watch and like your every video. Plz make 18650 4X or 8X discharged capacity tester using Arduino, I would really really appreciate that.
This is another excellent video. Your productions are clear and you get right to the point, even if it's over my head! Would you consider creating a project that would measure many recycled 18650s at once? Although many people buy a fake battery or two for their vaping, others of us have 100 or more cells from eBay we are trying to recover. I have been scouring the internet for ways to test many cells at once in order to create balanced packs out of these kinds of cells. Thanks. I will keep watching!!!
Sorry but the energy of a battery changes in accordance to the discharge current: the higher is the discharge current, the lower is the energy you can get from the battery. I'm not an expert of Li-Ion batteries, but with VLA and VRLA batteries usually the capacity is referred to a current that discharge the battery in 8 hours (C8) or 10 hours (C10). If those batteries are discharged in 2 hours the Amp-h you can get are lower.
Many years ago I bought some AA NiMH batteries from China and when I got them they were suspiciously light. So I did something similar and found that the capacity was closer to an AAA battery. So that was probably what it was. When I first complained to the seller he of course refused to acknowledge that the batteries were fake but when I told that I have measured the capacity by integrating the current over time I got my money back.
I'm not quite sure thats an opamp or a comparator. In the first case, I think the output will try to mantain 1V constant on R1, thus the mosfet will be working in its active zone with a voltage drop of Vbat-1V. That will disipate a lot of heat on the mosfet. In the second case. The comparator will be switching the mosfet on and off creating a pulse through R1 with an average voltage drop of 1V in R1 and 0V in the mosfet. Therefore you are drawing 500mA but at 1V, not Vbat, therefore the battery will last longer than if you just connect the battery in series with R1. Therefore real capacity would be much less that 900mAh
It's the first case with the circuit shown in the video. It's working as a non-inverting buffer with a gain of 1, not as a comparator. The mosfet needs a heatsink, for sure.
can you make a tutorial on how to charge a Ni-Mh battery and have it automatically turn off when at full capacity? also if it were possible to charge them while the batteries are connected in their primary circuit such as a stereo. 3 pin negative, charge, and discharge
+GreatScott! Would be awesome if you show how to use this constant load with arduino to automate this process, thank you and sorry for the bad English Hahaha
To be honest I am surprised they are above 600 mAH. An easy way to see if they are the real deal before purchase is by asking the seller how many grams a single battery weighs. They are usually well below 25-30 grams which is a dead giveaway that they are purposely mislabeled. A real 18650 battery weighs around the 35-40 gram mark
perfect timing. just got somthing in with some rather shady battery that had some pretty interesting warnings on the side. i need battery's, whats the best brand name for Li-Ion?
What can we add to this to make it better ? it doesn't seem very practical in this form with the requirement of using external measuring stuff, how can we add something which is going to show when the circuit is operating properly and also when the battery is drained ? comparator linked to an LED ?
You could have used a TP4056 and a resistor load and a usb doctor that measures capacity to do all of this but without having to stay there and watch the voltage and do the calculating. For me this is better because I don't want to stay there watching while my batteries are getting tested. I know it's a bit more money than your build probably, but could you show it to us too?
The snag with that is the TP4056 uses an internal linear regulator to do the charging, so efficiency is uncertain if you're measuring the capacity by charging. To measure the capacity by discharging, the TP4056 isn't doing anything. You could use one of the chips that boost the li-ion voltage to 5V and use a "usb doctor" with a resistor to measure the discharge, but then the efficiency is even more uncertain. If you don't want to stay watching, you'll need the extra protection circuit that cuts off the output when the li-ion gets low.
I am wondering if the amount of current that you continuously withdraw can affect the measurements i.e., if withdrawing 250 mAh will last for 3.6 hours
![โรงเรียนตอนตี 2 โคตรหลอน..!! [โรงเรียนไดเม่]](http://i.ytimg.com/vi/2MAEMwqVYh8/mqdefault.jpg)
![The Wall Song ร้องข้ามกำแพง | EP.213 | กรีน อัษฎาพร | 3 ต.ค. 67 [1/5]](http://i.ytimg.com/vi/3KrlLbCsadI/mqdefault.jpg)
UltraFire: Not the best name for a Lithium battery.
heyhococo ha
the ultrafire flashlight and batteries I got have protection ic's (batteries) and yes the flashlight (9000lum) can get over 100c under max brightness but the cheap ass batteries have an internal resistance that limits the light from going thermonuclear and even mid quality batteries do exactly the same fire may never happen but I was so worried it would I added arctic mx-5 between the led pcb and the case edge and it smoked a bit before it dried a little and I do like it as it is a great short term spotlight/worklight and seems to pull a stable 3.8a from a pack of 4 18650's
😂😂😂😂😂😂
Lol... True
Like 99% of those are fake with up to 9000mah ratings. That's how you can get free batteries from ebay if you tell the seller that the capacitance doesn't even close to stated one (and risk to burn your house with those).
The highest capacity 18650 cell ever made (afaik) is 3600 mah from Panasonic. But you cannot find these anymore. So currently the highest capacity is 3400 mah again from Panasonic. With the current chemistry its not possible to squeeze more energy into a 18650 cell. Anything above that is automatically a lie, unless it's from one of the top manufacturers, its genuine and a new chemistry has been used.
I believe you probably knew all these and I understand this video is for educational purpose. My comment too.
Mandragoras thank you :)
Vláďula Vaverka :) You're welcome.
Mandragoras Good news. LG is now selling a cell with new chemistry and 3500mah capacity.
goo.gl/ElR7jA
+Mandragoras Thanks for the info. I was looking for a battery for use with arduino and this info as really helpful.
Tomas Schertel
You are welcome. My comment is a little bit old and now there can easily be found new 3500mah cells, available from reputable manufacturers.
i'm gonna use this circuit to test all my liion batteries, thanks
Got the same result using my charger that tells how much mA was charged. Thanks for this test. Now i know I can trust my charger
finally a guy that knows what he is doing. Very good. There is ao much shit around this subject. Thanks a lot. Peter from Denmark
You're welcome ;-)
@I killed that beard guy Hello. How do you do ?
@I killed that beard guy I am really happy to hear that.
@I killed that beard guy Thank you so much . You are the kindest person I dont know. Health and success to you.
Don't forget to report the idiots that sell these counterfeit products on Ebay.
+TransfixedBunnies There is literally dozen of them popping up everyday on eBay, nothing will happen if you report them, in fact I've seen some of the better sellers will say it's not genuine, but what do you expect for $1 dollar? Most people will buy the battery pop it in the flash light and it will run for an hour and they will be satisfied. Your other option is to spend $11 bucks and buy something like Keeppower with Japanese protected cell so you get what you pay for.
Dima Prok My problem is not that you get what you pay for, that part is a property of the market economy. My problem is that you do not get what you pay for, or to be more precise, what you are promised. There is no gray area or matter of opinion in scientific matters. 9000mAh is 9000mAh, not 1000mAh + imaginary saved cash flow.
Sadly I have to at least partially share your pessimism about the end result of reporting fakery in ebay. That is just ebay for ya.
What do you expect of Chinese sellers? I once read a question on Quora, I don't remember exact wording but it was something about why do Chinese copy (steal) and sell cloned technology. And the answer was by Chinese guy is that for them it's a way of life and they don't consider this wrong. So I think this exact same mentality applies to grossly exaggerating specs, it's just part of marketing, how they do business. But you know this is not the worst, I was shopping for Power Supply just few days ago, and I see a big seller with 98% feedback and I am concerned about this seller's reputation, so I started reading complains and there was literally over a hundred complains in the last 6 months. #1 was seller was listing that he was located in CA while the items were shipped from China, then there were up to 10 day delays in shipping stuff out, there were people who never received stuff and didn't get refund, there were people who got wrong /broken / used items and it made me wonder how is this guy still in business? Well I guess if you sell 20 little junk items that people won't complain and then screw someone over a bigger item the feedback looks fine and ebay doesn't care because they get their piece of pie.
+Dima Prok Forgot to mention, I try to look for 99.5% or better feedback with big sellers. With small sellers I look how many negative feedback there was in 6-12 months and what kind it is. If the item arrived damaged, then most likely it's shipping company fault but most people will blame seller.
+Dima Prok you make good points. Having read entrepreneur blog and talked to people who have worked in China I gather contracts are much more abstract and flexible (as in not trustworthy in same degree) than in Western countries.
19% performance - That's pretty good for some Chinese knock off equipment!
This is the best explanation video out there in the internet about constant current for battery discharge test. Thanks GreatScott!
Should read 4800 China-amphours
lol
Bob Brown: "China-amp/hours"!!!!
Reminds me of the Hp ratings of Chinese motors.
There are those little bitty Chinese horses and then
there are those big, strong American horses.............
DD_Ranch Texas most US HP claims are BS. In Japan they under-rated their engines for a while (they started focusing on handling instead of an HP battle, which gave them much faster cars in real life, so they put an upper limit rating of like 275 HP or something even if the engines were 300-400 HP). In the US, we were rating older cars with stupid high HP which is easily disproven, we claimed crap like 1980’s cars with 300 HP that took 18 seconds or more to hit 60 mph... and they were cars, not literal battleships... and the new ratings are also BS, my dads Dodge Dart claims a higher HP rating than his old 1987 Mercedes (which is not even running right), and yet the Mercedes outruns it (and that car has like 200 pound electric seats, I felt like I was breaking my back trying to remove them from another car at the junkyard...). We still claim BS on most of the engines, and of course it’s BS rated on top of not using wheel power (somehow every car has the worlds most inefficient transmission ever produced right? That must be why 240 HP dropping to 170 seems reasonable...)
There are a few US cars that are actually rated correctly (typically dyno tested crate engines), and even a few that are under-rated. But then you also have BS like Dodges Hellcat engine bragging 777 HP or whatever, but then require a “race key” (in red obviously), to “unlock” that power (which is what you paid for already...), basically DLC from video games in real life... What a scam to advertise it as such and then lock it to 500 HP! So that’s beyond stupid.
As an engineer, I’d call it a faulty reject if it doesn’t hit the spec, if I buy a car and it’s out of tolerance, it should go in the scrap bin. The US marketing teams are about like the Chinese (which is super frustrating, I don’t pick a parts spec out of thin air, if that electronics piece smokes itself or melts at half the power, then why put that rating sticker on it?!?). Anyways, US drag racing is also flawed, it doesn’t measure time until the front wheel has already passed the laser, which means it takes off like 1/10 of a second or more (and yet posting crap in the thousandths of a second like it’s accurate or something...), so you could by their standard hit 60 mph in 0.000 seconds! Unfortunately you can’t get a negative time to 60 mph, but obviously if another car hit it 80 mph it wouldn’t have a lower 0-60 time). I guess it makes tanks with really long tracks look super good or something, it’s pointless, they are purposely measuring it wrong, just have a second laser, but oh no that would screw up marketing!
Fuck China
Is like PMPO Watts.
Fantastic! This is exactly what I was looking for. I tried making a constant current load with an LM317 to test battery capacity, but the voltage was too low resulting in the whole thing falling out of regulation. I think I have most of these parts lying around. Thank you so much, I can't wait to try it out!
Why don't you use a tp4056 charging and discharging module. It will automatically cuttoff you battery connection with your lm317 tester when it discharge alot....
I know I'm little bit late here,😅
When I tested my li-on battery for power tool, I just measured voltage over the resistor load at different times. For example I measured (18v at t=0, 17v at t=30min and 16.5v at t=1h) then I just do polynomial interpolate u(t) in matlab and integrate the function (p = u(t)^2/R) over the test time. This should give you stupendously accurate results since u(t) is changing slowly monothonically over time :)
Btw it is easier to measure voltage accurately than current
Good video GS..!
Just a few small points to add for all......
1. The constant load circuit is simple and neat but there are alternatives which will eliminate the ripple. This is relevant if we wish to to test capacity precisely.
2. The relationship between discharge current and capacity is non-linear. The lower the current, the greater the measured capacity.
3. I suspect you will find the Ultrafire is rated for a discharge current less than your 0.5A, which means you will generate a pessimistic result.
4. Low voltage cut-off at 2.5V will not damage the battery. All one can say is that life measured in discharge cycles will be lower compared to 3V cut-off.
Keep up the good work..!
The way you used to speak 4 years ago 😀 it's really great how you progress in many ways
I once bought a pair of Ultrafire 14500's rated at 1200mAh, and yeah, real capacity was about 300mAh.
Thank You, Greatscott, for all your constant output vids!Robert Murray-Smith has a large vid library devoted to DIY paper/carbon megacapacitors for use in motive apps. The voltage drop on a megacap is like a rock slide. I dunno if homebrew (raw materials start-point) development is your idea of fun, but I propose that megacaps constructed of low-ESR kilocaps would be the solution to this competent alternative to li-Ion, if efficient solid-state power-switching can be done to sequentially boost volts at the expense of amps as discharge progresses. RMS ( a phD chemist) is trying to compete with his professor-brother for the kiddie-crowd and sticking to paper bacitors. I further propose that nickel foil and PET film starting materials would guarantee low ESR and high cycle efficiencies. If you know anybody who wants to wow 'em with high-power motive demos (electric motorbikes, etc) this approach looks ripe-- but I dunno electronics, so maybe you could comment on the utility of the above switching scheme. Yours is the most awesome DIY electronics channel I've found in many years of YT searches. Thanks again, for being so cool!
I would like to have seen a direct comparison of a correctly marked battery with this example. To show how a proper battery should hold up.
*takes exact same ultrafire cell and writes "500mAh" on it* :D
yeah ... not very scientific
Fantastic Experiment.
I also made the same circuit to test a couple of fake lithium ion batteries that I had.
The actual capacity of those batteries proves out to be only 418mAh.
And not 2500mAh as mentioned on them.
Thank you very much for letting me know this capacity test experiment.
900mah Capacity at 2.5V discharged is even worse than the average 3.0V/2.9V most smart dischargers use. But those batteries are known for this. 900mah is actually higher than you see when other test the same batteries. I actually purchased 6 once and the capacity said 3000mah, but they actually tested 2200mah at 3.0V(I was floored). But the basic wear and tear(charge and discharging) cause it to lose Mah quickly. After 3 or 4, the MAH dropped to around 1100. I just put them in my recycled 18650 bin and called it a day! Great Video Great Scott!!
I am following your channel from Iraq and I found the videos very useful, educational and practical for hobbies and professional engineers, keep it up, many thanks.
When it comes to 18650 Li-Ion battery, the valid ranges are from 2,000 to 3,500 mAh depends on its drain characteristics.
Avoid the batteries claiming anything below or above this range.
Btw, higher the drain current(>30A), the smaller the capacity (close to 2,000 mAh), and of course, lower the drain current(
and I will see you next time... love the thought
Leaving a portuguese message for this vídeo:
foi bom pra caramba!
Ai Caramba!
2:18 "this guy will get really hot...literally."
The capacity of a battery shall be measured at 1/10C current (hence, you must have something near10 hours of discharge); if the drained current is too high the measured capacity of the battery become lower than expected (and, on the contrary, if the drained current is too low the measured capacity become higher than expected)
I also suggest the use of some smoothing & decoupling capacitors to remove the oscillations.
teslacoiler But 1/10C is far away from reality. I will do some further test with other big brand batteries to see whether they can supply their rated capacity with 500mA.
Li-ion batteries are pretty good at supplying most of their rated capacity a higher discharge rates than C/10, so the error is unlikely to be great. I have a "Multi Smart 3000" battery charger/tester from 7-Day Shop and it can measure battery capacity through a full charge-discharge-charge cycle at a choice of several set currents. I haven't seen much difference when testing the same 18650 batteries at any of the test currents, so I'm fairly confident that the capacity I've measured is close to its value for C/10.
I do strongly agree with the need to properly decouple the supply to the opamp to remove those oscillations. I'd suggest a 100nF ceramic capacitor between pins 4 and 8 of the opamp, as close as possible to it, along with a 10μF - 100μF electrolytic capacitor across the incoming 5V supply.
Love your work.. I wouldn't mind seeing the same test being done on a highly regarded battery brand, such as Duracell or for those people in the RC hobbies maybe turnigy.. Thanks for your informative video keep them rolling.
I Made a constant load by watching your video, works great . But I used a 12V power supply because of irf3205 . The MOSFET doesn't even heats.
Who here is happy that they got all this information?
Just found your channel today, and after watching few vids, subbed. Keep 'em coming!
Just a simple and dumb question, but only for being sure, in the constant curent load diagram, you use the logic level z44n, but i have the normal irfz44n, if i supply it with anyting between 10 to 12 v i suppose it will work as well, the only question is if the potentiometer to regulate the current pat has to change a lot or it can be the same 500k.
Thanks.
jotasgasco Potentiometer value stays the same. But you may need to increase the supply voltage for the LM358 and potentiometer. From 5V to 12V should do the trick.
GreatScott! BIG Thanks! that did the trick for me, here in Taiwan i only can get the IRFZ44V and the C358C, with 5V i was getting only 200mA max load. With 12V i can increase the load to 1A or even higher, but 0.5A is enough for testing LiIo's... i finally have a constant current load i always wanted, now i just need to dig out the arduino + gnuplot thingy i made a while ago to make real accurate testings. Besten Dank und.... Ohren steif halten und weiter machen!
tt
What do you mean, my UltroFite Li-ion cells do not have 8800 mAh?
Can you eliminate the oscillation in the op amp circuit by adding a capacitor in the feedback path?
Maybe by lowering the gain of the OP. Was surprised that he had no feedback loop.
Isn't the oscillation the point? It's effectively making a PWM load instead of literally constant current, isn't it?
@@reubenkriegel7639 No the oscillation is not the point, and it's a linear circuit (a non-inverting buffer), not PWM.
@@gamingSlasher It has 100% negative feedback already, so you can't lower the closed loop gain any further. It's a non-inverting buffer with the mosfet simply adding extra current capability (which it happens to draw from a separate supply - the battery under test).
To eliminate the oscillation, I'm pretty sure the solution is just to decouple the supply properly. Add a 100nF ceramic capacitor between pins 4 and 8 of the opamp, as close as possible to it. It wouldn't hurt to put a 10μF electrolytic capacitor across the incoming 5V supply, especially if you have long supply leads.
Can i ask why are you choose 2.5V threshold?
(I'm not a natural english speaker so if theres ugly errors about right writing , please do not account . Im still learning) So than , to the real comment. I just spent like a hole day watching your videos. Excellent channel and videos !! If you can , please make some video about using ir reciever with raspberry pi ( using php , or the languages that you use as well .. they are beeing very usefull to me , to learn some things and , get some ideas for my own project) .
Thanks , and again awesome channel !
João Eduardo Saad Thanks ;-) I will put your idea on my to-do list.
Hey Great Scott! I would like to know, why You disregard the transistor consumption power? You said that it's really getting hot... Hot means energy... I think that You don't need constant current, but constant power. I would choose buck-boost for constant voltage on the resistor. Constant voltage on resistor means constant current means constant power consuming minus buck boost efficiency.
Isn't the current used by the heating process? That's where what you are measuring goes if I understand correctly.
Yes a very good video from someone who is far more technical than me. I am not anywhere near as technical as this guy but he certainly knows what he is talking about.
I know a better way though.
Don't buy cheap Chinese batteries.
I buy only the Panasonic enelop rechargeables or the Optimum Power 18650's from Japan. These have never let me down and I have had them for years and still going strong.
2:34
With a negative feedback, you do not have a comprarator.
I may be wrong, but here's what I think:
Through the feedback path, the OP is trying to get a 0V difference in between its inputs. (Comparators do not have feedback, just two inputs. )
So, it is taking the voltage on the potentiometer as its reference, and regulates the mosfet in its linear region (that's why it gets hot), in order to let so much current pass, that
I_R1•R1 = U_ref
at potentiometer.
In orher words; The voltage drop at R1 were the OP is measuring, is equal to the voltage at its non inverting input (+), due to the current which it allows to pass through the transistor and through R1.
Meaning, you have an amplifier with the resistance of the Mosfet dictating the -let's say- degree of amplification.
Said degree of amplification changes with the voltage of the battery, because it takes a smaller resistor to allow equally as much current to pass, as the voltage of the battery drops.
The other way of looking at it is that the circuit is an opamp with a mosfet in source-follower mode to buffer the output current to a much higher value than the opamp is capable of supplying. Then, as you say, the opamp action under negative feedback will ensure that the voltage at its inputs is the same, leading to the mosfet sourcing a constant current (Vref / R1) to the resistor R1. Since that current must come via the drain of the mosfet, it will draw a constant current from its supply, i.e. the battery under test. It will do so regardless of that battery's voltage until the battery no longer has enough voltage to keep the mosfet's source at Vref.
There is no inherent oscillation or ripple in the circuit - that's just a misunderstanding of how a feedback loop works, along with poor layout.
Great video! Could you modify this circuit to use the current for charging an empty battery? Basically, run a battery discharge test while charging an empty battery. Might be a great idea for future video!
I've made a couple of discharge tests on a small 12v 1.3Ah battery.
I took a rough average of the discharge current.
I was annoyed that I only got 60% of the rated capacity.
Does anything from China do what it supposed to do or do they just write anything they want on the label?
Great :-) . I´ve checked an 9800 mAh LiIon Cell for a friend, it has had exactly the same: 900 mAh. And thanks for the circuit idea. Ich finde es übrigens auch gut dass die Videos - für alle - in englisch sind, und nicht nur in deutsch. So haben mehr Leute etwas davon :-) .
I did the same test a while ago, but i used a diy usb powerbank where you can insert your own 18650 (1€ from ebay), an adjustable load (3€ from ebay) and a usb charger tester (same price) and tested some vander batteries with it. The promised 6.000 mAh cells only took about 1.2 Ah..
I know that this video is more than 4 years old, but I just bought 9800mah UltraFire Li-Ion cells. They were really cheap when purchased in a lot of 50. I have a bicycle headlight that the 8.4v 3000mah battery pack was not powering the headlight more than 20 minutes when 3 years ago lasted as much as 1 hour. So I assembled 2 rows of 6 cells of these UltraFire with a BMS expecting this battery pack to last more than 2 hours. I charged it and it lasted about 35 minutes on the headlight.
I charged one of the 9800mah cells using a Li-Ion battery charger and then discharged it using the same charger/discharger and only got 1300mah. You can bet I was disappointed. But when purchased in a lot of 50, they were much cheaper than other 3000mah cells. I guess you get what you pay for.
Two thumbs up. A great demo.
All good people are left hand writers Great video very interesting Thank you
haha novadays you can find 2x 6800mah 18650 batteries for ~3 dollars on ebay.. seems legit
And you can see "650 sold" on the corner.
Gotta feel sorry for them.
What's worse is when a Chinese seller sells an item more quickly, they have an algorithm which automatically increases the price. So the ones who bought later are scammed even more than earlier buyers.
Not sure if it true, but I've had a couple people tell me that the "cheep" batteries need to be discharged at a very low rate. For example a good battery can 35amps at 500mah. While a "cheep" 6amps at 100mah. And that's how you see 9800mah batteries.
Hello GreatScott! I have made this dummy load several years ago, all by your instructions. It really works great and I have found it to be more than suited for my use for many DIY projects. Still, I would have 1 question please. What is the max voltage that I can apply to the dummy load input? So far it worked for me with 25V, but I dont know if I can go any higher. Also, what is the max current that you would suggest - I try to avoid anything above 1A (I do have a large heat sink, so heat is not an issue). Many thanks for all your videos and for a possible reply. Dont forget to make more of those easy projects for us dummies ;)
Kind regards from Slovenia
The max current would be the continuous current limit of the Mosfet. The max voltage will be the op amps limit, if it is lower than the Mosfet's drain to source voltage limit. Also, consider the resistor and mosfet's power limit or you'll have some interesting smoke. You can use a beefy heatsink with fan for the mosfet if it does get toasty.
I used an IRFZ44N instead of IRLZ44N, my circuit not works. Why? Please can you help me???
---UPDATE---
I solve the problem. It was the mosfet that have some problems 😂. I changed it with an IRF3205 and it seems to work perfectly. Now i can test my batteries! Thanks!
Best channel i ever seen about electronics...
Pls sir can you make a video about IR proximity sensor using LM358. And can you show how to make it??????
A better way about this would be buying an lipo charger/balancer. Imax B6 can be had for 15-20$ and using cheali-firmware it can do capacity check with a single click.
You can also set constant current or constant current and variable current with fixed voltage, when it drops to cutoff like 3V. So you can get a decent idea what the ideal capacity is.
Clever. But wouldnt be easier to use one transistor instead opAmp and mosfet. By establishing base current you will obtain practicaly constant current Jc in CE part, even though the load is changing. No current ripple, no heated resistors.
I have checked it with TIP35C. It works.
Another possibility was to use a microcontroller circuit that monitor and saves the current in a SD card. Then, use some computer program to plot the current x time and do the integral.
Attention. FET IRLZ44N is critical. It has Threshold Gate voltage (VGS) 1.0 to 2.0V. This is important.....Usually, Power FET have VSG from 3.0 to 4.0. Those common FET will not work. 🙂
How about npn transistors?
In this schema this mosfet is not working in impulse mode. The opamp keeps it in semi-open state, so his DS resistance is keeping high enough to provide needed voltage drop. So its not impulse mode, and not a key mode.
Could you explain how clamp meters work? Specifically when measuring amps? Thank you!
A clamp meter or clamp tester is a DMM with an in built current transformer. The CT can measure an ac current without having to connect the terminals of the meter or without having to cut open the cable whose current flow needs to be measured.
A CT (current transformer) is a device that measures a large current with an instrument having restricted range of about 5 Amps.
good job man! nice usefull project, the interaction between mosfet and opamp was simple and great at same time.
>be me
>theese cost 1 euro where i live
>go to ebay
>buy a empty power bank that fits 6 of theese for like 2 euro
>buy batteries for 6 euro
> assemble in my homemade workshop
>got an great power bank for 8 euro
> my friends ask where can they get that beast that can charge your phone like 4 times
>tell them i made
>you sell them?
>idea.jpeg
>tell them its 15 euro
>buy shitload of batteries
> buy some cases
> sell them
>everybody in school finds out
>everyone wants to buy
> sell like 15 more
> get cash
>there is an empty room in my house
>really small like some clostet or something
> use cash to buy a nice workbech and tools
>making projects like this guy
>make something cool and go to some science competition
>win
>everyone knows
>my face when i just sold some shitty batteries and now im here
Nice and simple circuit. I made it work, but mine is quite inaccurate. I suspect the voltage drop in the traces between the mosfet and the shunt resistor is the reason.
why N-channel mosfet is used at the high side? Is there specific reason to use N-channel instead of P-channel mosfet?
The mosfet is being used as a source-follower inside the feedback loop of the opamp which is arranged as a non-inverting buffer with a gain of 1. That results in the voltage at the source of the mosfet being forced to be equal to the voltage set by the potentiometer. Because we want that voltage to appear across a resistor that is referenced to ground, we need the source-follower to be operating in its linear region with the source at the negative potential. That dictates the use of an N-channel mosfet. We can't "flip the circuit" and have the resistor connected to the battery positive because the LM358 can't accept inputs any close to the positive supply rail. The LM358 can however include ground in its input range.
3 years before, the guy had some problems to speak english, today it is much better.
Have you ever tested a good Ultra Fire Battery? I think they are all crap.
The are all fake
I recently purchased and tested a couple UltraFire 5000s and a couple UltraFire 3800s. The 5000s test to 1000mAh. The 3800s tested at 450mAh.
every single one is fake.
Its their business to sell fake batteries.
Google Plus Blows Chunks
so one multiplied the capacity by 5, the other just basically added a 0 to it.!
that is quality.
most chinese ultrafire are laptop recicled batteries so they have less than half life cicle
This circuit also works very well as an LED driver.
What is it with Germans and safety. The batteries are called Ultrafire, what do you expect? A small amount of flameless warmth?
Greatscott you change way I previously think about electronic. Thanks bro .......😃😃
Shouldn't the voltage output of the Op amp be 5V not 4v?
Yeah, I dont understand why he said 4V
Voltage drop over shunt resistor?
Resistance is not entirely constant during discharge. Internal resistance is in series with the load resistance. Internal resistance is related to both temperature and state of charge (STOC). You can easily compare the capacity of batteries without a constant current source. Just charge the battery to its maximum voltage and discharge it over a fixed resistor (4ohm). The time it takes to reach the minimum voltage of, let's say 2.5 VDC, is relative to the battery's capacity. Obviously, a battery with the highest charge will last longer under equal conditions of which temperature is important.
The internal resistance of the battery doesn't affect this circuit's measurements because the load is a constant current. We want to know the numerical value of the battery's actual capacity in mAh, and that is most conveniently done by keeping the discharge current constant and measuring the time it takes for the battery's voltage to fall to a predetermined value.
Your method will tell you that the battery that lasts longest has the highest capacity, but we still won't know what that capacity is.
Jung was hast du denn gelernt?
Einfach Klasse was du da immer hinzauberst xD
UltraFire batteries are interesting. I bought a set of "4000mAh" red UltraFire's a few years back, and have been getting a fairly consistent capacity of around 2000mAh out of them when charging from around 3.2v to 4.2v which is very surprising considering.
Despite my fairly good results with a set of UltraFire, I still wouldn't recommend them for anything of importance as I doubt they will handle much of a load and despite being labelled as protected mine do not have protection circuits in them. Also I think I was extremely lucky with the set that I bought as 2000mAh is very high compared to what others seem to get. UltraFire is comparable to playing the lottery :D
Great video !!!
Its surprise to see it only works 900mAH
Hi, my mi power bank take 8 hours to charge by 2amp charger but discharge only charge my phone around half , the capacity of phone battery is 3300mah , now my question is when the capacity of cell of power bank is low then why it take longer to charge full almost 8 hours via 2amp charger... reply soon... Your videos are awesome keep it up...
i got 6 of these ultrafire 4800 mah in parallel on a diy power bank. as soon as i charge my phone the digital indicator in my powerbank drops from 100% -1% every seconds until it gets empty under heavy load of 1.2 amp lol. but noticed it last a little longer when drawing around 0.3 amp max like charging bluetooth headsets. these batteries are extremly light weight, i dont have a scale so i dont have figures. i charge each of them from 3.4v to 4.2v and my usb voltmeter calculates around 400 to 500 mah each
Hi, I'm not clear where did you get 1.8 h of charging if you stopped your stopwatch at around 1.48 h? Taken that into equation you'd multiply the discharge current of 500 mA * 1.48 h = 740 mAh (a bit more) of that battery's actual capacity. Am I wrong? And one more thing: if you had the digital balance charger like Turnigy Accucell 6 that I use, you could just set it to Li-ion and discharge mode, set the desired dicharge current and after it finishes, just read the mA value on the right bottom of the display and multiply that with the time (in h). Cheers!
There's 60 minutes in an hour, not 100.
1 + 48/60 (4/5) = 1.8 hours.
was suggested this video, your accent has improved so much since 2015, you don't even sound that German anymore
these usually are filled with sand and inside is a tiny nimh cell (knowing from experience) you should take a worn out cell from a good manufacturer and short circuit test it. also do the same with a china cell thats not sand filled.
Nice video GreatScott! If something looks too good to be true it probably is. They seem to do similar things (but not so bad) with big leisure batteries. Either re- badge car batteries &/or quote inflated AH figures at the 100 hour (or even longer) rate. Ordinary people can get confused. I've tested several that don't meet their stated capacity.(Anybody watching, never run a lead acid anywhere near flat. 10.8 V with a C/20 load (5 a Have a nice day.
Sorry error in 1st reply, should read- 10.8 v for a 12 volt battery should be considered flat with a load of 5 A drawn from a 100 AH battery. Don't even go this low, especially with a car battery. Best to stop when down to 11.5 v or so, if you want reasonable life out of a 12v leisure battery.
GreatScott. I watch and like your every video. Plz make 18650 4X or 8X discharged capacity tester using Arduino, I would really really appreciate that.
Thank you for your effort on saying the reality! Much appreciated. Some advices about good baterries?
This is another excellent video. Your productions are clear and you get right to the point, even if it's over my head! Would you consider creating a project that would measure many recycled 18650s at once? Although many people buy a fake battery or two for their vaping, others of us have 100 or more cells from eBay we are trying to recover. I have been scouring the internet for ways to test many cells at once in order to create balanced packs out of these kinds of cells. Thanks. I will keep watching!!!
Great to see another lefty ;)
tanks for keep sharing, i was planning to buy a liito kala that can do NOR test, the problem is i only have a dozen off cells, not worth LOL
You should make a tutorial on how you make such straight wire bridges for your protoboard :D
***** Super simple. Just grip the two ends of your wire with a plier and pull. That's it.
GreatScott! which wiregauge are you using? 0.8mm?
***** Yes
GreatScott! Excuse my ignorance, what kind of wire is it? It doesn't look like wire, rather like rigid metal. Can you give a link?
Mandragoras loks like tinned copper or silver wire
Sorry but the energy of a battery changes in accordance to the discharge current: the higher is the discharge current, the lower is the energy you can get from the battery. I'm not an expert of Li-Ion batteries, but with VLA and VRLA batteries usually the capacity is referred to a current that discharge the battery in 8 hours (C8) or 10 hours (C10). If those batteries are discharged in 2 hours the Amp-h you can get are lower.
Many years ago I bought some AA NiMH batteries from China and when I got them they were suspiciously light. So I did something similar and found that the capacity was closer to an AAA battery. So that was probably what it was. When I first complained to the seller he of course refused to acknowledge that the batteries were fake but when I told that I have measured the capacity by integrating the current over time I got my money back.
I'm not quite sure thats an opamp or a comparator.
In the first case, I think the output will try to mantain 1V constant on R1, thus the mosfet will be working in its active zone with a voltage drop of Vbat-1V. That will disipate a lot of heat on the mosfet.
In the second case. The comparator will be switching the mosfet on and off creating a pulse through R1 with an average voltage drop of 1V in R1 and 0V in the mosfet. Therefore you are drawing 500mA but at 1V, not Vbat, therefore the battery will last longer than if you just connect the battery in series with R1. Therefore real capacity would be much less that 900mAh
It's the first case with the circuit shown in the video. It's working as a non-inverting buffer with a gain of 1, not as a comparator.
The mosfet needs a heatsink, for sure.
Would it work to reduce the swing by adding a negative feedback capacitor to the opamp, to produce an analog gate voltage and change on resistance
can you make a tutorial on how to charge a Ni-Mh battery and have it automatically turn off when at full capacity? also if it were possible to charge them while the batteries are connected in their primary circuit such as a stereo. 3 pin negative, charge, and discharge
So only one terminal is necesary for the battery, other one is in paralel just for measuring?
+GreatScott! Would be awesome if you show how to use this constant load with arduino to automate this process, thank you and sorry for the bad English Hahaha
Can i use 2 z44 in perallel because single one get so hot??
To be honest I am surprised they are above 600 mAH. An easy way to see if they are the real deal before purchase is by asking the seller how many grams a single battery weighs. They are usually well below 25-30 grams which is a dead giveaway that they are purposely mislabeled. A real 18650 battery weighs around the 35-40 gram mark
perfect timing. just got somthing in with some rather shady battery that had some pretty interesting warnings on the side. i need battery's, whats the best brand name for Li-Ion?
What can we add to this to make it better ? it doesn't seem very practical in this form with the requirement of using external measuring stuff, how can we add something which is going to show when the circuit is operating properly and also when the battery is drained ? comparator linked to an LED ?
You could have used a TP4056 and a resistor load and a usb doctor that measures capacity to do all of this but without having to stay there and watch the voltage and do the calculating. For me this is better because I don't want to stay there watching while my batteries are getting tested. I know it's a bit more money than your build probably, but could you show it to us too?
The snag with that is the TP4056 uses an internal linear regulator to do the charging, so efficiency is uncertain if you're measuring the capacity by charging. To measure the capacity by discharging, the TP4056 isn't doing anything. You could use one of the chips that boost the li-ion voltage to 5V and use a "usb doctor" with a resistor to measure the discharge, but then the efficiency is even more uncertain. If you don't want to stay watching, you'll need the extra protection circuit that cuts off the output when the li-ion gets low.
Really "HOT" literally got me dying 💀
Since we are ALL familiar with ohms law..... I am but not ALL people know. I love this video though😊
Ok with this tips i think many people will stop to buy Chinese battery, you must create a tutorial to make our own 4800 mA battery ^^ Great job :)
Ultra fire, cuz they just start burning, got it?
You should re-do this video but add on the same test using a genuine Panasonic 3400mAh li-ion.
Nice trick to check battery capacity
WHOO FOR BEING LEFT HANDED!!!! MY NAME IS EVEN SCOTT L0000TTTT!!11!1!1gf
well im also left handed
I am wondering if the amount of current that you continuously withdraw can affect the measurements i.e., if withdrawing 250 mAh will last for 3.6 hours