Actually the test mode charges the battery first, then it records how far it can discharge, this is the actual capacity and it is measured not calculated. Then it charges them back up for your convenience. 8:55
They can measure individual currents (using one shunt) of the batteries if they switch off discharging for all but one battery. Just do it fast for all batteries once a second etc. It's all in the firmware.
Just because you *can* charge at 4C does not mean you should :-) I try to keep my charge-rates down to 1C, even for cells rated at 10C charge rates and my reward is a much improved life (in terms of charge/discharge cycles).
+xjet I understand that charging too slow can also do other kinds of damage, but that gets into the chemistry of the things. I do know the biggest offender is hot spotting when charging, and only when that can be addressed should the other factors with the crystallization be addressed. And even this would vary depending on the chemistry the batteries use. I like nitecore smart chargers as they have optimized charging for each type of battery chemistry, and are pretty good at auto-detecting the right chemistry.
+Richard Smith a definite issue with charging NiMH too slowly is that the -dV/dt moment will be much harder to measure because the voltage jump will be a lot less severe. With most decent quality chargers common accepted guidelines on candlepowerdforums was 0.5C at least or the termination of the charging might not happen reliably. Obviously if you can get the noise down in your measurements and have enough resolution it will work on lower currents too. For NiMH AA batteries a generally accepted overcharge current that isn't terrible (the heat can be handled with passive cooling) is about 0.1C. A lot of old "dumb" chargers would kinda use that as the maximum. I don't know about the consequences of long term overcharging on the chemistry but it's claimed to be relatively safe hazard wise. Otherwise those old chargers that don't turn off when forgotten would have caused more fires in the past. I'm kinda curious what the limit is where the battery can't safely get rid of the excess energy without fan based active cooing.
ToKeepTrackOfRandomSubs Ah, i see. So the issue with slow charging has more to do with charge level triggering rather than other factors? I read a paper not long ago about how many people were actually reducing the life of their batteries by going overboard and charging too _slowly_, some doing more damage to their batteries than quick charging, and although I do not exactly remember the details, i don't remember it having anything to do with technological limitations, like detection sensitivity, but rather, a limitation in the actual chemistry. It may just be my memory playing tricks on me. That paper did go on to talk about how quick charging does have to do with technological limitations, and batteries will get better at quick charging as they get better at consistent charge distribution and consistent thermal conductivity.
Richard Smith Well there are 3 accepted methods of charging. Method 1: Slow charging at 0.1C until they are full (overcharging with this will cause minimal harm to the cells, it has been the traditional method). Method 2: -dV/dt once the battery is at the point where its full there will be a slight drop in voltage of a few mV. This effect is stronger in NiCd compared to NiMH and the higher the pulsed charging current is the more noticeable it is. It's best to have a backup termination here as well in case it does miss that moment, both voltage and temperature ideally. Method 3: High current charging up to a certain cell voltage and after that point switch to a lower current trickle charge to top it off. In the case of the low self discharge batteries my own preference goes to method 2 without any trickle charge afterwards. With older batteries and normal "high capacity" ones rather than low discharge a maintenance charge can be acceptable (they lose about 10% to self discharge on the first day and then up to 5% or so per day afterwards, both heavily dependent on ambient temperature and the battery itself). I've got my charger set to a bit below 1C this way it picks the termination point up reliably and its still fast enough while giving good cycle lifetime without overcharging them (beyond the few moments where the -dV/dt moment triggers). I suspect where damage from slow charging happens is if you go overboard on the slow charging and you don't know exactly if the battery is full or not yet you keep charging. The temperature won't rise significantly and nothing really extreme happens with the voltage either because not enough energy goes into it, I don't know what happens on a chemical level I'm afraid.
That is of course obviously the case. To rephrase the question a bit. If you had been using your Varta 15 minute charger for the past 10 year exclusively charging it only with that device and also discharging it within it's described data sheet. How long on average did you find that the cells lasted in terms of productive usable working cycles until they were less than the 50% of people will disregard and repurchase. or How long do the batteries last.
I can think of two ways to monitor the charge for each battery: 1) Use a constant current charge circuit so all "active" batteries are being charged at the same rate, coupled with using the duty cycle of each battery being charged adjusting it's "share" of the 8A (two batteries at a time maximum). 2) Use a split charge/sense duty cycle. In the charge cycle, all (or half*) of the batteries are "on" (not shunted), but in the sense cycle, each battery is sensed using time division multiplexing (all but one battery is shunted at a time). This would also make it possible to sense the batteries' presence and adjust the duty cycle of batteries based on their voltage/temperature. A scope could figure out if either of these are being used...
Hey Dave, I think You have it wrong at some point. The charging circuit has to be able to cut out a single battery. If You put in on battery with 20% charge and one with 80% charge the already full battery would blew up if the other is charged. Maybe some ''magic'' with the special type of the mosfet (using out of spec)?
+WuWo Ah, yes, good point. In that case it's easy, the bottom MOSFET turns on and the top one turns OFF. Top body diode is reversed bias so no conduction. Even cleverer than I thought.
+EEVblog ...but Dave, you will not be able to turn the top MOSFET off - at least if the compliance of your current source is high enough to overcome the forward voltage drop of the body diode, there will always be current flowing through the top n-MOSFET, independent of the applied gate voltage! If on the other hand the compliance voltage was low enough, then the current could be switched off... ...ok, continued watching and you concluded that yoursefl - of course
+EEVblog Also, since you have a microprocessor in there, measuring the charge current on a single, common point (the single current shunt) is easy. Every second or so, switch to charging a single battery, step through all batteries and tada it's done. As for regulating the charging current: Do you actually have to do that? The constraint seems to be temperature, so couldn't you use pulse width modulation to set an "average" charging current? Maybe hook up one of your multi-channel osciloscopes to see if they are doing something like this?
+Rene „cavac“ Schickbauer Thy always charge and discharge all batteries with the same current. You simply can't select different currents. This single current can be measured by the bottom shunt and controlled via the DCDC. If one battery is full while charging or empty while discharging they'll switch the corresponding bottom FET to low impeadance and by that take the battery out of the chain. And btw: yes, you can't discharge and charge at the same time.
+Rene “cavac” Schickbauer That makes a lot of sense. That would be an ingenius software hack to simplify the overall hardware design. Interesting theory.
+radoinc Well, the mainland Europe standard typically (I have some older stuff that is quite shallow) have the socket recessed enough that the pins don't make contact until the recess and body of the plug 'meet' making it impossible to get your fingers on the pins while they're in contact with the conductive parts of the socket. images.duckduckgo.com/iu/?u=http%3A%2F%2Fi00.i.aliimg.com%2Fimg%2Fpb%2F453%2F748%2F466%2F466748453_494.jpg&f=1 Plugging a 'slim' euro plug into a round socket would offer more opportunity, but then the 'upper ' part of the pins are isolated. images.duckduckgo.com/iu/?u=http%3A%2F%2Fwww.enjoykalkan.com%2Fwp-content%2Fuploads%2FEuropean_Plug.jpg&f=1 Of course there's the always the crap adapters with fully conductive pins... images.duckduckgo.com/iu/?u=http%3A%2F%2Fi.ebayimg.com%2F00%2Fs%2FNTAwWDUwMA%3D%3D%2Fz%2F6QgAAOxyUrZS34xc%2F%24_35.JPG&f=1
+SDG Electronics In argentina we use the same kind of plugs and it really is a pain in the ass. Good thing there are some of these with holes in between so you can plug the European standar
+SDG Electronics Those ears also restrict you from swaping phase with neutral, there is only one way to plug it while in Europe you can plug it inverted and there is some safety issue with that that I can't recall.
+sukhoy I haven't heard anything about any safety issues with reversability of phase. In fact, the wiring for the sockets is _explicitly_ unspecified. So you can't tell whether the left or right pin is live.
I actually bought the new varta charger after watching this video and I have to say its great. I agree that it probably reduces battery cell life but its a small price to pay for such quick charging cycles.
Why don't you make a video where you scope the charger and see what exactly it is doing. Presumably it is pulsing the 8A current rather than a continuous DC. You could also measure the terminal voltages to see how accurate the charger reading is.
maybe they discharge just only one at a time, as it's only 2A discharge rate -> 8A for .25s and 0A for .75s so it's basically a PWM with 25% duty cycle and you only need the one shunt
Several years ago there was a 15 minute charger from Rayovac that required special NiMHs. The cells contained an internal presure sensor. They were available in AA and AAA format. The charger did work with other cells, but only at a lower current.
It's 2.1mv per degree C; and it's very precise; nothing sloppy about it at all. it's based on physics. Silicon diodes make superb temperature transducers.
+EEVblog Measureing cell volatge: They'll most propably switch the DCDC source to 0A, measure the cell volage as the difference between the sense points and resume (dis)charging. By this you eleminate the contact resistance between the battery holder and the chemestry insde the cell. With crusty contacts on the battery this contact resistance could easily be 500mOhm. Plus this resistance varies with every so little vibration or if you fart in the opposite corner of the room. You simply HAVE to do it with no current flowing. Discharging: Set the output of the DCDC to -2A or whatever you want and discharge all batteries in series. Then you can measure the current over the shunt resistor at the bottom. The different mAh numbers while discharging are just because they startet at different values and then count down. If you watch closely you will see all batteries decrease at exactly the same rate i.e. are discharged with the same current.
I wonder if the charging and sensing of currents is being "time sliced" through the stack? That way you only need a single sense resistor. Disable all the batteries and enable one at a time to do the measurements, then go back to full charge. They don't need to be monitoring the current all the time, a short measurement sequence through all the batteries every time the display updates. Can you stick some scope probes across the some of the batteries? might be very illuminating..
You talked about not being able to stop the upper FET from conducting because the body diode is there, but you didn't revisit the subject after finding out the battery slots are wired in series. Obviously it can't shut off the current source if other cells are still charging, but simply turning on the lower FET prevents current from going into the corresponding cell.
A fast-charge rate is safe for NiMH, as long as temperatures are monitored and the possibility of thermal runaway is controlled. Most chargers simply limit the charge-rate to minimize the risk.
They are regular NiMH batteries, and they'll get through about 200 charge/discharge cycles before you start to notice a big decrease in capacity, at least according to my non-rigorous testing. Also, charge current is more limited in most battery chemistries for the same reason that you can discharge them so quickly. It's harder to get the ions to flow backwards than it is to get them to flow forwards. They really, really want to be where they aren't at a full charge, the only reason they can't make the jump is because they can't shed electrons until you connect a load to them. There's a correlation between charge vs. discharge current and power density, at least in modern chemistries, it's most striking in lithium ion batteries, where some of them can discharge at upwards of 100c, while the fastest charging cells I can find will take 5C at most, and that's for 200 charge/discharge cycles as opposed to over 2000 at .3C.
It's now August 2018. I checked Amazon UK where the Varta charger is sold and most of the 1-star reviews were saying Eneloops are not able to be charged in it.
25:59 “And I can see some glue on that board but, nobodys home” hahahaha that cracked me up. Great video Dave.... I learned a lot from the good detail and description of everything you showed. Im starting to do micro soldering on phones and laptops. I have a lot to learn!!
Dave, I don't know if it would break their business, but it would be great to watch an inside out kind of video where you will explain the working principle of this kind of intricate product. Adding your scope probes will just reveal some of the interesting facts and it would be learning as well as great fun to watch such video. Please give a thought.
You mentioned Eneloops. I read user feedback on another site that this charger didn't work with Eneloops and some other types of battery; it apparently displayed a fault message.
Dave, the reason you body diode doesn't conduct is the cascade arrangement. To turn off charge to the top cell in you schematic they turn off the upper fet and turn on the lower fet. The current passes through the lower fet to the remaining cells. The source voltage on the upper fet is equal to the neg battery terminal meaning the upper fet diode is reverse biased. They most likely measure the current by selecting one cell at a time for a short period and let the micro average the value for that cell then move on to the next cell. I want to know what device is dumping the heat on the discharge -- 2A x 1.2V x 4 cells is 9.6W. Didn't see a heat sink on the fets. Craig
Keeping the charger from only running one mode at a time (charge/discharge) and disallow different currents between cells, this arrangement appears good. Only one CC power source, and even just one dummy load, just switch in/out various cells based on charge state. Seems to simplify the design quite a bit.
Hey Dave! What I would love to see is your version of a "proper good" charging circuit. In particular what components would you choose, and how would you lay out the discharge sensing
At 33:55 you were asking for data. I once shorted an Eneloop with a multimeter for a few seconds. It *sparks* and puts out about 15A. Remember kids, I tried it so you don't have to. It'll ruin your batteries. I also tested some cheap noname LSD AA cell and got only 5A or so, IIRC. So, quality matters.
Just picked up this fancy charger for $24.95 from Jaycar new in the packet with 4 x AA's. Ebay was $200 aud LOL. My original Varta 15 min charger and batteries are still going strong after more than 10 years of use. Mine looks older than your silver one. Got it from Bunnings back in the day.
The key to fast charging is when do you cut back ? charging at 4C is nothing when the battery is down at 15% capacity; but for instance taking the 15 minutes literally, that's a .25 hour; you're not going to put back more than about 1.75 Amp-hrs. in the cell; because you have to back off as the charging process goes on. So you can "charge them in 15 minutes"; but not if they're completely down.
Ive got one of the old 15min chargers. And i only ever use it as a 'I need batteries charged right now' sort of thing. The rest of the time i use a big 8 bay smart charger which takes around 2h for each cell. I do a fair bit of sound recording and my recorder and radio mics etc all use AA nimh batteries. Also you can run the old one off a car too. Cars tend to be about 13.5-14v with the alternator running and at least my charger doesnt seem to care at all.
What an elegant charging circuit. I was a bit surprised on the lack of current measurement for each battery like dave ... 8A charge is insane - i would be afraid of them blowing up - I don't even charge my RC LiPo-packs at more than 5A unless i'm in a really hurry.
My uneducated guess is that it has to do with the internal resistance of the batteries. The lower the resistance the less heat charging or discharging. Those 40 milliohm of the varta look pretty good.
Interesting video. This thing seems well designed. A small correction though: sense-FETs don't have an internal current sense resistor. It's basically a current mirror. One FET cell is used for current sensing, and a few hundred identical cells in parallel make up the power FET. The app circuit showed an opamp that makes a virtual connection between the source of the single cell and the source of the multi-cell power FET, so the VGS is the same on all cells, and therefore the current should be the same in all cells.
Don't confuse yourself: Constant current source -> all akkus in series -> there is no individual current, only the series flow. With individual voltage sense this is enough to get all data. Greetings from Kirchhoff and his circuit laws:) Obviously the discharge path has to follow a different way that makes individual current measurement, discharge and switching possible.
10:45 - You can buy long security screw sets from eBay for a few quid. I bought some the other month and they're pretty good quality. They have most types and sizes of security, including: Torx hole, Hex hole, those ones (in your vid), triangle Phillips style and offset quad Phillips style.
15min charger is just for people who totally can't plan things. The 4C charge is going to destroy the batteries in a dozen cycles. It also charges them to maybe 50% capacity only. They can't absorb the energy well at such an extreme rate.
DiodeGoneWild it's a good thing my batteries do not know this! Oh, my batteries only have maye 30 cycles on them but still have more capacity than they are rated. And they run my wireless mics >6hrs independent of how they were charged. And yes, as soon as you have stage tech around, plant Ng is great but you always have to be prepared! And being able to charge in 15mins can be the bit that just saves the day (and if you are paid for your work a fast charger saves more time/mine than it costs to replace the batteries a bit earlier)
Dave, my first impression about the battery charging is the possibility of sequential charging of the batteries during a quarter of the time. Since the Logic knows which position is filled with a battery and which position is empty, the mosfets and the constant current unit are switching the current in only, when the there is a battery n that position. In all cases, the current Will be monitored by the bottom shunt as suggested by yourselve. So in all, each battery will receiver a 8 amp puls during a quarter of the time....e.g. Is 2 amps averaged over time.....greetings Walter.
I'm pretty sure they are being multiplex, switching which battery is being charged, temp sensed, and voltage checked very quickly.you'd easily be able to differentiate in software what's happening!
About life of batteries.. high current usually damages the electrodes inside, even when kept cool. With each cycle the electrodes deteriorate. For optimal performance the electrodes have a high surface area. The Varta batteries might have the same Nickel-MH chemistry as any old battery, but there might be vast differences in electrode architecture that make them suitable for rapid discharge and charge. So no, I don't believe any type of battery can be charged in this device without major effects on number of cycles
I agree with Dave, it's is a mystery how they discharge. I think the two power FETs for each slot are just used as a SPDT switch: you turn on the upper FET when there is a battery in the slot, otherwise you turn on the lower FET to bypass the empty slot. I don't think they can be using the upper FET as the discharge load, because it seems like it would get too hot. (No heat sinks and no thermal vias.) Maybe there is a load circuit that hooks up via the sense lines, but I don't see any circuit that looks like a load (no heat sinks). Could there be another small DC/DC converter on the board somewhere that draws current from the battery that is being discharged, steps up the voltage, and feeds it to the input of the main converter? That wouldn't need a heat sink, but I don't see anything that looks like it.
+Steve Robbins I wonder if they somehow dump the current through the current sensing shunt.. If you somehow short the top, current will be dumped through the sensing shunt. The circuit is too clever to dump it just through the mosfets I think.
There are some shops which mention I-C3 technology. According to wiki this is a method patented by Rayovac (related to Varta) which consists of a pressure switch built into the cell.
Looks really similar to the lacrosse BC 9009 battery chargers, they even have the same lcd readings. But this is pumped up to 8A charging with a fancy Fan.
NiMh and NiCd batteries are not like Lithium Ion cells, in that you can actually leave them trickle charging basically permanently, the cells reach a sort of chemical equilibrium and they'll just sit there happily trickle charging away constantly. Big Clive has a few videos on this.
Fantastic vid again, Loved it! :) and you're quickly becoming my favourite videos in my inbox! Thanks for sharing - once again I went away learning something...
Charger must be cycling through the batteries by turning On the low side MOSFETs during discharge and sampling on shunt resistor. Given the + of first battery can be shorted to ground with some MOSFET.
I have an Energizer charger that can put out 7.5A and charges Eneloop batteries in 15 minutes. Model CH15MN. The power brick is rated at 16V 4A output while the charging unit can take in 11-16V at 5A. The energizer batteries it came with gave up the ghost after only a dozen charges or so. But I have Eneloop batteries that have more than triple that number of charges and are still running strong.
since your using a constant current source and current is the same in all points in a series circuit, there for if your constant current source is at 8 amps and you have 2 batteries in series then you have 8 amps in passing through both batteries, cant hardly have 8 amps in one and only 4 in the other. so you don't actually need current sense on each battery. but you do need voltage sense so you know when to take the battery out or circuit. pretty cool battery charger, wonder if i can get on here in greece.
What about charging batteries in a freezer? I've always wondered if you could crank up charging even further by recharging batteries with more power in a freezer? I've never had a disposal-able freezer to experiment on, worst case failure I am thinking a person could catch fire to their freezer.
We have a rechargeable lithium-ion weed whacker by Black and Decker. Has a 15 minute charge time or at least use to. Only had it a year and the battery now discharges about 80% faster then it use to, also takes about half an hour for the "quick charger" to actually charge it up. My money is on the charger dumping twice to 3x the amps the battery is rated for into the battery and slowly kills it.
Maybe they calculate the discharge current by measuring the voltage drop on the battery, when the discharge-mosfet is turned on/off. The internal resistor of the battery is known. Am I wrong? What do you think?
I have one of these fast chargers from Energizer, must've got it around 2005 or so. I don't use it anymore, using my trusty old Rayovac charger from the 90's instead because it seems every battery I put through the Energizer charger was ruined in a very short number of charge cycles. It looked very similar to your old Varta charger (never heard of Varta outside your channel, maybe they don't sell in the US).
Charging in 15 min? Very fast compared with mine. Mine charges with 80mA and it will take 14-18 hours. It is "for Ni-Cd battery's only". Before you ask, the charger is more than 10 years old. Update: The paper with the charging time was with tape on the back, and after removing, a sticker showed "89/4". So much older than i thought.
If all four batteries are in series, then the current will be the same across all of them, so they only need one sense resistor. current is the same everywhere in a series circuit.
The title should be "How The Varta 15 Minute Battery Killer Works" Im sure that this charger cut lifetime of nimh batteries even half! EDIT: Ok, Dave mentioned about it at end...
At first glance, the totem pole arrangement on the battery charge/discharge circuit seems to be a bad ideal in case someone installs a battery backwards, as it would cook the body diodes in the MOSFETs... However, the current may be limited due to the diode voltage drop, assuming .6v/diode, the double diode drop would nearly match the 1.2v nominal cell voltage... May be an interesting follow-up...
+icesoft1 True. The battery might discharge to a point, but it won't damage anything I guess. Voltage will drop to below 1.2 relatively fast under that load.
Although we have different style sockets in the US, polarized or grounded plugs turned so it fouls with things next to it truly irritate me. And I have a bag of short 1 foot(30cm) extension leads to deal with them. Irritating.
+Marcel Timmers No, i mean the orientation of the packaging so that if you plug it into a power strip, it fouls other plugs. If you take a bench grinder to it, youll saw off half of the power supply lol. Some of the fancy power strips have rotating plugs to deal with those annoying wall wart power supplies oriented sideways.
Marcel Timmers Some are polarized yes, but on those that are not, if they are sideways, itll foul other plugs regardless of the way you plug in it, at least on power strips. Therefore I have short extensions.
those ears on the side of the plugtop PSU are meant to give spacing from adjacent PSUs to aid heat dissipation. Except that they prevent airflow between them - duh. They should be more spindly to allow airflow while keeping spacing if that's what they need. Or, just supply a higher wattage PSU like the older one that doesn't try to melt into the earth's core when used at full blast on a hot day.
I have similar 15Min but Energiser charger very similar to your first one, got it 7 yeas ago-ish, so I can tell that batteries charged on max current did not survive even 100 charges, I think something like close to 50, I still have them but capacity is crap, maximum I can use them in wireless keyboard, they lasting for bit :)
18:04 I have seen that in a computer PSU before. They placed nearly a dozen of links on the other side of the board even after putting the solder layer on the trace. And the trace was over 1 cm wide anyways, so thex made extra-extra-extra-sure with that :D
+Krisztián Szirtes Yeah, computer PSU's these days can have some truly monstrous 12v rails - up around 70 Amps available on a single rail is not uncommon on good ones. Gotta keep that resistance down way low when you're at those kinds of possible currents! Would be cool to see Dave tear down a modern high-power, high quality PSU. They pack some huge power into those things for a low price.
+Mythricia I would also like to see how much it would change if someone were to remove them. We know how the solder did on the pcb, since there was a video about that, but what about jumpers?
Krisztián Szirtes Jumpers are probably quite powerful in terms of adding more current capability. The copper on a typical PCB is flat and only ~0.035 millimeters thick, so the actual "volume" of the traces is really small. Compare that to a jumper link, much more material I think. But yeah it'd be interesting to see an actual comparison!
+Mythricia problemo is that they are getting more powerful but their size doesnt change because well its a standard it cant really change. so a 150W and 600W psu can be the same size, and that is because they pack it with alot of white glue and you cant see inside, it is going to be hard for dave to really look inside those powerful psu but as much that I like psu teardown they are all basically the same just higher frequency higher current switching transformers and mosfet for all the computer psus.
They are all connected in series so current always the same for all of them (ignoring voltage sense input impedance). The only thing they need to measure is voltage. Individual cell current measurement is not needed at all.
From my experience with a 15 min charger it works with any of the batteries for my camera and Wii units and while they charge quickly the life is cut short on the batteries.
a similar charger used to be sold by raiovac here in the usa. it gave a "shallow" charge and killed many batteries before I stopped using it. had a few leaking cells during it's life as well. not safe, not in the slightest.
Doesn't say to charge at 210mA on the battey? Also I had a Energizer 15 min charger. It worked but it killed its batteries fairly fast. They would last like 6 months with light use.
The N channel mosfet output stage is good engineering. It shows a surprising lack of "let's save five cents per unit and screw the customer"; must be a good company. (I'm not familiar with it).
I'm using these batteries for a weather station that eats them. (3xAA of those varta 2100mAh rechargeable Batteries every two weeks.) after 4 recharge cycles the varta batteries can only supply it for five days as opposed to two weeks when they were new. Meaning that in 4 cycles they've must lost about half their capacity... Bit extreme for my taste. I think I'm going to mod the weather station to run from an USB supply.
Bloody rippa! I won one in a Varta newsletter competition it seems, also have a BC1000 that loves to roast expensive batteries so could be broken, a Sanyo 2 bay charger and an Xtar 1 bay USB charger all good intelligent chargers of compact size.
+Knuckles the Echidna Check online, many people have had their LaCrosse chargers catch fire and/or melt down ... I'll stick with my old and slow but reliable eneloop charger (takes like 6 to 8 hours), thank you XD
So why do those traces with the extra solder applied on top of them have segments(You can see the bare copper in between the solder covered parts)? Seems to me that it will cause a bottleneck at the segment joint.
Are they current sensing on the PCB trace? I was struggling to find this charger in the UK but finally found it under a slightly different model number on feepay. Interestingly it is described as an 'ultra fast charger' not 15 min charger. Other makers are selling them too, Duracell for one. Be interesting to see a tear down on other makes.
could it be that they are turning all but the one discharge element they are measuring off for a very short amount of time, and determining individual cell current this way (of course periodically cycling from top to bottom)?
15 mins to charge a battery? This is madness??... no.. this.. is.. .VARTA!!!!
+Walter Steenvoorden Oh dear....
+Walter Steenvoorden With special Clip-on frame and delivered to your doorstep by GEFCO. (fancy seeing you here ;))
Wouter Weggelaar Nerds always flock together... ;)
Oh deer
+Walter Steenvoorden oh that is a good comment.
Actually the test mode charges the battery first, then it records how far it can discharge, this is the actual capacity and it is measured not calculated. Then it charges them back up for your convenience. 8:55
They can measure individual currents (using one shunt) of the batteries if they switch off discharging for all but one battery. Just do it fast for all batteries once a second etc. It's all in the firmware.
Just because you *can* charge at 4C does not mean you should :-) I try to keep my charge-rates down to 1C, even for cells rated at 10C charge rates and my reward is a much improved life (in terms of charge/discharge cycles).
+xjet If you need them in 15 minutes you should :-> But yeah, almost certainly shortening the life.
+xjet I understand that charging too slow can also do other kinds of damage, but that gets into the chemistry of the things. I do know the biggest offender is hot spotting when charging, and only when that can be addressed should the other factors with the crystallization be addressed. And even this would vary depending on the chemistry the batteries use. I like nitecore smart chargers as they have optimized charging for each type of battery chemistry, and are pretty good at auto-detecting the right chemistry.
+Richard Smith a definite issue with charging NiMH too slowly is that the -dV/dt moment will be much harder to measure because the voltage jump will be a lot less severe.
With most decent quality chargers common accepted guidelines on candlepowerdforums was 0.5C at least or the termination of the charging might not happen reliably. Obviously if you can get the noise down in your measurements and have enough resolution it will work on lower currents too.
For NiMH AA batteries a generally accepted overcharge current that isn't terrible (the heat can be handled with passive cooling) is about 0.1C. A lot of old "dumb" chargers would kinda use that as the maximum. I don't know about the consequences of long term overcharging on the chemistry but it's claimed to be relatively safe hazard wise. Otherwise those old chargers that don't turn off when forgotten would have caused more fires in the past. I'm kinda curious what the limit is where the battery can't safely get rid of the excess energy without fan based active cooing.
ToKeepTrackOfRandomSubs Ah, i see. So the issue with slow charging has more to do with charge level triggering rather than other factors? I read a paper not long ago about how many people were actually reducing the life of their batteries by going overboard and charging too _slowly_, some doing more damage to their batteries than quick charging, and although I do not exactly remember the details, i don't remember it having anything to do with technological limitations, like detection sensitivity, but rather, a limitation in the actual chemistry. It may just be my memory playing tricks on me.
That paper did go on to talk about how quick charging does have to do with technological limitations, and batteries will get better at quick charging as they get better at consistent charge distribution and consistent thermal conductivity.
Richard Smith Well there are 3 accepted methods of charging.
Method 1:
Slow charging at 0.1C until they are full (overcharging with this will cause minimal harm to the cells, it has been the traditional method).
Method 2:
-dV/dt once the battery is at the point where its full there will be a slight drop in voltage of a few mV. This effect is stronger in NiCd compared to NiMH and the higher the pulsed charging current is the more noticeable it is.
It's best to have a backup termination here as well in case it does miss that moment, both voltage and temperature ideally.
Method 3:
High current charging up to a certain cell voltage and after that point switch to a lower current trickle charge to top it off.
In the case of the low self discharge batteries my own preference goes to method 2 without any trickle charge afterwards. With older batteries and normal "high capacity" ones rather than low discharge a maintenance charge can be acceptable (they lose about 10% to self discharge on the first day and then up to 5% or so per day afterwards, both heavily dependent on ambient temperature and the battery itself).
I've got my charger set to a bit below 1C this way it picks the termination point up reliably and its still fast enough while giving good cycle lifetime without overcharging them (beyond the few moments where the -dV/dt moment triggers).
I suspect where damage from slow charging happens is if you go overboard on the slow charging and you don't know exactly if the battery is full or not yet you keep charging. The temperature won't rise significantly and nothing really extreme happens with the voltage either because not enough energy goes into it, I don't know what happens on a chemical level I'm afraid.
I love my Varta 15 minutes charger!!! Have mine for 7 years... Still works perfect!
+Infinite loop i had mine for 10 yrs. but it let the smoke out recently.
+Infinite loop How long do the batteries last?
chimp3376 thats an impossible question to answer. it depends on their capacity and the current draw.
That is of course obviously the case. To rephrase the question a bit. If you had been using your Varta 15 minute charger for the past 10 year exclusively charging it only with that device and also discharging it within it's described data sheet. How long on average did you find that the cells lasted in terms of productive usable working cycles until they were less than the 50% of people will disregard and repurchase. or How long do the batteries last.
I can think of two ways to monitor the charge for each battery:
1) Use a constant current charge circuit so all "active" batteries are being charged at the same rate, coupled with using the duty cycle of each battery being charged adjusting it's "share" of the 8A (two batteries at a time maximum).
2) Use a split charge/sense duty cycle. In the charge cycle, all (or half*) of the batteries are "on" (not shunted), but in the sense cycle, each battery is sensed using time division multiplexing (all but one battery is shunted at a time). This would also make it possible to sense the batteries' presence and adjust the duty cycle of batteries based on their voltage/temperature.
A scope could figure out if either of these are being used...
Is the big graph paper "Dave CAD Pro" ?
They're obviously not diodes as there is no polarity mark
+mikeselectricstuff And the NTC on the PCB kinda gives it away too :)
+orbiter8 (John) Yeah, don't notice these things when watching the camcorder screen and yapping.
+EEVblog.. Lol, same here Dave 😉
Hey Dave, I think You have it wrong at some point. The charging circuit has to be able to cut out a single battery. If You put in on battery with 20% charge and one with 80% charge the already full battery would blew up if the other is charged. Maybe some ''magic'' with the special type of the mosfet (using out of spec)?
+WuWo Ah, yes, good point. In that case it's easy, the bottom MOSFET turns on and the top one turns OFF. Top body diode is reversed bias so no conduction. Even cleverer than I thought.
+EEVblog ...but Dave, you will not be able to turn the top MOSFET off - at least if the compliance of your current source is high enough to overcome the forward voltage drop of the body diode, there will always be current flowing through the top n-MOSFET, independent of the applied gate voltage!
If on the other hand the compliance voltage was low enough, then the current could be switched off...
...ok, continued watching and you concluded that yoursefl - of course
+EEVblog Also, since you have a microprocessor in there, measuring the charge current on a single, common point (the single current shunt) is easy. Every second or so, switch to charging a single battery, step through all batteries and tada it's done.
As for regulating the charging current: Do you actually have to do that? The constraint seems to be temperature, so couldn't you use pulse width modulation to set an "average" charging current?
Maybe hook up one of your multi-channel osciloscopes to see if they are doing something like this?
+Rene „cavac“ Schickbauer Thy always charge and discharge all batteries with the same current. You simply can't select different currents. This single current can be measured by the bottom shunt and controlled via the DCDC. If one battery is full while charging or empty while discharging they'll switch the corresponding bottom FET to low impeadance and by that take the battery out of the chain.
And btw: yes, you can't discharge and charge at the same time.
+Rene “cavac” Schickbauer
That makes a lot of sense. That would be an ingenius software hack to simplify the overall hardware design. Interesting theory.
The "ears" are to meet UK regulations for the distance between the phase and neutral pins to the edge of the plug
+radoinc Well, the mainland Europe standard typically (I have some older stuff that is quite shallow) have the socket recessed enough that the pins don't make contact until the recess and body of the plug 'meet' making it impossible to get your fingers on the pins while they're in contact with the conductive parts of the socket.
images.duckduckgo.com/iu/?u=http%3A%2F%2Fi00.i.aliimg.com%2Fimg%2Fpb%2F453%2F748%2F466%2F466748453_494.jpg&f=1
Plugging a 'slim' euro plug into a round socket would offer more opportunity, but then the 'upper ' part of the pins are isolated.
images.duckduckgo.com/iu/?u=http%3A%2F%2Fwww.enjoykalkan.com%2Fwp-content%2Fuploads%2FEuropean_Plug.jpg&f=1
Of course there's the always the crap adapters with fully conductive pins...
images.duckduckgo.com/iu/?u=http%3A%2F%2Fi.ebayimg.com%2F00%2Fs%2FNTAwWDUwMA%3D%3D%2Fz%2F6QgAAOxyUrZS34xc%2F%24_35.JPG&f=1
+SDG Electronics In argentina we use the same kind of plugs and it really is a pain in the ass. Good thing there are some of these with holes in between so you can plug the European standar
+SDG Electronics Those ears also restrict you from swaping phase with neutral, there is only one way to plug it while in Europe you can plug it inverted and there is some safety issue with that that I can't recall.
+sukhoy I haven't heard anything about any safety issues with reversability of phase. In fact, the wiring for the sockets is _explicitly_ unspecified. So you can't tell whether the left or right pin is live.
I think here they explain it pretty well.
www.physicsforums.com/threads/what-happens-exactly-when-ac-neutral-and-active-lines-are-switched.445490/
I actually bought the new varta charger after watching this video and I have to say its great. I agree that it probably reduces battery cell life but its a small price to pay for such quick charging cycles.
Why don't you make a video where you scope the charger and see what exactly it is doing. Presumably it is pulsing the 8A current rather than a continuous DC. You could also measure the terminal voltages to see how accurate the charger reading is.
maybe they discharge just only one at a time, as it's only 2A discharge rate -> 8A for .25s and 0A for .75s
so it's basically a PWM with 25% duty cycle and you only need the one shunt
Several years ago there was a 15 minute charger from Rayovac that required special NiMHs. The cells contained an internal presure sensor. They were available in AA and AAA format. The charger did work with other cells, but only at a lower current.
It's 2.1mv per degree C; and it's very precise; nothing sloppy about it at all. it's based on physics. Silicon diodes make superb temperature transducers.
+EEVblog
Measureing cell volatge:
They'll most propably switch the DCDC source to 0A, measure the cell volage as the difference between the sense points and resume (dis)charging. By this you eleminate the contact resistance between the battery holder and the chemestry insde the cell. With crusty contacts on the battery this contact resistance could easily be 500mOhm. Plus this resistance varies with every so little vibration or if you fart in the opposite corner of the room. You simply HAVE to do it with no current flowing.
Discharging:
Set the output of the DCDC to -2A or whatever you want and discharge all batteries in series. Then you can measure the current over the shunt resistor at the bottom. The different mAh numbers while discharging are just because they startet at different values and then count down. If you watch closely you will see all batteries decrease at exactly the same rate i.e. are discharged with the same current.
I have been one of these. I've been able to bring back to life batteries with the refresh mode on numerous occassions.
I wonder if the charging and sensing of currents is being "time sliced" through the stack? That way you only need a single sense resistor. Disable all the batteries and enable one at a time to do the measurements, then go back to full charge.
They don't need to be monitoring the current all the time, a short measurement sequence through all the batteries every time the display updates.
Can you stick some scope probes across the some of the batteries? might be very illuminating..
I know a lot of li ion chargers do that... pretty much guaranteed to be using the same methodology here.
You talked about not being able to stop the upper FET from conducting because the body diode is there, but you didn't revisit the subject after finding out the battery slots are wired in series. Obviously it can't shut off the current source if other cells are still charging, but simply turning on the lower FET prevents current from going into the corresponding cell.
A fast-charge rate is safe for NiMH, as long as temperatures are monitored and the possibility of thermal runaway is controlled. Most chargers simply limit the charge-rate to minimize the risk.
They are regular NiMH batteries, and they'll get through about 200 charge/discharge cycles before you start to notice a big decrease in capacity, at least according to my non-rigorous testing. Also, charge current is more limited in most battery chemistries for the same reason that you can discharge them so quickly. It's harder to get the ions to flow backwards than it is to get them to flow forwards. They really, really want to be where they aren't at a full charge, the only reason they can't make the jump is because they can't shed electrons until you connect a load to them.
There's a correlation between charge vs. discharge current and power density, at least in modern chemistries, it's most striking in lithium ion batteries, where some of them can discharge at upwards of 100c, while the fastest charging cells I can find will take 5C at most, and that's for 200 charge/discharge cycles as opposed to over 2000 at .3C.
+Teth47 Thanks for the extra info.
EEVblog No problem! Also, greetings from Canada.
It's now August 2018. I checked Amazon UK where the Varta charger is sold and most of the 1-star reviews were saying Eneloops are not able to be charged in it.
25:59 “And I can see some glue on that board but, nobodys home” hahahaha that cracked me up. Great video Dave.... I learned a lot from the good detail and description of everything you showed. Im starting to do micro soldering on phones and laptops. I have a lot to learn!!
I remeber that my old varta 15 min charger said you could charge non varta batteries too, and it worked a charm :)
Charges all NiMh cells in 10 - 20 minutes
www.varta-consumer.ch/en/Products/Chargers-12/Power-Line/15Minute.aspx
Finally! I have been yearning for a new teardown. Thanks Dave!
Dave, I don't know if it would break their business, but it would be great to watch an inside out kind of video where you will explain the working principle of this kind of intricate product. Adding your scope probes will just reveal some of the interesting facts and it would be learning as well as great fun to watch such video. Please give a thought.
You mentioned Eneloops. I read user feedback on another site that this charger didn't work with Eneloops and some other types of battery; it apparently displayed a fault message.
Dave, the reason you body diode doesn't conduct is the cascade arrangement. To turn off charge to the top cell in you schematic they turn off the upper fet and turn on the lower fet. The current passes through the lower fet to the remaining cells. The source voltage on the upper fet is equal to the neg battery terminal meaning the upper fet diode is reverse biased.
They most likely measure the current by selecting one cell at a time for a short period and let the micro average the value for that cell then move on to the next cell. I want to know what device is dumping the heat on the discharge -- 2A x 1.2V x 4 cells is 9.6W. Didn't see a heat sink on the fets.
Craig
+Craig S Exposed pad on the underside of the FETs soldered to the board maybe? Not uncommon, but I can't say I've seen it on that type device package.
Keeping the charger from only running one mode at a time (charge/discharge) and disallow different currents between cells, this arrangement appears good. Only one CC power source, and even just one dummy load, just switch in/out various cells based on charge state. Seems to simplify the design quite a bit.
I use 15 min fast charger from energizer to charge eneloops for years, no batteries died so far.
Hey Dave!
What I would love to see is your version of a "proper good" charging circuit. In particular what components would you choose, and how would you lay out the discharge sensing
At 33:55 you were asking for data. I once shorted an Eneloop with a multimeter for a few seconds. It *sparks* and puts out about 15A. Remember kids, I tried it so you don't have to. It'll ruin your batteries. I also tested some cheap noname LSD AA cell and got only 5A or so, IIRC. So, quality matters.
Just picked up this fancy charger for $24.95 from Jaycar new in the packet with 4 x AA's. Ebay was $200 aud LOL. My original Varta 15 min charger and batteries are still going strong after more than 10 years of use. Mine looks older than your silver one. Got it from Bunnings back in the day.
I'm getting all sorts of water cooling ideas now.
The key to fast charging is when do you cut back ? charging at 4C is nothing when the battery is down at 15% capacity; but for instance taking the 15 minutes literally, that's a .25 hour; you're not going to put back more than about 1.75 Amp-hrs. in the cell; because you have to back off as the charging process goes on. So you can "charge them in 15 minutes"; but not if they're completely down.
Ive got one of the old 15min chargers. And i only ever use it as a 'I need batteries charged right now' sort of thing. The rest of the time i use a big 8 bay smart charger which takes around 2h for each cell.
I do a fair bit of sound recording and my recorder and radio mics etc all use AA nimh batteries.
Also you can run the old one off a car too. Cars tend to be about 13.5-14v with the alternator running and at least my charger doesnt seem to care at all.
What an elegant charging circuit. I was a bit surprised on the lack of current measurement for each battery like dave ... 8A charge is insane - i would be afraid of them blowing up - I don't even charge my RC LiPo-packs at more than 5A unless i'm in a really hurry.
My uneducated guess is that it has to do with the internal resistance of the batteries. The lower the resistance the less heat charging or discharging.
Those 40 milliohm of the varta look pretty good.
Interesting video. This thing seems well designed.
A small correction though: sense-FETs don't have an internal current sense resistor. It's basically a current mirror. One FET cell is used for current sensing, and a few hundred identical cells in parallel make up the power FET. The app circuit showed an opamp that makes a virtual connection between the source of the single cell and the source of the multi-cell power FET, so the VGS is the same on all cells, and therefore the current should be the same in all cells.
+Steve Robbins Yes, I showed the internal equivalent circuit in the video.
My vehicle charger spoilt recently. Hope to see a teardown for CTEK and Deltran Bettery Tender. Thank you.
Don't confuse yourself: Constant current source -> all akkus in series -> there is no individual current, only the series flow. With individual voltage sense this is enough to get all data.
Greetings from Kirchhoff and his circuit laws:)
Obviously the discharge path has to follow a different way that makes individual current measurement, discharge and switching possible.
Interesting that the battery icon in the display is so similar to La Crosse Technology chargers.
10:45 - You can buy long security screw sets from eBay for a few quid. I bought some the other month and they're pretty good quality. They have most types and sizes of security, including: Torx hole, Hex hole, those ones (in your vid), triangle Phillips style and offset quad Phillips style.
The big ear beside the transformer is for safety purpose, this is to aviod big hand guy to plug in while the main socket power is on.
15min charger is just for people who totally can't plan things. The 4C charge is going to destroy the batteries in a dozen cycles. It also charges them to maybe 50% capacity only. They can't absorb the energy well at such an extreme rate.
DiodeGoneWild it's a good thing my batteries do not know this! Oh, my batteries only have maye 30 cycles on them but still have more capacity than they are rated.
And they run my wireless mics >6hrs independent of how they were charged.
And yes, as soon as you have stage tech around, plant Ng is great but you always have to be prepared! And being able to charge in 15mins can be the bit that just saves the day (and if you are paid for your work a fast charger saves more time/mine than it costs to replace the batteries a bit earlier)
Put the 4C object in the freezer and relax. 15 min and dinner is cooked! Anyone tried if this works at cardiac arrest?
Great video Dave, some very useful info for us newbies and of course everyone else I expect
Use white heat sink compound on faded semiconductor number to partially restore enough to read.
Dave, my first impression about the battery charging is the possibility of sequential charging of the batteries during a quarter of the time. Since the Logic knows which position is filled with a battery and which position is empty, the mosfets and the constant current unit are switching the current in only, when the there is a battery n that position. In all cases, the current Will be monitored by the bottom shunt as suggested by yourselve. So in all, each battery will receiver a 8 amp puls during a quarter of the time....e.g. Is 2 amps averaged over time.....greetings Walter.
I'm pretty sure they are being multiplex, switching which battery is being charged, temp sensed, and voltage checked very quickly.you'd easily be able to differentiate in software what's happening!
love the vids you do from the UK, i will send you a divice this month for your eevblog...
About life of batteries.. high current usually damages the electrodes inside, even when kept cool. With each cycle the electrodes deteriorate. For optimal performance the electrodes have a high surface area. The Varta batteries might have the same Nickel-MH chemistry as any old battery, but there might be vast differences in electrode architecture that make them suitable for rapid discharge and charge. So no, I don't believe any type of battery can be charged in this device without major effects on number of cycles
I agree with Dave, it's is a mystery how they discharge. I think the two power FETs for each slot are just used as a SPDT switch: you turn on the upper FET when there is a battery in the slot, otherwise you turn on the lower FET to bypass the empty slot.
I don't think they can be using the upper FET as the discharge load, because it seems like it would get too hot. (No heat sinks and no thermal vias.) Maybe there is a load circuit that hooks up via the sense lines, but I don't see any circuit that looks like a load (no heat sinks). Could there be another small DC/DC converter on the board somewhere that draws current from the battery that is being discharged, steps up the voltage, and feeds it to the input of the main converter? That wouldn't need a heat sink, but I don't see anything that looks like it.
+Steve Robbins I wonder if they somehow dump the current through the current sensing shunt.. If you somehow short the top, current will be dumped through the sensing shunt.
The circuit is too clever to dump it just through the mosfets I think.
There are some shops which mention I-C3 technology. According to wiki this is a method patented by Rayovac (related to Varta) which consists of a pressure switch built into the cell.
Looks really similar to the lacrosse BC 9009 battery chargers, they even have the same lcd readings. But this is pumped up to 8A charging with a fancy Fan.
NiMh and NiCd batteries are not like Lithium Ion cells, in that you can actually leave them trickle charging basically permanently, the cells reach a sort of chemical equilibrium and they'll just sit there happily trickle charging away constantly. Big Clive has a few videos on this.
I had a problem with the older versions of these batteries, as they aged they started going open circuit as they warmed up.
Fantastic vid again, Loved it! :) and you're quickly becoming my favourite videos in my inbox! Thanks for sharing - once again I went away learning something...
I'm just reading: "Eneloop Plus cells have a PTC thermistor built-in that cuts the power when the batteries are overheating" (Wikipedia).
Charger must be cycling through the batteries by turning On the low side MOSFETs during discharge and sampling on shunt resistor. Given the + of first battery can be shorted to ground with some MOSFET.
I have an Energizer charger that can put out 7.5A and charges Eneloop batteries in 15 minutes. Model CH15MN. The power brick is rated at 16V 4A output while the charging unit can take in 11-16V at 5A.
The energizer batteries it came with gave up the ghost after only a dozen charges or so. But I have Eneloop batteries that have more than triple that number of charges and are still running strong.
These videos are fantastic. Keep up the good work.
dave use a voltimeter to check out the voltages on the batteries during charge
since your using a constant current source and current is the same in all points in a series circuit, there for if your constant current source is at 8 amps and you have 2 batteries in series then you have 8 amps in passing through both batteries, cant hardly have 8 amps in one and only 4 in the other. so you don't actually need current sense on each battery. but you do need voltage sense so you know when to take the battery out or circuit. pretty cool battery charger, wonder if i can get on here in greece.
What about charging batteries in a freezer? I've always wondered if you could crank up charging even further by recharging batteries with more power in a freezer?
I've never had a disposal-able freezer to experiment on, worst case failure I am thinking a person could catch fire to their freezer.
We have a rechargeable lithium-ion weed whacker by Black and Decker. Has a 15 minute charge time or at least use to. Only had it a year and the battery now discharges about 80% faster then it use to, also takes about half an hour for the "quick charger" to actually charge it up. My money is on the charger dumping twice to 3x the amps the battery is rated for into the battery and slowly kills it.
Always get a chuckle when seeing the FLIR E8 and knowing my E4 thinks it's an E8.
+BlownF150 Sweet!
Details please?
+BlownF150 Did you override the firmware or what?
Yup, downloaded a little hack that unlocks an E4 to have the resolution and features of an E8.
+BlownF150 Nice!
Maybe they calculate the discharge current by measuring the voltage drop on the battery, when the discharge-mosfet is turned on/off. The internal resistor of the battery is known. Am I wrong? What do you think?
I have one of these fast chargers from Energizer, must've got it around 2005 or so. I don't use it anymore, using my trusty old Rayovac charger from the 90's instead because it seems every battery I put through the Energizer charger was ruined in a very short number of charge cycles. It looked very similar to your old Varta charger (never heard of Varta outside your channel, maybe they don't sell in the US).
+CalcProgrammer1 Both Varta and Rayovac are owned by Spectrum Brands
Charging in 15 min? Very fast compared with mine. Mine charges with 80mA and it will take 14-18 hours. It is "for Ni-Cd battery's only".
Before you ask, the charger is more than 10 years old.
Update: The paper with the charging time was with tape on the back, and after removing, a sticker showed "89/4". So much older than i thought.
If all four batteries are in series, then the current will be the same across all of them, so they only need one sense resistor. current is the same everywhere in a series circuit.
Please use the thermal cam at least once on each video. thank you.
Had one in 2005 for servos to my hpi savage truck. Worked great, but none rechargeble had more volt and servos was faster and stronger
4C ... I feel like you have to pretty confident to design to charge at that rate.
The battery symbol and the charge modes are similar to the BC-700 charger
The title should be "How The Varta 15 Minute Battery Killer Works"
Im sure that this charger cut lifetime of nimh batteries even half!
EDIT: Ok, Dave mentioned about it at end...
the refresh mode will extend the life of them exp if you half charge them alot
I've got an even older model of that. Works great!
At first glance, the totem pole arrangement on the battery charge/discharge circuit seems to be a bad ideal in case someone installs a battery backwards, as it would cook the body diodes in the MOSFETs... However, the current may be limited due to the diode voltage drop, assuming .6v/diode, the double diode drop would nearly match the 1.2v nominal cell voltage... May be an interesting follow-up...
+icesoft1 True. The battery might discharge to a point, but it won't damage anything I guess. Voltage will drop to below 1.2 relatively fast under that load.
This has several things in common with the LaCrosse BC-900. I wonder if there's any connection.
Although we have different style sockets in the US, polarized or grounded plugs turned so it fouls with things next to it truly irritate me. And I have a bag of short 1 foot(30cm) extension leads to deal with them. Irritating.
+Richard Smith Hi mate. Don't you have a bench grinder?????
+Marcel Timmers No, i mean the orientation of the packaging so that if you plug it into a power strip, it fouls other plugs. If you take a bench grinder to it, youll saw off half of the power supply lol.
Some of the fancy power strips have rotating plugs to deal with those annoying wall wart power supplies oriented sideways.
My bad, I thought that you meant the plugs them self. As I understand it, you now have one wider contact-er so you can only stick the plug in one way.
+Richard Smith I agree, I hate it when they face the big part of the plug sideways, it's especially annoying on power bars.
Marcel Timmers Some are polarized yes, but on those that are not, if they are sideways, itll foul other plugs regardless of the way you plug in it, at least on power strips. Therefore I have short extensions.
You should have a look at the Maha MH-C9000.
NiMH batteries can be rejuvinated by draining them completely(like 0v) and then recharging them.
those ears on the side of the plugtop PSU are meant to give spacing from adjacent PSUs to aid heat dissipation. Except that they prevent airflow between them - duh.
They should be more spindly to allow airflow while keeping spacing if that's what they need.
Or, just supply a higher wattage PSU like the older one that doesn't try to melt into the earth's core when used at full blast on a hot day.
I have similar 15Min but Energiser charger very similar to your first one, got it 7 yeas ago-ish, so I can tell that batteries charged on max current did not survive even 100 charges, I think something like close to 50, I still have them but capacity is crap, maximum I can use them in wireless keyboard, they lasting for bit :)
18:04 I have seen that in a computer PSU before. They placed nearly a dozen of links on the other side of the board even after putting the solder layer on the trace. And the trace was over 1 cm wide anyways, so thex made extra-extra-extra-sure with that :D
+Krisztián Szirtes Yeah, computer PSU's these days can have some truly monstrous 12v rails - up around 70 Amps available on a single rail is not uncommon on good ones. Gotta keep that resistance down way low when you're at those kinds of possible currents!
Would be cool to see Dave tear down a modern high-power, high quality PSU. They pack some huge power into those things for a low price.
+Mythricia I would also like to see how much it would change if someone were to remove them. We know how the solder did on the pcb, since there was a video about that, but what about jumpers?
Krisztián Szirtes Jumpers are probably quite powerful in terms of adding more current capability. The copper on a typical PCB is flat and only ~0.035 millimeters thick, so the actual "volume" of the traces is really small. Compare that to a jumper link, much more material I think.
But yeah it'd be interesting to see an actual comparison!
+Mythricia
problemo is that they are getting more powerful but their size doesnt change because well its a standard it cant really change.
so a 150W and 600W psu can be the same size, and that is because they pack it with alot of white glue and you cant see inside, it is going to be hard for dave to really look inside those powerful psu but as much that I like psu teardown they are all basically the same just higher frequency higher current switching transformers and mosfet for all the computer psus.
i have not seen a 494 used in a switching supply in awhile!
charging Nickel Metal Hydride at 4C is perfectly legitimate. It will not reduce the life of the cell. It looks like an excellent piece of equipment.
They are all connected in series so current always the same for all of them (ignoring voltage sense input impedance). The only thing they need to measure is voltage. Individual cell current measurement is not needed at all.
From my experience with a 15 min charger it works with any of the batteries for my camera and Wii units and while they charge quickly the life is cut short on the batteries.
a similar charger used to be sold by raiovac here in the usa. it gave a "shallow" charge and killed many batteries before I stopped using it. had a few leaking cells during it's life as well. not safe, not in the slightest.
Doesn't say to charge at 210mA on the battey?
Also I had a Energizer 15 min charger. It worked but it killed its batteries fairly fast. They would last like 6 months with light use.
The N channel mosfet output stage is good engineering. It shows a surprising lack of "let's save five cents per unit and screw the customer"; must be a good company. (I'm not familiar with it).
I'm using these batteries for a weather station that eats them. (3xAA of those varta 2100mAh rechargeable Batteries every two weeks.) after 4 recharge cycles the varta batteries can only supply it for five days as opposed to two weeks when they were new. Meaning that in 4 cycles they've must lost about half their capacity... Bit extreme for my taste. I think I'm going to mod the weather station to run from an USB supply.
I meant to say 7 cycles
Don't you think about that we can do current measure by disconnecting other charge channels at a time of the measure lasts
Could you supply one slot in discharge mode with your constant current supply? Just to check accuracy of the display...
The charger is cpu controlled all of the 15 minute ones are to manage charge rate temperature etc.
Bloody rippa! I won one in a Varta newsletter competition it seems, also have a BC1000 that loves to roast expensive batteries so could be broken, a Sanyo 2 bay charger and an Xtar 1 bay USB charger all good intelligent chargers of compact size.
+Knuckles the Echidna
Check online, many people have had their LaCrosse chargers catch fire and/or melt down ... I'll stick with my old and slow but reliable eneloop charger (takes like 6 to 8 hours), thank you XD
So why do those traces with the extra solder applied on top of them have segments(You can see the bare copper in between the solder covered parts)? Seems to me that it will cause a bottleneck at the segment joint.
Are they current sensing on the PCB trace? I was struggling to find this charger in the UK but finally found it under a slightly different model number on feepay. Interestingly it is described as an 'ultra fast charger' not 15 min charger. Other makers are selling them too, Duracell for one. Be interesting to see a tear down on other makes.
could it be that they are turning all but the one discharge element they are measuring off for a very short amount of time, and determining individual cell current this way (of course periodically cycling from top to bottom)?