EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: cigmas on October 08, 2020, 12:01:24 am
-
The manual:
(https://i.imgur.com/dpzX2Lp.png)
The circuit board:
(https://i.imgur.com/fBVQ6Bf.jpg)
The circuit diagram:
(https://i.imgur.com/HGUIXQH.png)
Am I missing something, did I misread the circuit, or is there nothing that prevents overcharging? If not, is there a simple way to add overcharging protection?
This is a Remington trimmer, model BHT-2000. I opened it up to replace the 2X 730mAH NiMH batteries which don't hold much charge anymore after over a decade. The charger is rated at 5.3V DC 140mA output.
Also, regarding #7, what inhibits it from being used with the cord? I can confirm that plugging it in when the battery is dead does not let it run much at all. Is the power supply just not providing enough current to charge the batteries and run the motor? Any way to adapt it for corded use as well?
I'm guessing the answer is no on both fronts, but worth a shot right?
-
You said the power brick output is 5.3VDC @ 140mA...
Looks to my rather inexperienced eyes like it is float-charging your battery whenever it is plugged in, and it is relying on the low current (140mA) not to be able to over charge (at least not to overcharge by much and causes the battery to heat up).
Constant float charging an NiMH at such low current should not do much damage. Not a very good practice in my view but it is not the first time I see that. (I saw that "technique" on a cheap car jumper helper SLA from Walmart before. That cheapie was just that, a cheapie.) Since your old battery lasted a decade, overcharge doesn't seem to be a huge problem here.
As to #7: Assuming your battery is fully charged and plugged in. With just 140mA coming in, it is probably not enough to run the trimmer. So the battery will start draining and begin to want to recharge as well. The power source now will be forced to overdraw to supply the both to run the trimmer and refill the battery at the same time. It will either not run at all; or if it does, it will probably run hot due to over-load, die, or something like that. That is why #7 is there - it doesn't have the incoming power needed to run and recharge at the same time. It can't use a more powerful source - with that design, a power-brick with more power will overcharge and cook your batteries.
As to your question: Naahh... No easy way to add overcharge protection here. There are ways, but not easy. Since it looks like this PCB is also the battery holder? You can always redesign the PCB, but it is not "adding" to the existing circuit and surely is not going to be easy. Best overcharge protection you can do is to unplug it when it is fully charged.
EDIT: expanded the reply regarding your question to point #7
-
Ah, that's a plausible explanation of the design.
I don't keep it plugged in, charging only when the battery goes very weak. Though, sometimes it is left charging for 12 hours or more.
I usually charged it roughly once a week, maybe less often when it was newer, although lately it was dying quickly and noticeably weaker than normal even when charged. Over a decade (more like 12 years actually) that's on the order of 500 cycles which matches the expected life cycle.
And you're right there's no room for adding PCB or even many extra components. That PCB is also the "battery holder", i.e. the batteries are connected underneath it, and the whole package slips very tightly into an ABS plastic case that clips onto the other half of the plastic case seen in the pic above, which houses the motor. The rubber o-ring for the waterproofing can also be seen. All of that then gets secured inside the trimmer's outer shell.
I wanted to add an actual battery holder to avoid soldering onto batteries, but there's no room -- hardly 1mm around the batteries.
Thanks for the answer. I won't try any mods and just content myself with the new 900 mAh LADDAs I installed in there. Haven't used it yet, but it's running noticeably stronger now.
-
If you want to reduce the charging current a bit, then you could add an extra diode in series with D1.
-
If you want to reduce the charging current a bit, then you could add an extra diode in series with D1.
NO
the battery will never fully charge with extra regular diode
and that accu needs complete charge/discharge routine for long life expectancy
either replace the whole 'charger' with a real one or just leave it like this
-
It will work: the charging will take longer, and the trickle charge
will be lower (which is better for the cells). As for the "complete
charge/discharge", that was true for NiCd, not so much for NiMh.
Edit:
It should be fine without the extra diode: That looks like a 42ohm
resistor and assuming a 3.1V fully charged battery, the trickle
charge is ~35mA, which is fine.
-
It’s just traditional old-school trickle (or float) charging.
Most NiMH batteries really hate that, but they make special ones (for things like cordless phones) specifically designed to tolerate constant trickle charging.
At 10 years, the battery has simply reached the end of its lifespan.
When replacing the batteries, be sure to get the special ones for trickle charging, as normal ones will not last as long in this application.
-
An addition to my post above:
If it's not a regulated 5.3V supply, but rather a
transformer with rectifier and capacitor, then the
voltage and current will be higher. Maybe higher
than the recommended 0.05C trickle charge.
-
In 2008? Highly unlikely, as SMPS wall warts were already fully established as the standard by then.
-
Adding an extra diode in series to reduce charging current is a bad idea.
If you want to reduce charging current, then replace R1.
If you want to go fancy, maybe you can glue on some uC "dead bug style. TQFP packages are about 1mm thick. Together with a MOSfet you can make a charge timer, blink the LED when battery is full, blink it when the battery gets empty etc.
But is it worth it?
About half a year ago I bought a no-name trimmer for around EUR20, that none-the-less looked reasonably decent. It has a Li-Ion battery inside and enough chicken fodder to be a proper charging circuit. It also does all the blinking LED stuff so I assume there is a uC inside also.
Just a few weeks ago I had to recharge the battery. Had to go searching for the charger as it's a 5V thing with a round plug. Maybe next time I'll solder in an USB plug.
-
IMO either leave it as-is and hopefully get another decade from new batteries or remove all the silly circuitry and use some external smart charger connected to the same jack.
-
Even with a good charger you can't expect much more then 10 years out of some battery pack.
These things have a shelf life too on top of the charge cycle rating.
-
Today I learned "dead bug style" is a technique of putting chips upside down on a PCB! hahaha.. In this case, I don't think there's even room for that unless it's a very small chip with tightly trimmed leads.
If I can get another decade out of the replacement batteries with the existing circuitry, looks like that's the best and simplest outcome. I can't hope for more life than the cells are rated for right?!
When replacing the batteries, be sure to get the special ones for trickle charging, as normal ones will not last as long in this application.
What is this type called? I just replaced them with 900 mAh LADDAs. My main worry was not to damage them while soldering the contacts, since I don't have a spot welder.
-
I’ve mostly seen them sold explicitly as cordless (DECT) phone batteries. For example:
https://www.panasonicbatteryproducts.com/dect-phone-batteries/ (https://www.panasonicbatteryproducts.com/dect-phone-batteries/)
https://www.varta-consumer.com/en/products/rechargeables/overview/device-oriented-specials/recharge-accu-phone-aaa-800-mah (https://www.varta-consumer.com/en/products/rechargeables/overview/device-oriented-specials/recharge-accu-phone-aaa-800-mah)
I haven’t found a really good generic term for them. The closest might be “high durability” type, as opposed to the “high capacity”, “high current”, and “low self discharge” characteristics of an Eneloop (which a white LADDA almost certainly is). In essence, the higher capacities and currents come at the expense of durability (service life/charge cycles). So a “phone” NiMH battery should not be the highest capacity you can find.
-
I wouldn't spend too much time worrying about it though, a single AA cell is cheap enough just put in whatever you can find and see how it holds up. If you're not leaving it on the charger all the time anyway then it's not really an issue.
-
@tooki Thanks for pointing those out. I figured most batteries named for an application are just marketing inventions to charge more.
@james_s Yes, for under $5 for 2X AAAs, my problem is solved. Correct, I don't leave it charging all the time, though when I do charge it, it might stay on more than 6 hours. There is no "charge complete" indicator.
It would have been nice to add the ability for corded usage, but oh well. Haircuts will just require some planning to have it fully charged.
Now the next decade-long test begins. Stay tuned, but don't hold your breath. :)
-
If you really wanted to get fancy you could simply bypass the internal circuitry and build an external smart charger for the cell to replace the existing charger cord. I don't think it's worth it though unless you're looking for a project to challenge yourself with.
-
Yeah, I don't think it's worth it either, especially for something I use daily and don't want to mess up.
-
NiMH actually is rather tricky to charge. From my reading of various NiMH characteristics, what I gathered is the best charge-full detection for NiMH is "NDV" - negative delta voltage. So when you detect the voltage goes from positive delta (voltage going up) to suddenly but brief and small negative delta (voltage going down), that NDV detection tells you the charge is done/full. At that point, the power going in becomes room-warmer.
To detect that brief and small NDV is not easy. Many commercially available consumer chargers just use a timer and call it a day. Some would have temperature detection to limit over-charge. Your problem is easily solve with an alarm clock - set a 12 hours charge time (140mA current for a 2x700mAH battery in series, perfect efficiency would be 700/140 hours charge time. Figure double that for imperfection in efficiency, 10 hour charge time. 12 hours will give it a bit of extra margin.)
If you must do something... I would think converting it to easy-to-open so battery can be swap without much hoopla would be best. So, you can get a good quality NiMH charger and have a pair of spare LSD batteries on standby. When it goes hungry, swap it out and swap in the pair of fully-charged LSD's in waiting.
If you are a fun seeker and wants a challenge... Test and see if LiIon (4.2Volt) would work - my guess is, it probably will. If it does, rework the PCB and battery holder with a TP4056, with a micro USB port. Or may be just get one of those TP4056 charge board, figure out how to mount it with the micro USB port exposed and visibility for the indicator LEDs... Of course, you need to shop for a LiIon cell that will fit in the remaining space. That would be kind of fun to do.
But with a LiIon cell, your trimmer will run at super-speed, cut carelessly, a hair trim will turn your hair into a Captain Picard style shiny head hair cut... Also, you are over-driving the motor comparing to designed condition, so, fun would come at a cost of possibly shorter life expectancy for your trimmer... Should probably use the TP4056's temperature sense pin to limit the risk, along with a 2A fuse in-line. Don't use RC LiPoly-soft packet type cells wthout fuse -- they can dump a huge load of current. You'll need to do your hair trimming/charging with a fire extinguisher near by... Better add this warning: do it at your own risk - possibility of adding an unknown level of risk of causing head injuries and possibly a fire.
-
I read about chargers before getting my simple but effective Panasonic BQ-CC17 smart charger. I'm not going to try to build one myself. Plus it seems any decent mod will involve losing the water resistance, which I rather appreciate for ease of cleaning.
What is typical for charging efficiency? I'm guessing the 900 mAh ( / 140 mA = 6.4 hours) batteries won't be much different than the original 700 mAh, and/or I'll notice the difference if it's not fully charged, and/or it doesn't matter if it's not at 100%.
If you are a fun seeker and wants a challenge... Test and see if LiIon (4.2Volt) would work - my guess is, it probably will. If it does, rework the PCB and battery holder with a TP4056, with a micro USB port.
Now that would be nice, since it is what I was looking for in possible replacements, but didn't find anything that satisfied my whole wishlist. But again since it would mean losing water resistance I'll pass... unless I find a way to expand the plastic case. Or you know, just keep going like the past decade with fresh batteries.
-
It will work: the charging will take longer, and the trickle charge
will be lower (which is better for the cells). As for the "complete
charge/discharge", that was true for NiCd, not so much for NiMh.
Edit:
It should be fine without the extra diode: That looks like a 42ohm
resistor and assuming a 3.1V fully charged battery, the trickle
charge is ~35mA, which is fine.
no, it won't.the diodes are kinda 'opened' with like 0.5-0.6v IF they have current flow, when you approach full charge this will never happen.
besides talking, TEST yourself and put the result, then we can talk seriously. what trickle charge when you have 2 closed diodes :)
you're saying the engineer that made the schematic is a stupid entity and you have a better aproach, have you considered TESTING first and talking later?
-
@tooki Thanks for pointing those out. I figured most batteries named for an application are just marketing inventions to charge more.
A sensible “default” reaction given how often that is exactly the case. (A great example of that in USA is generic diphenhydramine. I literally found it twice on opposite sides of the aisle in a pharmacy: on the cold remedy side it was sold as generic Benadryl. On the opposite side, in the sleep aids, as generic No-Doze Unisom. Same active ingredient, same dosage, same appearance, same inactive ingredients, same hours, but the ones labeled as “sleep aid” cost almost twice as much as the decongestant ones!) Or how identical razor blades are sold in pink for women. (Sure, the handles are different for very sensible reasons, but the blades are often identical, just cost more in pink.)
Anyhow, in electronic components, there often exist many, many, many models of the same component, with ever so slight optimizations for one thing or another. In many applications it makes absolutely no difference. But in some applications, it can have some advantage (or cost savings) that makes it worthwhile for the manufacturer to make various versions, since they’ll sell billions of each anyway. This is how we end up with a single manufacturer having two dozen lines of, say, resistors, all with identical key specs. Only a fine-toothed combing of the data sheets, application notes, and catalogs might provide insight into why one might use one over another. Often, that information is squirreled away at the company, really only coming out when you inquire about what’s best for your specific design.
As you learn more about electronics, some of the differences will become clear. I learned a lot about rechargeable batteries, for example, this year, when working on a project (a Bluetooth speaker kit for education, which first year apprentices will assemble as practice for various assembly techniques) that needed to be rechargeable.
-
no, it won't.the diodes are kinda 'opened' with like 0.5-0.6v IF they have current flow, when you approach full charge this will never happen.
The result of this very simple and inexpensive charger,
is that charging never stops completely. (And this is
possible because NiMH can tolerate a low tricke charge,
whereas e.g. Li-Ion cannot.)
-
A sensible “default” reaction given how often that is exactly the case.
As another example, how about this "special" battery for some devices which apparently have a contact touching the exposed lower side of the cell, so that regular batteries with a full wrap won't work: https://www.amazon.com/Philips-LFH9154-00-Rechargeable-Batteries/dp/B001AM2ESI/ (https://www.amazon.com/Philips-LFH9154-00-Rechargeable-Batteries/dp/B001AM2ESI/)
(https://images-na.ssl-images-amazon.com/images/I/61I4wzzEIcL._AC_SL1320_.jpg)
I learned a lot about rechargeable batteries, for example, this year, when working on a project (a Bluetooth speaker kit for education, which first year apprentices will assemble as practice for various assembly techniques) that needed to be rechargeable.
Oh, I'm currently trying to repair a small Bluetooth speaker that has lost its voice, and my next thread is likely to be about it! EDIT: Here it is: https://www.eevblog.com/forum/repair/help-a-(cute)-tongue-tied-bluetooth-speaker-recover-its-voice/ (https://www.eevblog.com/forum/repair/help-a-(cute)-tongue-tied-bluetooth-speaker-recover-its-voice/)
-
A sensible “default” reaction given how often that is exactly the case.
As another example, how about this "special" battery for some devices which apparently have a contact touching the exposed lower side of the cell, so that regular batteries with a full wrap won't work: https://www.amazon.com/Philips-LFH9154-00-Rechargeable-Batteries/dp/B001AM2ESI/ (https://www.amazon.com/Philips-LFH9154-00-Rechargeable-Batteries/dp/B001AM2ESI/)
(https://images-na.ssl-images-amazon.com/images/I/61I4wzzEIcL._AC_SL1320_.jpg)
Honestly, that’s an ingenious solution to the problem of allowing both disposable and rechargeable batteries in a device capable of in-device charging. That way you won’t accidentally charge disposable batteries, but easily detect and charge rechargeables.
-
A sensible “default” reaction given how often that is exactly the case. (A great example of that in USA is generic diphenhydramine. I literally found it twice on opposite sides of the aisle in a pharmacy: on the cold remedy side it was sold as generic Benadryl. On the opposite side, in the sleep aids, as generic No-Doze. Same active ingredient, same dosage, same appearance, same inactive ingredients, same hours, but the ones labeled as “sleep aid” cost almost twice as much as the decongestant ones!) Or how identical razor blades are sold in pink for women. (Sure, the handles are different for very sensible reasons, but the blades are often identical, just cost more in pink.)
Isn't "No-doze" a caffeine product meant to wake you up? Prior to more modern allergy medications coming out I used to take Benadryl and it certainly did not keep me awake, as the name "No-doze" would imply.
-
...
What is typical for charging efficiency? I'm guessing the 900 mAh ( / 140 mA = 6.4 hours) batteries won't be much different than the original 700 mAh, and/or I'll notice the difference if it's not fully charged, and/or it doesn't matter if it's not at 100%.
...
You have two factors affecting efficiency here (after the power brick). First is the electronics, second is the battery chemistry.
More important is, your question is not really about efficiency but about how fast you can actually stuff power into the battery from your 140mA power brick...
In the way things are connected in this case, the electronics would drop the voltage a bit, but what it drops to is largely not a useful piece of information (in answering how long it takes to charge). If you can pull the battery's voltage higher (but within acceptable range), you can indeed pump more current in, but you can't change that in your existing setup. (You could replace the inbound diode with one that has a lower Vf, but it will over charge more than existing setup).
The battery needs to convert the electrical energy to chemical energy. The energy conversion is not perfect and it varies by different battery-chemistry and varies as the battery ages and environmental factor like room temperature. From my not very extensive experiences, I have seen around 1/4 to 1/3 energy lost in that conversion for batteries in good condition. For batteries beyond its age, I have seen it way below 1/2.
With that background above, now you need to think about the three stages of charging:
Phase 1 - the battery is very empty thus it can take the whole 140mA coming in.
Phase 2 - the battery is getting full, so it is allowing less than 140mA to come in.
Phase 3 - the battery full and it is soaking - floating.
Note the choice of word I used in phase 2, your battery is not allowing the full 140mA to come in. As said, you can always increase the voltage to force it to some extend, but the important point is, you don't know how much current it is drawing at a particular time. So you can only evaluate the ideal case in Phase 1 and estimate the rest. The older your battery, the sooner you get to phase 2 and begin taking less current, and the older the battery, the less the energy conversion efficiency.
With 2x(900mAH 1.2V) battery, assuming they are indeed in series, it is a 900mAH battery at 2*1.2V=2.4V. If it is parallel, it would be like a 1800mAH at 1.2V. The picture you attached doesn't show definitively series or parallel since I can't see the other side of the PCB. But you can DMM measure the BT+ to BT- voltage when fully charge. I will continue to assume it is serial since it seem more reasonable in that setup.
So, if your battery stays in Phase 1 nearly all the time (best case)...
Hours = 900mAH/140mA
But since it will move from Phase 1 to Phase 2 and you don't know for sure when, I'd say use double the time calculated above for an approximation - that doubling takes care of Phase 2 and conversion efficiency, in approximation.
Side note:
With the existing PCB , you can actually measure the current the battery is taking by measuring the voltage of R1 and R2 simultaneously. You know the value of the resisters, you can measure the voltage and calculate using ohms law I=V/R
So, the current flowing through the battery is:
I(total) = I1 + I2
I(total) = V1/R1 + V2/R2
If you have only one DMM, you can measure both voltages in quick succession.
To monitor charge progress, measure R1 voltage. During phase 1, R1 voltage will stay relatively constant at max current, once it can no longer take the max current (phase 2) it will drop progressively. When full (phase 3, floating), R1 voltage will again stay relatively constant but at a much lower current.
EDIT: added bold and minor rewording to avoid confusion
-
@Rick Law Thanks again for the detailed explanation. The next time I open it up, I'll measure those voltages and resistances. Would be fun to know.
For now I'll go with your reasonable estimate of 2 X (capacity / power brick current) ~ 12 hours charging time. And yes, the batteries are in series.
-
A sensible “default” reaction given how often that is exactly the case. (A great example of that in USA is generic diphenhydramine. I literally found it twice on opposite sides of the aisle in a pharmacy: on the cold remedy side it was sold as generic Benadryl. On the opposite side, in the sleep aids, as generic No-Doze. Same active ingredient, same dosage, same appearance, same inactive ingredients, same hours, but the ones labeled as “sleep aid” cost almost twice as much as the decongestant ones!) Or how identical razor blades are sold in pink for women. (Sure, the handles are different for very sensible reasons, but the blades are often identical, just cost more in pink.)
Isn't "No-doze" a caffeine product meant to wake you up? Prior to more modern allergy medications coming out I used to take Benadryl and it certainly did not keep me awake, as the name "No-doze" would imply.
Ack, yes, of course! Complete and utter brain fart on the name brand’s name there! Googling, it was probably Unisom (not all their products are diphenhydramine, FYI).
-
NiMH actually is rather tricky to charge. From my reading of various NiMH characteristics, what I gathered is the best charge-full detection for NiMH is "NDV" - negative delta voltage. So when you detect the voltage goes from positive delta (voltage going up) to suddenly but brief and small negative delta (voltage going down), that NDV detection tells you the charge is done/full. At that point, the power going in becomes room-warmer.
To detect that brief and small NDV is not easy. Many commercially available consumer chargers just use a timer and call it a day. Some would have temperature detection to limit over-charge.
FWIW, my understanding of the process is this: though heat is produced throughout the charging process, heat production rises rapidly once full. The increased cell temperature in turn reduces the cell’s internal resistance, which in turn reduces the voltage at a constant current. So -dV monitoring is actually a proxy for monitoring the cell temperature! (And thus, increased heat production is actually the only true external indicator that charging is complete.) That’s also why good chargers monitor temperature with a thermistor as well. (And why some chargers have a cover over the battery compartment, which would help stabilize the temperature and reduce the effects of ambient temperature change if there’s a breeze or something.)
You’re incorrect in describing it as a “brief” drop in voltage. Yes, it’s just a few millivolts, but that’s easy to detect, and the drop isn’t like a little negative pulse — the voltage will continue to fall as long as the temperature continues to rise. So basically, you monitor the voltage long enough to be certain it’s not noise. Typically, this is done over the course of a few minutes. But basically, the duration of the voltage drop is determined by you, the charger designer. The cleaner, more stable and noise-free your design, the less measurement error you have, and thus the shorter a measurement interval you can use to confidently determine that voltage is falling. Commercial charger ICs typically only sample the voltage in the range of a few times per minute, to only every minute or two, resulting in perhaps 30 secs to 5 minutes to confidently decide the voltage is falling, since they usually compare more than just 2 samples. If you measured it, say, once per second (which is well within the capabilities of modern cheap technology), you could probably determine charging is finished within a few seconds of the voltage starting to drop. (You can easily see this with a multimeter, by the way, as the drop isn’t subtle in the mV. It’s only subtle compared to NiCd.) The only reason I can think of that this isn’t done is simply that all the development in battery charger ICs nowadays is for lithium based cells, not NiMH.
Even so, I don’t think -dV/dT charging is as rare as it used to be. Only the cheapest chargers these days are dumb timer ones.
I’ve thought about building a NiMH charger myself, using an Arduino, an INA226 voltage/current monitor, and temperature sensor, just for the fun of it, since I’ve learned so much about battery charging anyway! :p
-
In 2008? Highly unlikely, as SMPS wall warts were already fully established as the standard by then.
That's true, but some really cheap switched mode wall warts have abysmal regulation. Some consist of a high frequency oscillator and transformer, with a rectifier and capacitor on the secondary. One of the phone charges I looked at from the early 2000s, was no more than a blocking oscillator.
-
...
...
You’re incorrect in describing it as a “brief” drop in voltage. Yes, it’s just a few millivolts, but that’s easy to detect, and the drop isn’t like a little negative pulse — the voltage will continue to fall as long as the temperature continues to rise. So basically, you monitor the voltage long enough to be certain it’s not noise. Typically, this is done over the course of a few minutes. But basically, the duration of the voltage drop is determined by you, the charger designer. The cleaner, more stable and noise-free your design, the less measurement error you have, and thus the shorter a measurement interval you can use to confidently determine that voltage is falling. Commercial charger ICs typically only sample the voltage in the range of a few times per minute, to only every minute or two, resulting in perhaps 30 secs to 5 minutes to confidently decide the voltage is falling, since they usually compare more than just 2 samples. If you measured it, say, once per second (which is well within the capabilities of modern cheap technology), you could probably determine charging is finished within a few seconds of the voltage starting to drop. (You can easily see this with a multimeter, by the way, as the drop isn’t subtle in the mV. It’s only subtle compared to NiCd.) The only reason I can think of that this isn’t done is simply that all the development in battery charger ICs nowadays is for lithium based cells, not NiMH.
Even so, I don’t think -dV/dT charging is as rare as it used to be. Only the cheapest chargers these days are dumb timer ones.
I’ve thought about building a NiMH charger myself, using an Arduino, an INA226 voltage/current monitor, and temperature sensor, just for the fun of it, since I’ve learned so much about battery charging anyway! :p
Now this is a proof of value of this forum... Unlike reading books in the library, here, I have a chance of being corrected when I am wrong. Without this, I will take all my uncorrected mistakes to the grave... Between me, my ego, and my mistakes, I'd need an oversize coffin to rest comfortably when the time comes.
I would be remiss if I forget to say thanks - Tooki, thanks for the correction. Unless other experts coming around to dispute the correction, I will take your explanation and stand corrected. I've understood it to be seconds with range in single digit mV, and took in the reading unverified. Minutes and double digit mV would certainly change my thinking here.
I was taking issue with the word "proxy" there because it implies an indirect effect. But so is temperature. So that got me thinking... Temperature is probably the definitive indicator in general - the energy, if not converted to chemical energy, got to go somewhere and heat is where it goes. But NDV is a better way of detecting it sooner. It is rather like a temperature probe into the inside of the battery, whereas, a thermistor or something like that requires the heat to transmit to the outside of the battery, plus environmental factors and such.
The INA226 certain looks great even in the single digit mV range. My "go-to" solution would probably be the INA219 or the ADS1115 which I already have in my stock. They too discern single digit mV comfortably. I should play with that sometime too...
-
Wow, Rick, thanks for the high praise!
[...] I've understood it to be seconds with range in single digit mV, and took in the reading unverified. Minutes and double digit mV would certainly change my thinking here.
I was taking issue with the word "proxy" there because it implies an indirect effect. But so is temperature. So that got me thinking... Temperature is probably the definitive indicator in general - the energy, if not converted to chemical energy, got to go somewhere and heat is where it goes. But NDV is a better way of detecting it sooner. It is rather like a temperature probe into the inside of the battery, whereas, a thermistor or something like that requires the heat to transmit to the outside of the battery, plus environmental factors and such.
Well, the word proxy means a representative of sorts, so necessarily implies indirectness, but not necessarily an indirect effect. What I mean is, the measurement is indirect, but the effect that allows us to use indirect measurement is not: the internal resistance rises directly because of increased temperature, and internal resistance directly affects charging voltage, allowing voltage change to be a reliable proxy for temperature change.
Anyhow, once one knows that the voltage drop is caused by heat, it suddenly becomes obvious that high sampling rates aren’t needed, since the thermal mass of the battery will severely low-pass filter it anyway. Very good observation about it being more responsive than an external thermometer!
I highly recommend carefully reading the MAX713 datasheet. It explains a lot of the nitty-gritty of NiMH charging really well. (And other than the rushed removal of switch-mode operation — leaving instances of “when in linear mode...” which is now the only mode described in the datasheet — it’s arguably the best datasheet I’ve ever used.)
https://www.maximintegrated.com/en/products/power/battery-management/MAX713.html (https://www.maximintegrated.com/en/products/power/battery-management/MAX713.html)
The INA226 certain looks great even in the single digit mV range. My "go-to" solution would probably be the INA219 or the ADS1115 which I already have in my stock. They too discern single digit mV comfortably. I should play with that sometime too...
Definitely!
Take a look at my experience with the INA226 here: https://www.eevblog.com/forum/projects/ode-to-the-ina226-voltagecurrent-sensor/ (https://www.eevblog.com/forum/projects/ode-to-the-ina226-voltagecurrent-sensor/)
-
I don't know why this thread devolved into long discussions about overcomplicated solutions-looking-for-a-problem, but the answer to the title question is "yes". NiMH and NiCd before it are fine with being slowly and continuously charged.
-
I don't know why this thread devolved into long discussions about overcomplicated solutions-looking-for-a-problem, but the answer to the title question is "yes". NiMH and NiCd before it are fine with being slowly and continuously charged.
:palm: Except that most NiMH actually don't tolerate it well, which is why they have to make special versions specifically for that scenario.
This thread is hardly "devolved", it moved into nuanced and informative discussion of the exact topic at hand. Odd thing to get your panties in a twist over.
-
It's pretty much what happens to most threads. The original question gets answered and then discussion goes off on various tangents, I see nothing wrong with that. Anyone who is no longer interested can drop out easily enough.
-
Hi,
I hope it is not a problem to jump in here because I have a problem with the same circuit board.
My batteries were bad so I tried to change them with new ones. That worked. So after soldering the new battery the device works. Unfortunately it does not load the battery when plugged in. But the LED is on when plugged in.
I assume that I maybe damaged the trace. You can see this on the attached picture where the arrow is. Am I right? Or can someone see any other problem which leads to this problem?
I am not an electrician but sometimes I do repair some of my devices - or at least I try ;)
Sorry for my mediocore englisch - I am from Germany .
Thanks for your help
-
If you have a multimeter, measure continuity.
-
Can you tell me briefly how to do that? I have a multimeter.
This one here:
Do I need to plug it in to check or is it just continuity with no current applied?
EDIT: And of course I have the batterie soldered. Is it possible to check continuity while battery is attached and current applied?