Solar charge controller for lithium batterys

Hi Ron @zander and  @nurderfch1846 ,

Thanks Ron for that additional insight.

Overnight, I realised that one point I hadn't directly addressed was different use cases require different solutions ... and having just seen Ron's latest contribution, I think he too is illustrating different scenarios require different solutions.

I don't think we have a clear picture of nurderfch1846's use case, so I will propose two imaginary use cases to try to illustrate an important consideration:

• At one extreme, might be a solar powered burglar/fire alarm with a battery for the hours of darkness and to ensure there is enough power available to run sirens, etc. when triggered. On a 24 x 7 basis, it only consumes milliWatts for the processor and sensors, but in the event of it being triggered, the power consumption may rise to several Watts for a period of up to 30 minutes.
•  
• At another extreme, one might have a few, or maybe many, square metres of solar panels, capable of producing enough power to run power hungry equipment (ovens, washing machines, ..), but the battery provision might be relatively small, aiming at (say) low power lighting at night.

In the first case, the solar panels, during the hours of daylight, are used for both powering the alarm and charging the battery, but can only make a modest contribution to the power required when the alarm is triggered .. in this case the primary, maybe only, responsibilty of the charger is to maintain the battery in a good state of charge, particularly whilst the alarm is not triggered. When the alarm is triggered, the battery is the primary power source, and if it is at night, the only power source, so it must be capable of meeting all of the demands of that scenario. During the day, the solar panel might sometimes produce more power than the battery could safely absorb, but the charger only needs to control the charging rate, and simply reject the excess power.

In the second case, not only may the solar panels might be capable of producing more power than than the battery could accept whilst charging, but all of the power must be capable of being directed to 'real-time' loads. This can obviously become a complex situation with active changes to prioritising power usage.

.....

A simple power charger of the type discussed previously, based on the TP4056, when combined with the DW01, might be considered for the first case. Assuming it works as described, which may include adjusting the value of the current limiting resistor, it will prevent the cell from being over charged, over discharged or charged at too high a rate. However, the apparent lack of cell temperature monitoring is a deficiency that should be addressed.

Remember the accepted limitation that the charger will never pass a higher current to combined load and battery, than the maximum specified charge rate for the battery, even if the load is demanding a higher current than the charger output current limit, and the  power source (solar cell) is capable of producing more power, resulting in the battery being discharged.

Obviously, the TP4056 limitations of 1 Amp and 1 cell, preclude it from being considered for the second case. However, even a suitably uprated unit with the same "capabilities", would not meet the second case requirements of being able to direct a power level greater than the maximum battery charging rate, directly from the source to the load.

--------

Best wishes, Dave

@davee Don't know about where you live, but your case 1 would result in an excessive noise charge.

Although the second case might be doable, no responsible solar designer would build one like that.

Let's wait for the OP to re-post or update his requirements, then we can respond with useful replies.

@davee I neglected to specify a point regarding Dave's 2nd example. There is a relationship between solar panel input and charge controller output. There is also a battery manufacturer-specified charge rate. If you violate the battery charge rate enough for long enough, you will destroy the battery and likely have a fire. The proper way to think of a 'solar system' is a battery in the middle to act as a buffer, with solar panels on one side of the battery and an inverter on the other. In the case being discussed, there is NO inverter, which is ok, but the charge rate must still be honoured. I can't recall seeing more than a 1C charge rate anywhere, and 0.5C is more common. That being the case, my back-of-the-envelope calculation is as follows. If we want 4 x 18650s in parallel, then a voltage of 3.2 or 3.7, depending on chemistry and a capacity of roughly 4,800mAh then using 0.5C gives us a max charge current of 2.4A and a rough power calculation is 9W. If we connect 2 5V solar panels in series that produce about 1A then we have a balanced system, If you over panel then at peak conditions you will produce more heat from the charger but the batteries will not charge at faster then 0.5C or 2.4A at 3.7V. The charge controller is basically a transformer, so input POWER and output POWER are roughy the same, you simply pick a 'transformer' that matches the voltage and then divide the output power by the voltage to determine the current. Divide the battery capacity (C) by the charge current to determine the charge rate. 

I hope that is clear.

@zander @davee,

I´m sorry if I asked some stupid questions or didn´t get some important points. I´m not very experienced with many of these topics (I´m still in school...) and I am not a native English speaker.

To now restate/ update my objectives:

I have two solar panels (6V; 0-0.58A; 3.5W). They should charge a battery or battery pack that fits my project. My main power consumer are two of these motors ( https://cdn-reichelt.de/documents/datenblatt/C160/GM82.6_6v_db_en.pdf) and an ESP-32.

I found this protected battery:

(I found two datasheets:

1. https://www.jauch.com/downloadfile/5e200d2f610958478c59ddb760bb61954/3350%20mAh,%20PROTECTED%20LI-ION%20RECHARGEABLE%20BATTERY-18650,%20Mat.%20250669.pdf

2. https://asset.conrad.com/media10/add/160267/c1/-/en/002281109DS00/datenblatt-2281109-jauch-quartz-li18650jl-protected-spezial-akku-18650-li-ion-36-v-3350-mah.pdf)

A few questions to the battery:

1. The battery claims to be protected. I guess I should not rely on it completely protecting itself and therefore should still try to get a charger with battery protection if possible?

2. One of the datasheets says the battery has a standard current of 670mA, the other one says 0.2C, altough Coulomb is not a unit for the current. How can both be the standard discharging current?

3. Assuming the standard current is 670mA: Is it safe/ recommended to discharge the battery with 5A or will it heat up by a lot? (Max. discharging current is 5000mA on one datasheet and 1.5C on the other)

If one cell cannot give me the current safely, I am looking for another solution for getting enough current. For whatever type of battery/ batteries I end up using, I have to find a solar charge controller that works with them and the solar panels described above.

Again, I´m sorry if I asked some stupid questions before and thank you so much for your help, I really appreciate it.

Hi Ron @zander and @nurderfch1846,

   Ron, I agree with your points about my examples, but please remember my 'extreme imaginary examples' were just that ... hopefully technically OK in illustrating the points about charging and powering loads, in that different use cases require different approaches. I appreciated, even as I was writing it, that you might decide a burglar alarm siren might be time limited to 10 minutes or so.. but may be an accompanying flashing light might not, and so on ...and maybe you specify a battery that can still perform adequately after aging, and only being half charged. I was only considering the details of the charger/power controller functions, not designing the perfect burglar alarm.

In reality, over the years, I have known the odd neighhbour's alarm or two keep going for a long time .... but double glazing, etc. mitigates most of the problem. 🤨  I doubt if anyone got charged.

Similarly, my second example was completely contrived, again to illustrate an extreme, not a practical, sensible solution.

----------

Ron, I think I agree with your worked example .. the only thing I would add, is that professional systems, with inverters, etc, are a much more controlled environment in terms of voltages and currents, than a couple of small cheap solar panels and a bare module or two, as typified by those labelled TP4056.

In particular, any device that limits current flow by acting as a linear regulator, which applies to the TP4056 chip, and the FETs attached to the DW01 on the boards discussed, is susceptible to overheating problems if the current flow is 'high' at the same time as the input voltage is 'high'.

The situation is a bit like the small linear regulators on the Arduino boards, that can theoretically cope with input voltages up to about 20V, but if the Arduino board is talikng an appreciable current, the regulators get cooked. In this case, we have both recommended putting an additional buck regulator, to drop the voltage to around 7V.

I do not have enough data to know if this will happen in this case, but if I was building such a system, I would be making measurements, etc. as I prototyped it, and modifying the design accordingly.

--------

@nurderfch1846 .. sorry, I haven't got time to look at your references at the moment, but I notice an understandable mistake.

"C" in this case is not Coulomb, but a term battery people use. "1 C" means charging (or discharging) at a rate that would take the battery from zero charge to full charge (or vice versa). Simplistically, for a 2 Amp-hour battery, then 1 C (maybe written as just C) would refer to charging at 2 Amps.

Maybe Ron can add any subtle notes to this explanation, but this should help to explain the principle.

Best wishes, Dave

@davee

Posted by: @davee

↑

"C" in this case is not Coulomb, but a term battery people use. "1 C" means charging (or discharging) at a rate that would take the battery from zero charge to full charge (or vice versa). Simplistically, for a 2 Amp-hour battery, then 1 C (maybe written as just C) would refer to charging at 2 Amps.

Ok, that makes sense. With 3350mAh and 0.2C standard current it would indeed be 670mA then, as in the other datasheet

@nurderfch1846 I can't open the first file.

1. I would be ok with a battery that is protected. I might not try to charge it in very cold conditions and of course only charge it per specs and use it per specs.

2. Coulomb is valid, google it. It is not really used though in day to day calculations. See below for the definition. Both can be true as they are different units of measure. 0.2 x 3350 is 670mA. 

3. Again, 1.5C is the same as 5,000mAh

Your question now appears to be 

 I have to find a solar charge controller that works with them and the solar panels described above.

As far as the charge controller, it's just simple math, (POWER OUT) * eff%) = POWER IN. For your tiny setup, an MPPT controller is overkill so look for a 'reasonable' PWM controller. Bogart Engineering makes the best PWM and when you combine their Charge Controller with their Battery Monitor you have a really good system. I used to have one when I had a smaller system but now I use Vicrtron MPPT for my over 1 Kwatt array. Check out these 2 links LINK1 LINK2. Bogart is at HERE

As I said in my earlier post, you already seem to have what you need I terms of about 20Watts of solar panels (6V X 0.5A) x 2. I recommend a series connection for reduced wire size and earlier turn on so 12V @ 0.5A is 6Watts. Since you need to convert that to roughly 4V that gives you roughly 1.5A or 75% of max charging current for 2 batteries, 38% for 4 batteries if you need that much. That will all work, just be aware that each battery is getting about 375mA which means 9 hours of high noon sun. We use 4hrs per day to estimate the area under the curve so that setup will do ok for full sun days as long as you start the charge cycle at 50% DOD. Of course you can NOT be also using the batteries, I suggest you buy double the batteries, and use one set down to 50% while charging the second set.

I think I have answered all your questions and have shown you what you need. Perhaps a summary:

1. 2 6V @ 0.58A solar panels. (a)

2. 2 sets of 4 18650 batteries connected in parallel although two or one might work depending on how many amps are being drawn for how long. One set is in use down to 50%, and one set takes all day to recover from 50%. (b)

3. Solar charge controller that is at least 10W. (c)

4. Misc - Wire, marine grade connectors, breakers, possibly mounting boxes etc. A marine supply shop is a good place to source quality (but very expensive) goods. I say marine grade because I assume the equipment is NOT in a conditioned space but is in an outdoor or unconditioned space is therefore subject to humidity. Double-wall glued heat shrink is your best friend. (d)(e)

5. Optional - Mains powered charger to 'catch up'.(f)

NOTES: 

a - More of these would help a lot, 2S2P connection would yield 12V@1.16A and stepping down to 4V would yield 3.48A or 0.87A per cell of a set of 4 and almost fill up in 4 hrs.

b - Your problem is you probably can't employ a shunt, so the capacity numbers are a guestimate.

c - 15W if you go 2S2P.

d - All wire is 100% copper, multi-fine strand-like welders cable.

e - Remember to provide a shutoff or breaker between the solar panels and the charge controller as well as between the battery and the charge controller so you can isolate each component. In a small low-voltage system such as this, the breaker can serve as the disconnect as well. My system can see as much as 350+ amps under the right conditions so I employ both in the inverter side.

f - If you are going with 18650, this charger is adequate. 

@zander Thank you for all that information. I´m going to read through it later. To the first link: It included the bracket at the end. here´s the correct link: https://cdn-reichelt.de/documents/datenblatt/C160/GM82.6_6v_db_en.pdf

@nurderfch1846 BTW, there are NO stupid questions unless you count the question NOT asked as stupid, then there is one. Keep on asking, I might bark, but I don't bite unless it's @davee. Dave

@nurderfch1846 Um, that's a motor spec, I don't see how that relates to solar and batteries. The power requirements of the motor are well within spec of any battery you might use.

@davee C is Capacity, in this case, a 'claimed' 3,200mAh, but I doubt it is more than 2,000mAh and even as low as 1,200mAh. I even have batteries labelled as 5,000mAh that are still 18650 form factor, so the label is a fake unless they are totally different in chemistry. I bought them on the basis they were at least 1,200mAh. This is why you want a decent mains charger that can also tell you the real Capacity so you will know how long the battery will last. That motor, coupled with a single battery that you posted the spec on, will last 3250 / 160, which is roughly 20 hrs unless you stall the motor very much, which uses 20x the current, so 1 hr if always in stall (not likely)

@zander Stall current is 3200mA, so for two motors 6400mA. That´s also at 6V, so it would be even more at 3,7V. The battery I found has a maximum discharge current of 5000mA, so that wouldn´t be enough, right?

@zander Yeah, I was also surprised about how good these seemed to be...

@nurderfch1846 Correct, you would need 2 batteries in series to give you the voltage, but then for 2 motors a second set for the other motor.

Hi Ron @zander and @nurderfch1846,

  To tidy up a confusion:

re: If we want 4 x 18650s in parallel, then a voltage of 3.2 or 3.7, depending on chemistry and a capacity of roughly 4,800mAh then using 0.5C gives us a max charge current of 2.4A and a rough power calculation is 9W.

This statement is correct, but only if we assume that 0.5C applies to the charging rate.

There is no 'standard rate' definition... maximum charging and discharging rates may be different for a given battery ... it is essential to clarify which is being specified.

---------------------------------

Coulomb is valid

This statement is also valid ...Coulomb is the SI unit for a quantity of charge, but it is not a quantity of current ... so 'x' Coulombs  cannot be equated to 'y' Amps (or milliAmps mA).

That is like trying to measure a distance in 'metres per second'

---------------------------------

However:

One Coulomb can be equated to amount of charge transferred past a point
 in a conductor in a time of One second, when the current is One Amp.

It follows, that 4.8 Ah battery, would have a capacity of 4.8 * 3600 Coulomb = 17,280 C

(as there are 3,600 seconds in an hour)

-------------------

Hence, the latter part of

0.2 x 3350 is 670mA. 

3. Again, 1.5C is the same as 5,000mAh

is nonsense.  1.5C relates to a current of 5,000 mA

----

Consider two batteries A & B, both 3,350 mAh capacity, but described as capable of discharge rates of 0.2C and 1.5C

They both hold the same amount of energy, so if discharged at 0.2C rate, which is a current of 670 mA, they will both take 5 hours to go from full to zero charge.

However, if you tried to discharge them at a 1.5C rate, which is a current of 5A, then the B will be discharged in 40 minutes, whilst A may fail to deliver 5A, overheat, etc...  who knows? 🙄 

 ----------

So in summary, the majority of what stated in this section is correct, but a few confusions crept in.

Best wishes, Dave