[Chhanda Das]

Update 1 :

Increased the Lithium battery's capacity to 35 Ah

It has been over a year since my son made the 12v 25 Ah Lithium battery pack and I thought about letting you all know about the present situation.

First, let me share the information about a temporary setback. The LiitoKala branded 18650 Lithium cell charger that we were using died all of a sudden due to an ant infestation inside it

When we started looking at 18650 Lithium cell chargers, we were appalled at the sky-high prices and decided to make one on our own. After a little bit of research, we decided that the cheap TP-4056 printed circuit board (PCB) module would suit our requirements perfectly. Please note that this TP-4056 PCB module is available in a few different variants. We got the 1 ampere-rated ones with current protection and micro-USB input. This would allow us to power it from almost any power source like computers, solar charge controllers, smartphone power banks, etc as long as the source current did not exceed 1 ampere. We also got a USB combination ammeter-voltmeter with a maximum current capacity of 3 amperes and a voltage range of 3.5-7 volts to measure the amount of power being drawn.

As you can see above, the TP-4056 PCB modules' input ends have been connected to each other in parallel. This was done solely to minimize costs since we got just one USB combination ammeter-voltmeter and coincidentally had just one spare micro-USB male to USB male cable lying around. The wires were salvaged from the dead/defective laptop battery packs from which we got the 18650 cells for refurbishing/salvaging. The 4 bay 18650 cell holder in the image above was originally intended for a 4S configuration but we didn't have any use for it anymore. So my son turned it into a 2 bay 18650 cell holder. He didn't remove the original connections in the 4 bay cell holder so that he could use it again in a 4S/3S configuration if necessary.

Now he could finally increase the capacity of the 12v 25 Ah Lithium battery pack. Please note that we had not noticed any degradation of the 12v 25 Ah Lithium battery pack in the last year despite its continued usage almost every single day. The original configuration was 3S11P. He added 5 sets of refurbished 3S cells (Total 10 Ah) in parallel to that, resulting in an increase in capacity by approximately 40%. I won't bore you with the details of the procedure since that has already been discussed in detail near the beginning of this thread.

First, the BMS was removed from the battery pack.


Then the 15 cells were added in the proper orientation for matching the 3S configuration of the older cells.


Spot welding completed

BMS attached and connected

The present configuration is 3S16P and the total capacity is 12v 35 Ah. This 12v 35 Ah Lithium battery is approximately equivalent to a 12v 70 Ah Lead-Acid battery when it comes to the amount of usable power. Mission accomplished !

Disclaimer : We followed all precautions/procedures mentioned above and earlier. If you want to copy all this then please do so at your own risk.

I hope that you all found this interesting and I wish you all happy and safe drives ahead

[Heisen]

Very nice post.

I'd like to point some of the things from safety point of view.

1. There is concept of lithium battery temperature monitoring while the battery is in the state of charging, usually that is the time when it generates the most heat and chances of something going wrong are extremely high.

The temperature monitoring is done via thermistor which is usually placed deep in between the cells or sometimes on the cells with the help of sticky tape, also there could be multiple thermistors spread over the battery pack to monitor each zone.

The thermistor is then connected to the BMS which offer this functionality or there could be a sperate circuit for that. When the reading of the thermistor reaches some critical threshold value, then the monitoring circuitry just cuts power to the battery.

Thermistors are used for this job because they have very fast response to the change in temperature, which is very important in this application.

2. Don't let the cell voltage go below 3.3 volts, these cells cannot be deep discharged. In off the shelf BMS's this cut off voltage is usually set very low, the worst I have seen is 2.5v, which is very bad for the cell. From 4.2 to 3.3v most of the cell energy is gone, there is no point going any lower for slightly high battery backup.

3. Be careful of TP-4056 modules, they should charge the cell exactly to 4.2 volts but since these are cheap, sometimes there can be some offset. The 4.2v value is very critical, one should not charge even slightly beyond that. I usually prefer to charge only to 4.1 or 4.15 volts.

Stay safe.

[tsk1979]

You are a lot braver than me. For my astro power station, I first decided to get 2S cells and wire my own BMS. But being in the USA I was able to get a 1.4KW (100ah*14V) battery for slightly more, and that has built in low temp protection, as well as a 100A BMS

Regarding your setup, try sourcing a current shunt and add to the system. That way you will always be aware of charge and discharge currents.
You will also need high temp cutoff as Heisen said. Lack of high temp monitoring during charging is the main reason poor quality Electric scooters etc., go boom.

Secondly, your system looks to be exposed. Get a plastic box, and build an enclosed unit so accidental shorts or other accidents are less probable.

But I feel your pain in doing DIY. Looks like the Aliexpress ban has not caused a boom in electronics manufacturing as promised. So getting anything niche is a big pain, and you often have to compromise on quality.

[Reinhard]

Originally Posted by tsk1979

But I feel your pain in doing DIY. Looks like the Aliexpress ban has not caused a boom in electronics manufacturing as promised. So getting anything niche is a big pain, and you often have to compromise on quality.

<<Apologies for the Off-Topic Rant!>>

AliExpress ban - has only added to the miseries of us DIY folks. What more - while nobody ever expected the imports to be replaced by local manufacturing in less than a decade - what has happened is - now there is a select specific group of people, that are importing in bulk, and selling absolutely the same products at at least 10X the price, on the same old online sales channels like Amazon.

So a select bunch of people (will be interesting to see how many of these importers/license holders are related with the "authorities") - are making millions, at the expense of the retail customer. Interesting way to beat China!

I'm all for curbs on Chinese imports and reprisals for their misadventures into our motherland. But - they should actually mean something more than just further pain to the common man.

[Chhanda Das]

Everyone, please accept my apologies for replying so late. Somehow I missed the notifications for your posts

Originally Posted by Heisen

Very nice post.

1. There is concept of lithium battery temperature monitoring while the battery is in the state of charging ...

2. Don't let the cell voltage go below 3.3 volts, these cells cannot be deep discharged. In off the shelf BMS's this cut off voltage is usually set very low, the worst I have seen is 2.5v, which is very bad for the cell. From 4.2 to 3.3v most of the cell energy is gone, there is no point going any lower for slightly high battery backup.

3. Be careful of TP-4056 modules, they should charge the cell exactly to 4.2 volts but since these are cheap, sometimes there can be some offset.

Thank you very much for your kind appreciation of our efforts and your very valid concerns. However, let me try and explain the situation as follows :-

1. You are absolutely and undeniably correct here. However, the maximum charging current capacity for the battery pack before my son increased its capacity was above 22 amperes (A). This is when the battery pack would start getting warm to the touch. However, the adapter for charging it with our home electricity was only rated at 3 A. I had solar panels rated at a total of 150 watts which have an I (Max.) or maximum current rating of a total of 7.8 A. For charging in the car, the boost converter is calibrated to deliver not more than 3 amperes of current at 14.2 volts. Since the charging current was way lower than the threshold of 22 A in every case, we did not feel the need to monitor the temperature then. But we have recently procured a 180 watt solar panel with an I (Max.) of 9.91 A and now we definitely need a temperature monitor (preferably with a high-temperature current cutoff). However, the sad part is that we have been unable to find a suitable digital thermostat controller with a high enough current rating at reasonable prices neither online nor at regular shops here in Kolkata. So till we are able to procure it, we will have to stick to our earlier self-imposed low charging current limit.

2. I would beg to slightly disagree here. An overwhelming majority of the cells that we used for/in this battery pack are branded cells from LG, Sony, etc and all of them have an officially declared fully discharged voltage rating of 3.00 volts (V). So for a battery pack like ours involving rows of 3 cells in series, the low voltage cutoff would be 9.00 V. You are absolutely correct about these cheap battery management systems or (BMSs) having substantially lower voltage cutoff limits than 3 V for each cell. However, that is not a problem but an advantage for our setup. This is because all of our charging (home/solar/car) and discharging of this battery pack happens through the solar charge controller which has a low voltage load cutoff exactly at 9 V. As a result, the BMS still remains in the ON state and the solar charge controller does not turn OFF. Charging the battery pack through the solar charge controller would have been impossible if it had turned off at 9 V due to the BMS. About the high voltage cutoff, the solar charge controller cuts off all current to the battery pack once it reaches 12.6 volts and does not start the charging again until the battery pack reaches 12.3 volts. Also, we have not seen any battery degradation in any of these cells as of today, touchwood.

3. You are again absolutely and undeniably correct here. And that is exactly why we only use the TP-4056 modules when we can closely monitor the individual charging cells by touch especially since we do not have a suitable thermostat controller as of right now.

I hope that I was able to assuage your very valid concerns.

Originally Posted by tsk1979

Regarding your setup, try sourcing a current shunt and add to the system.
You will also need high temp cutoff as Heisen said.

Secondly, your system looks to be exposed. Get a plastic box

But I feel your pain in doing DIY.

Thank you very much for your sympathies and your kind suggestions.

We have procured an ammeter with a 100 A rated shunt but didn't get around to installing it yet partly because of laziness and partly because of the fact that there will be some loss in efficiency/power because of it. However, with the increase in capacity, hopefully, you will see it in one of the updates in the near future. The same goes for the high-temperature cutoff as well if we can find a suitable one.

Your concerns about the exposed system are absolutely valid here. However, we do not intend to get a plastic box for it because such battery boxes are very rare in India. The ones available here are too flimsy to say the least and more suitable for lead acid batteries. Additionally, a box would prevent capacity upgrades to the battery pack beyond a certain point. Our eventual plan is to make the battery pack as long as the width between the B-pillars' plastic trims inside our first generation Creta. So we feel that a 4 inch wide piece of plastic pipe would suit our requirements better. That would also allow us to install a small cooling fan at one end of the pipe for some cooling. But this is all in the pipeline for future updates

Originally Posted by Reinhard

AliExpress ban - has only added to the miseries of us DIY folks.

You have reiterated my thoughts and sentiments exactly. The situation is indeed quite frustrating, to say the least.

[Chhanda Das]

Update 2 :

Added a bidirectional ammeter module to the setup

Ever since the prodding by our fellow Bhpian tsk1979 about adding an ammeter to our setup in one of the posts above, I started thinking seriously about it and would often nag my son for the same. He finally got tired of it and got around to making a bidirectional ammeter module.

Problems galore

Firstly, we had to get a digital ammeter since an analog one would not be able to suffer the rigours of a moving car regularly. But which one to choose ? Since our maximum charging current would not exceed 15 amperes (A) of direct current (DC), the closest-rated digital DC ammeter above it would be at 50 A. However, we could not find one like that either here in Kolkata or anywhere online that we could access. I still feel that the ban on Aliexpress continues to hurt us to this day. The next larger capacity DC ammeters that are available to us are rated at 100 A. These are available with a 75 millivolt (mV) shunt for around INR 350.

However, there was a problem. These ammeters measure the current flowing only in one direction. This meant that we could either measure the amount of current going into the solar charge controller or the current being drawn by the load from the charge controller. Our aim was to measure the net amount of current going into or out of the battery when the solar charge controller is receiving a charging current while running a load all at the same time. Hence, we started looking for bidirectional DC ammeters but we were shocked to see that these cost approximately 5 times what the 100 A DC ammeter with a 75mV shunt costs.

We realized that it would be much cheaper to use two separate 100 A DC ammeters with two separate 75mV shunts to measure the amounts of current going in and out of our system at the same time. This seemed like a suitable idea to me but my son was still not satisfied.

The brainwave

After pondering over the wiring diagrams and technical manuals of the cheaper 100 A DC ammeters for a few days, he suddenly had a brainwave. Why don't we use a manual switch to reverse the direction in which the ammeter measures the current ? Oh yes of course, why didn't I think of this ? He decided to use a double pole double throw (DPDT) switch with a center-off to reverse the direction of the current measuring wire with respect to the ground wire. The center-off was important since it would allow the meter to turn off which is very important for maintaining efficiency without removing the entire module from our setup, especially on cloudy/rainy days when there wouldn't be much harnessable solar energy from our 150 (100+50) or 180 watt solar panels.

The wiring

Did I mention that it also measures the voltage ? Well, it does. The wiring of the 100 A DC voltmeter ammeter with a 75mV shunt can be done in either one of two ways. One way is to use an independent power source to power the ammeter while in the other case, the ammeter draws the power from lines it is measuring. If we use an independent power source, the measurements will be more accurate to the tune of 0.1 A but in the other case, the accuracy drops to 0.5 A. Please note that in the latter case, the voltage has to be in the 4-28 V range. Since it would have been impractical to use an independent power source in our specific use case, we chose the other way as you can see in the wiring diagram below :-

In the image below, you can see that the DPDT switch is connected electrically between the shunt and the meter. There is a single pole single throw (SPST) switch connected electrically on the current sensing wire between the DPDT switch and the meter. This SPST switch has to remain off or has to be turned off before changing the position/state of the DPDT switch or else the meter can get burnt/destroyed. My son used some double-sided adhesive tape to insulate the shunt from the exposed metal terminals/wires. I will not bore you with the nitty-gritty of the rest of the wiring which is pretty much evident in the images/video above/below. Since this meter setup will be connected electrically between the charge controller and our DIY Lithium battery pack, it will show the net amount of current flowing through the wires while the position/state of the DPDT switch will show the net gain/loss of charge of the battery pack.

The total cost for this update came to less than INR 400 and here is a short video of the last test done by my son :-

https://www.youtube.com/watch?v=jbpU3bhG3JE

I hope that you enjoyed going through this update on our portable power station project and I wish you all happy & safe drives ahead

[Chhanda Das]

Update 3 :

Added temperature-based controls to the DIY portable power station

As mentioned by our fellow BHPians Heisen and tsk1979 above, temperature monitoring and control is very important for any Lithium battery pack. This is especially true since we have substantially increased the capacity of the original DIY 12v 3S Lithium battery pack. So we did just that.

But first things first. The small 2 ampere (A) rated barrel connector port on the solar charge controller was chosen to power the two thermostat controller modules that we intended for our setup and hence a male to two female (all 2.1 mm x 5.5 mm) barrel connectors was prepared as you can see in the image below.

Then one 6010 model 12v 0.15a rated brushless cooling fan was stuck to the back of the battery pack with some double-sided adhesive tape. The fan was then connected to a 2.1 mm x 5.5 mm male barrel connector.

The entire battery pack, excluding it's back and the sides of the first cells closest to the BMS, was covered with Kapton tape for electrical insulation and heat insulation up to an extent since the BMS module and the nickel strips were exposed as pointed out by tsk1979. The back of the battery pack and the sides of the first cells closest to the BMS were deliberately left uncovered by the Kapton tape.

We know that the flow of air may not be much but the intention here is to compensate for the reduction in passive cooling due to the liberal use of the Kapton tape. Essentially this small fan would prevent the buildup of micro heat islands within the battery pack up to an extent.

Next we needed a way to start running the fan automatically once a particular user-settable temperature was reached and it should turn/remain off below that temperature. For that we needed a thermostat controller module that could run on 12v direct current. We found that the cheapest thermostat controller for our particular requirements was what is called a W1209 thermostat controller. This is rated to handle upto 10 amperes of current and comes with a negative temperature coefficient (NTC) thermistor sensor. This cost us ₹130 including the transparent acrylic plastic case that can be purchased separately. Here is a fun fact, the W1209 thermostat controller is widely used in DIY egg incubators here in India.

As you can see in the image above, a male 2.1 mm x 5.5 mm barrel connector is used for powering the W1209 module while the female counterpart is for powering the fan. There are also 3 buttons on the module. The left (SET) button is for setting/choosing the parameters, the middle one (+) is for increasing the value of the chosen parameter and the right one (-) is for decreasing the value of the chosen parameter. The wiring of the W1209 module was done in such a way that the positive is common for the W1209 module and its relay input while the negative is common for the W1209 module and the fan. This module has to be set in "cooling" or C mode so that its relay turns on when the set temperature is reached and the relay turns off when the lithium battery pack cools down below the set temperature.

The thermostat displays the current temperature (in °C) by default. When in any other mode, making no input for approximately 5 seconds causes the thermostat to return to the default display. Here is a brief guide to the settings :-

1) Short press the SET button and the display will start flashing. We can now set a trigger temperature (in °C) using the ‘+’ and ‘-‘ buttons in 0.1° increments. If we don’t press any buttons for approximately 2 seconds the trigger temperature gets stored and the display returns back to the current temperature.

2) To set any parameter first long press the ‘SET’ button for at least 5 seconds. The display should now show P0. This represents parameter P0. Pressing the (+) or (-) buttons will cycle through the various parameters (P0 to P6). Pressing the (SET) button whilst any of these parameters are displayed will allow you to change the value for that parameter using the (+) and (-) buttons. When finished setting a parameter press the set button to exit that option. If no buttons are pressed for approximately 5 seconds the thermostat will exit the parameter options and will return back to the default temperature display.

3) The parameter P0 has two settings, C and H. When setting to C (default) the relay will energize when the temperature is reached. Use this setting if connecting to a cooling system. As mentioned earlier, we chose this setting for our use case here. When set to H the relay will de-energize when the temperature is reached. Use this setting if controlling a heating system.

4) The parameter P1 sets how much change in temperature must occur before the relay changes its state. For example, if set to the default 2° C and the trigger temperature has been set to 25° C, it will not de-energize until the temperature falls back below 23° C. Setting this hysteresis helps stop the thermostat from continually triggering when the temperature drifts around the trip temperature.

5) The parameter P2 limits the maximum trigger temperature that can be set. It is useful as a safety to stop an excessively high trigger temperature from accidentally being set by the user.

6) This parameter P3 limits the minimum trigger temperature that can be set. It is useful as a safety to stop an excessively low trigger temperature from accidentally being set by the user.

7) The parameter P4 allows you to make minor corrections to the temperature reading if you find a difference between the displayed temperature and the actual temperature, (for instance, if the temperature probe is on a long run of cable).

8) The parameter P5 allows for delaying the switching on/off of the relay after reaching the trigger temperature. The parameter can be set in one-minute increments up to a maximum of 10 minutes.

9) The parameter P6 will cause the relay to switch off when the temperature reaches this setting. The display will also show (—) to indicate an alarm condition. The relay will not re-energize until the temperature falls below this value. The default setting is OFF.

The list below shows the default values (within brackets) and the range of the individual parameters :-
P0 Heat C/H (C)
P1 Backlash Set 0.1-15 (2)
P2 Upper Limit 110 (110)
P3 Lower Limit -39 (-39)
P4 Correction -7.0 ~ (0)
P5 Delay Start Time 0-10 mins (0)
P6 High-Temperature Alarm 0-110 (OFF)
Long pressing “+/-” will reset all values to their default values.

For the purpose of the demonstration in the video below, the user-settable temperature limit was set at 29.9° C but we intend to keep it set at 43° C since that is the highest ambient temperature that we have experienced here in Kolkata.

https://www.youtube.com/watch?v=auD1WCCfK2k

The NTC temperature sensor was then stuck (image below) between the first and second cells towards their positive ends closest to the BMS on one side of the battery pack using some adhesive tape for now. We will be adding some thermal compound paste later. This was done because these are the cells that become the warmest first during intense usage.

The next order of business was to figure out a way to stop the charging of the battery pack whenever its temperature exceeded the user-settable temperature and to restart the charging once the battery cooled down below that temperature. For that, we could have used another W1209 module like the one above by setting it to the heating (H) mode. However, there was a minor issue. The W1209 module can handle direct current only upto 10 amperes but we needed something that could handle upto 15 amperes of direct current since that is what the 12v 180w rated sockets in my 2017 Hyundai Creta can provide, theoretically speaking. We found that the W3230 thermostat controller (image below) module's 12v version would suit our requirements very nicely since it is rated at 20 amperes of direct current. This module is also available in 10 ampere 110-220v AC (alternating current) variants and 20 ampere 24v DC (direct current) variants. The W3230 module that we chose is very similar to the W1209 module except that the W3230 module also shows the user settable temperature limit on the default screen (just like a W1219 module) and this module has a dedicated "Restart" button apart from the obvious higher current handling capacity.

As you can see in the image above, a male 2.1 mm x 5.5 mm barrel connector was used for powering the W3230 module while regular 2 pin connectors were used for the relay controls. One very important aspect to note here is that for some reason the W3230 module's relay has to be on the negative line since some people seem to have faced a few issues after connecting that relay to the positive line. This module has to be connected electrically between the charging source (220v AC to 14v DC converter at home/ 12v solar panels / 1.2 kilowatt boost converter in the car) and the solar charge controller.

For the purpose of the demonstration in the video below, the user-settable temperature limit was set at 29.9° C but we intend to keep it set at 43° C since that is the highest ambient temperature that we have experienced here in Kolkata. You can see the charging symbols turn off/on once the user-set temperature is reached. Obviously, we can also use this W3230 thermostat controller module while discharging the battery by connecting it electrically between the charge controller and the load. After capturing the video below, the NTC temperature sensor was then stuck between the first and second cells towards their positive ends closest to the BMS on the other side of the W1209 module's sensor on the battery pack using some adhesive tape for now. We will be adding some thermal compound paste later. This was done because these are the cells that become the warmest first during intense usage.

https://www.youtube.com/watch?v=25_df6zr0UE

I hope that you found all this useful. Please feel free to share your criticisms as well as ideas for further modifications and/or upgrades. Wish you happy and safe drives ahead

[Redex]

Hi,
Good effort. Love the problem solving. I know practically nothing about electronics. To me it doesnt feel right to mix 5 batteries with different voltage and capacity ratings and mix new with used batteries. Was there a reason for this?, will it cause any problems over time.
I would have preferred to see identical batteries, or is that just my lack of knowledge.
I also appreciate the need for insulation and cooling. Wouldn't it be better to enclose the battery pack in a close fitting container with forced ventilation that can flow freely between and around the individual cells. Just think it may provide a stable temperature throughout the entire pack ?
Keep on creating !!!
Regards Neil

[Chhanda Das]

Originally Posted by Redex

To me it doesnt feel right to mix 5 batteries with different voltage and capacity ratings and mix new with used batteries. Was there a reason for this?, will it cause any problems over time. I would have preferred to see identical batteries, or is that just my lack of knowledge.

Thank you very much for your kind appreciation of our efforts. Your hunch is absolutely correct here from a theoretical point of view. However, there is a catch. We can indeed mix and match cells (up to an extent) by following a few basic precautions. For example, in case of the maximum voltage, that was maintained at 4.2 volts which is what even the lowest capacity cell can support even though some of the cells can support a maximum voltage of 4.3 volts. Then in the case of the minimum voltage, that was maintained at 3 volts which is what all the cells support even though some of the cells can be drained up to as low as 2.55 volts. Similarly, the charging voltage of 3.7 volts each is supported by all of the cells even though some of the cells can be charged at an even higher/lower voltage. And all of this was done with the BMS and solar charge controller. However, we do need to ensure that the exact same type, age and capacity of the cells are used in the same series of 3 cells to ensure the similarity in each parallel row for the BMS to function properly.

The capacity ratings do not matter too much as long as their summation is the exact same in each parallel row. Also, we never used new cells anywhere here. All the cells are refurbished ones with varying ages. We used refurbished cells since they cost us around INR 9 or GBP 0.085 per AH (ampere hour). In stark contrast, similar "new" cells would have cost us ₹ 100 or GBP 0.95 per AH at the very least. I mentioned "new" because we usually don't get new cells here in India because an overwhelming majority of them are professionally refurbished cells. So as long as we follow the proper process of refurbishment ourselves, we can get good quality cells of reputed brands like Sony, LG, Toshiba, etc at very cheap rates. Still, if one can afford them, one can definitely go for the more preferable identical cells.

My son made the 25 AH battery pack first in February 2022 and later increased its capacity to 35 AH by adding more cells after around 6 months. Touchwood, we have never noticed any issues with it to date despite its nearly 24x7 usage

Originally Posted by Redex

I also appreciate the need for insulation and cooling. Wouldn't it be better to enclose the battery pack in a close fitting container with forced ventilation that can flow freely between and around the individual cells. Just think it may provide a stable temperature throughout the entire pack ?

Again you are absolutely correct here and that is indeed going to be one of our next updates. However, till then the 6015 model fan shown earlier would have to take care of the forced ventilation job.

I hope that I was able to assuage your very valid concerns

[Redex]

Hi,
I am a vehicle mechanic, but aged 67 have started to gain skills in electric vehicles. I am trying to make sense of the battery packs used, layout, benefits of various types, fast and slow charging etc etc. Exactly the problems you are solving. I have completed the first 3 stages of "E.V and Hybrid Electrical Safety". Working with battery packs and capacitors up to 800v was a bit daunting and stressed the need for safety at all times.
I will continue studying and Qualify in "EV and Hybrid Diagnosis and Repair" in March.
Today I am travelling to Goa for 3 months, arriving at the New Mopa Airport. An annual treat, usually 4 months, to escaped from the U.K. winter. I ride my 12 year old "modified" Gen 1 Suzuki Access 125 with my wife riding pillion.
Thank you for your detailed explanations.

Regards Neil

[Chhanda Das]

Update 4 :

Added another DIY battery pack to bring the total battery capacity to 109 Ah

Our initial plan was to increase the capacity of our original homemade lithium battery pack but we ran into a logistical problem there. The thing is that we had procured the initial lot of the interlocking plastic cell spacers for the 18650 cells from a seller on Aliexpress in China just before it was banned here in India. We ran out of these spacers after the previous capacity upgrades done to our DIY battery pack. We got new "Made-in-India" spacers from a local shop here in Kolkata.

As bad luck would have it, these new Indian spacers couldn't be interlocked with the old Chinese ones since the ridges were off by approximately half a millimetre for each one. Additionally, these new Indian spacers had a minor problem. These didn't interlock with each other as tightly as the older Chinese ones.

As a result, we have had to make a new 12v lithium battery pack. All 3 DIY 12v lithium battery packs have their own individual 3S40A BMSs and are now connected in parallel bringing the total rated battery packs' capacity up to (57+35+17) Ah = 109 Ah. Despite the substantial difference in capacities, we have not noticed any issues while charging/discharging them since their voltages are the exact same.

I hope that you enjoyed this. I wish you all happy and safe drives ahead.

[Chhanda Das]

Request for help with your suggestions :

We are so bummed right now. We were looking at a few maximum power point tracking (MPPT solar charge controllers (Victron, Ep ever, Growatt, etc) for our DIY 109 Ah (total) 12 volt lithium battery packs made of 18650 model cells (images above and earlier). But the prices are absolutely shocking here in India. For example, the basic Victron 100 | 20 model which costs less than US$ 90 there costs a whopping ₹14550 = US$ 174.75 here in India on Amazon India. And this is the situation with all US/Canada sourced products.

Then we thought that we would check out the best value-for-money option and started looking at the PowMr branded MPPT controllers from China. The PowMr 1230 model MPPT controller seemed perfect for our use case but alas, Chinese imports by ordinary citizens are totally banned in India. In fact, as you all may already know reputed websites like Aliexpress, Banggood, etc too are either banned here or their products here are extremely cost prohibitive. The few importers who import Chinese MPPT controllers here, rebadge and sell them at around 300-400 per cent markup which makes them even more expensive than the US/Canada sourced products. The few Indian manufacturers who actually make MPPT solar charge controllers within India only make them either for lead acid batteries or for LiFePO4 (LFP) batteries or both but not for Li-ions/LiPos unless the model costs at least 4 times of what it costs in China. LFPs here are priced similarly to the pricing structure of the imported MPPT controllers from China.

In such a woeful scenario, we do not seem to have any reasonable or economically viable option left. As you may have noticed above and earlier, we are using a cheap 30 ampere PWM controller right now and want to get a 30 ampere MPPT charge controller for increased power extraction from my solar panels. The PowMr 1230 model MPPT controller would have been ideal for our DIY project but alas, we can't get it.

What would you all knowledgeable folks suggest ? Would buck converters work ? We can easily get upto a 300 watt 20 ampere rated buck (step down) converter here in India. Our intention is to have/make an economically viable option compared to the expensive portable power stations from the likes of Jackery, Growatt, Bluetti, etc. Thank you in advance

[prasanth DIYer]

Dear @chhanda das, First i would like to appreciate the work you have done so far with your DIY expertise, its quite good work you have done.

Here are some observations i had.

1) Mixing lead acid with lithium in same circuit might have very high negative consequences in regards to safety and life of the pack, the Internal resistance of both are very different and the voltage ranges of these are not inline. i would recommend not not to mix any different chemistry together in a single pack, i would also would not recommend different part number or type of capacity cells in same pack.

2)Adding fan on side of pack for cooling might compromise the IP rating of the pack , as it can attract metal particles and start depositing on the cell welded areas which could lead to shorting.

3)you do not have to worry about temperature of pack if your average discharge rate is less than 0.5C and charging rate is 0.3C and ambient temps are between 15*C to 35*C.

4)I have not seen any balancing leads going to BMS, which is critical for battery pack life.

if you are interested you can DM me , i can share few Pack manufacturing contacts for 12V pack in various chemistries and smart BMS with Bluetooth app monitoring with all the fancy protections.