18650 DIY powerwall 48v

ALL NEW - Battery Finder Search for 12/24/36/48v or by capacity www.batteryfinder.net
New & used Batteries, Solar, accessories, LifePo4 cells & more 5% Coupon "Powerwalls" www.batteryhookup.com
kje said:
Liitokala has max 1A
Click to expand...

This is true only for charging. Unfortunately discharge is max at 500mA so not a good test of V drop.
Also if you depend on the LiitoKala for IR measurements it is not accurate at all. It is a makeshift DC IR reading extrapolated with some programming trickery and tries to take into account the contact resistance of the sliders. The only way to accurately test IR is with a 1kHz AC impedance meter such as the YR1035 or the RC3563 or equivalent. Check out 4 wire Kelvin measurements.
Also look at manufactures cell spec which indicates all measurements are to be performed with a 1kH Impedance meter
"The impedance meter with AC 1kHz should be used."
Most will state that initial internal impedance measured at AC 1kHz after standard charge. ( Initial internal impedance is ≤ XXmΩ )

Please see an extensive review of the LiitoKala here:https://lygte-info.dk/review/Review Charger LiitoKala Lii-500 UK.html

Quote from the review
"Fast test Li Ion (Discharge & charge)
This test measures the charged capacity. It might be a fast test, but it takes about 10 hours!
The possible discharge currents are 500mA and 250mA, they are selected depending on charge current:
1000mA and 700mA charge current will select 500mA discharge current.
500mA and 300mA charge current will select 250mA discharge current."

Also Vdrop is measured in a different way You measure the voltage of a circuit at rest i.e. a battery or cell with no load.
You add a load such as a 1Ω resistor and measure the voltage of the cell during the load, that is your voltage drop. Vdrop is not measured over time but at the moment of the application of the load.
Example
Cell at rest Voltage 4.0V
Cell with 1Ω loads3.74V
Vdrop is 0.26V

Wolf
 
Wolf said:
This is true only for charging. Unfortunately discharge is max at 500mA sonot a good test of V drop.
Click to expand...
Yes, I forgot. I remember now, from the manual it's 500mA discharge at max.

Wolf said:
Also if you depend on the LitoKala forIR measurements it is not accurate at all. It is a makeshift DC IRreading extrapolated with some programming trickery and tries to take into account the contact resistance of the sliders. The only way to accurately test IR is with a 1kHz AC impedance meter such as the YR1035 or the RC3563 or equivalent. Check out 4 wire Kelvin measurements.
Also look at manufactures cell spec which indicates all measurementsare to be performed with a 1kH Impedance meter
"The impedance meter with AC 1kHz should be used."
Most will state that initial internal impedance measured at AC 1kHz after standard charge. ( Initial internal impedance is ≤ XXmΩ )
Click to expand...
Yes I know, you told me this before. That's the reason why all my cells is tested with the RC3563 😊 I never use the IR numbers on the Liitokala.

20241201_233046.jpg


Wolf said:
Also Vdrop is measured in a different way You measure the voltage of a circuit at rest i.e. a battery or cell with no load.
You add a load such as a 1Ω resistor and measure the voltage of the cell during the load, that is your voltage drop. Vdrop is not measured over time but at the moment of the application of the load.
Example
Cell at rest Voltage 4.0V
Cell with 1Ω loads3.74V
Vdrop is 0.26V

Wolf
Click to expand...
I see, it's measured at the moment of the application of the load. Thank you. My mistake, I waited too long.

Yes, I know I compare the resting voltage to voltage under load.

What charger/software do you use to make discharge curves?
 
Last edited:
kje said:
What charger/software do you use to make discharge curves?
Click to expand...
I have 3 different ways.
1733632228707.png
One is the SKYRC MC3000 The 3 units on the right https://www.skyrc.com/MC3000_Charger
1733632019206.png
The second way is the MegaCell Chargers 4 units
1733632140682.png
The third is my CBA V100V Pro https://www.westmountainradio.com/product_info.php?products_id=cba5
1733632272726.png1733632104616.png
The CBA V no matter what version is the standard in accuracy and the price reflects it.
When I want to know how accurate a tester is I will compare it's results with this unit.
Wolf
 
Wolf said:
I have 3 different ways.
View attachment 32773
One is the SKYRC MC3000 The 3 units on the right https://www.skyrc.com/MC3000_Charger
View attachment 32770
The second way is the MegaCell Chargers 4 units
View attachment 32772
The third is my CBA V100V Pro https://www.westmountainradio.com/product_info.php?products_id=cba5
View attachment 32774View attachment 32771
The CBA V no matter what version is the standard in accuracy and the price reflects it.
When I want to know how accurate a tester is I will compare it's results with this unit.
Wolf
Click to expand...
Nice! I'll check them out. 👍
Thank you Wolf.
 
The last half year I've charged the power wall to 4.05v and wonder if I benefit to a more conservative float voltage at 4.00v in the long run? Or is it not so much difference in this voltage range between 4.00v and 4.05v?
 
kje said:
The last half year I've charged the power wall to 4.05v and wonder if I benefit to a more conservative float voltage at 4.00v in the long run? Or is it not so much difference in this voltage range between 4.00v and 4.05v?
Click to expand...
Early on (7 yrs ago) I ran across this page - https://batteryuniversity.com/article/bu-808-how-to-prolong-lithium-based-batteries - and on this page are some charts that suggest the relative effectiveness for Lithium-ion (as apposed to LifePo4) to trade low stress for more cycles.

In particular, I built my system around this specific chart on that page. The goal was to operate at the purple 75-45% SoC line.... and get 7000+ cycles! Exciting!
1738683496825.png


In reality, I've achieved a 35% lifetime average DOD but I allow a wider operating voltage range of 4.0v hi and 3.54v low. A lot of the year (maybe 2/3) I don't have the perfectly matched PV/Load combination to charge/discharge over the whole range - so even though I allow a wider voltage range I still have a lower 35% DOD average as in winter the end of the lower voltage range is prominent and in summer the higher end.

I post cycles now and then but at 300 (or so) cycles per year it takes A LOT of years to get 3000 or 6000 cycles. But I'm still hopeful that chasing this chart will cause the battery to actually reach 7000 cycles with significant capacity. :)

I've come to believe that LOW STRESS in general (modest ma/cell, modest voltage ranges, modest temps) correlate with with longer life for older 18650/Nissan (non-LifePo4) chemistries. But I'm only at 2,000 cycles on my powerwall - I would say I need to reach in the neighborhood of 4,000+ cycles (another 7 years) to start any kind of celebrations.
 
Last edited:
OffGridInTheCity said:
Early on (7 yrs ago) I ran across this page - https://batteryuniversity.com/article/bu-808-how-to-prolong-lithium-based-batteries - and on this page are some charts that suggest the relative effectiveness for Lithium-ion (as apposed to LifePo4) to trade low stress for more cycles.

In particular, I built my system around this specific chart on that page. The goal was to operate at the purple 75-45% SoC line.... and get 7000+ cycles! Exciting!
View attachment 33010

In reality, I've achieved a 35% lifetime average DOD but I allow a wider operating voltage range of 4.0v hi and 3.54v low. A lot of the year (maybe 2/3) I don't have the perfectly matched PV/Load combination to charge/discharge over the whole range - so even though I allow a wider voltage range I still have a lower 35% DOD average as in winter the end of the lower voltage range is prominent and in summer the higher end.

I post cycles now and then but at 300 (or so) cycles per year it takes A LOT of years to get 3000 or 6000 cycles. But I'm still hopeful that chasing this chart will cause the battery to actually reach 7000 cycles with significant capacity. :)

I've come to believe that LOW STRESS in general (modest ma/cell, modest voltage ranges, modest temps) correlate with with longer life for older 18650/Nissan (non-LifePo4) chemistries. But I'm only at 2,000 cycles on my powerwall - I would say I need to reach in the neighborhood of 4,000+ cycles (another 7 years) to start any kind of celebrations.
Click to expand...
Talk to you after 7 years and 4000+ cycles. 😂

Well I now lowered the floating voltage to 4.00v (56v) and bulk voltage to 4.007v (56.1v).

I'm still slowly working on battery #2. I got a lot of solid core wire from the scrap yard and am removing the isolation. They are in different gauges so I bought a new digital caliper (the old was broken), and use https://www.wirebarn.com/combined-wire-gauge-calculator_ep_42.html to calculate how many copper wires in each bus bar.

I'm also half finished pre-tinning cells. 800ish out of 1680 cells. This time I made a fan system to avoid the soldering fume. I'm also reusing the old ebike cellholders. 😊

20250205_115654.jpg
 
New power record from my solar panels: over 2kw ☀️😀 Exciting!

20250212_113835.jpg


Everything working flawlessly charging my 18650 battery #1 at 13kwh. The powerwall is powering my heatpump.

One cellpack is lower than the rest, but service is simple: desolder the cells in the cellpack, wait a day or two, then take out the bad cell(s) and change with a good one. 👍
 
You must log in or register to reply here.
Back
Top