Comparing Panasonic A, Panasonic B, Sanyo, and Ultrafire cells March 11 2016, 8 Comments

Cells deconstructed from the top, Panasonic A, Panasonic B, Sanyo FM showing various lengths of electrolyte foils.

This blog article will explore the differences between a sampling of Panasonic, Sanyo, and Ultrafire 18650 batteries.

To do so I will exclusively refer to a paper published here in 2015.

Background

Lithium-ion was first made available in 1991. Since then it has become more apparent that it is the best solution for energy storage. However one problem with lithium-ion and the 18650 battery is the difficulty in getting reliable, honest ratings.

It’s hard to know whether the specifications are true because:

  1. Possible rating inflation from manufacturers
  2. Person rating is not educated on how to accurately test cells
  3. Difficulty obtaining authentic cells and proper equipment

Overview: Important physical characteristics

Cells tested

18650 cells tested

The cells tested from left to right are:

  1. UltraFire TR18650 4200 mAh
  2. UltraFire TR18650 4900 mAh
  3. Sanyo UR18650FM
  4. Panasonic NCR18650A
  5. Panasonic NCR18650B

The most important physical characteristics that distinguish the 18650 are:

  • size
  • weight
  • shape

The 18650 (18mm by 65mm) has evolved as the best trade-off between capacity and charge/discharge ability in terms of these characteristics.

Other 18650 battery characteristics

There are many conditions that will alter the performance of your cells.

  • Max and min voltage
  • Capacity of cell at beginning of life
    • Capacity at different charge / discharge rates
    • Capacity at different temperatures

The max and min voltage can be reduced from the extremes.

For example instead of charging a battery to 4.2 V and discharging to 2.5 V, one might only use a range between 3.2 V and 3.8 V. This can lead to an increased cell cycle life.

The capacity at the beginning will be greater than any proceeding time because of natural cell degradation.

An increase in charge or discharge amperage will decrease the working capacity and life cycle of a cell.

Furthermore, any deviation from the testing temperature of 25 degrees Celcius the cells will result in reduced performance. This is especially true under 0 degrees and above 60.

Other testing variables that should be accounted for when accurately testing cells:

  • cell chemistry
  • capacity
  • formats
  • charging, discharge profiles
  • temperature
  • test set-up

Initial results

P 3100 P 3400 S 2600 UF 4200 UF 4900
Typ. Charge Capacity (mAh) 3070 3350 2600 4200 4900
Min. Charge Capacity (mAh) 2950 3250 2500 - -
Meas. Charge Capacityb (mAh) 2862 (2961) 3166 (3267) 2480 757.1 579.4
– (% of typical) 93.2 (96.4) 94.5 (97.5) 95.4 18 11.8
Meas. Energy Capacity (Wh) 10.19 (10.61) 11.33 (11.78) 9.25 2.74 1.96
Nominal Weight (g) ≤47.5 ≤48.5 ≤48 - -
Measured Weight (g) 44.66 46.7 45.62 32.24 37.98
Nominal Length (mm) ≤65.3 ≤65.3 64.9±0.2 - -
Measured Lengthd (mm) 65.08 67.34 64.96 65.74 65.44
Nominal Diameter (mm) ≤18.5 ≤18.5 18.3±0.2 - -
Measured Diametere (mm) 18.28 18.57 18.28 18.27 18.22
Specific Energy (Wh/kg) 228.3 (237.6) 242.6 (252.3) 202.8 84.8 51.6
Energy Density (Wh/l) 596.9 (621.3) 621.3 (646.2) 542.6 158.7 114.8
Cost per Cell (AU$) 11.48 13.48 5.31 1.45 1.63
Cost/Meas. Capacity ($/kWh) 1125.9 (1081.8) 1189.7 (1144.0) 574.1 530.1 831.7
Cost/Meas. SE ($kg/kWh) 50.29 55.56 26.19 17.09 31.59

 

How the test works

Step 1: Charge and discharge

Charge to 4.2 V at 1A and then taper charge to 50 mAh and terminate.

Then,

Discharge at 1A

  • First 5 cycles discharged to 2.8 V
  • For discharge 6 and 8, the Panasonic cells were discharged to 2.5V (because these cells offer a lower cut-off voltage, allowing the researchers more comparability)

Stage 2: Weight and dimensions

Taking measurement of weights especially enable comparisons of 18650 battery energy density. Weight is a critical constraint in electric vehicles especially as it reduces the km driven per charge.

Stage 3: Impedance

Checking the impedance - cells tested at a frequency range of .1 Hz and 100 kHz.

Step 4: Aging and life cycle of cells.

Testing to conclude after 100 cycles, or by the end of the month whichever happened first - how much the capacity has faded and impedance has changed.

Step 5: Rest and test

Let the cells rest for anywhere between a couple days to a few weeks. And then take a final capacity test in the same conditions as the very first test. That will show how much the cell degraded between first and last test.

Initial testing:

Dimensions

All 18650 cells are actually slightly larger than 18mm by 65mm. The paper found the diameter is 3.2% and the length is 3.6% larger than the 18650 standard. It’s important for those who want to create large battery packs with thousands of cells.

Interestingly, it also skews the “lithium-ion energy density measurements” that everyone uses as the size is a critical component of the formula and seldom is it accounted for.

Capacity

Battery mAh Cycle Av. Capacity Reduction per 100 Wh Throughout (%)
Panasonic 3100 partial 0.43
Panasonic 3100 full 0.7
Panasonic 3400 partial 0.57
Panasonic 3400 full 1.51
Sanyo 2600 partial 0.75
Sanyo 2600 full 0.66
Ultrafire UF4200 partial 3.71
Ultrafire UF4200 full 2.24
Ultrafire UF4900 partial 16.6
Ultrafire UF4900 full 67.9

 

Panasonic cells achieved over 93% of their advertised rated charge capacity. The Panasonic cells did not reach their full capacity because they did not use the proper cut-off voltage (2.5V) and they used a “fast” charge current of 1A where they might use .5A or lower. Furthermore optimizing the temperature would have likely allowed them to reach their full potential.

Ultrafire cells were between 12% and 18% worse. There is no surprise here.

Full Cycle Life

18650 full cycle discharge test
Note, the solid lines are for full cycles of 4.2 V to a cut-off voltage of 2.8 V. The dashed lines are partial cycles of 4.0 v to 3.2 V.

View full-size chart here.

There is clear downwards aging trend after 50 - 130 cycles

Partial Cycle Depth of discharge study
partial discharge test
Again, the dashed lines are partial cycles, and the solid lines are full cycles. Note how crazy the Ultrafire cells cycle.

View full-size chart here.

Charge and discharge from 3.2 to 4.0 V, to decrease the depth of discharge. This should offer an increase in the cycle life of the cells.

Note, the Depth of discharge, or (DOD) differs from the state of charge (SoC) by being an alternate way of measuring charge, albeit somewhat reversely. When one goes up, the other goes down.

The comparison between cells fully cycled and cells partially cycled is interesting because it can help you decide between two options:

  1. To buy expensive high capacity cells and do partial discharge
  2. Or buy cheap low capacity cells and do full discharge

If we look at the chart we can see that the Panasonic cells to degrade slower when having a decreased voltage range. Unfortunately, the Ultrafire cells do something not expected for 18650 batteries - they degrade much faster.

Unfortunately, this makes the data (because of the cells) unreliable - calling for more cells to be tested for conclusive comparisons.

However if you just focus on the Panasonic and Sanyo cells you can see by limiting the depth of discharge there is some improvement in the life cycle.

Impedance Study

Panasonic Cells

  • The main difference was charge transfer capacitance - it decreased by two orders of magnitude
  • At the same time, the electrode diffusion capacitance increased by two orders of magnitude.

What is charge transfer?

Now most readers have no idea what charge transfer capacitance is, but if you are curious you want to know. It begins with something called the charge transfer coefficient.

In lithium-ion cells, charge transfer is defined as the fraction of over-potential that affects current density.

Overpotential is a measure that directly translates to voltage efficiency (using more or less energy than thermodynamics require). Inefficiencies are seen as a loss of usable energy in the form of heat.

Sanyo Cells

The same thing happened to these cells (charge transfer capacitance decrease, diffusion increase).

The difference between the Sanyo and the Panasonic cells are an increase in ohmic resistance (how much is the cell opposing the flow of current). This is due to the battery chemistry type, notably the electrolyte formulation which is outdated.

Ultrafire Cells

The same trend continues here.

The big difference is the Ultrafire cells had a much higher increase in ohmic resistance. This could have been predicted - these cells are terrible.

Paper Conclusions

Initial capacity testing found the Panasonic and Sanyo cells performed within 8% of their rated capacity.

Panasonic and Sanyo cells:

  • Capacity: Within 8% of rated capacity
  • Life Cycle: 2-6% degradation after 50 - 80 cycles (700 Wh)
  • Resistance: Low rise in ohmic resistance

Ultrafire:

  • Capacity: Less than 20% of rated capacity (one cell died)
  • Life Cycle: 6% (300 Wh) - 20% (30 Wh) degradation
  • Resistance: High-rise in ohmic resistance

More on the Ultrafire cells

The authors of the paper found that actually inside many Ultrafire batteries are smaller batteries. It’s not Russian dolls, it’s cheap 18650 batteries!

Counterfeiters purchase blank 18650 steel tubes and then don’t fill them up all the way. It’s the classic potato chip bag bait-and-switch.

The electrolyte foils were found to only comprise 73% to 74% of the space the Panasonic or Sanyo cells did.

CT scan of Ultrafire batteries.

Last picture - the Ultrafire 4200 from this article cut in half.

Source:

Vyroubal, P., et al. "3D Modelling and Study of Electrochemical Characteristics and Thermal Stability of Commercial Accumulator by Simulation Methods." Int. J. Electrochem. Sci 11 (2016): 1938-1950.