FMA Direct : LiPo Handbook

HOW TO EVALUATE AND SELECT CELLS

Important definitions

1. Ohm’s law: E = IR where E = volts, I = amperes of current, and R = resistance in ohms.

2. R/C models require watts to fly. If you know the wattage required, you have a leg up.

3. Power = watts. For reference, one HP = 750 watts. 1 V X 1 amp = 1 watt.

4. Watts = volts X amperes, e.g. 12V X 10 amps = 120 watts.

5. Amperage is determined by the voltage a cell can maintain at a given discharge rate and the resistance of the circuit.

6. An important part of the resistance of the circuit is the internal resistance of the cell.

7. The wiring, motor, and ESC add resistance to current flow.

8. We are concerned only with cell internal resistance.

9. The length of time a cell can deliver a given wattage is called watt-hours

10. Battery capacity is defined in milliampere – hours (mAh) or ampere-hours (A-H).

11. Ampere-hours X volts = watt-hours.

12. Discharge curves are used to display the voltage of a cell as a function of elapsed time or capacity delivered.

13. Capacity is defined as the mAh or AH a cell can deliver when depleted from full capacity to a specific cut-off voltage at increasing current draw (amps).

14. Dividing the capacity in mAh or AH that a cell can deliver at a given average current gives the run time the cell can deliver.

What defines cell performance

1. What distinguishes one cell from another

• Life cycle cost

• Performance and reliability

• Pack weight, not cell weight

• Pack complexity and reliability

• Vendor knowledge and service

• Backing of a major supplier and a dealer network

• Safety

2. What should not distinguish one cell from another

• Old - wives tales

• Unsubstantiated vendor claims; if data isn’t available, it may not be good or he may not have it

• Anonymous claims and critiques from someone you do not know

Your ultimate success and enjoyment of electric models can be enhanced by taking the time to study and understand the principles discussed below. If a model builder elects to try to start and run a glow engine with no understanding of the engine or the fact that he can ruin a new engine with one lean run or by loading it with too big a prop, then he can well expect to ruin a perfectly good engine that may have cost him upwards of $150.00. By the same token, if one refuses to learn how to use Li Pos properly, then we can expect to see people report in forums that they ran a 3S1P pack 1020 mAh cells in a Belchfire Heli and the cells swelled after two runs. They don’t know that that heli draws 25 amps. Same thing.

1. Basic cell performance is defined by internal resistance.

2. Internal resistance determines cell voltage depression. Voltage depression is directly proportional to internal resistance per Ohm’s law.

E = IR

Where E = voltage drop (depression)

I = current flow in amps

R = cell internal resistance

As an illustration: if a specific cell has internal resistance of 3 milliohms (.003 ohms),

For a 60 amp load, E = 60 X .003 = 0.18 volt. Another cell that has internal resistance of 30 milliohms would have depression of E = 60 X 0.03= 1.8 V at the same current.

3. The amount of heat that the cell must dissipate is also directly proportional to the internal resistance of the cell but is proportional to the square of current flow; i.e.,

P = I X I X R

Where P = power dissipation in watts

For the same illustration, the low R cell has to dissipate 60 X 60 X .003 = 10.8 watts while the 30 milliohm cell has to dissipate 60X60 X .03 = 108 watts. This is significant.

4. If a dead short were placed across a cell and the voltage measured across the cell, voltage would drop per the above equation. The drop is called voltage depression. The practical measurement is to load the cell in a square wave at 1 KHz. The data is recorded as AC impedance at 1 KHz.

5. A discharge curve will be illustrated a bit later. Because voltage depresses with increasing load, the cell reaches cut-off voltage earlier as load increases, thus, reflects capacity depression proportional to voltage depression.

6. The amount of wattage a cell can deliver per unit weight is labeled specific power or power density.

7. The watt-hours a cell can deliver is labeled specific energy or energy density and defines how long a cell can sustain a load.

8. Discharge curves plot cell voltage against either time or capacity delivered.

9. Voltage Vs. time can be converted to capacity by multiplying by the time elapsed at constant current drain. (Amps X time in hours = AH)

10. Capacity can be converted to run time by dividing capacity by amps. (AH/Amps = Hours)

11. Cell temperature is driven by internal resistance as described earlier. Cell life is greatly shortened by high temperature.

How to determine the propulsion your airplane needs

1. Determine the wattage needed to fly the aircraft intended. The detailed aerodynamic analysis required is not within the scope of this document. However, if Electrocalc or Motocalc can be accessed, those programs will help. If the designer or a review has defined the wattage required, you are half way there. Rest assured; this is a question our service and sales departments receive many times a day!

2. If you have little information, at least find the weight of the airplane, use the table below for a ball-park figure, then refine the results later.

- Indoor and micro flyers 25 w/lb

- Park and backyard flyers for Speed 280 to Speed 400 25 to 50 w/lb

- Speed 400 up to brushless sport-scale trainers 50 to 75 w/lb

- Maneuverable, higher performance sport-scale & 3D 75 to 150 w/lb

- AMA/FAI pattern competition & F5A gliders 150 to 250 w/lb

3. The low end of each category is flyable and sedate.

4. The higher end of each category is much hotter and demands skill.

5. Within each category, wing loading has an effect. Wing loading is the weight divided by wing area, usually lbs per square foot. Higher wing loading imposes higher power loading to fly properly and pushes the requirement toward the upper end of the scale.

6. Once the power loading (w/lb) is known, determine what motor and ESC can deliver the needed wattage. Both motor and ESC specs will give a range of voltage, usually as cell count. Take care not to mistake Ni Cd cell count for Li PO. One Li Po cell = three Ni Cd or Ni MH cells. CAUTION: Some ESCs specify upper voltage limits and upper amperage limits. In some cases both may not be permitted simultaneously. Check the specifications and demand to know.

7. Pack requirements are by peak wattage that may be demanded only for a few seconds each flight, while overall pack capacity sets flight time. Li Pos are almost never run at peak load for a whole flight. If they are, the pack is under-designed and may be damaged.

8. A rule of thumb for success and long life is to allow for a 30% margin. Because Li Pos are so light, this is not really a burden.

9. Divide peak watts by planned nominal voltage of 3.7V X the number of cells in series, e.g. a 3S pack is 11.1 V, nominal to determine peak amps..

10. If you can, measure the amperage in a short static run using a watt - meter or an ammeter to confirm the estimate. NOTE: This can save you the cost of a pack and is well worth - while.

11. The capacity required depends on desired flight time per charge. You may elect to have a pack with a lot of capacity and fly multiple flights per charge or use a smaller pack and recharge between flights.

12. The standard most recommend with modern 4^th generation cells is the “ten minute flight”. Ten minutes is about the average flight time for glow or electric. Most club fields have some limit on flight time after which one must relinquish the frequency pin.

13. FMA/Kokam HDR packs charged with our cell balancing chargers may be recharged in 20 minutes and this makes the “10 minute pack” very practical. Two packs then allow almost continual flying.

14. The capacity required = desired flight time in hours X average current drain in amps.

15. The capacity for the 10-minute flight = 0.167 X average current in amps.

16. You may enter peak current and voltage in LI POCALC II, available in the Li Po Compendium at www.fmadirect.com. LI POCALC II will provide the specification for the pack you need. Then reduce throttle setting to approximate the average current anticipated to see what flight time will be.

The chart below summarizes the above step-by-step procedure.

A look at performance parameters in greater depth

It all begins with the discharge curve.

1. A lot of discharge curves are published based on use of the West Mountain Computerized Battery Analyzer (CBA). The above is typical of curves that can be plotted with that excellent little device. However, the discharge curve means little unless you know how to use it to compare cells or to select the pack needed. Imagine trying to stack up a half dozen separate sets of curves for comparison; almost impossible. So, most get published with no real explanation of what they mean.

2. There are some things on the curve itself that can be interpreted directly:

a. The amount that the average voltage drops as load is increased is called voltage depression. It is the difference in average voltage from, e.g. the red to the yellow curve to the green, etc. A better way of displaying will be described later. Most casual students of Li Pos have a sense of the importance of voltage depression.

b. The “flatness” of a given curve is an indicator that pack voltage will not keep falling rapidly so that throttle adjustments are needed continually. By the same token, when voltage does drop, it is rapid and cut - off can occur abruptly.

c. Capacity depression is reflected in the spread between end points at the right end as current load increases The KOK 3200 illustrated is a particularly “stiff” battery, i.e. it loses little voltage or capacity as load is increased.

d. Note that casual observation of the curves above as a standalone bring little sense of how good the cell is. Comparison with other cells in its family below will illustrate the importance of being more thorough.

3. The four key data points listed form the basis for a great deal more analysis that can be done if one is interested in knowing how to do a true evaluation. The key data may be entered in the spreadsheet below for analysis.

The data presented in this handbook are as of June 2005. Li Po performance changes almost daily and new cells and suppliers are arriving. Careful study of this section will permit you to incorporate and evaluate claims made.

Simple math for calculating performance

The following equations are built in to the EXCEL spreadsheet to calculate performance parameters once the basic data are entered.

1. Run time in hours = capacity in AH/amps discharge (X 60 to get minutes).

2. Average watts delivered = discharge rates in amps X average volts.

3. Watt-hours delivered = average watts X run time in hours (X 60 for watt-minutes).

4. Specific energy (energy density) = watt-hours/kilogram = watt-hours/weight in kilos (multiply kilos by 1000 for grams).

5. Power density = watts/ weight in kilograms (multiply kilos X 1000 for grams)

6. %-capacity delivered = capacity measured in AH or mAh at lowest discharge rate divided by capacity at discharge rate of interest. Note: If the manufacturers rated capacity is used, the % may exceed 100% at the lowest discharge rate measured. For best results, use the international standard 0.2C as the initial discharge rate.

7. Number of cells required in each parallel pack = maximum expected discharge rate / C - rating for the cell of interest, e.g. a 10C, 2100 mAh cell would have to be paralleled 3 times to carry a peak load of 63 amps.

8. # of cells in series desired = voltage permitted by the motor and ESC of interest/ 3.7 V per cell.

9. Total # of cells in the pack = Eq 7 X Eq 8.

10. Pack weight = cell weight X # of cells in the pack X 1.05 (Estimate 5% for wiring, interconnects, and heat shrink or protective covering)

Note: If the cell has high voltage depression, you may want to use capacity delivered at the expected peak discharge rate instead of the manufacturer’s nominal capacity.

Using EXCEL CHART WIZARD, any of the columns may be plotted against each other as desired. Some key curves are illustrated in the sections that follow.

The first real indicator: voltage depression

1. From the spread - sheet: Plot cell average voltage as a function of discharge amps.

2. More depression means that the cell gets hotter as amps increase. See discussion below.

3. High voltage depression precludes achieving high discharge rate.

4. Voltage depression impacts other parameters as shown in charts below.

5. The only way to go to higher current drain is by paralleling packs.

The cost of light - weight is high internal resistance and overheating

Rather than clutter the chart of voltage depression, a separate chart shows that the final discriminator is the temperature rise in cells. Thus, it is important to know what the temperature curve for any cell is and that sometimes is not shown in test data. Data from tests made in April 2005 of a number of cells in the 2AH range by a test lab in the UK is shown below to illustrate. Please note the date and do not assume that such figures will remain the same later; be sure to check. Things do change with time. The solid lines with markers are voltage Vs amps. The dashed lines are temperature measured during each run. Each line represents a minimum of three discharge rates. That is, the KOK 2000, e.g., was run at 2 amps, 10 amps, 20 amps, and 30 amps continuously from full charge to a cut - off voltage of 3V.

Experience accrued as the FMA Li Po systems described earlier were developed has shown that the determinant of cell life and performance is the temperature the cells reach during discharge. A cell run continuously at the maximum C rate will lose capacity to 80% in as little as 25 cycles. Properly managed, the same cells run with bursts of < 10 sec and average current of half the maximum allowable discharge rate can last 500 cycles.

1. The impact of light - weight and high internal resistance limit cell performance severely.

2. Cycle life may be very short for light - weight and high internal resistance cells.

3. While some light - weight cells are claimed to be capable of 10 to 12C discharge rate, it is clear that they cannot be used safely above 10 amps continuous.

Lower temp and HDR capability = long service life

1. The race car current profile is tougher than for airplanes.

2. That profile was recorded for a number of races.

3. Kokam is cycling cells to validate life cycle.

4. The results after 200 cycles shows 92% of capacity.

5. 80% is international standard cell life.

6. The projection is for 500 to 600 cycles.

7. Packs that run at high temperatures won’t do that.

8. Kokam packs now in use in F3A airplanes at up to 60 amps run < 125 deg F.

Capacity depression mirrors voltage depression

1. Plot the data from the spread-sheet capacity Vs discharge amps to see the effects on capacity depression. Now, it is much easier to see and to compare when all are on one plot with little extraneous information.

2. Just because a cell is heavier does not mean anything by itself.

3. If a light - weight cell is loaded lightly, performance can look good.

4. Paralleling packs curtails any advantage of lighter weight.

5. Paralleling adds pack weight and cost and reduces reliability. For example, a 3S 8P pack for larger airplanes has 24 cells and 48 solder joints compared to an 8S1P pack. Reliability of such a parallel pack is, by definition, going to be 1/3 that of the 8S1P pack that can do the same job.

A design example illustrates the effectiveness of the method

1. This is the example of an 11 - Lb F3A pattern airplane such as Icepoint illustrated earlier. The method is equally applicable to any example.

2. The givens:

· F3A Airplane weight can - not exceed 12 lbs, battery and all.

· The usual 1.8 cubic inch glow engine produces a stated 3 HP when the pilot can get it to run and peak out reliably, so that is the target.

· 3 HP = 3 X 746 watts = 2238 watts.

· It is desired to produce a minimum of 2238/11 = 203 watts / Lb of airplane

· Airplane weight sans the battery pack is 8.5 lbs.

· The brushless motor and speed control are able to handle up to 10S = 37V nominal and up to 60 amps. NOTE: The motor manufacturer limits voltage to 10S, so that sets the upper design limit. Thus, the 12S1P design proposed could not be used to reduce current draw and extend flight time for practice.

· The pack voltage was chosen to be as high as permissible to keep current consumption as low as possible.

· Average current consumption was estimated at 25 amps, including short bursts of 60 amps or more during acceleration from vertical maneuvers.

3. Primary analysis:

· Watts available from the spread - sheet is calculated from average watts = average volts X fixed current; thus voltage depression has an effect.

· 2236 watts is the peak design point that will be demanded of the propulsion package. This is the “burst capability” that is demanded for but a few seconds periodically.

· The pack is sized by the wattage demanded.

· Peak amperage is 2236/37 = 60 amps.

· The KOK 3200/20C can deliver up to 64 amps continuously with adequate cooling and can deliver short bursts up to 100 amps as it does in the FMA Scorpion packs for race cars.

· A 10S1P pack of KOK 3200/20C meets the peak power requirement.

· The KOK 3200/20C delivers 98% capacity at 29 amps, the average current draw. Thus, flight time = 3.136 AH/ 25 amps ~ = 7.5 minutes and enough to complete an efficient flight pattern.

· Weight of the pack is 1.88 lb. Thus; total airplane weight is 10.4 Lbs. He savings in current by reducing all-up weight by 0.6 lbs was not estimated. It was preferred to make an empirical determination by test flying.

4. Alternatives

· Three alternative pack configurations are examined in the Watts Available chart above.

· The results are self - explanatory.

· It takes a 2P pack of KOK 2000/15C to deliver the peak 60 amps.

· It takes a 3P pack of TANIC or TP cells to meet that requirement.

· The two latter cells will be operating too close to the temperature limit to have long life.

· The lightest weight pack is the KOK 3200 followed by the KOK 2000/15C, the TANIC, then the TP.

· Note that voltage depression, even at 20 amps, drives the 2000 mAh packs below the reference KOK 34200 line. See discussion below.

5. How to do an analysis

· Find out the wattage or horsepower needed

· Define the voltage planned

· Determine how many packs in parallel are required to meet the peak amps

· Calculate the weight and cost of the pack required

· The data in this curve can be replicated for any range of cells desired, using the spread - sheet and plotting from Chart Wizard.