Battle of the Batteries - E*** vs D*** vs Bargain

This project build and subsequent test of batteries was inspired by shabaz who several times in the last couple years, has brought up the question of which battery is better, E***, D***, or some other less expensive option. I was also eager to use my newly built DC Load unit for something more important than rack decoration.

The project build itself is an adapter that uses a bench power supply to power its internal circuitry, the new DC Load unit to provide a constant current load, and an external battery that is under test. The Adapter allows us to measure from the time when the battery is put under load to the time when its voltage level drops to a preselected level. I have used the circuitry of the adapter to automate this process so I do not have to physically sit and watch the millivolts drop off on the battery while holding a stop watch.

I will begin by posting the schematic of the adapter.

Let's work across the schematic to explain what I am trying to do. U2 is a 12 volt linear regulator that allows the unit to be powered by a non-critical power supply of 15 volts to 21 volts. The regulator is needed to provide stable 12 volt power to the LM339 comparator, a 10K variable resistive voltage divider, and a 2K2 resistor feeding 3 diode drops to produce a 1.5 volt DC supply that powers a small 1.5 volt battery operated clock. Because the current draw is quite low the 12 volt regulator does not need a heat sink. The battery under test provides the voltage to the inverting input of the LM339 comparator. The battery is also directly under whatever load we have decided to dial up on the DC Load unit. There are 18 gauge wires running from the battery clips to the inputs of the DC Load unit. VR1 is a 10 turn 10K potentiometer that provides the non-inverting input of the LM339 with a reference voltage. Turning the VR1 control we can select the voltage which we want to be the termination target of the test. Small Chinese Voltmeters monitor the voltage from the battery as well as the reference voltage on the non-inverting input of the comparator. As long as the battery voltage is above the reference voltage the output of the comparator is pulled low. The output of the comparator is also wired to the gate of an N channel Enhanced MOSFET IRF511. As long as the gate is held low the MOSFET is open and current flows from the 12 volt rail through R5 and produces a 1.5 volt drop across the junctions of the 3 diodes. I am using diode junctions to produce the 1.5 volts for the clock since they act as regulators at this voltage for the clock. This regulated 1.5 volts in turn runs a small 1.5 volt travel clock which is timing our test.  We have set the clock to 12:00 before we start each test. Incidentally one of the 3 diodes is a Schottky Diode which is necessary to make the combined drop of the 3 diode junctions equal 1.5 volts.

When the battery voltage drops below the target voltage set by VR1 the comparator outputs a high and this causes the MOSFET to go into saturation and shorts out the 3 diodes and the clock, effectively stopping the timer. R3 provides a mild positive feedback to the non-inverting input which stops any oscillation between on and off.

This arrangement allows us to pick a current draw from the battery and then measure the time between a start voltage and a target voltage. If we multiply the current by the time we get a rough estimate of the power that was provided by the battery in this time period. If we subject different brand batteries to the same loads we will have the data we need to determine which battery is the best value of energy per dollar cost.

As usual for me, the project was begun with a rough approximation of the schematic posted above and then the circuit was breadboarded for testing and proof of concept..

This project actually had to be breadboarded twice as after building the finished unit there were some anomalies that needed to be addressed and it was going to be easier to experiment on a breadboard than on the finished unit.

The physical construction of the unit begins with finding a project box. For this unit I chose one of my favorites, a plastic Pencil Box sold by Advantus Corp of Jacksonville, Florida.

http://shopadvantus.com/shop/pc/showsearchresults.asp?pageStyle=H&resultCnt=20&keyword=Pencil+Box&ctl00%24ctl03%24imbGo.x=0&ctl00%24ctl03%24imbGo.y=0

I like working with the plastic and they usually look presentable when finished. I use a diamond bur in a high speed drehmel to cut the holes that I need in the plastic. Experience has shown that drilling with a standard bit almost always results in cracking the plastic. For this project I used a totally clear plastic box. Here is a picture of the assembly of the mechanical parts under way.

Here are pictures of the completed unit in operation testing a battery and a close up of the completed unit.

In this last picture you can see that the battery we are testing is at 1.11 volts. It has been providing 500 mA for 56 minutes so far. Our target is actually 1 volt even though the target meter says 0.96 volts. The calibration of the unit was a real challenge and I found that it was not linear. I chose to have the calibration as accurate as possible at 6 volts which left it necessary to fudge a little at 1 volt. In the area of the 1 Volt cut off I must set the target about 60 mV below my desired cut off voltage. 

For the actual test I chose the 2 major brands of battery E**** and D**** as well as a Panasonic battery that I purchased in a 3 pack at the local Dollar Store. The cost break down was Brand E**** $1.20, Brand D**** $1.20 and Panasonic $0.33.  Since my old camera calls for new batteries whenever they drop below 1.4 volts I decided the first leg of the test would be from New out of the package to 1.4 volts with a 250 mA load. This first test was followed by a second test where the load was increased to 500 mA and then timing the battery until it reached 1 volt. The results of these two experiments are summarized below.

Battery Test Report

August 16, 2015

Disclaimer: This battery test was conducted with loose controls and non-precision equipment so the results are perhaps anecdotal at best. For this test “AA” batteries were chosen. The Name Brand  batteries, D**** and E**** that were tested were purchased new at a reputable retail outlet for $4.79 for 4 batteries and were dated 2014 with a guaranteed shelf life of 10 years. The Panasonic Batteries were purchased at The Dollar Store paying $1.00 for 3 batteries. The Panasonic Batteries were not dated but also guaranteed the shelf life to be 10 years. The E**** battery weighs 23.6 grams, the D**** battery weighs 24.1 grams, and the Panasonic battery weighs 23.0 grams.

Test #1 is a 250 mA load from brand new condition until the battery reaches 1.4 volts which is the point at which my camera calls a battery low.

Bat.     START V        LOAD             V UNDER LOAD       TAGET V        TIME to TARGET

E****  1.62              250mA          1.53                           1.4                  20 minutes

D****  1.62             250mA          1.51                           1.4                  22 minutes

PANA  1.67              250mA          1.53                           1.4                  19 minutes

Test #2 is a 500mA load from the previous 1.4 volt level down to a 1 volt level.

Bat.     START V        LOAD             V UNDER LOAD       TAGET V        TIME to TARGET

E****  1.4                 500mA          1.33                           1.0                  1 Hour 45 Min.

D****  1.4                500mA          1.33                           1.0                  2 Hours

PANA   1.4                500Ma          1.33                           1.0                  1 Hour 40 Min.

Battery Output analysis:

In the range of brand new to 1.4 volts at 250 mA all three batteries had similar performance. If 1.4 volts is considered the low level point then time wise we would rank the batteries:

#1 D*** with 22 minutes of usable power.  ( 0.092 Ah)

#2 E*** with 20 minutes of usable power. ( 0.083 Ah)

#3 PANA with 19 minutes of usable power. ( 0.079 Ah)

Reevaluating with cost as a factor we obtain:

#1 PANA with $4.18 per Ah

#2 D*** with $13.04 per Ah

#3 E*** with $14.46 per Ah

In the range of 1.4 volts to 1 volt at 500 mA I found the bulk of the power stored in each battery. This test demonstrates that a lot of battery potential is being thrown away when equipment is unable to tolerate battery levels down to 1 volt. Here is the battery ranking in this range.

#1 D*** with 2 hours of usable power. ( 1.0 Ah )

#2 E*** with 1 hour 45 minutes of usable power. ( 0.875 Ah )

#3 PANA with 1 Hour 40 minutes of usable power. ( 0.833 Ah )

I will now integrate the two test runs and calculate the cost per Ah for each battery:

#1 PANA with 0.912 Ah of power between new and 1 volt costs $0.36 per Ah.

#2 D*** with 1.092 Ah of power between new and 1 volt costs $1.10 per Ah.

#3 E*** with 0.916 Ah of power between new and 1 volt costs $1.31 per Ah.

I can’t speak for anyone else but if you can find an alkaline battery in a discount store that can supply 84 % of the power of a top name brand at 28% of the cost of the same name brand I would call it a no-brainer.

John