Battery tester Schematic and code for Brian Grimm

Dave W

2 files attached

This is the best way to get this info to Brian, I think.· For those of you who are interested, I am using a BS2 to discharge AA NiMH cells to the 1 volt level and measure the time it takes to do that.· This is so that I can determine if the batteries in a given group of 4 are pretty much equal in their ability to deliver power.· I use them in a rather power hungry digital camera and some of my batteries seem to die too soon.· I have built and tested this circuit and it works fine.· My next effort will be to increase the number of cell stations to 4 so that I can do 4 cells at a time.· I don't believe that the BS2 can drive that many relay coils, so I will need to add drivers for each relay.· The Stamp should be able to run 4 independent timers and give me the results for each cell.· That approach cuts the total time over the current design by 75%.

If anyone has any suggestions for improving my approach, I'd like to hear them.· Keep in mind that I don't want to spend money to do it, at least little money ($15 or so).· I have 2 Basic Stamp Homework boards which belong to a local middle school.· I lead a technology club for 6th through 8th graders and these boards were purchased for them to experiment with.· This project is also helping me to learn the Stamps.

Dave W

Brian Grimm

Thanks Dave!· I appreciate all the effort you put into the .pdf.· Assuming my wife doesn't find any additional jobs for me this weekend I'll start playing with it.

Thanks again,

Brian

Beau Schwabe

Do you absolutely need to drop 1 Ohm across the battery?
Could you go with a 2 Ohm resistor to lower the current demands within the relay specs?
I'm afraid that the relay contacts could "spot weld" themselves under the right conditions.
Have you considered a MOSFET transistor with low on resistance? <-- about the same cost as a relay
If you must use a relay, be sure to place a reverse biased diode across the relay coil - This will help protect the Stamp from back EMF

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔

Beau Schwabe
Parallax IC Layout Engineer

Dave W

Beau,

Thanks for the·suggestions.· I'm a retired product developer, both analog and digital designs.· I know about protecting both against arcing of contacts and back emf from the coil.· I was just too lazy to do it, but now you have me worried, especially about protecting the stamp.

The reason for the 1 ohm resistor is that I want to simulate the load of a digital camera as closely as possible.· Those buggers are power hogs, especially during exposure and recording to memory.· To be completely realistic, I should use a pulse load, but, again, that is too much trouble and will surely kill the relay contacts (which are rated for 1A by the way).· To illustrate just how much current my camera draws, I'll relate a story to you.· The camera is a Minolta Dimage 7i.· Normally, the sales people in the camera store demonstrate the camera to the customer by opening a fresh pack of alkaline batteries and popping them into the camera.· The saleman in Ritz Camera told me that when he did that with the D7i, the camera took one picture and reported dead batteries.· Apparently, the current draw is so high that the battery voltage is pulled below the minimum level for the camera and it shuts down.· NiMH batteries can put out a lot of current.· Even so, my latest sets of 2200 mAh batteries quickly reach the warning level in the camera, but then seem to last a fairly long time.· My guess is that the internal impedance of these particular batteries is slightly high, even fully charged.· I have 2 sets of them and both behave the same.· Older sets of lower capacity batteries don't do this.· This is why I built the setup for discharging them.· The 2200 mAh batteries are lasting 117 minutes with the 1 ohm load.· So the batteries are okay.· The camera must actually exceed 1 amp peak current to show a nearly exhausted battery.· Incidentally, I have a lot of voltage drops in my circuit, so the voltage across the 1 ohm resistor is about 950 mv or about a 950 ma load.· When I rebuild this thing, I'll use a lower guage wire, short runs, and solder everything.· Right now, I have everything plugged into the Stamp breadboard.· Not the best way to do it, I know, but it is quick and dirty.

In my career as a design engineer, I never used power MOSFETS.· Can they switch an amp with a low saturation voltage?· What is the gate voltage required to drive them?· The relay contacts in my circuit drop 55 mv which is too high.· I'm thinking of getting a higher current (lower contact resistance) relay.· I suppose I could use a MOSFET if I knew that the saturation voltage was constant from FET to FET (I want to use 4 discharge positions simultaneously).

Can you give me part numbers and sources for the MOSFETs?

Dave W

Dave W

Beau,

I've been thinking about the FET idea to replace the relay.· It occurred to me that I can even elliminate the 1 ohm resistor if I can design a constant current drain using a FET, or at least a fixed resistance.· Here are the specs of a Radio Shack FET which sells for $2:

N-Channel Field Effect Transistor
(276-2072)················· Specifications··········· Faxback Doc. # 31806

Absolute Maximum Ratings

Drain Source Voltage.................................................100 V
Drain Gate Voltage...................................................100 V
Continuous Drain Current...............................................4 A
Pulsed Drain Current..................................................16 A
Gate Source Voltage...................................................20 V
Max. Power Dissipation................................................20 W


Electrical Specifications

Gate Threshold Voltage............................................2 to 4 V
On-state Resistance.........................................0.6 Ohm (Max.)
Forward Transductance................................................1 mho
Input Capacitance............................................150 pF (Max.)
Output Capacitance...........................................100 pF (Max.)
Input and Output Capacitance calculated at 1 MHz

Since this transistor can dissipate 20W, I should be able to use it as·the load and kill two birds with one stone, so to speak.· Can you refer me to any application notes which show how to do what I want to do.· I'm afraid that I will need more voltage than a single cell battery puts out to operate such a circuit, but I'd like to look into it.· I want to avoid drawing high current from anything other than the battery under test.· If I had to use the 5 volt supply to drive the load, I'd be dissipating far too much power.· By the way, notice that the max saturation resistance of this FET is 0.6 ohms.· That's way more than any relay and is the reason for my thinking about just using the FET as the load.

Any suggestions?

Dave W

Beau Schwabe

DaveW,

I would keep the heat away from the MOSFET... in other words use a MOSFET with a lower RdsOn and sacrifice a couple of cents on a resistor.

In a former life when I used to work at a prosthetic facility in R&D, I had a hobby of building fast electric RC cars.· The speed control circuitry for
forward and reverse on an RC car is not entirely different than the control circuitry used in a prosthetic arm for moving up or down.

The MOSFET of choice that I used was an ECG2395 or STP50N06...· These have an RdsOn of about 0.022 Ohms.· There are lower RdsOn MOSFETS,
but you pay the price, and it is usually very steep.· I remember the STP50N06 at just under $5.



STP50N06 Datasheet: http://www.ortodoxism.ro/datasheets/SGSThomsonMicroelectronics/mXsrurs.pdf

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔

Beau Schwabe
Parallax IC Layout Engineer