- Hi Rick,
I guess soon I will need to work on it since I sell lately many DC to AC inverter/charger sets for house usage. I had to include in their design all possible protections and regulations since only foreign products don't have to be guaranteed while mine are supposed being 100% reliable for life and stand all conditions :)
The main problem my assistants have is convincing a consumer that his battery is weak. Even a car battery having relatively high internal impedance gives 12V but cannot be charged properly even at C/10 current or provide its expected load power. The consumer sees only the good looking of his battery and is proud that he has a voltmeter to prove it is working. I hope with this simple measuring tool, the few LEDs will be enough for showing the internal difference between two batteries; fresh and old for example.
If I will finish it, it will be my pleasure to present it here (hardware and firmware). I am afraid that I usually include on my boards my favorite small MCU from ATMEL, AT89C2051. Obviously LM324 will take care of the analogue part. The trick is doing the measurement at a 'constant' current (likely not low) in a voltage range say 8 to 13V (battery in standby) and for a period of time which is long enough to stabilize the drop and short enough to minimize the load dissipation. But perhaps it is better replacing the load current regulation by measuring both maximum and minimum voltages of the battery and by calculating (software) its internal resistance based on the known load resistance.
I envy people who have time to design what they like and are also allowed getting any component they heard of and know it suits their application. On the other hand, sanctions push one's mind to work much harder hence to have more fun ;)
--- In Electronics_101@yahoogroups.com, "Rick Sparber" <rgsparber@...> wrote:
> I think the circuit could be modified to do what you want. Rather than use
> comparator "C", you would need an amplifier with sample and hold
> capabilities. The monitor should work with any reasonable load on the
> battery so could run continuously. I played with this idea back in 1973 when
> I developed an in-circuit ohm meter.
> -----Original Message-----
> From: Electronics_101@yahoogroups.com
> [mailto:Electronics_101@yahoogroups.com] On Behalf Of Kerim F
> Sent: Saturday, June 30, 2012 1:31 AM
> To: Electronics_101@yahoogroups.com
> Subject: [Electronics_101] Re: question
> --- In Electronics_101@yahoogroups.com, "Rick Sparber" <rgsparber@>
> > I have posted two articles in the last few weeks that deal with analog
> > circuits. The response has been underwhelming. Not a single comment.
> > Are there any analog circuit designers left on this BBS?
> > Rick
> > rgsparber@
> > rick.sparber.org
> Your work reminds me an old small project I was working on but I didn't
> finish it. I liked to measure approximately the DC internal resistance of a
> car acid battery by measuring the small voltage drop on its terminals for a
> known load (DC current).
> Let us assume the battery voltage is 12V and its internal DC resistance is
> about 5mR. If the load current (actually a current pulse with a proper short
> period) is 10A (via MOSFET and current regulator) the voltage drop would be
> 50mV. In this case, the ratio of the high side voltage is 11.95/12=0.996.
> This ratio depends on the battery size and its age, for a few.
> I didn't have time yet analyzing attentively your interesting circuit so I
> am not sure how much it may be related to the high side voltage measurement
> that has a very small difference to be measured even approximately. After
> all, the project is just to build a tool for comparing the strength of some
> big batteries under test. I am afraid it will need a few LEDs (instead of a
> display for simplicity).
It is funny what consumers latch onto for importance. You work hard to get
the circuit board down by 0.1" and then have to bolt it to a block of steel
to get any credibility. Going with a simple resistive load current should
work fine since delta Vbattery is small.
I look forward to hearing more.
[mailto:Electronics_101@yahoogroups.com] On Behalf Of Kerim F
Sent: Thursday, July 05, 2012 7:10 PM
Subject: [Electronics_101] Re: question
In my free time (since I work on other projects), I started designing its
analogue part on LTspice. I will include the digital part using voltage
sources having each a proper timing list. Since AT89C2051 is an old and
simple MCU (20-pin), it doesn't have an ADC feature. Instead, it has an
internal comparator and its two inputs are P1.0 and P1.1 (its output is
P3.6, an internal flag). So when I need measuring a voltage, I measure for
example the charging time of a capacitor till its voltage reaches the
measured voltage. Obviously an external opamp or comparator can also be
You are right about the display; a number looks more professional. After
all, implementing either one is done mainly by software. The irony is that I
noticed most consumers are impressed more by the size (if not the weight
too) if it is larger ;)
To reduce the circuit, I removed the load current regulation. So the current
will be I_load = V_batt / R_load when R_load is on.
Measuring V_open (I_load=0) will give
R_batt = (V_open - V_batt) / I_load
R_batt = R_load (V_open - V_batt) / V_batt
The circuit on LTspice is almost finished. When it will be ready, I will
upload all its necessary files in case some members have LTspice and like to
play with it :) The software will be next. For instance, since I cannot get
windows C51 compiler, assembler, debugger and/or emulator, I used (for 30
years) writing my MCU (or CPU in the past) programs using a text editor and
DOS assembler (via command line) ;)