Re: (ET) How NOT to check your batteries... Battery Charge / Monitor / maintainer

["Jeremy Gagliardi com" <jeremy gagliardi com>]

Think modular.  For what you suggest, it would perhaps be better served by 
different modules.  If the maker or the buyer didn't want to go to the 
expense 
or cost of implementing everything, make a base kit that handled just the 
charger, no readouts, no leds, no buzzers, etc., but put jumper blocks in 
the 
circuit at key points.  For example, let's call these jumper blocks A, B & 
C.

At all times, the unit passes through voltage to jumper block A.  You can 
either plug in a cheap voltmeter to block A for easy readout, or you can 
make 
your own with a simple 2-digit readout.  Some of the more sophisticated 
off the 
shelf voltmeters have logging capability so you can put all of the polled 
results into a spreadsheet and graph it out on your PC.  If you rig a plug 
that 
fits into block A, you could log the voltage during the entire charge and 
later 
see how well things worked or didn't.  If you see wild up & down voltage, 
spikes, or dramatic dips, then you know you have a problem.  Of course, 
you 
could do this anyway, with connecting it to a controller.  Just connect 
your 
voltmeter to the batteries during a charge cycle.  Still, having block A 
would 
be convenient.

Now, say the conroller detects a failed charge.  I think in this case, 
when 
the unit starts the charger, it would just trip a simple 5-10 minute timer 
(doesn't have to be real time).  If after the timeout occurs, the voltage 
has 
not risen above a certain sweet spot setting, then it stops charge and 
sends a 
continuous on/off signal to a built-in LED and to jumper block B.  At 
block B 
you could add an optional buzzer for alerting.  This could either be a 
small 
thing that stays close to the main unit, or it could extend off on long 
wires 
to outside the tractor for easier detection.  Hell, I suppose you could 
even 
wire block B into your home's alarm system, or at least a module inside 
your 
house, if you really wanted to.  I'm sure ThinkGeek.com might have a 
gadget or 
2 that might appeal to the geekier members of this list.  You could 
probably 
modify the "Scrolling LED Name Badge II" with an alert message, I'm sure.

I'm not sure what temperature and current monitoring would do for you.  
Perhaps those would better be served by off the shelf components?  If 
you're 
referring to an overheating condition that would damage the batteries, 
then you 
could possibly add this as an add-on module to jumper block C.  If nothing 
is 
on block C, then the controller does nothing with it, but if that module 
is on, 
it will simply break the circuit (turn off the charger) if the temperature 
exceeds a certain setpoint, say 125, 150 degrees (again, what's good here?)

So far, this is all academic.  Nice thinking and good ideas, but are we 
going 
somewhere with it?

Regarding your comments about who can do it, I don't think you need a 
computer 
science degree or even programming skills such as C or assembly.  It's 
really 
about logic diagrams and boolean algebra - http://en.wikipedia.
org/wiki/Boolean_algebra_(logic).  The hard part is working out which AND, 
OR, 
NOT, XOR gates go where, keeping the voltage, resistance, capacitance, 
etc. 
just right, and so on.

>If we had nothing else to do ... <grin>

Yeah, I second that emotion.

On a related note, my 7 year old son has "Snap Circuits".  Maybe I can use 
those to relearn a thing or 2 to do this.  Ha ha  ;-)

--
Jeremy Gagliardi
E20
Potomac, Md

>----Original Message----
>From: reinharj frontiernet net
>Date: Aug 7, 2007 17:00 
>To: <Elec-trak cosmos phy tufts edu>
>Subj: Re: (ET) How NOT to check your batteries... Battery Charge /     
>Monitor / 
maintainer
>
>We need to find an 'inspired' Microcontroller Engineer (meaning EE & CS 
>skills) who wants to design PIC (or similar) microcontroller into a 
>charge 
>controller with real time clock, alarm (piezo buzzer, etc), voltage 
readout,  
>and maybe temperature and current monitoring.  [adding more features adds 
>some cost in hardware, and tedious software testing]
>
>I know bits and pieces of this (strong enough in hardware part), but am 
>far 
>too weak in C or Assembly language programming to get it done in timely 
>manner.
>
>Maybe we could form an 'Open Source' team to design this - similar to the 
>project for PWM motor controllers called OSMC (Open Source Motor 
>Controller).
>In that case, a few guys collaborated on electrical hardware and software 
>design for a microprocessor controlled, Pulse Width Modulated DC motor 
>controller, for hobby Robotics competition.  
>
>If we had nothing else to do ... <grin>
>
>
>John
>
>
>
>
>On Tuesday 07 August 2007 4:30:49 pm Jeremy Gagliardi com wrote:
>> I used to dabble in digital electronics myself, back in college in
>> *cough*92*cough*, but I'm wayyyy out of practice.  Still, I think a 
>> compact
>> enough unit could be made to handle this.  I just don't think I could
>> relearn enough to get it done in a timely manner myself.
>>
>> There are really 3 things to be aware of, IMHO...
>>
>> (1) The original:  If the voltage dips below the setpoint, activate the
>> charger.
>>
>> (2) Timer & voltage detection:  If the voltage does not rise or even
>> *drops* over a set period of time, stop the charge and make an LED blink
>> incessantly, or just leave a simple 2-pin jumper block for something the
>> buyer can insert later, whether it be an LED, a buzzer, or what have 
>> you. 
>> Activation would simply be to send voltage to jumper "xyz" in a
>> repeating-on/off pattern. Stopping this action would be to reset the 
>> unit
>> by simply removing power to the controller.  Hopefully, upon restoring
>> power, the situation will be corrected, and the unit will detect normal
>> voltage or it will be desireable to activate the charger once again 
>> (such
>> as after adding water to the batteries).  By the way, it bears noting 
>> that
>> the way I have my controller wired is not direct-to- batteries, but 
>> rather
>> to honor the power cutoff switch on the side of the tractor.  I
>> accomplished this by tapping into the leads that go to the 36V power
>> accessory outlet on the left, under side of the tractor.  That way, if 
>> I've
>> tripped the cutoff, the controller will lose power, and it cannot 
>> activate
>> the charger.  I've often used this to reset the controller, based on the
>> below condition (charge starts, cutoff power, correct situation, restore
>> cutoff switch, charger stays off)...
>>
>> (3) Not mentioned yet:  If the voltage dips below the setpoint, wait a 
>> few
>> seconds and check again; if it's still below, then activate the 
>> charger. 
>> This simple, check-pause-check-activate/ignore logic would prevent the
>> charger from activating for brief power drops.  I've often had these 
>> when
>> the ET is in the garage, plugged in, and I activate some feature like 
>> the
>> front lift.  The current controller activates instantaneously after the
>> power dips.
>>
>> If it would save you time, effort, money, etc., I'd be perfectly 
>> willing to
>> build the circuitry myself, if you provided the parts list, diagram &
>> instructions (for a fee?).
>>
>> --
>> Jeremy Gagliardi
>> E20
>> Potomac, Md
>>
>
>_______________________________________________
>Elec-trak mailing list
>Elec-trak cosmos phy tufts edu
>https://cosmos.phy.tufts.edu/mailman/listinfo/elec-trak
>