Use a bi-color LED as an ammeter?

What do you think of this idea to make a simple ammeter for my
motorcycle - Tap into the positive lead in the battery in two spots, so
the battery cable itself is like a shunt resistor. Run wires from
these two spots to a bi-color LED mounted on the dash (and have a
current limiting resistor in-line). The LED glows green when the
battery is charging, and red when it's discharging because I have too
many accessories turned on.

Would this work? Crude ascii below.


|---LED---|
| |
| *<-Resistor
| |
----|+ ---------------------
B |
A |
T |- ---GROUND
----|

Thanks,
-Ryan

 

he battery cable itself is like a shunt resistor

No.
A LED requires ~1.6V to light.

Use Notepad to draw your ASCII prints
then cut & paste them into your browser.
A monospace font is required (Courier font ).

 

Yes but they are wired into 12v sources all the time with the use of a
resistor. Are you saying the LED would not get enough power to light
up?

-Ryan

 

Perhaps for _your_ motorcycle. The most common color deficiency
experienced by humans, which occurs most often in men and boys, is an
inability to distinguish green, red and gray.

So you can do this (after adding a suitable transresistance amplifier to
address the 1.6V threshold problem), but if you loan your motorcycle to
me I won't be able to tell charge from discharge.

 

The resistance of the battery cable is so low that the voltage drop
across it will never be enough for the LED to illuminate.

 

Not to mention the fact that the current vs. voltage characteristic of
any junction diode is exponential, so you want to drive the LED with a
current proportional to the battery current, _not_ a voltage
proportional to battery current.

You _could_ sense the voltage in the manner stated, then amplify it with
a transresistance amplifier that'll drive the LED with a controlled
current that depends on your sense voltage -- but you'll need split
grounds and a good amplifier and all that fun stuff. It's doable, but
see my other post about the suitability of multi-color LED's for
colorblind folk.

 

I don't think you want to have 2 volts of drop across your battery cable.
That would be a terribly inefficient battery cable.

 

I read in sci.electronics.design that Tim Wescott

I have worked with a number of red-green blind people who had no
problems with resistor colour codes. Can you really not tell a red LED
from a green one? Not even as different shades of what you perceive as
'yellow'?

 

The LED glows green when the battery is charging,

That would be pretty radical,
but I have seen techs who put in a 1M resistor
where a 10M was supposed to go.

 

I went to school with a guy who was red-green colorblind. He bribed me with
a 6 pack to go through his resistor collection and hand him five of each
value. He would go to labs with a piece of styrofoam with rows of carefully
labelled resistors stuck in it. Without that, he was helpless.

Bob

 

Green/blue problem is a very unusual defect. 15 ohms is much easier.
Crossed eyes. (;-)

Purple/brown can be difficult, especially under incandescent light. But
it doesn't normally cause problems with resistors, because 2.1 and 4.1
aren't preferred values. 75 ohms/15 ohms is a possible confusion.

 

Maybe that's foveal tritanopia - where the very center of vision is weak
on blue. The brain usually manages to fudge things other than the
brightness of really pure deep blues since the perceived brightness of
most blue things is from stimulation of green receptors and maybe red
ones. But blue bands of resistors, especially 1/4 watt ones, all too
often look like green bands.

I have heard of it. I do not know what causes it or if this is
something normal.

- Don Klipstein ()

 

I'm more or less OK with color codes if it's a 5% type. Four-band color
codes throw me off completely. Even so I always check (and I have a
well-organized set of parts drawers) I can usually get close enough with
an ohmmeter, even in circuit -- and if not I have my wife and children
trained.

And no, I can't tell the difference between green and red on a bi-color
LED -- this absolutely drives me up the wall when some clever SOB
designs it in, like my Yaesu FT-23 where red means "squelch" and green
means "transmit" (or visa versa).

 

One of them -- and I think it was the green/blue deficiency -- was very
highly prized for infantrymen during WWII because they tended to see
camouflage easier. Ordinary folk would be so distracted by the color
match that they wouldn't notice the tones being slightly off, the folks
with this deficiency would then be able to see the mismatch in the pattern.

 

Here is the relevant context from JeffM's post:

You replied:

OK, Ryan, you are confusing me, and probably others as well. Can you
please draw a schematic of exactly how you intend to connect these LED's
to your electrical system?

The schematic in your original post didn't make any sense. The
written description makes sense but won't work for the reason JeffM
expressed above.

The bottom line answer is that you can sense whether your battery is being
charged or discharged, and light up LED's in response to the direction of
charge, but doing so will involve more than just LED's and resistors.
Also, it may or may not be practical to use a length of cable as the shunt
resistor in this application.

--Mac

 

I concur with other posters that the circuit doesn't look plausible
whichever way I try to allow for ASCII character shifting.


No that definitely won't work.

Regarding coloured LEDs, about 10% of males are red/green colour blind.
Actually, 'blind' seems a bit of a misnomer.
One of that 10% said he found it hard with dark greens and reds but bright
ones were okay.
Everyone fades into monochromatic vision at low intensities, so I would
guess that such people just have a brighter threshold for those colours.

10% is a significant percentage of chaps so I always add extra stimuli.
Such as steady green = okay, flashing red = warning.


In an application such as described, I would have two separate LEDs.

You can get triangular body LEDs, which would be nice for indicating the
direction of flow.

 

After solving the amplifier problem, just add two extra green LEDs.
Configure the bi-color LED to produce a horizontal red or green bar and
the auxiliary green LEDs to add green squares above and below the bar.
Now you have a red '-' sign for discharge and a green '+' sign for
charge.

 

I read in sci.electronics.design that Tim Wescott

Those are usually called '4-band'. Brown, black, brown, gold.

I think you mean 5-band. They confuse practically everyone. Brown,
brown, black, black, brown? Or is that brown, black, black, brown,
brown?

Noted. What colour do you see? The same colour as [insert name of
familiar coloured object].

 

OK, Ryan, you are confusing me, and probably others

OK I've put up a crude drawing at
http://myweb.cableone.net/zirconx/LEDammeter.gif

LEDs are available that run off as little as 2ma of current.

If I had two wires to run power from the battery to the bike, current
would run through both of them. This is essentially what I am doing
here, but one wire has much less resistance than the other and the
other has an LED in it. Could I get 2ma would flow through the other
wire?

-Ryan

 

I read in sci.electronics.design that wrote (in

You could, but the problem is that you need around 2 V (minimum) voltage
drop across the wire carrying the charge/discharge current. That is far
too high; you battery won't charge properly from the alternator and the
starter won't work at all, probably.