On Sun, 21 Dec 2003 16:17:13 -0500, "Mark Zarella"
wrote:

With the numbers you're using, 0.003 ohms resistance between the
battery and the headlamps, and 0.008 ohms between the cap and the
battery, I'd consider that "no voltage drop". In this case, your
headlamps would have to draw 300 amps of current before they saw a
1-volt drop between the battery and the headlamps, and they ain't ever
going to draw that much. Even putting the cap back in the trunk, and
using the combined .011 ohms, the headlamps would have to draw 90 amps
before the wiring caused a 1-volt drop. This still isn't going to
happen.

I still say that if putting the capacitor closer to the headlamps
versus back in the trunk with the amps makes a big difference in how
effectively it reduces flickering, then there's something wrong with
your headlamp wiring.

Using those numbers, I was able to demonstrate this in Multisim's (crappy)
simulation software. I've attached a copy of the simulation circuit. It
consists of a battery with a finite output Z. An amplifier being treated as
an oscillating source (the voltage, mind you, is always below battery
voltage so it acts as an oscillating sink). The wires. And the headlights.
The values were chosen such that the amp draws roughly 50A peak at 14V,
there's about a 1V drop at the battery with 50A draw, and the headlights
consume about 120 watts.

With no cap, there's an 820mV drop at the amplifier and 720mW drop at the
headlights. Putting a 1F cap at the amplifier terminals, the drop at the
amplifier is 31mV and at the headlights, 28mV. If the cap is put on the
headlight terminals, it's 300mV at the amp and 20mV at the headlights.
Hence a reduction.

This is hardly a large difference. This is due to a number of factors.
First, the wire resistances were the perfect values. In the real world,
they'd be much higher. Increasing the resistances uniformly yields an even
greater discrepancy between the two setups. To wit, multiplying the wire
resistances by 10 gives a ratio od 12:5 instead of 7:5. Secondly, capacitor
ESL was ignored. That should reduce both cases equally though, but helps to
explain why the dimming on both sides was reduced so dramatically. Most
importantly, wire inductance was neglected. This would result in a it
for*huge* benefit for the shorter wire, which would be the headlight wire
most of the time.


Mark,
This isn't a binaries newsgroup. I'm not bothered by your
posting the .GIF here, but others might complain.

So in your circuit simulation, relocating the capacitor from
the trunk to right next to the headlamps gave the headlamps an extra 8
millivolts (at the expense of almost a quarter-volt drop at the subs).
You've made my point. I agree that your simulation program made some
optimistic assumptions, but multiply that 8 millivolts by a factor of
ten or twenty, and you STILL won't be able to see a difference.

So I go back to my original assertion - if moving the
capacitor from the trunk to the headlamp terminals results in
noticeably less flickering, then your headlamp wiring is cocked up
somehow.

Scott Gardner