Halogen light and generator

I want to replace the conventional incandescent bulb in my headlamp with

a Halogen or Krypton one to get better illumination. The generator is 6 volt, 3 watt, and I would
replace the bulb with the same voltage/wattage. Current bulb is 2.4 watt 6 volt screw base. Is there
any electrical reason why this cannot be done? Has anyone out there done

this? What physical changes are necessary (i.e. different bulb base etc.) Any info would be
appreciated.

 

John Henderson wrote:
>
> You have the option of dispensing with the old tail light and using a battery-powered flashing LED
> tail light instead. You'd then put the full 3W into a correctly-rated halogen bulb for extra
> brightness. This removes the risk of the rear bulb blowing without being noticed, and taking the
> front bulb with it some time later through its being overloaded.

FWIW, I usually use a 2.4 watt bulb without a taillight. I'm sure I'm overvolting it somewhat (but
not much, since I've got the zeners for a regulator). Brightness is somewhat better than with the
taillight, and bulb life has been fine.

--
Frank Krygowski [email protected]

 

John Henderson wrote:
>
> "Frank Krygowski" wrote:
>
> > FWIW, I usually use a 2.4 watt bulb without a taillight. I'm sure I'm overvolting it somewhat
> > (but not much, since I've got the zeners for a regulator). Brightness is somewhat better than
> > with the taillight, and bulb life has been fine.
>
> A properly designed zener-diode regulator should prevent overvoltage. At a regulated 6.0V, your
> 2.4W bulb will draw 0.4A only.

An opposed zener pair does not regulate the voltage as precisely as you seem to think. The zener
spills current it the applied instantaneous voltage exceeds its rated voltage. But remember, bike
generators are AC devices. As a first approximation, the zeners cause the AC sine wave to transform
into something vaguely like a flattened sine wave, with its peak shaved off. But if the sine wave
being input to the zener pair has a higher voltage, the RMS value of the output from the zeners is
still higher. The waveform fills in more of the area below the flattened peaks, and the zeners can't
do anything about that.

IOW, a bike generator with an _ideal_ AC voltage regulator would show RMS voltage rising with speed,
from zero mph to whatever speed gives 6 volts RMS. The voltage would stay precisely 6 V RMS at any
greater speed. But with a zener pair as a regulator, voltage rises as above, but then continues to
rise, only at a lower rate. The voltage-vs-speed curve is sort of a rising dogleg curve, with a high
slope followed by a much lower one.

At least, that's what I recall. My data's in my other office.

> Most bicycle dynamos are more total-power-limited than voltage-limited, capable of delivering 3W
> max. (but with poor voltage regulation, depending on the load to do this). With a "hot" resistance
> of 15 ohms, a 6V, 2.4W bulb being fed with such a 3W source will stabilise the voltage and current
> at about 6.7V and 0.45A (while it lasts).

No, bicycle dynamos are current-limited, not total-power-limited (unless, perhaps, they are
tire-driven units that slip, which would be a design fault). If you present a bike generator with a
higher output resistance, it will develop a higher voltage. It tries to maintain a constant current.
It doesn't succeed precisely, of course (any more than a battery always delivers a constant voltage)
but under reasonable conditions, it comes pretty close.

On one of my bikes, I've got the generator driving two 2.4 watt lamps in series. (The second one can
be switched off at low speeds.) Above roughly 13 mph, the generator puts out enough voltage to run
both lamps at only slightly reduced current. I'm getting about six watts out of that "three watt"
generator.

--
Frank Krygowski [email protected]

 

On Mon, 23 Dec 2002 11:55:27 -0500, Frank Krygowski <[email protected]> wrote:

>Keep in mind, generator sets are AC devices. You'll have to rectify to put a battery in that
>circuit. That means you'll be eating the voltage drop over the forward-biased diodes.

I did once build a set where the rechargeable batteries were charged separately, but cut in when the
dynamo voltage dropped (a sort of external standlight, before standlights were invented). Now
standlights exist, I haven't bothered to rebuild this arrangement

Not the parameters under discussion, I know.

Guy
===
** WARNING ** This posting may contain traces of irony. http://www.chapmancentral.com (BT ADSL and
dynamic DNS permitting)
NOTE: BT Openworld have now blocked port 25 (without notice), so old mail addresses may no longer
work. Apologies.

 

On Mon, 23 Dec 2002 11:55:27 -0500, Frank Krygowski <[email protected]> wrote:

>[email protected] wrote:
>>
>> I've missed much discussion on this topic, but haven't seen a simple suggestion to put a small
>> gel-cell in parallel with the system. In the above 6v system, a 4x2x1" 1.3ah battery would serve
>> quite well to moderate the generator voltage outputs, would also provide light(s) when the bike
>> is stopped/slowed -- maintain a steady voltage.
>
>Keep in mind, generator sets are AC devices. You'll have to rectify to put a battery in that
>circuit. That means you'll be eating the voltage drop over the forward-biased diodes.
>
>Still, what you describe can be (and has been) done. However, most people wouldn't need it. And I
>think most of those that do need it could use a much smaller battery than you've described.

Alas, foiled by the bicycle world's classing an alternator as generator; similar to cleat-clip
pedals as clipless. Everyday a simple storage battery stores/moderates/ballasts 0-20vdc photovoltaic
output to serve sensitive electronics within +- 0.5 vdc.

Bruce Ball

 

Frank Krygowski wrote:
>
> The Pomeranian wrote:
> >

> > So-called "dropless" bridges can be constructed -- I've done it. ...The 1.2 V diode drop in a
> > conventional bridge is a bit much for a low voltage power system, a dropless system is more
> > appropriate.
>
> I don't know what a "dropless" bridge is. Can you elaborate?

They can be made of regular bipolar resistors and the essence is avoiding a good deal of the voltage
drop w/ regular diodes. Some _functional_ diagrams are a little misleading in that they leave out
some stuff other than the main rectifying transistors and their bias resistors. More is required and
this is due the the fact that bipolar resistors can be run "backwards" as poor beta transistors with
the collector and emitter exchanged. This "reverse mode" is important in a dropless bridge because
they will see the reverse mode and something must be done to keep them from conducting in that state
(things would get hot, otherwise). Some extra "stuff" is involved (more transistors and resistors).

Dropless bridges are used in most modern telephones. Note that the modern simple phone gets all its
power from the phone line. Because the voltage drop from the CO to the demarc can be large, the DC
voltage applied the phone can be rather small. Also, to avoid the need for hooking up the phone with
"correct polarity" (DC is present), a bridge rectifier is used and it is usually a dropless bridge
for the case of low available voltage. I don't know if the dropless bridges used in phones can be
used for a bike generator -- they likely don't have the power handling capability. I assumed this
and constructed my own. It seemed to work for the way I happened to evaluate it: which means it did
work better than regular diodes. Since I made it in my garage, I never hooked it up to a scope or
did much data taking.