variable power battery pack

Hi guys,

I'm building a bicycle light and I've decided to go with the 7.2V battery pack (for RC) as you suggested. I'm using is 16W xenon bulb. How can I change the intensity? I assume I have to reduce the voltage? How could I get my pack to switch voltage to give me say 5W ,10W and 16W at the flip of a switch?

These questions might have been asked before... but I'm new at this thing... so feel free to just point to the answer...

Many thanks

Reply to
John
Loading thread data ...

Yes, but what you don't want to do is just drop the voltage through something that'll just turn it into heat like a resistor. Instead, what is often done is to send a square wave with an adjustable duty cycle to a light bulb. The lower the duty cycle, the lower the RMS value of the voltage the bulb sees and the less power the bulb will use. This is readily done using a 555 timer IC and a MOSFET as the switch. I'd suggest running the 555 at somewhere between 100Hz and some kilohertz, nominally -- lower values can start to allow you to _see_ the switching action (this is based on the thermal inertia of the filament -- how quickly it cools down), whereas higher values will make the switching action slightly less efficient (although until you get to megahertz, this is probably pretty negligible).

Note that your 16W bulb is going to become a lot less efficient at 5W, because most of the filament's thermal energy will be emitted as infrared radiation that isn't going to help you, as a human, one iota. (You might make the comparison between a single 2 'C' bike light, which often use 2.4W bulbs, and you setup running at 2.4W -- you'll find that your light will be much redder and not as good. There isn't much of a way around this problem with blackbody radiation sources -- if you get ambitious someday, you can try high intensity discharge bulbs or LEDs which don't exhibit this behavior.)

What I've described here is on open loop system in that, if the battery voltage drops a little, the bulb's power consumption will change a little as well. NiCads have a pretty flat discharge curve so this won't be a problem, but if you actually want _calibrated_ wattages regardless of battery input voltage, you'll need a rather fancier design with (most easily) something like a switching regulator controller IC.

---Joel Kolstad

Reply to
Joel Kolstad

Which brings to mind the much simpler method. Just use multiple 2.4W bulbs. You can switch them on as needed. And you have the safety of never losing your only light source, since if one burns out, you still have the other(s).

You could go with three 5W lamps, or some other combination. The nice thing is that the smaller prefocused flashlight lamps are avaiable at hardware stores and Rat Shack, too.

Another plus that I noticed when I rode my bike is that a single light source causes dark shadows behind objects. With multiple light sources spaced a foot or two apart, this isn't so much of a problem. This is probably the reason why car headlights are not in the middle, but at the left and right corners of the car. And again here you have the issue of reliability, so if you lose one headlight you still have the other.

--
@@F@r@o@m@@O@r@a@n@g@e@@C@o@u@n@t@y@,@@C@a@l@,@@w@h@e@r@e@@
###Got a Question about ELECTRONICS?   Check HERE First:###
 Click to see the full signature
Reply to
Watson A.Name - 'Watt Sun'

Make sure your bulb can handle the vibration and stress of being mounted on a bicycle. Bulbs that can are significantly more expensive.

Adam

Reply to
Jacobe Hazzard

Do you mount you lights on your handlebars or on your helmet? Have you tried combining the two?

Thanks

Reply to
John

is

light

the

using

Can you buy this ready made or do you have to make ti yourself? Also, wouldn't bypassing the wave generator be better when you want maximum power?

as

problem,

I'm going with NiMh by the way.

Thanks again!

Reply to
John

No, the bulb will fail rapidly. If you can, to keep a reasonable bulb life, and brightness, you really want to keep the RMS voltage to within 5% or so.

--
http://inquisitor.i.am/    |  mailto:inquisitor@i.am |             Ian Stirling.
---------------------------+-------------------------+--------------------------
 Click to see the full signature
Reply to
Ian Stirling

want

You mean 5% less than rated voltage? Why would the bulb die quicker with the rated DC? So when you underun your lights, they last longer?

Thanks a million John

Reply to
John

I know this is a really basic questions, but where could I find the schematics for this? I've found many for wave generators but I'm not sure how to include the MOSFET. Why is the IC more efficent than just a physical switch?

Thanks again!

Reply to
John

It is not really necessary to include the MOSFET if it is the only thing you have difficulties with. In this case a (big) BJT transistor like the 2N3055 (NPN) will also do. Be prepared however that it gets hot. In this case, use a 6 Volts lightbulb because 1) There are always losses in the transistor, somewhere around half a volt. 2) It will be hard to get an almost 100% duty cycle, so that some 0.7 Volts will get eaten up by the converter. Make sure that the signal is really square-wave, not sine neither sawtooth, this will increase the efficiency.

To the second question: An IC is never as efficient as a physical switch, but the switch itself is not the cause of the losses. It is the resistors (or transistors) that are used to drop voltage that heat up and make the least efficient part of the circuitry apart from the light bulb. Using a switching circuit you can avoid the high voltage drop thus conserving battery power when it is run at a lowered wattage.

Reply to
Dimirtij Klingbeil

[...snip...]

You can't pulse the H.I.D. xenon's like this. His only solution is to use multiple bulbs at lesser wattages and he may need multiple ballasts too.

Reply to
Fred Bloggs

I'll be using lights on the mountain so I've decided to use 2 lights for safety reaons. One will be on the helmet and I'll play with the second one to see where it works best. I'll be using H.I.D. Xenon bulbs. I'll use the

555 wave generator to pulse the lights and have variable power. I'll include a bypass switch so I can use the lights on max power without having to feed the 555 circuit. I want the switches to be on the handlebars so I can change lighting as I ride.

I want both lights to be on seperate battery packs to improve reliability and so I can take just 1 light on beer runs and for commuting. I'll be using rechargeable NiMh C or D cells for power. This way I can use regular batteries if I have no way to recharge the spent batteries.

I can build this 'ideal' (at least for me) system for about the same price as a decent system. So I think I'll go for it....

Any thoughts appreciated!

Thanks John

Reply to
John

No, he's saying that you want to be close to the rated voltage of the bulb, and the PWM regulator he tells you how to build should be designed to do so.

5% accuracy is a good number to aim for with incadescent bulbs, you don't have to better than that.

Running the bulb below rated voltage will lengthen life but will dramatically hurt efficiency. And the assumption is that if you're using batteries, efficiency is a prime goal.

Running the bulb above the design voltage will greatly shorten the life.

Tim.

Reply to
Tim Shoppa

If your lamps come with a ballast regulator then you cannot pulse battery power into it. These bulbs are very touchy- they MUST be run at rated power- too much by ~10% and they burn up/explode and too little causes electrode damage. The ballast is there to put power consumption right on the money.

Reply to
Fred Bloggs

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.