Step voltage down to 9.6 from 12

[kxvett]

This seems the best place to post this, so thought I would give it a shot.

Background:

I have installed a stator with light coil on a kx450, and charging challenged bike to say the least. This stator is SMALL, about 2" in diameter, so it doesn't put out much power. I am running two 10 watt LED lights, which draw 1 amp each. Tried a capacitor, but the stator will not put out enough to charge the capacitor and run the lights, so switching to battery power.

Problem:

The problem I have is that it takes a lot of RPM to get to 12 volts on the stator, and I really only need 9 volts to run the LED lights. So, I thought I would just build a NiMH battery back that is 9.6 volts instead, but need to step down the voltage from the regulator/rectifier from 12 volts to 9.6 volts.

I really don't have a problem running a 12 volt battery pack, but I just don't think the stator will keep it charged enough to run the lights at idle, so the batteries would be constantly undercharged. As a result, I have my doubts that 12 volts will work, so the 9.6 volts seems a better option.

Question:

Anyone know what resister I need to reduce the voltage from 12 volts to 9.6 volts? Any other problems with stepping the voltage down? I plan to use the Trailtech regulator/rectifier, and they have an overcharge sensor that will probably not work since I am only charging 9.6 volts.

[brewster]

I'm confused at your aproach. You say that the system has a hard time generating 12 volts but you want to use a 12 volt regulator and then drop the output from the regulator to 9.6 volts. If you can't generate 12 volts at idle, how is the regulator going to be able to put out 12 volts?

A different approach would be to rectify, filter, and then regulate to the desired 9.6 volts, not 12 volts.

Ride on

Brewster

[rayivers]

You can drop 2.4 VDC (12 - 9.6) at 2 amps with a 1.2 ohm 20 watt resistor, or a 2.4V 20 watt Zener diode. These wattage ratings are intended to increase reliability, not to save money.

Are your voltages all DC? What is the AC voltage/current output of the stator coil at idle and max rpm?

If the rectifier can supply at least 9 VDC to the lights at idle, you might consider a small-value resistor to the battery bypassed by a diode, along with a separate 'steering diode' from the rectifier output. This would send a current-limited charge to the battery when rectifier output voltage is high, and then allow the battery to power the lights when the rectifier output drops too low or the motor is off.

Ray

[William1]

You do not need to drop the voltage at all. To charge a battery, you need a higher voltage than the battery is. You need to have more volts going in than come out to charge. This is why normal charging systems are 14.5 volts for a 12 (actually, 12.8) volt battery.

[grayracer513]

I think he's primarily concerned with overpowering his 9 volt lights with the full 12 volt output of the system. Given that the battery is NiMH, and not LiPO, I think William1 is probably right in saying that it's not a big deal to have a 12 volt DC charge on a 9.6v battery. Even with a LiPO, you'd have (if you were smart) a charge controller card in the pack and you'd be OK at 20% overvoltage.

Beyond that, I think the LED's can handle 12v, and you could still use a resistor as suggested.

[kxvett]

Wow, I didn't expect this many replies so quickly. Let me provide a little more info that may help.

I am running two 10 watt LED lights with a 9 to 50 volt input rating. As a result, all I need is 9 volts to get them to ignite.

I have been able to get 11 volts AC out of the stator (just under 10 volts DC), but that's at pretty high RPM level. As a result, I don't think I can ever charge a 12 volt battery. Since the LED lights are fine at 9 volts, I started thinking out of the box and came up with using a 9.6 battery pack. With this, I have enough voltage to run the lights at idle, as well as enough voltage to charge the battery.

Currently, I am using a bridge rectifier with a 9 to 50 volt rating. I have read that the voltage drop is less with a bridge rectifier compared to a full wave rectifier, but this is really not my area of expertise. Here is the link to the Radio Shack rectifier:

http://www.radioshack.com/product/index.jsp?productId=2062584

Brewster, I was going to use a 12 volt regulator, as I don't think anyone makes a 9.6 volt regulator. If someone can tell me how to just regulate the voltage to 9.6 volts out of the bridge rectifier, I'm all for that approach.

William1, If I understand what you are saying, then I can just run a 12 volt regulator/rectifier with a 9.6 battery pack without blowing the thing up. I expect to have to replace the NiMH pack about once a year, as I expect I will charge them to death over time.

To all, I originally use a LiPoly battery pack without the charging system, but my night rides can last much longer than the battery. If I can get this lighting coil worked out, I will use the extra battery on my HID helmet light, which should give me almost four hours on my head.

To pull this all together:

1. Can I just run a 12 volt rectifier/regulator without damaging the 9.6 batteries?

2. If number one is not possible, can someone tell me what I need to purchase/install to cap the voltage out of the rectifier to keep from destroying a 9.6 battery pack?

Thanks for all of the great info so quickly. As I said, I have some knowledge here, but obviously well below many of you who have posted.

[rayivers]

Thanks for the additional info. It sounds like you really don't have a lot of voltage to work with right now. To have a reliable LED lighting system that doesn't cut out or eat batteries, you need more than enough voltage at almost all engine rpm, since most voltage regulators work by reducing excess voltage to the desired level.

If this were my bike, I would proceed as if I had a 6 VAC stator (even that may be high), which will give a much more accurate representation of real-world results. Your 11 VAC reading sounds like absolute max or close to it, which would only really be useful for circuit design if the motor ran continuously at high rpm.

Without knowing the AC current capability of the stator winding, it's hard to know which solution (voltage doubler, transformer, lower-voltage LED's/battery) would work in your situation. Sorry if all this sounds overly complex, but I'm an OEM-or-better kinda guy, especially when it comes to lighting for frequent night riding.

It seems to me the simplest guaranteed-to-work solution would be lower-voltage LED lights (3V would be best, if they're available) and matching battery. This way, you would probably have enough rectifier voltage to work with even at low rpm; idle might still be a problem. LED lights are unforgiving of very low voltages; they simply shut off. A stator+battery system is a very good idea.

Could you take AC and DC voltage readings with and without the 2A lighting load, at maybe 4,000 rpm? This would pretty much eliminate guesswork.

A side note on battery charging; it's important to limit charge current to prevent overcharging, especially when lots of regulated charge voltage is available.

Ray

[brewster]

If there is enough voltage and current available coming out of the bridge rectifier at lower RPMs, you could use something simple like an LM338 adjustable voltage regulator:

http://www.national.com/mpf/LM/LM338.html#Overview

FYI: The bridge rectifier drops 1.4 volts.....2 diodes per phase.

Ride on

Brewster

[grayracer513]

Currently, I am using a bridge rectifier with a 9 to 50 volt rating. I have read that the voltage drop is less with a bridge rectifier compared to a full wave rectifier, but this is really not my area of expertise.

Also note that a bridge rectifier IS a full wave rectifier:

http://en.wikipedia.org/wiki/Bridge_rectifier

[kxvett]

Thanks for the additional info. It sounds like you really don't have a lot of voltage to work with right now. To have a reliable LED lighting system that doesn't cut out or eat batteries, you need more than enough voltage at almost all engine rpm, since most voltage regulators work by reducing excess voltage to the desired level.

If this were my bike, I would proceed as if I had a 6 VAC stator (even that may be high), which will give a much more accurate representation of real-world results. Your 11 VAC reading sounds like absolute max or close to it, which would only really be useful for circuit design if the motor ran continuously at high rpm.

Without knowing the AC current capability of the stator winding, it's hard to know which solution (voltage doubler, transformer, lower-voltage LED's/battery) would work in your situation. Sorry if all this sounds overly complex, but I'm an OEM-or-better kinda guy, especially when it comes to lighting for frequent night riding.

It seems to me the simplest guaranteed-to-work solution would be lower-voltage LED lights (3V would be best, if they're available) and matching battery. This way, you would probably have enough rectifier voltage to work with even at low rpm; idle might still be a problem. LED lights are unforgiving of very low voltages; they simply shut off. A stator+battery system is a very good idea.

Could you take AC and DC voltage readings with and without the 2A lighting load, at maybe 4,000 rpm? This would pretty much eliminate guesswork.

A side note on battery charging; it's important to limit charge current to prevent overcharging, especially when lots of regulated charge voltage is available.

Ray

Ray,

On the possibility of using smaller LED lights, that's really not a solution, as I want all the light I can get at 60 mph in the dark. Crashing sucks, but crashing at night is even worse, especially if my HID helmet light comes unplugged, then I'm searching around in the dark for a big green motorcycle. At the 24 hour race last year, I passed tons of folks trying to use a glow worm as a light, so have little interest in going smaller. Just no safe.

On my 4 hour test night ride, I was able to keep the two LED lights running by just keeping the rpms up in every corner. Since the LEDs will work on 9 volts, that's somewhere in the 2500-3000 rpm level, so wasn't that hard. That's what makes me think I am close to a solution, as I just need help below that rpm level. Oh yes, since the KX450 is a short stroke motor/close ratio transmission, the rpms live on the high side, so I spend a lot of time well over 6000 rpm.

As for the LED lights being either on or off, I love that about LED lights, as they really don't dim at lower voltage. In my mind, that makes them the perfect choice for a dirt bike.

I will take the AC/DC reading at the rectifier connections with both lights on and post here tomorrow. I'll have to estimate on the rpm levels, as I really don't have a way to measure.

On battery charging, as I understand, if I use a 9.6 pack and overcharge, it will simply shorten the life of the batteries, correct? I don't have a problem replacing the battery every year, as they are rather inexpensive, just don't want to cause a explosion in my air box caused from over charging.

Brewster,

Love the idea of an adjustable regulator, but it looks like I have to purchase lots of them to get one. Do you know of a source where I can just purchase one unit?

To all,

I have been posting install pictures and progress pictures for several weeks in the Kawasaki kx450f forum, as no one has found a charging/lighting solution for these bikes, so we simply run everything straight from batteries. I decided to jump in this forum with the more technical questions, so thanks for helping out your green brothers....

[Beyond The Pale]

Any KX450F stator is going to be pretty low powered. The thing is tiny. Since your LED lights can take 50 volts and you only measured 11 volts AC at high RPM, I would consider running it with no regulation. Just use the bridge rectifier with a cap straight to the LED lights.

Regulating down to 9.6 volts is not a good idea. You are already low on power and this will just throw more away.

[rayivers]

On the possibility of using smaller LED lights, that's really not a solution, as I want all the light I can get

Not smaller - lower voltage. Believe it or not, a compact 2.1 VDC high-intensity LED array could be constructed that would cause temporary blindness in 5 seconds or less, and (with the proper focusing setup) outline a deer at 1/4 mile. Put enough LED's together and you could light up a football stadium, or maybe the moon.

Each LED type has its own conduction voltage, from infrared (1.2 VDC) to blue (2.5 VDC). Practical LED lamps use a series-parallel arrangement to get a higher working voltage for real-world applications, but there's no reason why an all-parallel configuration couldn't be used, resulting in a very low-voltage, high-current LED light that would be every bit as bright as a similar unit wired in series-parallel or all-series. Whether or not this would be practical for your situation will depend on the AC current capabilities of your stator winding, which I imagine are somewhat limited.

On my 4 hour test night ride, I was able to keep the two LED lights running by just keeping the rpms up in every corner. Since the LEDs will work on 9 volts, that's somewhere in the 2500-3000 rpm level, so wasn't that hard.

OK, cool - this is just the kind of info I was looking for. It could be that two 6 VDC 10 watt lights and a 6 volt battery would completely solve your problems, with the battery kicking in only at idle and/or when the motor's off. If you wanted even more light and could supply the power, 6 VDC 20 watt lights would surely do the trick. If you didn't want to overcharge the battery, the resistor/diode arrangement I mentioned in an earlier post could be used.

On battery charging, as I understand, if I use a 9.6 pack and overcharge, it will simply shorten the life of the batteries, correct?

Overcharging basically overheats the battery and eventually degrades its current capacity. Extreme overcharging can do extreme things. The guy I used to work with built his own laptop charger, which sent his molten battery down through the computer, table top and carpet; the only thing that stopped it was the steel deck plate (this was on a ship). In your case, though, I'd primarily be worried about not having decent battery power when you needed it.

For electronic stuff, Mouser is about as good as it gets:

http://www.mouser.com/Search/Refine.aspx?Keyword=lm338

Ray

[jsantapau]

This seems the best place to post this, so thought I would give it a shot.

Background:

I have installed a stator with light coil on a kx450, and charging challenged bike to say the least. This stator is SMALL, about 2" in diameter, so it doesn't put out much power. I am running two 10 watt LED lights, which draw 1 amp each. Tried a capacitor, but the stator will not put out enough to charge the capacitor and run the lights, so switching to battery power.

Problem:

The problem I have is that it takes a lot of RPM to get to 12 volts on the stator, and I really only need 9 volts to run the LED lights.

would it be a possibility to add more wire to wrap around the coil in the magnetic field? that would bump up voltage but would drop amperage.

also if you are using Light Emmiting Diodes do you really need a rectifier? could you just build a full wave rectifier using the led's for the diodes?

[SnowMule]

I have been able to get 11 volts AC out of the stator (just under 10 volts DC), but that's at pretty high RPM level. As a result, I don't think I can ever charge a 12 volt battery. Since the LED lights are fine at 9 volts, I started thinking out of the box and came up with using a 9.6 battery pack. With this, I have enough voltage to run the lights at idle, as well as enough voltage to charge the battery.

Is this with the lights on? See what you get without the lights.

Currently, I am using a bridge rectifier with a 9 to 50 volt rating. I have read that the voltage drop is less with a bridge rectifier compared to a full wave rectifier, but this is really not my area of expertise. [/QUOTe]A Bridge rectifier is a full-wave rectifier. Passes the positive side of an AC waveform, and inverts the negative side. (End up with camel-humps instead of a sine wave.)

A single diode will half-wave rectify it, that just cuts off the bottom half of the signal.

1. Can I just run a 12 volt rectifier/regulator without damaging the 9.6 batteries?

2. If number one is not possible, can someone tell me what I need to purchase/install to cap the voltage out of the rectifier to keep from destroying a 9.6 battery pack?

1. No. You need a charging circuit to charge 9.6v NiMH cells. Applying 12v to any NiMH battery pack will make it go boom.

2. A bigger stator, and probably a new voltage regulator. The 12v battery you have is probably OK.

[SnowMule]

The problem isn't with voltage... it's with power (wattage). You simply don't have enough power to run the lights you have. You are -NEVER-, no matter how hard you try, going to get more energy out of a system than you put into it (Conservation of energy - fundamental laws of physics). The system, in this case, consists of a stator as a source and a lightbulb as a sink. The battery stands as a buffer, in an ideal situation it neither sinks nor sources energy.

If you're measuring voltage off the stator, you're getting AC volts out of it. The lights most likely run on DC, so you're going to need to rectify and regulate that voltage off the stator for the lights to be happy. You do NOT want to bring the system down to a 9-volt system to "fix" this problem. All that's going to do is increase the current load on the wiring (V=I*R, Ohm's Law), causing more problems.

The real solution here is to either lighten the load on your electrical system, or give it more generation capabilities. How you do that is up to you, I recommend a stator rewind and bigger voltage regulator if the one you currently have can't handle the rewound stator's output.

You can drop 2.4 VDC (12 - 9.6) at 2 amps with a 1.2 ohm 20 watt resistor, or a 2.4V 20 watt Zener diode. These wattage ratings are intended to increase reliability, not to save money.

Bad idea. That resistor's going to get very hot, and that heat is lost energy. I'm not even sure how you'd use a 2.4v zener to drop the voltage, all that's going to get you is a 2.4v output. And running that much current through a zener diode will melt the diode. These are both the wrong parts for dropping voltage for a big current sink. A switching power supply is the most efficient, a transformer and bridge rectifier/capacitor is the easiest way (least parts) to drop an AC voltage. A small voltage regulator circuit will get the job done (It'll drop the 12v to 9.6), but you're going to be losing a lot of power to heat in that conversion.

You do not need to drop the voltage at all. To charge a battery, you need a higher voltage than the battery is. You need to have more volts going in than come out to charge. This is why normal charging systems are 14.5 volts for a 12 (actually, 12.8) volt battery.

Yes, this. For a lead-acid system, exactly. Don't overvolt a NiMH/Li-Ion battery. There's a reason you need charging circuits for those - it's so they don't explode.

Not smaller - lower voltage. Believe it or not, a compact 2.1 VDC high-intensity LED array could be constructed that would cause temporary blindness in 5 seconds or less, and (with the proper focusing setup) outline a deer at 1/4 mile. Put enough LED's together and you could light up a football stadium, or maybe the moon.

Not really... voltage in an LED isn't as important as a lot of people think. Lumen intensity is how "bright" the LED is. And at the same time, you're not going to get more energy out of the LED than you put in. 1 watt input to the LED (Watt = voltage x current, as long as voltage and current are within operating specs of the LED the two don't matter as long as they multiply to the wattage) can generate NO MORE THAN 1 watt of equivalent light energy at 100% efficiency.

Each LED type has its own conduction voltage, from infrared (1.2 VDC) to blue (2.5 VDC). Practical LED lamps use a series-parallel arrangement to get a higher working voltage for real-world applications, but there's no reason why an all-parallel configuration couldn't be used, resulting in a very low-voltage, high-current LED light that would be every bit as bright as a similar unit wired in series-parallel or all-series. Whether or not this would be practical for your situation will depend on the AC current capabilities of your stator winding, which I imagine are somewhat limited.

The voltage you're talking about has nothing to do with the wavelength of the light output (see: electromagnetic spectrum) - That's determined by the chemical composition of the semiconductor. I have no idea where you got those voltage/color correlations from.

I think what you're talking about with the "series-parallel arrangement" has more to do with the current limiting resistors and power dissipation in them than what you're going for. Don't think of it so much as one huge series-parallel circuit, but rather as a smaller series of nodes.

[SnowMule]

also if you are using Light Emmiting Diodes do you really need a rectifier? could you just build a full wave rectifier using the led's for the diodes?

LED's don't handle reverse voltage very well at all. They're designed to light up when they're forward-biased, ordinary diodes are designed to block that reverse voltage and pass current in the forward direction. An LED on an AC signal should have either a second LED in the opposite direction in parallel, or a protection diode across it.

So yes, you need a rectifier.

[rayivers]

SnowMule,

Guess we'll just have to agree to disagree on some things. You might want to poke around the Web for info on the many uses and ratings of Zener diodes and the different conduction thresholds and other characteristics of the various LED types; lots of great stuff out there.

Here's something to get you started - note the first chart in the 'Colors and Materials' section (the graph I used had some voltage differences and predated the violet LED spectrum):

http://en.wikipedia.org/wiki/Light-emitting_diode

Ray

[jsantapau]

LED's don't handle reverse voltage very well at all. They're designed to light up when they're forward-biased, ordinary diodes are designed to block that reverse voltage and pass current in the forward direction. An LED on an AC signal should have either a second LED in the opposite direction in parallel, or a protection diode across it.

So yes, you need a rectifier.

so you don't need dc power as long as 2 leds are wired in opposite direction of each other but in parrallel ? If I could use ac power why would I need to rectify.... I got lost

[SnowMule]

SnowMule,

Guess we'll just have to agree to disagree on some things. You might want to poke around the Web for info on the many uses and ratings of Zener diodes and the different conduction thresholds and other characteristics of the various LED types; lots of great stuff out there.

Here's something to get you started - note the first chart in the 'Colors and Materials' section (the graph I used had some voltage differences and predated the violet LED spectrum):

http://en.wikipedia.org/wiki/Light-emitting_diode

Ray

That voltage in the table is voltage drop across the diode when it's forward-biased. An LED with a 3.0v drop could be anywhere from green to UV. All that's saying is you need to have that much difference between the leads in order for that diode to behave as designed.

I'm well aware of zeners and avalanche breakdown properties... i work with this stuff frequently - and i still trust my degree in electrical engineering and the shelf full of books on semiconductor design and analysis over a wikipedia article. :dunno:

Take a look at a datasheet for any LED (the component, not a light engine). In there it'll have a forward voltage (this is a spec in mouser/digikey/allied's searches), reverse current (excessive reverse current destroys the semiconductor, this value is usually quite low hence the reason you need a protection diode on LEDs on an AC circuit), DC forward current (use this value to figure out what value current-limiting resistor to put in there using ohm's law), and a reverse voltage (go over that and magic smoke comes out). Each LED is different, and similar designs still vary manufacturer to manufacturer. It's up to the engineer designing the assembly to get the right functionality for the inputs the assembly's designed for using the guidelines in those datasheets.

[William1]

I second snowmules comment about LEDs, AC/DC and reverse voltages.