A bike generator for the winter...

[Griff]

. . .Or maybe even make something using magnets glued on to the flywheel as a rotor and and a bunch of home-made coils mounted around the edge as a stator. A few bike gen / wind power web sites I've seen have detailed making PM alternators from bits of plywood and epoxy resin!

I don't know much about these things, but have you considered a lawnmower engine flywheel and stator? I don't know how the RPM requirement would compare to your current set up, but they also produce up to 14.7 volts, and depending on the type of stator, can produce up to 16 amps. The weight of the flywheel might also help to maintain momentum once at speed.

Just a thought. . .

Griff

[AccordGuy]

Cruising another car boot sale I scored an old washing machine motor! It's amazing what you can find at these things. There was even a matching grooved drive belt in the same junk box. The guy let me have them for 50p

Not a permanent magnet one but it has the advantage over the car alternator that there are no pesky electronics on the rotor which is configured as a 16 pole DC motor. It seems to work best if I use some battery power to energise the stator coils on the outside. At first I tried energising the rotor and collecting power from the stator but the rotor is quite high resistance and the shape of the coils meant they didn't radiate much field out into the two opposing stator saddles.

I used a light bulb to limit the current in the stator which is surprisingly low resistance considering the motor is meant to be mains operated. It has few windings of a fat gauge and measured about two Ohms. At least that means I can energise it easily from a battery and not have to mess about with mains voltage. The rotor has much finer wire with lots of turns and just turning it with the drive belt by hand it made ~10V at an Amp or so with 600mA in the stator. It also has the advantage of having a much smaller diameter pulley than the alternator so the gearing will be more like 25-30:1 rather than the alternator that was only 14:1

With some fiddling about with different size light bulbs I should be able to find a happy medium where I can pedal against an acceptable load and still make a few Amps net. I might even be able to make it self-excite as the steel core is already magnetised a bit and turning the rotor makes a few mA just by itself with no stator current. If I glue a magnet on the laminated core and feed back part of the rotor output to the stator it might be enough to get it going by itself. I'd need the blocking diode to stop the battery being discharged through the rotor.

[Griff]

Thanks for responding, AccordGuy!

I've been following you & Sharkey's discussions on this and greatly appreciate the updates and experiment results. Please keep 'em coming as I am trying to soak all of this in for my own avant-garde experiments.

Thanks again!

[AccordGuy]

When you get going on something I'd like to follow your thread.

There's a thousand ways you can make some juice and mostly from old junk you can find laying about the place.

We love going to car boot sales as in itself it is a form of recycling. The economy is bad here (as everywhere) but people haven't stopped spending money - I see thousands of people at these sales compared to the almost deserted retail park in our town. People just aren't buying NEW stuff these days. With a bit of hunting you can find a second-hand anything you need for peanuts. I almost bought a 1.4kW genset as well as the washing machine motor. With it I could have charged the solar batteries on cloudy winter days without using utility mains. It was a 2-stroke engine though so no good for converting to run on LPG. I wouldn't run it on petrol as it's too expensive but LPG is cheap. Ideally, I'd want a diesel genset as then I could run it on cooking oil but diesel gensets are usually only in the high power (and expensive) class.

As Sharkey said, it's not about saving money... I've spent loads on the solar system and it makes pennies in power. It's just my hobby and I've a grudge against utility companies so I want to explore ways of doing without them.

[Griff]

Thanks, I will certainly post when I have something to offer and/or questions.

Right now I'm still in the education and research stage. I have some ideas, but need to learn more about circuit boards and their components to see if the ideas will pan out.

As I am full-timing off-grid in my bus, my ultimate goal is to meet my power requirements without having to run my gennie so much. Fortunately, my requirements are much less than others, but still a little on the heavy side for A/C, and refrigeration of perishables.

Thanks again for your input!

[AccordGuy]

Some more playing around with the washing machine motor...

I discovered it will self-excite without a magnet . Just wiring the field stator in series with the generating rotor is enough to make it work (as the rotor is 16 pole DC it always has some current output with no gaps in the waveform)

If I put a 21W 12V car brake bulb between the rotor output and the stator field coils and spin it up on the bike, it quickly makes up to 15-20V at 2 Amps. The only problem now is that all the current is going into the field bulb (it can glow so bright I worried it might blow). But if I use a smaller bulb it makes less current and then when you put a load on the "output", the field current gets shared and output collapses.

If I was just running a bulb as the load it would work best with the bulb being the load and the resistor for the field current. I thought about putting the battery in the place of the bulb so that the field current is actually used to charge the battery... The only problem is that a battery is going to push current the wrong way into the field coil until the generated voltage is higher than the battery voltage and then there will be a split second where the field current is zero and the thing won't work at all.

I think I might be able to fudge something with a bulb and a diode... If I put a blocking diode on the battery so it can only accept field current and use the bulb to bypass the battery, then the rotor will self excite the field coil using the bulb and when the generated voltage is higher than the battery it will start accepting current via the blocking diode. If I put a momentary switch on the bulb, I can then turn the bulb off once the thing is charging. When I stop pedalling the blocking diode will stop the battery discharging through the field coil.



I might still have to use a resistor or bulb to limit the charge current as otherwise the battery will try to draw a huge current and the pedalling will get really hard suddenly.

[AccordGuy]

Actually that plan worked sort of...

In the end I had to replace the bulb with a 20W 8 Ohm resistor and a 40W 4 Ohm resistor for the battery. It kept blowing bulbs. It actually worked better with a 21W bulb as it is a non-linear resistor (when the filament gets hot its resistance increases) but it has a limited overload capacity before it blows. At least the resistors have a big short term overload capacity.

I might still end up using a small 6V NiMH solar battery to feed the field current as it is quite tricky to try and get the speed right so that it makes enough field current to start charging the main battery bank while not over-doing it and wasting power in the current limit resistor. I also worried about the power spikes that blew the battery current limit bulbs. Although a lead acid battery makes a fairly good voltage regulator it is a bit slow to absorb spikes as it is a chemical beast. As the inverter is permanently connected to the battery,I wouldn’t want any voltage spikes from an unstable positive-feedback driven generator to break my inverter.

At least if I feed the field coils a steady current from an independant source the thing will work in a more predictable manner.

Reposted with Japanese unicode removed I was using my sister-in-law's pc and accidentally pressed some button on the IME language toolbar

[Sharkey]

??? Am I the only one seeing a bunch of ASCII characters instead of text in that last post?

[Griff]

I can read the post, but the font is definitely not the default font of this board. . .

[Sharkey]

In the database, the post text looks like this:

Code: Select all

Actually that plan workeｄ　ｓï½�ï½’ï½â€

[Phil Feinstein]

I can see text as well and I was going to tease about posting in all caps like a Noob. Cut and paste and edited from Ariel Unicode to something a bit more human. Text as follows:

Actually that plan worked sort of…

In the end I had to replace the bulb with a 20w 8 Ohm resistor and a 40W 4 Ohm resistor for the battery. It kept blowing bulbs. It actually worked better with a 21W bulb as it is a non-linear resistor (when the filament gets hot, its resistance increases) but it has a limited overload capacity before it blows. At least the resistors have a big short term overload capacity.

I might still end up using a small 6V NiMH solar battery to feed the field current as it is quite tricky to try and get the speed right so that it makes enough field current to start charging the main battery bank while not over-doing it and wasting power in the current limit resistor. I also worried about the power spike that blew the battery current limit bulbs. Although a lead acid battery makes a fairly good voltage regulator it is a bit slow to absorb spikes as it is a chemical beast. As the inverter is permanently connected to the battery, I wouldn’t want any voltage spikes from an unstable positive-feedback driven generator to break my inverter.

At least if I feed the field coils a steady current from an independent source the thing will work in a more predictable manner.

[lemmiwinks]

It may be to do with posting from Japan via a Japanese keyboard. Just a wild uneducated guess (which doesn't make all that much sense since the first couple of words came out fine).

[AccordGuy]

Now that it's been getting a little warmer, I've been playing with the bike again.

For a laugh, I bypassed the diode to see if I could pedal against the battery current. Despite the motor being designed for mains voltage, it ran pretty good backwards on 12V. Not too fast but it could turn the flywheel and pedals and with so much torque that there was no way I could stop it by trying to cycle "against the flow". It actually made the drive belt slip.

Then I put the diode back in the circuit and started pedalling but nothing happened, no matter how fast I went. With the momentary switch closed, I could feel the current going into the 8 Ohm start-up load but when I released the button - nothing! Just free-wheeling...

So I measured the voltage. To my surprise the voltage was backwards. Then it occurred to me. I'd let the battery push current through the field windings backwards while fighting it. The core was only very weakly magnetised, which allowed the generator to self-excite with lots of positive feedback. I'd accidentally reversed the weak magnetic field in the core and so the generator was now self exciting in the wrong direction.

So, I hooked up the battery the other way round and let it run the generator as a motor again but this time in the direction I wanted. Sure enough, the core residual field was reversed again and now when I pedal, it generates current in the right direction.

With the car boot season starting up in earnest from this weekend, I'm still on the lookout for a decent permanent magnet motor to use on this thing.