Pedal-Powered Home Office: Charging a Battery & More

[samsonmustang]

So I want to charge a battery from a pedal-powered dynamo, and use that to power my PC, my screen, and pretty much everything else in my home office.

This whole thing is a little project to improve my own health while I'm working; I'm not here expecting to make major miracle savings on the electric bill or anything like that.

I'm still thinking about where to buy/how to make the dynamo (and much gratitude to everyone who's already helped me to better understand how they work.)

But when it comes to the battery, there's a lot I still need to understand.

My reading gives me the impression that charging a battery (whichever type, still not sure what's best) is an awkward task, and a battery charger has a precise voltage/current curve it follows, and if it doesn't, the battery explodes. Or at least, suffers and loses time out of its operational lifespan.

Meantime, the current coming out of the dynamo is going to be all over the shop, and cutting out irregularly. (My software/networking background inclines me to "always assume malevolence": ie, imagine that the user of the pedals will be actively TRYING to wreck the battery - therefore I should make that physically impossible to do.)

So what do I need in between the dynamo and the battery? At a guess:

- something to rectify the current coming out of the dynamo (because I can't see how you'd charge a battery with AC.)
- something to buffer it/smooth it out so the battery charger can be as constant as possible. (Batteries take what, multiple hours to charge? Meanwhile I might want to pedal thirty minutes and pootle off to have my dinner. We can afford a lot of wasted effort, since it's only a toy. But still, it would be satisfying to be as efficient as possible.)
- something to actually charge the battery, with its precise current curves and shut-off circuitry to prevent overloading.
- then presumably something to invert the power coming out of the battery, so I can wire it up to the standard 13amp plugs that everything runs off. (I live in the UK, so that's 240V at 60Hz I need coming out.)

I'm a bit naive about all this, so I haven't really got any numbers on how long the battery will last, or how much work it will be to charge it up all the time. (It's only the for office, I'm not going to try running the fridge and washing machine by pedal power.) But all the same, maybe I also need something to automatically switch it all over to the mains once the battery's dead.

Can all / any of these stages be bought as off-the-shelf components, or am I building them all myself? (Do things like UPSs already have to do a lot of this already?)

Also, are there any important stages I've forgotten, or problems I'm not seeing?

 

[Baluncore]

Do not use a dynamo. Use a second-hand alternator from a car. The alternator will have an internal three-phase rectifier that will produce a DC current.

Make the field current to the alternator manually adjustable from operating position so you can select optimum RPM and torque for your training. That field control method works because the voltage generated is the product of RPM and field strength. If you pedal faster than optimum RPM the torque will increase as more voltage causes more current flow into the battery.

The battery voltage will be reasonably well fixed by the chemistry of the battery. You must accurately detect over-voltage to prevent damage by overheating or gassing of the battery. That over-voltage detector must be designed to reduce the field current, or to spill excess generator output through a load resistor to maintain optimum RPM and torque for your training.

You will need to gear up the alternator speed from the pedal sprocket. You may need to use a two stages of chain and sprocket speed increase. The higher speed second stage will need good lubrication in an oil-bath. Avoid 'V' section belts as they are very inefficient. Consider stepped belts if you can get stepped pulleys and belts at a good price.

 

[CWatters]

You can actually buy the whole thing ready to go...

http://www.mygreenerhome.co.uk/inde...=export_feed&gclid=CJfV2oSDp8sCFQso0wod8jIE8w

but read an understand the specs.

If you want to build your own I would use a 12V lead acid battery as these are easier to charge than other types. There are also plenty of inverters around that will turn 12V into 240V AC 50Hz (it's 50hz in the UK not 60hz). They are less hazardous than say Li Cells but I suspect not as efficient.

As expected the tricky bit is the actual generator and charge control side. Googling for Pedal Powered Battery Charger finds this vid and a link to a charge controller..

http://lifehacker.com/build-your-own-bicycle-powered-battery-for-emergency-po-1456439155
http://www.mdpub.com/555Controller/index.html

The latter is a "design" not a "kit with instructions" so you would need some electronics/assembly knowledge to put it all together and perhaps modify it to suit your motor/alternator.

As for performance... The system comprises..

You
A bike
An alternator
A charge controller
A battery
An inverter

Start with you...

https://en.wikipedia.org/wiki/Human_power

...adults of good average fitness average between 50 and 150 watts for an hour of vigorous exercise. A healthy well-fed laborer over the course of an 8-hour work shift can sustain an average output of about 75 watts

Then let's assume each part of the system is 85% efficient. It might be possible for some parts to be more efficient but for now let's make that assumption, so..

Overall efficiency = 0.85⁵ = 0.44 or 44%

So if you generate 100W for 1 hour you would put enough into the battery to deliver 0.044KWH (44W for 1 hour) to the equipment. The battery and inverter would allow higher powers to be drawn for shorter periods (88W for 30 mins, 166W for 15 mins etc).

If your laptop and other equipment can work off 12V that would be better than using an inverter.

So your first step is to be realistic about how much power you need and how much effort you can put in. If you needed a lot of power you might have to pedal for many hours just to run the kit for one hour. Your mileage may vary.

 

[anorlunda]

I note that spin cycles used for exercise supply their own electronics with pedal power. They dump excess power into friction via the "resistance" adjustment. Adapting that to charge a battery might be the shortest path to your goal (if charging the battery is your #1 goal).

If exercise is the #1 goal, then forget the battery and just buy a spin cycle.

 

[samsonmustang]

Wow, thanks everyone for the helpful responses.

I like the idea that I can change my torque up and down by adjusting the field current. Presumably I use a variable resistor for doing the adjustment - but wouldn't that be wasteful because the 'spare' current just goes into heating the resistor?

It seems the thing to do is set your torque/resistance, then cycle at a set speed to give a constant power out. What if I want to be able to cycle at any speed, with any torque, (changing either at any time,) and harvest all possible power from that?

My mental model is saying that power should accrue like water in a tank - so even one turn of the motor should slosh a little more water in there. I know to be useful you need to get that water out at a constant rate and pressure (ie current and voltage) - so is there an electrical analogue to a float-operated valve?

I get the concept of over-voltage - look at the voltage coming out of the battery, take action if that value is too high. But it seems going into the battery, it doesn't matter what voltage/current you have? (So long as you're 'pushing' in harder than the battery's pushing back.) Is this true, or is that a dangerous way to behave?

Also, coming out of the battery, I assume the voltage drops over time as power is used up, but the inverter compensates somehow so I'll always get my 240V out? If I want to have other DC outputs as well (eg, a 12V plug for a laptop, or a microUSB to charge a smartphone, etc) presumably there's voltage regulators I can buy to do the same thing (which are hopefully more efficient than just plugging the laptop power supply into the 240V output).

As I see it now, it goes:
- I spin my crank, which turns my alternator. (Or, my motor+3ph rectifier, if I can't get an alternator.)
- a dial on my handlebars controls a variable resistor that limits the field current to the motor
- as the motor turns faster than it should, that extra power sloshes out of the motor and into the battery.
- my difficulty in turning the motor is governed by how much the motor's magnetic field is fighting against me, plus how much power is already in the battery, pushing back.
- voltage sensor across the battery, shut it off when it's at full voltage. (I could even have an array of batteries, and use an arduino to automatically move from one to the next as they fill up.)
- coming out of the battery, pipe to an inverter and/or a bunch of transformers to give me common and convenient plugs - laptop power supply, usb power, etc. etc.
- some heads up display to let me know how much power is left in the battery (at a guess, I find this by reading instantaneous voltage and looking it up on a curve?)
- one last switch, looks at voltage sensor and swaps to mains power once the battery's dead.

Is that about right, or am I horribly off the mark?

My apologies for all these dumb or obvious questions. Things are resolving into focus, but I got to test my understanding to make sure I've got it right. (Hence the baby-simple phrasing. I can't be sure I understand the concept 'power' if I only ever use the word 'power' to talk about it, can I?)PS anorlunda: yes, exercise is the main goal - but so is learning about power electronics. I've had a look at a couple of spin cycle machines and got my head around how they work; but I think it would be rather unsatisfying to just buy something in.

Once again, thanks for all your help.

 

[Baluncore]

Things are resolving into focus, but I got to test my understanding to make sure I've got it right. (Hence the baby-simple phrasing. I can't be sure I understand the concept 'power' if I only ever use the word 'power' to talk about it, can I?)

Terminology is important to understanding and communication.
Power is the rate of energy flow, watts = joules per second. Energy is stored, not power. If 1A flows into a 12V battery, that is 1A * 12V = 12 watt. If the 1A flows for one second then 12 joules of energy will have been stored.
A current of one amp is a flow of one coulomb of charge per second.

If the battery is a “12V lead acid car battery” then it will need the charger voltage to exceed 13.5V before it starts to accept energy. The charger voltage should never exceed 14.5V. Battery voltage when idle after charge will gradually fall to 12.5V. Under load the battery voltage should be just over 12V.

The 12VDC to 240VAC inverter will produce a regulated output so long as the battery voltage is between 12VDC and 14.5VDC. You can run 12V chargers as used in cars directly from the 12VDC battery without needing an AC inverter.

 

[CWatters]

My mental model is saying that power should accrue like water in a tank - so even one turn of the motor should slosh a little more water in there.

Correct, provided the pressure (aka voltage) is high enough to lift the water over the rim of the tank. If not then nothing will go in.

Likewise it could be dangerous to put too much energy into a tank of limited size or put it in too fast. That's the job of the charge controller to manage.

I know to be useful you need to get that water out at a constant rate and pressure (ie current and voltage) - so is there an electrical analogue to a float-operated valve?

Yes it's called a regulator although in this case you want a power inverter...

https://en.wikipedia.org/wiki/Power_inverter

A regulator essentially takes water at one pressure (one voltage) and either reduce or boost it as necessary. Inverters not only regulate but also convert from 12V DC to 240V 50Hz AC.

 

[sophiecentaur]

If you are supplying Pedal Power then you really don't want to waste any of the energy you are putting in. Take a clue from how a good solar PV system works. A good charge controller uses absolutely all the available energy with a buck-boost converter, which will give the most power it can to the battery, whether the light is high or low, by boosting low levels of volts to the required charging voltage and will let bursts of charge into the battery even when the PV output voltage would be too high.
I know people favour car alternators but they are very clunky devices that will not be very efficient with the sort of power transfer that a pedal drive can deliver.
You have to use a commercially available bicycle alternator - or even two or three at the same time and have direct drive (no friction drive is efficient enough). The rectified output from each alternator should, I think, have its own regulator ( they should coexist quite happily). Charge controllers are not hard to build and you should look for a pretty sophisticated design. A good buck controller will limit the power you can put in and stop you 'overdoing it', it will just allow you to pedal faster - easier and find a speed that suits an individual pedaler.
You have to be prepared to be pretty knackered whilst charging your battery and 100W is about the maximum you could expect to deliver over a useful length of time. You need to make the job as easy as possible or you will grow to hate the task. How many hours / minutes are you prepared to pedal each day? Will that supply your energy requirements over the whole day?

 

[samsonmustang]

Okay, I got my hands on an old DC motor and I used the battery out of an electric hand drill.

- Connecting the battery to the field windings made the motor start spinning. Result.
- Swapping the connections swapped the direction of the motor. Putting a diode in there ensured it would only turn in one direction, and not the other.
3/3 So far the laws of physics are doing what they're supposed to.

I connected the windings to the battery against the flow of the diode - so the motor wasn't turning - then put a voltmeter across the diode ends. That reads 13.44V - sounds about right? (At a guess, that means the battery is about halfway charged, since it's supposed to fall from 14.4V to 12V?

Then when I turned the motor by hand, I saw that voltage increasing to about 16V. Turn in the other direction, it fell to nearer 10V.

Right - so all I have to do is turn fast enough to drop that 10V down to 0V, and then once I go faster then that, current will start flowing into the battery, from the motor. Correct?

Going as fast as I could, I could get it to 6V. But that was using my fingers on the fiddly little motor shaft. Next step is couple it to a proper handle - and actually I may be able to get hold of a gearbox. Not sure when that will be, though.

My predictions/questions/thoughts for next time:
- when I get the voltage across the diode to 0V, and begin pushing down into negatives, at that point the battery should start charging.
- this presumably will make the battery start acting as a resistor somehow, and I will find it harder turn the handle as a result.
- what happens as I increase the speed? Will the voltage stay constant and the handle get tougher and tougher to turn (as the battery charges faster and faster) - or will the voltage keep rising, the battery charging at the same rate, with the extra energy going towards heating up the battery until it explodes? I'm assuming the second one.
- I thought I'd need a rectifier/alternator, yet the DC motor seems to be giving me a DC output already. Presumably (when I get to that stage) I can have some basic switch to swap the connection if I start pedalling the wrong way, letting me harvest power from both directions. Unless I'm using the wrong bit of kit? The thing I've got says 5000RPM, 95V and 3.9A on it. I have no clue if there is a right or a wrong make/model of motor for me at this point.PS. sophiecentaur: thanks, that answer was a fantastic lead. Spent a couple hours reading about solar panels, now to learn more about buck-boost converters. They seem like the exact solution to the problem of just harvesting all the energy you can. Question: how long can that inductor/capacitor pair store energy for? If instead of an hour of genteel pedalling, I want to beast it for ten minutes then walk away, can this circuit take that energy and feed it in slowly over the whole hour? (My understanding of how this works may be off, still got reading to do.)

You need to make the job as easy as possible or you will grow to hate the task.

I had a thought about this: once I know what I'm doing, I might try to add a kind of proportional mains-assist setup - where for every joule I put in, nine more are drawn from the mains. (I don't know how this would work as an electronic circuit, but I'm happy to go digital on it if need be.)

That way I have to exercise every day or I can't use my laptop (which is the goal) but the gruelling reality of how much power I really need is hidden from me. But that's just a thought for the future, I'm putting it to one side for now.