2
\$\begingroup\$

I'm working on a project that allows a user to crank a hand generator while current and voltage sensors measure the output. After the user finishes cranking, software displays information about how much power was output.

This is the generator we're using: http://windstreampower.com/Human_Power_Generator.php

It seems like it should be a simple problem. However, in the current setup, I've been using the provided 12V voltage regulator and feeding power into 12V incandescent bulbs, and once the rig gets under heavy use, the voltage regulator burns out, and the bulbs blow.

The setup is like this:

Hand Crank => 12V Voltage Regulator => 12V Incandescent Light Bulb (50W)

So, I'm wondering if there is an alternative approach to this configuration (we don't need light bulbs or anything visible). Maybe feeding the crank directly into a 12V battery with some type of charge regulator and a constant drain? Maybe connecting the generator to several bulbs in series?

What is the simplest, most inexpensive dummy load we can attach to the generator, just so we can get current and voltage readings of the user's output while they crank?

\$\endgroup\$
  • \$\begingroup\$ Sounds like you are describing a resistor, but it is hard to know what the software is measuring/assuming. So I am not sure if your readings would still be correct. \$\endgroup\$ – Matt May 7 '12 at 21:27
  • \$\begingroup\$ The software takes the instantaneous current and voltage readings (a few times a second), and just multiplies them together to get a power reading. The user cranks for 15 seconds, then a graph is displayed of their power output over time, and some basic real-world comparisons ("You could power an iPhone for X minutes"). I'm more of a software developer than an electrical engineer, so I'm definitely green when it comes to understanding the implications of different loads on those readings. \$\endgroup\$ – Alan Christopher Thomas May 7 '12 at 21:32
  • 2
    \$\begingroup\$ In that case, @Brian Carlton is correct. Depending on the resolution of the current reading, a small high wattage resistor could be better than a large value resistor (which would reduce the current). \$\endgroup\$ – Matt May 7 '12 at 21:43
  • 1
    \$\begingroup\$ We set up something similar to this at school, without the power measurement. The generator produced quite a wide range of voltage, possibly up to about 24v if you went for it. We just used normal automotive headlight bulbs as the load - they seemed to be fine with the short term overload. \$\endgroup\$ – Cybergibbons May 7 '12 at 23:15
5
\$\begingroup\$

You need power resistors (e.g. from Ohmite). Simple, not that expensive (~$5 US).

I wouldn't mess around with the regulator.

EDIT: As @Alan Christopher Thomas pointed out, this will get hot. For extended use (more than a few seconds) consider an additional heat sink.

\$\endgroup\$
  • \$\begingroup\$ So, maybe something like this? amazon.com/Power-Rating-Resistance-Aluminum-Resistors/dp/… Would that mean I should hook up a heat sink, just to make sure it doesn't get too hot? \$\endgroup\$ – Alan Christopher Thomas May 7 '12 at 22:32
  • 1
    \$\begingroup\$ I picked up one of these this morning, and it seems to work really well, with nice readings: i.imgur.com/WmLUc.jpg The torque feels the same as it did with a 50W bulb. The heat is DEFINITELY an issue. It gets hot very quickly, so I'll try to find a big piece of scrap metal and thermal paste to see how that goes, then report back here what I find out. Thanks! \$\endgroup\$ – Alan Christopher Thomas May 8 '12 at 17:28
  • 1
    \$\begingroup\$ Maybe run a 12v fan over the resistor(s)...I don't know if this will have any impact on your readings. \$\endgroup\$ – Hair_of_the_Dog May 9 '12 at 3:47
  • \$\begingroup\$ Here's the setup with the resistor: i.imgur.com/XKzjz.jpg All the stuff mounted to the wood on the right is just sensor equipment for the computer. We mounted the resistor to the aluminum box pictured. It still gets hot pretty quickly, but cools off a bit faster than it did without the extra metal. The setup has pretty mild usage on weekdays, so we'll watch it today and see how it performs. There's an AC outlet next to the setup, so we may just get a small fan, separate from the hand crank system, to add some additional cooling to the resistor. \$\endgroup\$ – Alan Christopher Thomas May 9 '12 at 14:04
  • 1
    \$\begingroup\$ Performed really well yesterday. The crank was used about 70-80 times, and seemed to work great! Resistor was the answer. \$\endgroup\$ – Alan Christopher Thomas May 10 '12 at 13:58
6
\$\begingroup\$

If you happen to have a car battery available, that would work fine. There is no way hand cranking can overcharge a car battery, and it will maintain the voltage fairly steady. You'll also have your muscle work charging the battery, but I don't know how relevant that is in your setup.

\$\endgroup\$
  • \$\begingroup\$ I've also considered this a few times as an alternative. We have a pretty good sized 12V battery on hand. I'm leery about it, though, because the setup has already been used 2500 times in the few weeks it's been open to the public, and it will remain open, unattended until August. Maybe if I kept an MR16 lamp attached to the battery also to be a constant drain so it never overcharged? \$\endgroup\$ – Alan Christopher Thomas May 8 '12 at 17:31
  • 1
    \$\begingroup\$ @Alan: Lead acid batteries can be charged indefinitely at a float voltage, which is usually around 13.6 V for ordinary 12 V car batteries. Such a battery will draw enough current at that voltage that a hand crank is unlikely to supply enough power to overcharge it. If you're really worried about it, you could put a circuit accross the battery that discharges it thru a resistor if it's over 13 V or so. Then it will always be ready to absorb extra power. Basically, a car battery is "big" compared to the power a hand crank can generate. \$\endgroup\$ – Olin Lathrop May 9 '12 at 15:22
2
\$\begingroup\$

Cement resistors (caution, may get hot with enough current):

cement resistors

Note that resistor value will affect braking torque, i.e. the less the resistor value, the harder it will be to crank your generator quickly (probably not much difference at slow speeds).

\$\endgroup\$
1
\$\begingroup\$

One rather obvious solution is to drop the regulator and just use an incandescent bulb (or network thereof) which can take whatever voltage the generator puts out.

As a bonus, the user will have to crank hard to get the light bulb to grow more brightly and and at a higher color temperature.

With a regulator, there is no additional feedback for faster cranking once the regulator has enough input to maintain the voltage clamped at 12.

(No additional feedback other than the voltage regulator blowing out, that is, haha).

\$\endgroup\$
0
\$\begingroup\$

I am currently experiencing the same problem with a similar generator http://windstreampower.com/443540_PMDCG.php. I feel that using a dummy load produces too much heat in my application, where the speed of the DC generator cannot be predicted or regulated by mechanical means. I am currently considering some PWM approach for current limiting but in my system, I have no stable voltage reference to drive the gate on/off. This is a problem I have not yet solved.

Another consideration is this: the generators are rated to produce more current for short periods of time, and of course, the body of the generator will heat up. However, during this period ( 10 min or 15 min ) the generator is quite usable, and to waste that usable current by a resistor seems wasteful. A better approach would be to allow the generator to run over-rated for an allowable time, and then apply limiting. A thermistor could be attached to the body of the generator and monitor the case temp. When the case temp gets to high, switch on the current limiting circuit.

How to realize that current limiting circuit is still unknown to me. I am charging up a bank of supercapacitors rated at 25V. To the DC generator, they appear as a very high load indeed, a dead short when V=0 on the bank.

\$\endgroup\$
  • \$\begingroup\$ Not a bad idea. I did consider using a charge regulator and a battery, so maybe the regulator would have some type of built-in limiting. Not 100% sure. We only had 15 seconds of power at a time, and we were able to attach an enormous heat sink and leave plenty of space in the assembly, so a resistor worked fine. But, it did get VERY hot. \$\endgroup\$ – Alan Christopher Thomas Mar 8 '13 at 21:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.