# Measuring DC output

Discussion in 'General Electronics Chat' started by jvanland, Feb 15, 2004.

1. ### jvanland Thread Starter New Member

Feb 15, 2004
2
0
This is a science fair question (yeah, and I'm the dad.) Our measurements came out the inverse and I am trying to figure out why.

Proposed study is to measure the change in power output of a simulated wind turbine propeller (fabricated from a 12" brass strip) as the diameter is reduced (by trimming the ends of the propeller.)

Propeller drives a small permanent magnet DC moter used as a generator, a box fan provides a steady wind. The load is a 10K ohm resistor and a solid state multimeter is used in parallel with the load. I had thought that as the prop was shortened, the power output would drop, indicated voltage would be lower, graph the results and you are done.

Except that the measured voltage increased as the prop got shorter until the prop lacked sufficient torque to carry the load. The measured voltages were in the neighborhood of 0.20 volts.

I have read in the tutorials here that the meter used this way (parallel with the load) participates in the circuit - my manual says my meter has 1 M ohm impedance but I'll be darned if I can figure out how to reconcile the measurements nor how to get a meaningful set of measurements. After my ten-year-old fabricates a new prop, that is.

Any help in sorting this out would be greatly appreciated. My only theory is that since resistance remained constant, as current dropped the voltage would have to increase but I can't get that out of the power equations.

John

2. ### Dave Retired Moderator

Nov 17, 2003
6,960
144
Thats quite an interesting problem John.

I have read in the tutorials here that the meter used this way (parallel with the load) participates in the circuit - my manual says my meter has 1 M ohm impedance but I'll be darned if I can figure out how to reconcile the measurements nor how to get a meaningful set of measurements.

This problem of insertion error is right on the edge of what engineers would consider acceptable. They normally recommend the meter is 100 times the impedance of the load measured and your meter is exactly 100 times the value of the load. However this shouldn't pose a problem such as the one you are describing.

My only theory is that since resistance remained constant, as current dropped the voltage would have to increase but I can't get that out of the power equations.

This wouldn't stand-up to theory because if the resistance of the load is held constant, an increase in voltage will corresppond to an increase in current, or more correctly, vice versa.

Is it not possible that the decreasing mass of the propeller is reducing the physical exertion on the motor itself. You may have to look at is the motor underperforming at the start of your experiment and the performance getting better as you remove bits of the propeller. This would explain why your seeing this rather bizarre behaviour.

3. ### jvanland Thread Starter New Member

Feb 15, 2004
2
0
Thanks for the answer about the metering arrangement. It removes one level of uncertainty. Everybody agreed with your assesment of power output - if the measured voltage increased, so did the current.

Collectively, I've cost my consulting engineer friends a fair amount of billable hours today.

One, (thermodynamic design and naval propulsion systems) offered an opinion along the lines of your speculation. The weight of the rotor (the brass strip) was causing the front bearing of the dynamo to bind and as the weight was reduced, so was friction in the system. Can be evaluated by changing the orientation so the gravity component is along the shaft axis instead of across.

Another, (electrical engineer in California) suggested the rotor geometry deserved another look - small scale models don't behave in textbook manner.

We'll probably pursue the latter and bring it to conclusion. Been a good opportunity for dad and daughter to learn about dealing with unexpected results.

John

4. ### Dave Retired Moderator

Nov 17, 2003
6,960
144
Another, (electrical engineer in California) suggested the rotor geometry deserved another look - small scale models don't behave in textbook manner.

This is a good point, one which I didn't think about. The only real way to know is to experiment.

Keep us posted on your conclusions, whatever they may be. Cheers