# DC non invasive power measurment

For one of my PhD project I have to measure power consumption from different electric wheelchairs during various tasks and recor these measurment for later studies. I have not much experience in electronics so I am not sure where to start! I will neeed to measure current and voltage in a non invasive way and have a system that can be moved faast among the different wheelchairs. Could someone give me some guidance in terms of what sensors could be more appropriate and how I could do something like this?

Thanks

• Is inserting a device between the battery and wiring too invasive? The connectors just pull off, so it is completely non destructive and very quick. Apr 24, 2015 at 14:53
• No that should not be a problem Apr 26, 2015 at 9:16

The typical way of doing it would be with a clamp-on power meter.

The clamp-on DC power meter works by measuring two things in parallel:

1. The magnetic flux in a ferromagnetic clamp that encloses the wire with a DC current flowing in it.

2. The voltage across the load (or source).

Inside the meter, a measure of the flux is converted (scaled) into current, and multiplied with voltage to provide instantaneous power. The instantaneous power is time-averaged to provide average, or RMS, power.

This isn't totally non-invasive, as you need access to the battery leads - you'd clamp the meter around one of the leads, and attach the voltage probes to the battery terminals. But that's the best that you can do, unless the wheelchair has a power/energy logging function that's available to the user.

The meter itself might have a logging function that totalizes the energy, or you might attach it to an external logger. Finally, instead of using a dedicated power meter, you could use a data logger, and an external voltage-output DC current probe. The logger would log the voltage and current on two channels, and you could do the power and energy calculations yourself.

• Thanks. Would you be able to suggest a clamp-on power meter? The one in the link doesn't seem to have a recording function and that will make it unsuitable for the job. I will be keen on trying to use the external probe with the datalogger, again would you be able to suggest me some options in term of what sensors and data logger would be good for me? Apr 26, 2015 at 9:53

You can measure the voltage on the batteries without causing any disturbance by simply attaching wires to the batteries (perhaps with alligator clips). You should add an inline fuse near the battery to limit the current in case of a fault. The battery voltage needs to go to some kind of appropriate ADC.

Measuring current from the battery can be accomplished with a Hall-effect current sensor that can be attached to the battery lead (the battery lead may have to be removed to thread it through, or some can be clamped on without removing the lead).

You will likely need to sample at a very high frequency to get very accurate results, as the controllers probably use a high frequency PWM. Average current and voltage (what you'd get if you had a low pass filter and sampled more slowly) will give a rough indication. Without the low pass filter, sampling may well give you weird results because of aliasing of the sampled data. Typical Hall current sensors can respond up to perhaps 100kHz, which is likely well above what a motor controller would use.

Of course instantaneous power is the product of voltage and current (remember that current can flow in either direction), and energy usage is the integral of that product. Current may flow back into the battery during regenerative braking, for example, if it's a 4-quadrant controller.

• Perfect, thanks, will you be able to suggest me a couple of sensors that will be suitable for the job? Apr 26, 2015 at 9:21
• Try the Hall sensors from LEM, available from distributors such as Digikey Apr 26, 2015 at 11:11