The waveform looks to be correct.
As a sensor
If you want to use this as a sensor, treat the piezo as a voltage source with a big series resistance. E.g. you can put a 100 kOhm resistor across the piezo output, and you will see that the generated voltage is much lower. For sensor usage, ensure that a high mechanical input will not produce a voltage which kills your processing device (e.g. a microcontroller) by using a series resistor of a few hundred ohms.
As a power source
If you want to use this as a power source, well, that is not a shiny way. You can get out very little electric power from a piezo. In order to get the most out of it, you have to have a mechanical input which is having the same frequency as the resonant frequency of the piezo (i.e. tapping with a pole won't work). For more info, refer to http://www.piezo.com/tech3faq.html#app4
You can measure how much power this produces by rectifying the output, and connect it to a resistor of known value. Measure output of the piezo, and current in this circuit, multiply it and you get a (very little) wattage. This is a non-periodic signal, so you may wish to connect the measurement to an excel sheet where you can mmultiply voltage and current at each measurement point and calculate some average. Alternatively you can feed your measurement to a buffer, and drive an averaging circuit.
Example measurement setup to measure generated power
A good mechanically triggered piezo shall produce a voltage of 4V and over. I'd suggest starting with 1k-2k loading resistor, which simulates a load. If you have a 2 ch digital oscilloscope, do the followings:
- measure the voltage across the pins of the piezo. One pin will be GND, another pin CH1.
- split the load resistor into two, i.e. have two 1k in series. The split point will be CH2.
Now you have a voltage signal on CH1, and a current signal on CH2. Make a measurement, feed these to Excel. Use I=U/R to convert CH2 measurements to current, i.e. 1V on CH2 will mean 1mA (current flowing trough 1k resistor between CH2 and GND, right?)
Once you have the CH1 and CH2 points in excel, simply multiply them at each data point to get a power at the given point. Note that even sometimes you have negative voltage on CH1, that also means power, since also the current flows in the opposite direction, this way you will always get positive values for power.
Now you have a column telling power (energy) generated in the timeline. Note that this will be bigger at the hit, and will be slower at the end. You need to average this out to get an average power.
If you want to use it for serious energy harvesting, you shall try to give a continuous mechanical input (i.e. use a vibrating motor to hit the piezo, or something like that during your tests). Otherwise it is very hard to measure anything if you can not repeat the measurement reliably.