In this case, the input voltage to the amp is inversely proportional to the total capacitance, so changes in cables affect sensitivity. As the cited papers discuss, the generated charge applied to the internal (transducer) and external (cable and amplifier input) capacitances in parallel results in a voltage, which can be amplified by a voltage amplifier. Jon, ALL piezo transducers are basically charge generators and can be amplified by either voltage amplifiers or charge amplifiers. I have also been involved in the application of piezoelectric transducers over a 40 year career in noise and vibration consulting, and I have been involved (with others) in the design of accelerometer amplifiers due to cost, weight and complexity of commercial units. Looks like you guys are making rapid progress toward a solution, but let me toss in some comments. I have a sketch of the circuit I've used which I'll dig up if it will help - it's hiding at the moment. To separate the AC part of the signal, a DC blocking capacitor of 10 uF is usually used.Īn ICP supply is pretty simple to make using 2 or 3 nine-volt batteries and an LM7805 configured as a constant current source.
The AC output signal is imposed on the resulting DC bias voltage. This type of sensor requires a constant-current power supply of 4 - 10 mA. I suspect the cost might be lower than Kistler's price if you don't have the sensors already. Check out their website and search for force transducers: Their sensors incorporate an intergal charge amp, commonly referred as ICP electronics (PCB's trademark, also called low-impedence sensors). You might consider force sensors from PCB Piezotronics. All those nice signals they had been analyzing were the result of cable flex! The only problem was that when the test was finished, somebody realized that none of the accelerometers had actually been connected. There have been some remarkable screwups in the past such as a aircraft wing modal test with hundreds of accelerometers installed all across the wing. The stuff is expensive and difficult to work with, as the graphite will kill the signal if it ends up in the wrong place when assembling connectors. Low-noise cable has graphite on the shield so that any charge caused by flexing is bled off. Since you're measuring charge, guess what this means. When a non-low-noise cable is flexed, motion between the jacket, conductors and dielectric generate a charge. The second requirement of using a charge mode transducer is low-noise cable. Be careful to remove solder flux and any other possible contaminants from around the high-impedence part of the circuit. For accelerometer connections, even touching the gold connector pin with your finger is enough to screw up your signal and get nothing but noise. No good ideas on a charge amp circuit, but having spent much of my career using charge-mode accelerometers, I have a few pointers.įirst, all of the connections made to the piezo transducer must be absolutely clean.
0 kommentar(er)
