Why would you use an accelerometer with low bias voltage rather than the standard?

Thanks,

Do you mean an accel that has a bias of say 9.5 compared to 12.5? I've noticed the difference between brands and quality. This interests me as well.

Ron,

That's right.

There some other differences such as the impedence, power requirements, etc. but I really don't know enough about electronics to tell from that why I would use one or the other.

Someone will come along soon and help us out.

The bias voltage is a function of the electronics inside the accelerometer. The electronics draw power from the constant currect source for operation and set the bias voltage.

Let's consider a typical ICP power supply. Most of them supply a constant current of a couple milliamps at about 24 VDC. A typical accelerometer biases the voltage to 12 VDC by the amount of current it draws. This voltage is important because the AC output of the accelerometer rides on this voltage. If we have a 10 g peak signal, a sine wave of 1 volt peak will be imposed on this 12V bias voltage; we end up with an AC level varying between 11 volts and 13 volts.

The signal level can't go above the open circuit bias voltage (actually a volt or 2 lower than that depending on the components used) nor can it go below zero (actually 1 or 2 volts again depending on the components used). A 100 mV/g accelerometer can handle about 50 g RMS maximum, resulting in about 70 g peak, or a voltage swing of +/-7 volts. This get imposed on the 12 volts bias voltage, resulting in a voltage swing from 5 volts to 19 volts, which is inside our working range of around +2 - 20 volts with an ICP supply that puts out a maximum voltage of 24.

Some ICP supplies in little blue boxes operate off 2 nine volt batteries, supplying a maximum voltage of a little less than 18 volts (there is some drop in the current regulator circuit). The above 50 g RMS signal biased at 12 volts results in peak level of 19 volts, but the ICP supply can only handle 18. The signal is clipped at 18, resulting in distortion! This would be a good application for an accelerometer with a bias voltage of around 10 volts. You could use the entire range of the accelerometer without clipping.

Bias voltage is a function of the components in the accelerometer, but it can also be affected by other conditions. It often is a function of temperature, both static and dynamic. One model of accelerometer I've seen has a bias voltage as low as 3 volts in colt conditions. For any but the smallest vibration levels, clipping will occur.

One final point. IF low-quality accelerometers with an epoxy seal are used, the bias voltage can be tracked as an indication of moisture infiltration. If the bias voltage has been 11.5 volts at room temperature and it now reads 10, the epoxy seal has failed, allowing moisture inside the case. This is a bad thing. Very, very bad.

Jon
Spintelligent Labs

Thanks Jon, nice summary.

Part of my external pwr supplies are 27V and 18V. I use the 18V for high frequency acels such as with a sens of 10 mV/g. And basically everything else either off the 2120 or the PCB 27 Volt.

Thanks Jon.

Thanks, Jon.

I almost just called you and asked, but I figured others might be as ignorant as I.

Danny,

I'm always happy to answer acceleormeter questions! Feel free to call or e-mail; asking them here is great too to provide information to the group.

I have a lot of experience with a variety of accelerometers over the years....(ok, ok....I didn't get the expected results a lot of times!) so I can often provide some insight into getting good measurements.

Jon
Spintelligent Labs

Good Job Jon, You helped me build my icp power supply a while back. It's great reading explanations like this. I find them very informative. It's great having people like you on board to help us out.

Joe