Acceleration signals for shakers or convert to dislacement?

Wondering if anyone has been here before and can share their experience?

I have chassis acceleration signals available from a sim for driving a shaker set up and this all works and can be seen on the micro controller DAC outputs as a simple waveform.

Should i double integrate (with op amps) to get a position signal to feed to the shaker coils instead or would simple low pass filtering of the acceleration signal be sufficient for a convincing signal at the low frequency range of the shaker?

The system is a simple chair shaker experiment with moving coil shakers.

Thanks

PArdon? got photos?

No photos yet, still WIP.

The issue i was pondering is that my sim telemetry supplies acceleration data in 'g', the shaker ideally should be driven by displacement if modeled correctly and i was wondering whether i should convert i.e.

for a 1g assumed sinusoidal acceleration signal ..
at 10 Hz this would be 5mm peak to peak displacement
at 20 Hz this would be 1.2mm peak to peak displacement
at 30 Hz this would be 0.55mm peak to peak displacement
at 40 Hz this would be 0.31mm peak to peak displacement
at 50 Hz this would be 0.19mm peak to peak displacement

So you can see the big difference in output displacement with frequency for 1g acceleration. A low pass filter slowly rolling off from a chosen frequency or DC would possibly be a simple approximation of the conversion (this was my chosen option in the end as its easy to build and modify). The correct way would be double integration, both can be accomplished with simple op-amp circuits.

The telemetry samples at 100Hz so 50Hz is the maximum frequency which is also ideal for the shakers.

Anyhow after finally getting my amp working (free - but numerous burned out components) which was a nice find ARCAM 2*150 watt, i have 2 100 watt 'puck' type shakers working and it seems OK with the acceleration signals , the 'boost' for higher frequencies seems to work well in this situation, this has also saved a lot of coding or electronic effort.

My setup is a PIC micro on USB driving MPC4725 I2C DAC breakout boards which gives 12 bit resolution. The signal is then passed through a low pass filter (50Hz cut-off at the moment with no attenuation across the band of output frequencies) and buffer to give anti-aliasing, filtering and amplitude output control via potentiometer as the power amp has no controls of its own.

Working quite nicely but i need to work on mounting to my chair and the isolation of the drivers from the chair base as apparently this has a big impact on the transmitted levels of vibrations.

I have found that for a 'vehicle crash' situation i have needed to attenuate the output in software otherwise you cannot have so much gain (vibration level) for normal driving without overloading the drivers in a crash.