Switched capacitor filter for known phase appnote?

[juanfhj]

virtual reality magnetic position tracker

Hi. I'm working on a virtual reality magnetic position tracker for motion capture. I wrote about it in Circuit Cellar 133, and I'm planning to improve it (you can check it out at Youtube tracker video https://www.youtube.com/juanfhj#p/u/5/W73qbm6ggG0 ).

Basically I have to measure very tiny magnetic fields. In the current tracker, the user is inside a cube with strong fields. For the next version I want to have a small magnetic field generator instead.

For this, I have to measure tiny magnetic fields with field sensors. Now, these sensors have a lot of noise, so I need to filter the signal. It's not straightforward, cos the circuit operates at 100 kHz, so no active filters here. I thought switched capacitor, which would be the first time I try my hand at it.

Sure there is the standard switched cap schematics, but these don't take advantage of the fact that the phase of the input signal is known. So I'm trying to figure out a custom switched cap circuit that uses this information. Some sort of variation on the LMF100.

Basically I want to charge the capacitors in tune with the generated magnetic fields. I thought a custom topology could achieve this, rather than going for a generic design.

For a rough draft on the tracker, see

tracker_draft_2010-10-26

Thanks

Juan Herrera
Grad student
Universidad EAFIT
Colombia

 

[juanfhj]

Virtual reality Magnetic position tracker

The project I'm working in is a virtual reality magnetic position tracker for motion capture. I wrote about it in Circuit Cellar 133, and I'm planning to improve it (you can check it out at Youtube tracker video).
YouTube - Captura movimiento feria createc 2007 monterrey

Basically I have to measure very tiny magnetic fields. In the current tracker, the user is inside a cube with strong fields. For the next version I want to have a small magnetic field generator instead.

For this, I have to measure tiny magnetic fields with field sensors. Now, these sensors have a lot of noise, so I need to filter the signal. It's not straightforward, cos the circuit operates at 100 kHz, so no active filters here. I thought switched capacitor, which would be the first time I try my hand at it.

Sure there is the standard switched cap schematics, but these don't take advantage of the fact that the phase of the input signal is known. So I'm trying to figure out a custom switched cap circuit that uses this information. Some sort of variation on the LMF100.

Basically I want to charge the capacitors in tune with the generated magnetic fields. I thought a custom topology could achieve this, rather than going for a generic design.

Any pointers to an appnote or something?

Here's a link to a brief summary I posted on scribd:

tracker_draft_2010-10-26

Thanks

Juan Herrera
Grad student
Universidad EAFIT
Colombia

 

[crutschow]

If you have the phase of the signal then you could perhaps use some form of synchronous detector. That would eliminate a lot of random noise. Such a technique is often used to detect very low S/N signals such as from deep-space craft.

I remember years ago seeing a type of synchronous switched-cap filter which consisted of numerous RC filters, each connected to an analog switch. The switches were driven at the desired filter frequency in sequence (such as with a shift-register), with a repeat period equal to the desired detection frequency. If the detected frequency and the switched frequency were synchronous, then the caps would charge up to the envelope of the input waveform. All out of frequency signals would average to zero.