Quadrature question

[golfnut]

In IQ system there are 2 a/d channels each is fired with 90deg time split. This gives amplitude 1 at time 1 and amplitude 2 at time 2.


Then I thought, could you use a single channel and one a/d to get quadrature channel information by reading 1, waiting 90deg phase time then doing reading 2 ???

Then invTan( reading 1 / reading 2) is your alloy phase

This would get rid of phase, gain, DC offset imbalance between channels - as both samples are on same CH. !!

Food for thought?

S

[Carl-NC]

Whether you do direct sampling (as you are implying) or sample the outputs of analog demods (as most designs do it) you can either use 2 ADCs or mux into a single ADC. Using 2 ADCs does add potential mismatches, but they can be calibrated out in software.

[Davor]

You can do that, provided your AD is fast enough. Why not.
There is however some additional bonus in IQ mixing and LPF prior to AD in sense of improving noise and reducing the bandwidth if you are using some pedestrian speed AD, but as you can perform this in digital domain as well - why not direct samples.

There is a "phase slip" method of establishing a signal phase by taking 4 samples in a wave period as described in a Circuit Cellar article on a proton precession magnetometer. It can also work for discrimination.

If you consider 4 successive time samples at ~4*f(Tx) as a single instantaneous phase sample, you'll have 4 voltages v0, v1, v2, and v3, and the phase sample is calculated as:
Φ=arctg((v0-v2)/(v1-v3))

It works.

[Carl-NC]

If you consider 4 successive time samples at ~4*f(Tx) as a single instantaneous phase sample, you'll have 4 voltages v0, v1, v2, and v3, and the phase sample is calculated as:
Φ=arctg((v0-v2)/(v1-v3))

It works.

Prizm 6T works this way.

[golfnut]

Reading thru A/D tech - is it true that if one ch is fed to the A/D input and one ch is fed to the REF input the resultant output word is the ratio of the two. (A/D needs access for External REF in).

If true, user has result of the division already, need only do invTan of the result for phase?

and you woudnt need a simultaneuous A/D or a mux to resolve IQ ch's

S