Vectrino Z1 and Z2 velocities

Hi All,

What exactly is the Z2 velocity and how does it relate to Z1 and the other velocities in the Vectrino device?

I am looking at wave velocities and there is an obvious difference between the two traces (see attached JPEG) with the red trace being Z2. It appears to be smoother than Z1 (no significant spiking) so is there some low pass filtering going on? In addition the velocities appear to be shifted. Is this due to instrument misallignment?

Thanks,

Pascal

Hi Pascal,

The Vectrino has four receiver arms as opposed to the more traditional three found on other acoustic velocimeters. Because of the arrangement of receiver arms, each pair (composed of two opposing receivers) can measure a horizontal component and the vertical velocity. Z1 and Z2 are the vertical velocity estimates associated with receivers 1&3 (X) and 2&4 (Y).

There isn't any hardware filtering on Z1 that wouldn't also be applied to Z2 since all four receivers will receive the same ob board processing. I think the differences you are seeing between the two velocities point to a rotation of the probe head away from vertical so the vertical velocities Z1 and Z2 are contaminated with a small component of the horizontal velocity.

You can apply a small correction (essentially a pitch or roll) to this data to guess the misalignment angle, but it's probably on the order of 5 degrees or less. I use a post level (like this: http://www.amazon.com/Johnson-Level-Tool-175-L-Post/dp/B0009WG3HI) to minimize the rotation of the head when mounting a Vectrino or Vector.

P.J.

Hi P.J.
I think I follow. This means that in theory Z1 & Z2 should match perfectly. In practice the discrepancies are due to the misalignment angle of the probe and the errors associated with the X & Y velocities (where Z1 is estimated from a component of X and Z2 from a component of Y). Is this approximately correct?
Pascal.

Hi Pascal,

I think you've basically got it. I'd suggest reading through the Coordinate Transforms section of the Pulse Coherent Primer (http://www.nortekusa.com/lib/technical-notes/tn-027-pulse-coherent-primer) and taking a look at the Vectrino manual (Chapter 4 of the most recent version) to gain a better understanding. The Snyder and Castro reference and my conference paper from 2006 (both available at http://www.nortek-as.com/en/knowledge-center/bibliography) talk about contamination of the XYZ velocities due to a probe head rotation as well.

The difference between the two velocities you see isn't due to an error in the X or Y velocity, it's because as the head is rotated the vertical (Z1 and Z2) velocities are now measuring a component of the horizontal velocities proportional to the sine of the rotation angle. So, in theory Z1 and Z2 should match almost exactly (it won't be exact because they are measured using two different sets of beams). The fact they are so different to me indicates a probe alignment problem, probably a slight rotation about the Y axis.

P.J.

Thanks, very helpful information P.J.

so basically, X and Y are the horizontal component of the velocity

Z1 and Z2 are the vertical component of the velocity, is it correct?

if it is the case, and if Z1 and Z2 (in theory) should be the same, then does it mean that X and Y should be the same too?

somehow, my X and Y velocity are very different, not sure why it is the case.

for example: for one case, my X velocity is 0.72 m/s my Y velocity reading is -0.055 m/s

does not look good

if it is the case, and if Z1 and Z2 (in theory) should be the same, then does it mean that X and Y should be the same too?
somehow, my X and Y velocity are very different, not sure why it is the case.

Thanks Peter for your comments