Official swirl meter discussion thread

[DeeEmm]

This discussion is for all things swirl meter related.

Regular visitors will know that swirl meter support was recently added to the firmware.

If you are wondering what a swirl meter is, or how it works, here's a video of the Performance Trends swirl meter where its units operation is explained.

https://www.youtube.com/watch?v=fgqvHwF3HhY

Our swirl meter differs from the PT unit in that the concept is based on a quadrature encoder (two channels with 90 degree phase shift) to detect both speed and rotational direction. The speed is displayed on the main dashboard as +/- rpm.

The swirl meter code uses the majicdesigns/MD_REncoder library and should work on any two channel (3.3v) quadrature encoder or on any encoder signal that has suitable level shifting to the ESPs 3.3v input level.

Whilst there is some basic code in place that will read the encoder data and display it on the GUI, the code is at this stage untested. I ordered (and have now received) two Polulu micro motor encoders which I plan to test with. The idea for trying the Pololu encoders is that they are very small and so only minimal support structure is required. This should hopefully reduce the mass of the mount and not make as much of an influence on the air flow.

https://www.pololu.com/product/2591

The Pololu encoder trigger wheels require a 1mm shaft, which is pretty small. I had initially assumed these were a little larger and had planned to mount the shaft on two bearings, but I think that given the size of the shaft and the low inertial mass, coupled with the driving force from larger vanes, this should happily run without bearings. My initial plan is to use three thrust washers under the vane end of the shaft to help address friction from the weight of the shaft and vanes but I actually expect this load to be nominal given the size of the shaft. 1mm bearings are actually available should the need arise, but following the KISS principle it is worth exploring the simpler design in the first instance.

The low speed and low load characteristics allow to mount the shaft directly into the 3D print, allowing the plastic to act as bearing material. Material choice may be a consideration with something like nylon or PTFE being optimal, but minimising shaft contact by using considered design will minimise friction and allow pretty much any material to be used.

The lightweight nature of this particular setup may however be an issue in as much as it might not be super robust, but testing should hopefully determine if this is a problem. I suspect that it will be fine in operation as the swirl meter sits inside the cylinder but may be fragile to handle and store. There's always opportunity to increase shaft size etc should this turn out to be a problem but this increases the inertial mass which in turn impacts the overall operation of the vanes.

I'm tied up with regular life for the next week but I'll sit down and put a 3D design together after I've got that stuff out of the way.

At the moment I'm thinking of three or perhaps four vanes rather than two. I suspect that the PT design has two vanes simply for ease of manufacture but we do not have such constraints with 3D printed parts