angryowl

Have not tried the chip the other way around. But I think it should work as that magnet is quite powerful and its magnetic field should penetrate the breakout board and the back of the chip. Curious to see if it actually does though.

My magnet is centred on the chip and found no issues whatsoever regarding accuracy, so maybe positioning the magnet's centre exactly above the sensor location within the chip is not crucial. Having it centred on the chip just makes it easier to mount in my opinion.

Yeah, I also found the breakout boards to be very helpful indeed. I had considered soldering them onto stripboard, but for the price of a SOIC board just wasn't worth it. Plus I find the legs that go into the board can be used to attach the small chip/board combination much better.

I wouldn't worry so much about the weight. Mine weighs roughly 30g for the vane itself plus 80g for the casing and then the M6 rod which weighs the most. Add to that the tip with the counterweights inside, and the sleeving on the rod and you end up with a fairly heavy vane. There's some intentional slack in the way my two bearings were mounted to ensure the vane is sensitive enough to pick up even a light breeze. I don't yet have an anemometer to measure the minimum speed at which it starts spinning but I can tell you that it's REALLY sensitive and even my smallest fan sets it off!