Cornelius van Tonder

Hi, Richard (What a coincidence: an hour ago I arrived back home in East Sussex, UK after a 3hour drive from Birmingham, West Midlands,UK, where my daughter is starting her 3rd year at university next week!) Yes, so, I simply ordered it through Amazon: the description is something like "Neodymium rectangular/bar/small /collar stay magnets for crafts etc etc" the spec size they give is 10x5x3mm but one of the photos show more precise measurements of 9.01x4.52x2.57mm but close enough i guess.

I have had the same problem for a few months and had basically given up and gone back to my old Saitek Pro yoke. Then thinking about replacing the Saitek yoke potentiometers with Hall effect sensors, I thought I'd open the TCA yoke and "see what I can see". It was out of warantee and I wasn't going to spend money sending it somewhere to get fixed - for some it didn't help anyway. While the whole thing was apart but the sensor bit plugged back in, I noticed that with a larger, slightly stronger magnet, it seemed to work better. It didn't take long for me to realise it didn't really... However, I did notice that instead of moving the magnet in a "spinning" motion under the sensor, it showed the correct reading on the "calibration grid" when I moved the magnet in a straight line in front of the sensor, from one edge of the magnet to the opposite edge, essentially. But how could that work in the design and positioning of the sensor. I tried MANY things, and concluded - it doesn't... I then hit on the idea, that I have subsequently read in a post related to some other Hall sensor mod, that a small bar magnet might work. It turned out that, when placed just off-centre of the sensor, it still measured enough of a movement from N pole to S pole, to read nearly perfectly! However, it didn't seem to register an even movement along the whole range - i.e. it was more sensitive near the middle range than near the outer edge (apologies for not using the correct technical jargon here...) After replacing it all (more on that below), I adjusted the resonse curve slightly and got a near perfect roll movement, with which I am extremely pleased! Now a few caveats: I did "dig" out the original (VERY small!!) round magnet which did a bit of damage to the plastic into with it was glued/cast. Also: I could not put my little bar magnet in the same place. There was a rough hole and I would not be able to get the positioning just right. I therefor glued a very thin metal plate/disc (blade of an old, cheap extendable "Stanley" knife, rounded with an angle grinder (yes! I know, right)) onto which I was then able to place the magnet in the right position relative to the sensor, first to get the centre, and then checking the roll by turning the yoke. I had to move it (with a plastic tweezer) a couple of times to get the desired repsonse. Lastly, once I was happy with the roll movement, I dropped a little bit of hot glue carefully just on the edge of the magnet to keep it in place, with the option of easily removing it if I had got it wrong. Happy to report it was right first go!! I don't know how to add photos to this but I am happy to oblige if someone finds this useful. Cornelius