My first thought for thruster placement was to have thruster 5 be CW and 3 and 4 be CCW. It has twice the contribution for the throttle and pitch so it made sense to me to cancel out the most unintended yaw. I’m just wondering why the bluerov1 design does it so that 4 and 5 are the same and 3 is opposite? am I missing something here?
Additionally, I wanted to check why 0.45 is used for thrusters 3 and 4 in throttle rather than 0.5?
Another additional question(sorry!): If I were to create a custom thruster configuration in ardusub with just the 5 thrusters and run it with mavros, would it just ignore any strafe commands? OR is it better to use the blulreov1 configuration table with no thruster 6 attached?
Both approaches are functionally equivalent. The BlueROV1 approach means you don’t need to build a custom firmware, but if you want to set up custom / more optimised thruster contribution factors for your vehicle then there’s no reason for you to define an extra motor that doesn’t actually get used (unless you think you might use it at some point).
I haven’t had a reply yet, but will let you know if/when I get one.
In the meantime I’d say go for it and see what happens. I don’t expect it to make a huge difference, but ideally your hardware design allows replacing the propellers anyway, in which case you could switch them if you want to try the alternative approach (without needing to change your wiring)
I went with the safe option which is the original bluerov 1 configuration for now. Since it changes the insulated wire length, it’s non trivial to switch but possible once we have a verdict
I’m unsure what you mean here - are you unable to change the propellers on your thrusters without fully disconnecting them from your vehicle? That may make cleaning and maintenance challenging.
If you swap a propeller between CW and CCW then you can just change the corresponding SERVOn_REVERSED parameter (or re-run the automatic motor reversal detection test) and it should continue to work normally (albeit with changed counter-rotating torque effects, as is intended in this case).
Yep you’re right, I forgot I can just change the propellors, I was thinking changing the thrusters entirely. Should be quite straightforward in that case, let me know whenever you hear back:)
Following up, Rusty has replied, and he doesn’t remember the exact reasoning that was used for the propeller rotation direction assignments on the BlueROV1 frame[1], but has drawn attention to the fact that roll and pitch/vertical control both use motors 3+4, but only one of 5 or 6. Depending on your relative use of those control axes (and/or their combinations), the optimal propeller directions for minimising secondary moments may differ.
He did also mention (in response to your initial post):
Perhaps not too surprising, given it was designed over a decade ago, as part of the initial Kickstarter that started the company ↩︎
