[quote=Nige321;9376615]PDR 1
Your elementary look at brusless motors is correct, but you are assuming that Dyson is using motors similar to your drone propulsion.
The basic principles are the same. The detail is different.
[quote]
So we're talking about a clever application, but Dyson claims it's a clever
motor, which it ain't IMHO.
There are, but not many in this application...
So you agree there's nothing inherently unique about the motor - it's a shame that Dyson't marketing pretty well all focusses on his "clever" motor, then.
The Dyson motors only have 2 (assymetric) poles, not the 9 to 12 used in 'normal' motors and as you note use Hall Effect sensors to sense armature position.
The two-pole motor config is simply the standard way of getting a high motor constant (high rpm/v number). Most applications for electric moptors are seeking to do the opposite and obtain a
low motor constant to get the torque to drive loads without the need for gears - hence the pole counts that go from 8 to as many as 200 poles to get Kv numbers in the more usually useful 2,000 to 100 rpm/v region. This specific application needs the opposite to get a small diameter turbine that will deliver the desired pressure/flow characteristics. It's not unique - even cheap, mass-produced commercial motors like the Mega-16EDF have two-pole configurations for similar reasons. This is not novel. Nor are kevlar/carbon wrapped rotors to provide magnet containment - these concepts have been in series production for over a decade in many places.
You don't. But you do need to be clever if you want to go from zero to 100,000 RPM in 0.7s, do it as the battery is dropping in V, and do it with a motor efficiency around 90%.
Getting the acceleration is simply a matter of motor power and having a controller that can track that sort of a speed ramp. 0.7secs might sound pretty fast when your normal experience is waiting for a big turbofan to respond to a throttle demand, but to a low-inertia electric motor it's quite a while, and to a microcontroller in a speed controller 700,000microseconds is a vast amount of time for doing the simple job of deciding when to switch poles on and off - especially when the chosen microcontroller has been selected to have a clock speed suited to switching poles at a bit under 2,000 times a second. These are not complicated electronic problems.
Can I suggest you search and read some of Dyson's patents...?
Can I suggest you do a little reading-up on the ambient state of the art in BLDC motors, and where it's been for well over a decade?
PDR