Eagle Tree Systems Brushless Motor RPM Sensor?

Tookings

Alright -- was looking for some pointers, and maybe some help.

I plan to use an Eagle Tree Systems Brushless Motor RPM Sensor or similar to detect RPM of a brushless motor. Very helpful if I can get it to work, as it requires only an electrical connection to motor power, and no encoders, etc. (And the motor/speed controllers are all RC gear and respond to easy servo type signals.)

Device is here: middle of www.eagletreesystems.com/Plane/plane.html or www.eagletreesystems.com/accessories/7.htm

I know it's power and pinout -- and that the signal requires a decent amount of software work to get an RPM, but that is all I'll be able to get.

What is the best way to start finding out how to read and process this signal? I have a few props, and a very old oscilloscope that I'm still trying to get to come back to life so I can take a first look -- but no luck yet.

If I don't have any luck, would anyone possibly be willing to trade a free RPM sensor sent to them in return for trying to figure it out?
(You would need a brushless motor and controller though of course....)

As for the actual project, I'm now working on a 4-prop flying vehicle (like the dragonfly) but with variable pitch and constant (mostly) RPM -- much like a constant speed prop on a real plane. (working with this at the moment for the props, and using the motor controller's fixed RPM governor for now: www.electrifly.com/miscproducts/gpmg4501.html) Once I have a stable hovering platform, there should be some interesting possibilities with negative pitch...maybe even some light acro, like flipping and flying upside down....especially being able to use pitch, instead of the delay in spooling up or down motor RPM.

Thanks for any ideas! (And if I figure it out myself...I'll post code, as it looks like a pretty useful sensor.)

deSilva

There have been two or three threads around here during the last week about BLDC; I think Joerg even gave the three phase power control needed for them.

I should assume, that this "tap" just catches the small changes of voltage in one of the lines and - if lucky - forms a clear square wave for each turn of the motor...

Tookings

Thanks deSilva. I used the wrong search again...

Fred Hawkins

http://forums.parallax.com/showthread.php?p=617610

Take a look at the pdf pointed to in Beau's post.

simonl

@Tookings: Without a datasheet it's going to be a hands-on discovery process :-( Here's a few thoughts on what it might be sending:

1. Fixed-width pulse for each RPM (or part thereof)
2. Pulse-width corresponding to RPM
3. Serial stream of RPM data

If it's either of the first two, I'd suggest you create a simple test that just logs the pulse-widths & CNT values.

If it's a serial stream, then hooking-up one of the serial objects should help.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cheers,

Simon

---

www.norfolkhelicopterclub.co.uk
You'll always have as many take-offs as landings, the trick is to be sure you can take-off again ;-)
BTW: I type as I'm thinking, so please don't take any offense at my writing style

ErNa

The problem with BLDC is, that You must know the commutation points (every 60°) to a precision of about +- 3°. The most simple way to get the information is: install 3 Hall-Element. Then You know at every moment, which phases have to be switched ON. Any sensorless mode is very hard to implement. It can be done, the first and main problem is to start the motor. All published methods are very complicated. Having reached 10-20% of top speed, you get a strong BEMF-Signal. The known method work then quit well, if the torque doesn't change abruptly. This should be true for a "flying circus". If there are no special reasons not to use Hall sensors (source: CD-Player), this is the preferred method and the programming can be done in SPIN.

Paul Baker

I'm with Simon, you need to get the datasheet from the manufacturer in order to figure out how to interface with it.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Paul Baker
Propeller Applications Engineer

Parallax, Inc.