Adding sensors to a Astro 3210 motor

[JEB]

What is the best way to go as to adding sensors to a Astro motor- hall devices, or optical sensors?

The first problem that needs to be addressed is the on/off timeing lenght, the purchaced brushless motor that have hall devices imbeded in them. Is there any design in the placement? or is it just pot luck? The sensors that are in the motors are latching type to keep the noise problem from a false trigger as a on/off could induce?. They turn on/off as to a fixed crossover point as to the rotation of the magnets from north to south, with no adjustment. Maybe the margin of error is in the "noise" (not importment) as far as the timeing is concerned.

If the way this happens to turn out is the optimum placement, and best triggering on/off point, I could attach a 200count endcoder (in quadradture, sp?) to the shaft of the motor which would give about a .008 thousandths error on a 2" circle (a 1000 X 4 would be better), if the turn on/off point is not in the center of the magnets location.

Constructing a timing disk, (external) using say a .250dia. ~ magnet on a .500 center line, on a say a 2"~ or so dia., (depending on the magnet count) would give say a 50/50 turn on/off point, keeping the same number of magnets but increasing/decreasing the dia. of the magnets placement would change the pulse lenght, could this be determined by anything other than trial and error as viewed of a scope? Increased no load current by having on pulse to long? too short would not give the maxium HP output?. Would it be better than adding a one shot device and a timing capacitor? more induced signal noise?

The optical disk interrupter system still requires a selection of the correct width of the open/closed vane. Using lower voltages, and amplifiers, could be more sensitive to extraneous noise, and dust/dirt.

Lets hear your input. Any experts out there on parameters of brushless motors?

Jim

[liteCycles]

Since this thread was initially started over in the "Astro 3210 group buy?" thread - viewtopic.php?f=28&t=14151 - I have a few questions regarding what was posted over there.

liveforphysics said "Combine with the IRF 4110 fitted 6-fet programable infinion controllers the size of a creditcard, and you've got yourself a 100v sensored drive solution for a fraction of the cost of a castle ESC."

So would we be looking at double the power out of this motor? Has this ever been done and what are the odds it could run at a continuous load of around 6000 watts?

I'm pretty green on this, but does that controller have a electrical rpm limit that could come into play? - viewtopic.php?f=28&t=12813&start=0&hilit=controller+timing+speed

[Miles]

So would we be looking at double the power out of this motor? Has this ever been done and what are the odds it could run at a continuous load of around 6000 watts?

I'm pretty green on this, but does that controller have a electrical rpm limit that could come into play? - http://endless-sphere.com/forums/viewto ... ming+speed

There's a maximum speed that it's safe to run the motor at, which is about 12,000 rpm. The maximum continuous power will go up in direct proportion to the speed so, theoretically, you could get about a 50% increase on the cont. power figure in the specification.

[johnrobholmes]

We need to figure out the sensored commutation speed limit on the infineon build.

[Burtie]

While waiting for Astro Bob to weave his magic on our group buy, I felt the need to make some stuff

Just before I got consumed by the rotating chuck (-----only joking!)

Mill.jpg (120.9 KiB) Viewed 3000 times

Bits of perspex

Disc.jpg (59.11 KiB) Viewed 2943 times

Bits of ally

Rotor.jpg (63.49 KiB) Viewed 2557 times

Add some small magnets and hall sensors

Components.jpg (66.24 KiB) Viewed 2932 times

Parts assembled.jpg (66.19 KiB) Viewed 2563 times

With these 8 pole motors I think a sensor spacing of 30 mechanical degrees will work the same as 120 mechanical degrees, but will make the wiring neater.

about 1 cm thick

Thickness.jpg (49.06 KiB) Viewed 2940 times

The theory:
Glue 8 small magnets to the rotor with alternating polarity
Screw or clamp the perspex plate to the back of the motor so that it can be rotated slightly to adjust the timing
Fix the rotor to the motor shaft with grub screws and dimples (to make re-location easy)

This should be useable with the 3210 or 3220 Astro motors. If the motor does not have a double shaft , I guess you could drill and thread the back of the shaft to add an extension for the rotor.

Now all I need is a motor to try it out

Burtie.

[etard]

UUUHHHH, Burtie, don't you have a Turnigy motor or the sort?

Seriously, nice work man! I can't wait to see this applied. Maybe you could connect a blue LED that turns on each revolution too!!
Yeah, hopefully after Astro Bob sees all these motors ordered from us, he will find the time to solve this problemo. In the meantime we have wise wizards like you spinning their magic.

Curious, what did you use for the magnet rotor?

[recumpence]

Burtie,

If this works out, give me the specs and I can replicate these parts in quantity on the CNC.

Bob is just so swamped with other stuff, the hall sensor issue is on the FAAAAAR back burner. Also, he is not swamped because of the group buy, he is swamped with all the other products they make along with military contracts.

Matt

[gwhy!]

Burtie,
That look the dogs dangley bits mate. There is no reason that this method could not be used for out runners also but mounted to the front of the motor but will also need a different number of mags in the rotor. So if anyone is gonna start making these in larger quantity's its worth thinking about making rotors for the cheaper outrunners also.

edit:
just a though wouldnt it have been better to go with non-round mags, I think you will get a more precise switch or is it a cunning plan so you can move the halls inwards and outwards which will give even more adjustment for timming and switching..

[liveforphysics]

While waiting for Astro Bob to weave his magic on our group buy, I felt the need to make some stuff



Burtie.

Proof that people who make weird faces in pictures make awesome stuff!

Very Very cool Burtie! This is going to open pandoras box for burning up some Astro motors. Yeah!!!!

[kfong]

Nice job Burtie, so simple of a solution. Great idea thinking out of the box.

[methods]

Burtie = Awesome for this picture alone.
Keep up the good work.

If you select the Astro model with a shaft poking out of the back you could also pick up the hall sensors in a similar way to the Cyclone motors. I posted pictures of the guts a few months back. Magring on the shaft w/pcb holding the sensors up around it.

-methods

[GGoodrum]

Very clever. There's about 1/2" of space at the end of the inside of the rear of the motor. I think your nifty little widget will fit inside there.

AF3210-12T-02.jpg (151.48 KiB) Viewed 2372 times

I have a couple extra 3210s. If you PM me your address, I send you one you can use to test this new gizmo.

-- Gary

[Burtie]

Gary,
You are right

1/2 inch is quite a lot of space to play with. We may even be able to see the flux through the end of the rotor, thus obviating the need for the sensor rotor.
I will PM you shortly.


Methods,
Nice idea. There are some off-the-shelf 8 pole ring magnets around, I will look into it.


Matt,
Cool, it would be great to get this into some sort of production when we have it sorted.

Edit update: check out this solution for using that 1/2 inch free space in the Astro:
viewtopic.php?f=2&t=28966&start=30#p434772

Burtie

[recumpence]

Burtie,

I am open to anything. We have shop time available to work with. Let me know if you want to go ahead with this and we can get on it.

I actually had the exact same idea (rotate-able magnet ring and sensor ring). I have no idea how the electrical portion of a sensored system works, though. So, I never persued it. I am also swamped with builds right now too.

We have CNC time available, however.........

Matt

[Burtie]

There is no reason that this method could not be used for out runners also but mounted to the front of the motor but will also need a different number of mags in the rotor.

I think positioning the sensors in a small arc around the edge of the can (as you have already done) would be an easier solution for the HXT style outrunners.

Given that these outrunners are usually 14 pole motors? the sensor spacing could be as little as 120/7 = 17.1 degrees.

I might start a new thread to discuss adding sensors to these motors

Burtie

[methods]

Unfortunately the change in forum settings means that you now have to click each picture - but here they are for reference:

Cyclone hall setup

-methods

[gwhy!]

There is no reason that this method could not be used for out runners also but mounted to the front of the motor but will also need a different number of mags in the rotor.

I think positioning the sensors in a small arc around the edge of the can (as you have already done) would be an easier solution for the HXT style outrunners.

Given that these outrunners are usually 14 pole motors? the sensor spacing could be as little as 120/7 = 17.14 degrees.

I think i will start a new thread to discuss adding sensors to these motors

Burtie

Yes it would be tight i didn't think of that, The arc method was pretty easy and do work well.

[Miles]

There's about 1/2" of space at the end of the inside of the rear of the motor. I think your nifty little widget will fit inside there.

Just measured mine:

The core starts 15mm in from the inside face of the end-cap.

The internal dia. of the core is 45mm.

The outside dia. of the bearing housing is 25.4mm and it protrudes 7mm from the inside face of the end-cap.

The rotor endcap is 3.5mm thick with an integral spacer 15.7mm dia. protruding 5mm from it (to the bearing).

Anyone want to see photos of the rotor?

[Burtie]

Yes please Miles!
Pictures are good

[Miles]

Ok. Here's the rotor:

[Miles]

Here's the stator:

[bigmoose]

Miles are the magnet poles made up of 10 thin magnets? What does the goo look/feel like between poles? Are the magnets captured under the aluminum end caps? If you have a macro lens, a couple more close ups of the rotor structure appreciated!

[Miles]

bigmoose,

That's about as close as I can get with my big camera. If I can get my old Coolpix 990 working, I'll take some close-ups

It looks to me like thin magnets on edge, bonded into slots in steel. I'll see if I can get a better shot and post the full-sized image.

[Miles]

Here's the full-sized version of the above rotor pic.

[bigmoose]

Thanks Miles! From the large photo, it almost looks like a pole composed of magnets; followed by an iron pole; then another magnet composed pole. Does it look like that in person?