RC Groups

RC Groups
    Scratchbuilt Indoor and Micro Models
        Rave brushless power to the micros

#1 Rajesh L G Aug 08, 2004 02:54 AM

brushless power to the micros
It started with Prashanth giving me a small brushless motor pulled out of a laptop HDD. There was a controller IC on the PCB attached to the stator. Got the datasheet of the BA6840 controller and found that it had torque control input, that gave me an idea that the speed might be controllable. I set up the reference voltage and a pot to control the torque and there it was working as I hoped. Ok, Prashanth started it, so I borrowed his JMP receiver too and set about interfacing the controller to JMP output. And here are the results.

Power source: 8 nicd cells
current: 750mA
Thrust: 38gm
Prop: GWS 5x3

Pictures and video here. Please excuse the poor quality of the video, I have only a cheap webcam.

The interface circuit is very simple. I didn't want to make any modifications to JMP rx (I am sure Prashanth would kill me if I did, so that option was ruled out :eek: ). So I derived the control voltage by low pass filtering the ESC output. The reference voltage was provided by a led in series with a current limiting resistor. I used a pot to set the control range, this can be replaced by a fixed resistor. Thus the whole interface circuit is 1 led, 1 capacitor and 3 resistors.

The whole motor + controller weigh 16gm. The back plate is huge metal piece as you can see in the pictures, so I guess the weight can be brought down to about 10-12gms easily by replacing it. The can alone weighs 6gms.

And there is scope for further improvements. The controller is working at its power dissipation limit at this current. The saturation voltage of the builtin bipolar output transistors is pretty high. Dual mosfets could be added to outputs to improve this and increase the current handling. This might add a gram of weight, but my guesstimate is thrust can be improved to atleast 60gms. More if the motor is rewound with less turns.

-- Rajesh

#2 Prashanth Aug 08, 2004 09:29 PM

hey, now its working, quickly get it up on a plane....

#3 Rajesh L G Aug 10, 2004 01:46 AM

Over 250 views and no comments!
May be I was not clear enough...Guys, this is about a cheap brushless controller for micro models. Especially for those guys who have been hacking cdroms and hdd's for motors and throwing away the controllers. Some of those controllers are good enough to be used in low power applications. This is a demo of one such controller used with a JMP rx.

-- Rajesh

#4 Zlatko Aug 10, 2004 08:19 AM

I heard you :)
I've got a old IBM 6 gig notebook hard drive that crashed. I've already pulled it apart, need to trim the motor from all the other aluminium and need to figure out the controller.
Would you like to put up a sketch of the way you connected the controller to the JMP?
BTW. It should be easy to make a controller with 1 PIC and 3 FETs .....
I'll post my details here once done.

#5 mcross Aug 10, 2004 09:51 AM

Hi Rajesh,
I have been working with Philips motor control chips. So far I'm not to the point of interfacing with a rx, but will be soon. Can you tell me how you did the conversion from PWM(from the rx) to analog voltage(for the controller chip)? values of components and circuit layout would help, if you could.
Mine are a lot smaller, motor is 7mm diameter and weights 3 grams. Controller weights about 1 gram and is good for about 1 amp without heatsink. No thrust figures yet, I'm still working on finding the best winding.

#6 epilot Aug 10, 2004 11:09 AM


Originally Posted by Zlatko
BTW. It should be easy to make a controller with 1 PIC and 3 FETs .....

Don't think you could be more wrong! Writing good code for brushless controllers is not easy at all. Besides there is more to it than just switching the winding on and off. You need to do this at the right time, hence the need for some sort of positional feedback system. I believe there is a good reason why the market is not swamped with cheap brushless controllers.


#7 Zlatko Aug 10, 2004 06:50 PM

I know. I am just throwing the gauntlet to JMP and Coural. They have made a lot of impossible products, they do a teriffic job for us. When I say easy I mean thats on my wishlist.
You can add that to the gyro based on a magnetic field sensor ( 2 cm away from the motor ) ;)
And a 3 phase motor thats lighter than a lightened 4mm Diedel ;) ..
Whoops, now you know my secret agenda :)
But back on track, If you know the maximum RPM the motor can handle and the spool up time then 1 pic a 3 fets will be enough but the code will have to change with different props ... . Basically the 3 phase drive for the micro lights will have to be written for each configuration and as such will be of value only to people that can play and program PICs. Which realy means that most of us will have to stick to brushed motors which are also lighter and cheaper. But I can dream ;)

#8 Rajesh L G Aug 11, 2004 04:42 AM

1 Attachment(s)
I derived the control voltage using a simple low pass filter. I have attached the schematic and the values I used. The JMP connection is to the mosfet (where negative lead of the motor goes to). The rest of the circuit is pretty much the same as in the datasheet.

You have got a very light motor there. Your controller has higher current capability than this chip, you should be able to get more power out of it. Is it a sensored or sensorless controller ? I am yet to try a sensorless controller, don't know if they will be able to keep the synchronization across the rpm range we need. I am still on the lookout for a sensorless controller with good enough output current capability. If you are working on such a chip, I would be interested to know the results.

As epilot pointed out, it is not very easy to implement this in a PIC. All the PICs I have access to don't have enough MIPS or the right peripherals for this application. There are other more capable chips that could be used. But for a low power application, an integrated, mass produced chip like this will win on cost and weight.

-- Rajesh

#9 Zlatko Aug 11, 2004 09:57 PM

Thanks Rajesh,
Fair enough about the PIC based brushless ESC. I guess I can dream ;)
The controller chip from my HD is a Texas Instruments TLS2251 , I googled and searched TIs site but no joy. Does anyone know more about this chip?

#10 luckyluc Mar 04, 2005 12:18 PM

[B]This controller is great but I search equivalent at low price

YGE4-BL brushless controller

Max continuous current: 4A
For use with 1 to 3 Li cells. Number of cells automatically detected.
Reduced power at low voltage. Low voltage detection can be disabled.
BEC: 5V at 1A
Timing adjustable in 5 steps (0 degrees to 30 degrees)
Switching frequency: 14kHz
Mass: 1.3g without wires

Price: $79.00

#11 lanre Mar 22, 2005 10:02 AM

Rajesh could you plz put up a schematic diagram of the controler you made and how to program it cos i have that type of ic?

#12 66tbird Mar 22, 2005 12:23 PM

If I could understand more of what your talking about I make one myself :D . I watched the video and it deos work.
Has the setup gone into a plane yet??

#13 Rajesh L G Mar 23, 2005 06:40 AM

The schematic is same as the one on the datasheet. There is no programming needed for this controller.

If you can follow the schematic, you can build it. It is not necessary to understand all the technical details. If you have any questions I would be happy to help. The chip costs about $2, hall sensors you can pick up from old crdom/hd drives, so not much to lose :D, go ahead!

It hasn't gone into a plane because I don't have all the stuff needed to make a micro model. Also lack of a good indoor place to fly didn't help motivate me to go down that path. Outdoors is not an option with the winds that we are having this year. Even my speed 400 planes are getting pushed around a lot.


#14 scrutor Mar 23, 2005 10:07 AM


I have only one question. Would this setup work with other receivers that has speed controlled output? not just the JMP rx.

nice idea :) poor little CD drive, I'm gonna kill one too :)))


#15 Rajesh L G Mar 24, 2005 12:20 AM


Yes, it should work on most speed controller outputs. If the PWM freq is very different, it might need some tweaking for R1 C1 values to get the full throttle range. But most of them have very similar PWM frequencies.


All times are GMT -5. The time now is 10:16 PM.