Thread Tools
Jan 04, 2004, 03:54 PM
Thread OP

CD ROM rebuild--Success!

I was inspired by the 1 Hour CD ROM conversion thread to start learning about CD ROM brushless motors and how others had rebuilt them for use in R/C aircraft. After reading the above thread and disassembling two CD ROMs, salvaged from old computers, I rebuilt my first motor. My intent here is to share with you what I learned and how I built a brushless motor for less than $10 which out performs the brushed motors I have been using and is lighter, quieter, and consumes less current (Amps).

Finding the right CD ROM to disassemble is the first challenge. You can do what I did and just start disassembling what ever you can get. Old computers, yours, your neighbors and friends, are a place to start and the price is right. Some have reported success contacting local computer stores or the IT folks at work. If they have old units removed for upgrades or service they will usually give you a few after they give you a funny look as you describe why you want them. The oldest CD ROMs (4X speed or less) are often little help as the drive motors are small and I even found one with a brushed motor. The best units are reported to be from the newer 48X and 52X high speed drives. These drives have stronger magnets, which are grey rather than dark brown, and may even have ball bearings. I have disassembled 5 drives and, like others, I have never found any two with the same motor. Similarly, I have no list of manufactures or model numbers which are best candidates for conversion to R/C use. Just keep looking. Taking old units apart will help to develop techniques for disassembling the metal parts that may be pressed, and/or swaged together and the even occasional plastic part ultrasonically welded together.

This photo shows some typical parts. At the lower left is a rotor with magnet ring. Above is the stator with the original windings and circuit board.
Last edited by bz1mcr; Jan 04, 2004 at 04:38 PM.
Sign up now
to remove ads between posts
Jan 04, 2004, 04:45 PM
Thread OP

CD ROM conversion Post 2

This next picture shows the backside of the circuit board. The bearing tube is visible in the center
Jan 04, 2004, 05:02 PM
Thread OP

CD ROM conversion post 3

During disassembly, the objective is to get a rotor (also referred to as a can or bell) with its magnet ring and a stator out of the drive without damage. Donít worry about the circuit board or the wire wound on the stator. These parts will not be used in the rebuilt motor. If it has ball bearings save them and try not to damage the center bearing tube. If it has bushing type bearings they may be reused to avoid the cost of ball bearings or if the stator center hole is too small to fit ball bearings inside.

If you do not have access to salvage CD ROMs you can consider buying a new one from suppliers like A advanced search on CD/DVD ROM Drives with a price under $25 recently provided a list of 22 drives. These are all new drives, and some include free shipping! You may also find new drives advertised locally with rebates which make the final price less than $20.

It takes more than finding a suitable drive to end up with a useful brushless power system. The most expensive item needed is a new ESC (about $55). No, there is no way to use a brushed motor ESC or make your own brushless ESC unless you are very good at making circuit boards and have a lot of time. The stock motors, I have seen, use a brass bearing carrier and two bronze bushings (bearings). Some stock motors have reportedly been equipped with ball bearings. The bushings will work but ball bearings have lower friction, require less lubrication and should last forever. You can salvage ball bearings from an old gear drive such as the IPS units or buy those made for RC cars and trucks by Dura Trax and others from major hobby suppliers. For example has Dura Trax 2X6mm and 3X6mm bearings in sets of two for $3.99. You will also need a 2 or 3mm shaft. You can purchase drill rod, carbon fiber rod or salvage a shaft from a brushed motor. Suitable brushed motors are available from or for less than $1. You may also need a bearing tube.

A new bearing tube can be made from Al or brass tubing or custom machined if you have access to a drill press or metal lathe. Tubing should fit over the outside of the bearings and if the fit is sloppy the tube diameter can be resized locally by rolling on a hard flat surface while pressing down with a piece of wood. The bearings can be retained with CA glue or the tubing can be crimped by applying pressure just past the edge of the bearing with a dull knife blade or the corner of a piece of wood as the tube is rolled on a hard surface. If you have access to a lathe, a new bearing tube is quickly made to fit the bearings and the stator. Start with a piece of Ĺ or 3/8 inch round Al stock from the hardware store. Drill a hole about 1 inch deep in the center of the stock. This hole diameter must be large enough to provide clearance for the motor shaft and small enough to allow a shoulder to retain the bearing after the bearing bore is drilled in each end. Finish the outer diameter (on at least one end) to fit inside the stator.

Below is a picture of a new bearing tube and a bearing shown on an enlarged drawing.
Jan 04, 2004, 05:05 PM
Thread OP

CD ROM conversion Post 4

The bearings and bearing tube are shown in relation to the other parts in the figure below. In the cross section, the motor is shown with the shaft vertical in the center. The shaft is supported at the bottom and a little more than half way up by ball bearings. The ball bearings are held in position by the bearing tube and the top end of the tube is fastened to the inside diameter of the stator.
Last edited by bz1mcr; Jan 04, 2004 at 05:07 PM.
Jan 04, 2004, 05:10 PM
Thread OP

CD ROM conversion Post 5

Below, the stator is pictured on the right, the part with the nine poles and windings on it. The bell or magnet housing is shown on the left. The twelve pole magnet ring is just inside the outer edge of the bell. So far I am using the stock grey ring magnet, but others have reported improved performance using high performance magnets. The shaft is press fit into a brass hub which connects it to the bell. The bell and shaft rotate together. The stator and bearing tube assembly are stationary and get fastened to the airframe.

The stator assembly and the bell-shaft assembly just pull apart. The shaft is a slip fit in the bearings and the magnetic force is the only thing holding the two together. The prop thrust will not pull the assembly apart at the highest loads tested to date.
Jan 04, 2004, 05:13 PM
Thread OP

CD ROM conversion Post 6

Here are pictures of the assembled motor, front (shaft side), and back (stator-bearing tube) side. The windings are 20 turns of #26 magnet wire. They are connected in the star or ďYĒ winding configuration.
Jan 04, 2004, 05:14 PM
Thread OP

CD ROM conversion Post 7

additional picture for post 6
Jan 04, 2004, 05:18 PM
Thread OP

CD ROM conversion Post 8

CD ROM vs. a brushed motor.
This photo shows the rebuilt brushless motor and the Cox brushed motor it replaced. The brushless motor has a mass of 18g . The brushed motor is 31g. Both were tested with the same 2 cell 1200 mAh Etec LiPo battery, and a GWS 5X3 propeller. The brushless turned 9400 RPM and consumed 1.8A of current. The brushed motor turned 9360 RPM and used 3.3A. The BL used only 54% of the power, was quieter, and more consistent in these static bench tests. Even with a controller which is 6g heaver than the brushed ESC the system mass of the BL is 24 vs. 31g. Thatís a mass savings of over 20% for the motor ESC system.

My Cox ElectriCub plane originally had a 5 cell 650 mAh NiMH battery. That provided near scale flying speed and 5-8 minutes of flight time. When I tried my first LiPo, a two cell Kokam 1020 mAh pack, I was amazed. The improved performance was much better. Simple aerobatics were now possible and flight times more than tripled to 18-25 minutes. The brushed motor was running almost too hot to touch. It was hot enough that the stock press on plastic propeller would come off if operated more than 30 sec. at WOT.

The brushless motor can operate at power levels which surely would over heat the brushed motor. I have bench tested the Brushless motor with a 12 V battery (simulating a 3 cell LiPo). It turned the 5-3 prop 11430 RPM on 2.5 A and 8360 RPM throttled back to 1 A. That is a 30W input (2.5X12) into an 18g motor. The 31g brushed motor was running hot with a 25 W (7.5 V X 3.3 A= 25 W) input. On larger props the brushless showed it would tolerate even higher power levels. 37 W on a GWS 6-3 prop (9540 RPM, 3.08 A), and 48 W on a GWS 8-4.3 prop (4430 RPM, 4.0 A). It puts out way more power than the brushed motor ever did and still runs cooler.

Wow! With the rebuilt CD ROM motor, Iím looking forward to flights of an hour on the 2S1P 1200mAh LiPo and I may even try a 3S1P!

WOW! I canít wait to get this in the air!
Jan 04, 2004, 05:20 PM
Thread OP

CD ROM conversion Post 9

Two final notes, first although the prop thrust will not pull the magnet assembly off the stator assembly it is sufficient to move it noticeably forward and at high loads the shaft oscillates in and out of the motor at least a 1/16 of an inch. Moving the magnet ring out of alignment with the stator has to be limiting output. I have ordered a wheel collar to install on the back end of the shaft to prevent this. And finally, the ESC for the brushed motor was glitching and shutting down when my multimeter was measuring current in the circuit. (Remember you are not supposed to have long wires in your ESC circuit). So the ESC was removed and the brushed motor data was run with no ESC. Since the ESC would have some loss even at WOT the brushed motor got an advantage because no ESC was used.

Edit 1/6/04 After installing a wheel collar to limit how far the prop thrust could move the magnet, the brushless RPM increased from 9400 to 9570 using the same 2S LiPo battery and 5X3 prop. Current increased from 1.80A to 1.88A.
Last edited by bz1mcr; Jan 06, 2004 at 05:58 PM.
Jan 04, 2004, 08:00 PM
Great information!
Jan 06, 2004, 03:33 PM
homo ludens modellisticus
Ron van Sommeren's Avatar
Replacing the original magnet ring with 12 N45 quality neodymium magnets will increase torque and lower rpm. Efficiency increases. It's also easier for controllers to synchronise, no more coughs, hickups and farts.

Magnet sources: (USA) (Switzerland)

Met vriendelijke groet Ron van Sommeren
ē diy outrunner brushless e-motor discussion.
ē Electric fly-in & lrk meet, June 27th, 2004, Nijmegen, the Netherlands.
Jan 06, 2004, 06:17 PM
Thread OP

Thanks for the information on magnets.

I have purchased 1X5X5 and 1.5X5X5 magnets from . I decided to try the stock magnets first partly because this was my first rewind and a chance to check out my new brushless controller. Second I am not sure there is room for the new magnets. It looks like they would not go without knocking off edges or reducing the rotor diameter. So I decided to wait.

I frist powered everything up on a 6v. supply and it was lossing control at about 60% throttle. It ran great up to about 1.1a. When I use the fresh 2S Lipo (about 8v) operation is rock solid all the way up. That is also true when using a 12v supply. I may loose full throttle when the 2S LiPo is below 7v or so. That has not been tested.
Jan 06, 2004, 10:04 PM

This is great info! Thanks for taking the time to make things so clear.

Quick Reply

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Mini-Review Hubsan H109S X4 PRO Unboxing, GPS Hold Test, FPV Test, RTH Test, Gimbal Test & Review maninblack Multirotor Drone Talk 2 Mar 12, 2017 10:11 AM
Gallery test test test vonpierre27 Test Forum 0 Jan 11, 2016 12:44 PM
Mini-Review 2W 2.4G wifi booster RANGE TESTS! (Sunhans VS Wii-link tested at ground level) nonoze FPV Equipment 22 Nov 25, 2015 04:39 PM