No this isn’t a CD-ROM drive that has been modified to scratch my head but rather a mechanical assembly in a CD-ROM drive that has me scratching my head wondering how it functions. A few weeks ago I ripped apart an old CD-ROM drive to get a DC motor for the VHS Tape Storage Drive, the circuit board was also cut up while testing the CNC machine that was just assembled (yes it was a very productive afterlife for the old drive).
Here is the question. The CD-ROM drive had 3 DC motors, one to open the drive door, one to spin the CD and one to move the read head back and forth. We normally see a small stepper or something similar that can provide accurate position moves of the read assembly. In this case the read assembly moves via a rack and pinion drive system powered directly from a DC motor. When there is a CD in the system I can imagine that the drive would be able to get accurate head position information by reading the CD tracks but how could it move with any accuracy? It has me puzzled, if anyone can fill in the blanks I would be real interested to find out what makes this tick.
Permalink
It doesn’t have to be very accurate, the lens assembly has the ability to adjust for any slight offsets. Plus because of the gear system, a small adjustment of the motor, becomes a very small adjustment of the read/write head.
Permalink
hes right about the lens assembly… ever notice the magnets and wire surrounding it? it almost floats there(not actually but is attached by small wires) thus making the lens able to compensate for possible “errors”… At least this is what i noticed in one i took apart.
Permalink
I’m just guessing, but I think it has to do with the fact that data on CDs and DVDs is stored in a spiral direction. Because of this, one the read head finds the right position, it doesn’t have to make large movements afterwards.
Permalink
You dont really need a stepper motor to get precision in the moviment, if there’s a way to know the velocity it can be controlled with very simple control laws.
And there are lots of ways for one to read the velocity from a motor. Like encoders and hall effect sensors.
Permalink
THe way I see it, there should be a hall-efect sensor somewhere and a magnet in one of the gears. I had an old cd-rom like that. Or any other feedback mechanism.
Permalink
Hmmm… I think its magic. Seriously. 😉
Permalink
I would suggest that it just uses simple PWM signal driving it which can act nearly as good as a proper stepper motor, but it does appear that it’s just straight dc running to the motor, no logic before hand? If so, I’d have to support the prior comments regarding the lens assembly, it must be compensating enough.
Permalink
Does the motor spin continuously when a current is applied? What I’m getting at is, could a stepper move forward or back one increment each time it receives a pulse?
Permalink
yep – the lens’s voice coils perform fine tracking in a closed loop feedback manner, you can sometimes hear the lens squeaking as the CD spins round.
Permalink
I believe the motor is used for coarse positioning (so it doesn’t have to be so accurate) and the lens assembly floats in a magnetic field controlled by the CD drive electronics. The variable magnetic field is how the beam is kept in focus even though the distance to the surface of the CD may be changing as it spins.
Part of every 2352-byte sector on a CD-ROM is a sector address, so the drive can read where it’s up to, and home in on where it wants to go.
http://en.wikipedia.org/wiki/Cd-rom#CD_sector_contents
Not sure about how drives skip tracks on music CDs.
Permalink
I’ve seen ones that did it the optical way, like in old computer mice…
Permalink
The head assembly has a tracking system that keeps the lens locked on, not unlike a line follower robot. Servo tries to move the caret so that the lens in it is centered over the current position, the lens in the caret is tracking the grooves. When it moves far enough, servo motor compensates a little. E.g. when the lens drifts 10% off center, the servo starts moving the caret to compensate until the lens is again within 1% range off center. Maybe a current sensing is used to “feel” that the caret has reached either end of its working area, so even the end switches are not really necessary.
Permalink
OK so if fine movements are being done with the head I wonder if the DC motor uses the feedback from the head while moving as a closed loop system and then using short circuit braking to stop immediately when the head is in the right position. I had a look at the board but of course the area that the traces go to were removed when the sign was cut… I guess there are so many things you can do when saving a buck or two will make a difference to the profit on a system.
You would think that the rack and pinion would be made more precise to help getting closer. It takes less than 4 turns of the DC motor to move the head from one side to the other!
Permalink
This is a good example of when good control systems (ie embedded micro controllers) are used to overcome bad (cost efficient) engineering.
Permalink
I wouldn’t call this poor engineering. The nature of CD is such that a closed-loop system is absolutely necessary anyway. Cost efficient it is, though.
Just btw, 3.5″ drives have nice little steppers and worm screws.
Permalink
@svofski: All of the 3.5″ blu-ray and HD-DVD drives I’ve taken apart have had a stepper motor with a worm screw. Here’s a picture of one of them (it’s the one with the worm screw) http://voidingwarranties.com/gallery/the-great-internet-migratory-box-of-electronics-junk/files/page7-1005-full.html
Permalink
@Jordan: curious! I was actually referring to 3.5″ floppy drives and yours looks exactly like some of them that i’ve got here. Could they be repurposing the oversupply of the old steppers? 🙂
Permalink
If it were good engineering, you would not need a control loop (-> precision mechanics). The cleverness is cutting down the cost while making it still work inspite all the compromises made..
Permalink
“just fix it in software” 🙂
Permalink
Why suddenly everybody starts dissing closed loop control and call it bad names?
Permalink
2 good things going for it, firstly; wear and tear becomes less of a problem, as the closed loop system will duly deal with it and secondly; I bet the heads can fly from one side of the disc to the other pretty quickly, leading to faster seek times
Permalink
Blue-ray and HD-DVD, both have more density, thus requiring more precision. I would suspect that early CD-drives used stepper motors as well.
Permalink
Engineering is “approximate science for profit” (probably the best definition I’ve heard), this satisfies that entirely, and is, by that reckoning, excellent engineering!
Permalink
Permalink
maybe it uses a similar technique as Woz used in the Disk II floppy drive for Apple II which from what i read was really ingenious and simple. To my understanding he drove the read head toward the center a certain number of times to insure it was at the beginning no matter where it started from(causing the infamous noise of the head hitting the stopper that made people thing it was broken)! 🙂