The Real Zack Morris

Fix Your DVD Drive for Free

My last few posts have been kind of heavy and I decided it was time for something completely different.

So right now I’m in the do-it-yourself computer repair biz bootstrapping my way towards being a successful indie app developer.  I fix a lot of older G5 iMacs and PowerBooks and the occasional MacBook.  So far the most failure prone devices I’ve found are the slim and super slim SuperDrive.  They fail within just a couple of years usually, whether you clean them or don’t even use them at all.  I think I’ve figured out why.

My drive wouldn’t spin up when I inserted disks.  If your drive is just spitting out disks with an error halfway through burning, then it most likely needs a DVD lens cleaner, and if that doesn’t work, then you may have to buy a new drive.  My fix is mainly for drives that make funny sounds or fail to see disks at all.  It should also work for Combo drives.  I haven’t tried it on a full size desktop drive yet though.

Pictured above is the inside of a slim SuperDrive from an old PowerBook.

Before going any further, be warned that you should NEVER look inside ANY optical drive when it’s on, because they have powerful lasers that can blind you instantly.  If you try the mod I did, only do it while the drive is off and disconnected, then put the cover back on for each test.  Also be warned that if the drive is under any kind of manufacturer warranty, then opening the cover will void it.

I tried lubricating the central motor with some light oil and that did get it to start spinning.  But it would only spin up a little bit, make some whirring sounds and then eject the disk.

I finally realized that the disk was rubbing too hard on the foam brake in the red circle.  So the first test I tried was to remove the brake altogether.  That caused the disk to spin up to full speed and then be ejected at high RPM.  So the brake is required, it’s how the drive reduces its speed accurately (otherwise the disk would keep spinning even without power).

Then I tried cutting the foam down a bit at a time until about half of it was gone, those are the remnants in the blue square.  That worked better but the disk was still spinning too fast and would be rejected half the time.  So I moved the brake to the left a bit until it touched the disk with just the right amount of area.

It worked!

But some disks wouldn’t seat properly on the spindle after insertion, so would either clunk around or not catch on the spindle as it began turning.  It turns out that the whole center of the drive is a lever that bangs the disk against the top of the drive housing to seat the disk firmly onto the spindle.  Over time, either the gripper springs on the spindle got weak or the aluminum case fatigued just enough to not quite seat the disks.  Your drive may or may not have this problem.  But if it does, you can lightly bend the cover inwards with your hands to bring it closer to the slap mechanism.  If you bend it too far, it will prevent the disk from spinning at all.  But it wasn’t too hard to get mine adjusted properly, and now the drive works like new again!

So what the heck is going on here?

I think that these drives use tiny cheap-o motors and their magnets lose their strength over the years until they can no longer overcome the brake.  So the drive tries to spin up but can’t get the disk spinning fast enough for the laser to see it.  It finally gives up and just spits out the disk.  Then one more drive goes into the landfill and you buy another one.

One other thing to note is that name brand disks seem to burn better in the drive than off-brands.  I finally figured out why.  The off-brand disks are slightly thicker, perhaps because they use cheaper polycarbonate that isn’t as strong (you can feel the difference if you put the disks side by side on a table and slide your finger along the top where they meet).  This causes them to rub more on the brake, and also to not seat in quite the same place on the spindle as a thin disk, and the laser seems to be sensitive to the difference.  Also the thicker disks seem to vibrate more, perhaps they are heavier so slight imbalances are magnified.


But what’s REALLY going on here?

I don’t know anything about how these drives really work but I’ll take a guess.  They seem to be missing two key components: 1) some kind of directional or phase controller like an H bridge to allow the motor to slow itself down and 2) a rotation rate sensor like a hall effect sensor or optical sensor or some kind of circuit (warning: PDF) to count the voltage ripples per second as the motor spins.

Without a control mechanism to slow itself down, the motor has to rely on a brake to do it.  This was probably done to cut costs, because it’s much easier to make a motor that just spins faster as more power is applied, and then use something like pulse width modulation to control the speed.  A controller would use more power when slowing down (H bridges also get very hot), and a brake would use more power when speeding up.  So it’s probably a wash power-wise if the drive is mostly accelerating and decelerating.  Of course a controller would be more efficient at constant speed, so the lack of one could help explain why DVD players wear down your laptop’s batteries inordinately when watching movies at a constant streaming rate (RPM).  Lasers use only milliwatts, so the rest of the power is probably used by the CPU to decode the MPEG.

The lack of an RPM sensor is intriguing though.  This made little sense to me at first, but, with new SuperDrives costing just $60 on eBay, every bit of clever engineering helps cut costs.  These drives use Reed–Solomon error correction to read disks even if they are scratched.  The head reads the most data at a high RPM (but not too high or signal noise defeats any gains), so the drive constantly tests the water by speeding up and slowing down in whichever direction minimizes the error coefficient from Reed-Solomon.  In this way, it automatically finds the optimum RPM.  It’s pretty clever, because then there is no need for a complicated algorithm to determine the proper RPM as the read head moves in and out radially like a record player arm.  If I’m completely off base here and drives don’t use a method like this yet, then they probably will someday as competition removes all unnecessary cost.

I find myself simultaneously in awe of just how amazing/inexpensive technology has become, and also how brittle/throwaway it is.  Consumer opinion and brand reputation seem to be fading away as price trumps everything else.  The exact formula is to minimize the number of returns within the warranty period, while maximizing profit.  If the part fails the day after the warranty expires and the customer buys another one, so be it.  So few companies can even make this stuff profitably, that the oligopolies ensure lack of competition.  A side effect of this is that to compete with the big companies at all, you have to do something completely different.  The logical next step is probably solid state storage, and Apple realized this years ago when it dropped the optical drive from the MacBook Air.  So while we may complain about computers, we know that some geeks in a garage somewhere will invent something revolutionary next year.  If technology didn’t stink, it wouldn’t get better.

If you successfully fix your optical drive, please let us know in the comments, and if not, well, let us know anyway!

Comments

blog comments powered by Disqus