CD Players, Vibration and Jitter
by Jon M. Risch     [email protected]  11-25-2000

I have found a tweak that can provide a great benefit on some CD players.  However, it involves taking the lid off, and placing objects inside, so if you are concerened about a warranty, or are not mechanically inclined, etc. don't attempt it.  If you damage your CD player, I will not be held responsible.

Jitter has become a hot topic for digital audio, and jitter is real.  It can creep in via the digital cable connecting a transport and DAC, it can creep in from the CD itself, and it can occur due to the master clock oscillator crystal inside the CD player or DAC.  Yes, the very device responsible for eliminating and controlling jitter, the one that is supposed to make the possibility of jitter intruding into the system impossible, is also one of the weak links in the jitter defense!  Crystals are piezoelectric, they generate an output when vibrated, just as they vibrate when exposed to a signal.  If the sound from your speakers vibrates the crystal, this will generate spurious signals that will introduce signal related jitter.

How can this be, you might ask?  This does not jibe with the common assumption that any corruption of the digital signal, must actually alter or lose a one or a zero.  Computer data does not depend on the timing accuracy beyond preservation of actual bits and bit values.  During most of it's journey, a digital audio signal theoretically is unconcerned with timing issues and subtle changes to the signal stream too.  It is at the DAC, or Digital to Analog Converter where timing accuracy is paramount to accurately reconstructing the analog signal.  If the timing is off, then the waveform has been corrupted.
Jitter is any variation in absolutely consistent and perfect timing, when the DAC converts the digital code into an amplitude value.

What causes jitter in a digital audio system?  Simple: anything that interferes with the digital logic circuitry detecting the logic transistions at the proper time will introduce jitter.  Noise, RFI, varying power supplies and ground references causing the detection threshold to shift ever so slightly, injected signals, etc.  It is ironic that the very digital logic chips that are necessary for the system to function also generate some of the interference that prevents it from reaching it's full potential.  Everytime a logic chip changes state, it draws current to send the digital signal downstream, a pair of transistor's is turned on, and the current drawn is sudden and not trivial.  How many logic gates are their in a modern CD player or digital audio system?  Thousands, maybe tens of thousands?  These are all being turned on and off in a complex sequence, some patterns which are unrelated to the signal, and some which are signal related.  Often, dozens of gates fire within moments of one another, compounding the power supply rail and ground bounce issues.  All of this activity shifts and modulates the timing of the detection threshold of a transistion from a one to a zero or a zero to a one by the variations caused in the power supply rails or the ground reference.  The popular audiophile press has referred to this as LIM, or Logic Induced Modulation.  Even CD's with excessive jitter recorded onto them from the glass master can induce this form of jitter into the system, as the tracking servo works harder and the optical read system has to pass along the jittery data it is reading, the transistions of this read data end up getting into the PS rails, etc. and making it's way to the DAC.

The one device that is supposed to make all this moot, and provide freedom from system induced jitter is the master timing crystal, usually placed very near the DAC, which is responsible for controlling the timing of the DAC analog data clock out.  The crystal is connected to a digital logic gate, so right there is a potential problem.  Worse yet, the crystal itself can generate spurious signals due to vibrations from the speakers or other sources of sound or vibrations.  Many CD players and DAC's use very cheap crystals, which are prone to external vibrations.

Placing sand bags on top of a CD player can help this, as can cones, platforms, air suspension systems, etc. but the most effective way to prevent crystals from contaminating the timing is to control the vibration problem at the source: place a sandbag on the crystal itself.

This needs to be done ever so carefully.  All power should be removed from the CD player or DAC, you should discharge any static electricity by grounding yourself before opening or touching anything inside the player, make sure the drawer mechanism will clear any thing you add to the player ( a huge bag of sand is not necessary for this to be effective), and that the cloth or bag of sand will not block ventalation or rest on a hot power supply regulator IC.   A cotton cloth, such as a handkerchief (or a piece of one) should be used as insulation from the plastic of the sand bag, then the limp "baggie" style (not zip lock with the thick walls, buit the ones with the thin walls) plastic bag loosely filled with sand is placed around and over the crystal, tucking the loose sand in around the crystal without getting too forcefull.

What does a crystal look like?  A small metal can, usually rectangular in nature, about 1/2" to 1 1/4" wide, from 1/8" to 3/8" thick by 1/2" to 1 1/4" tall, with just two leads.  It is usually near the DAC chip, which is one of the biggest IC chips in side a CD player.  Surface mount crystals can be solder so they are laid over on theirt sides, and so would not stick up from the PCB very much.  Hoever, they are stillusualy all metal cans.

Be sure that you do not force any of the parts mechanically, or bend them, If you want to make the tweak permanent, and not subject o falling off the crystal, some dabs of silicone or glue in areas where there are no components, can tack the bag down around the edges, plastic wire ties (not metal twistie ties) can be used to seal the bag, and to hold it in place using the chassis or local mechanical structures, again being sure that the ties are not joined to the drawer mechanism, etc.

Once you have placed the sand bag on the crystal, additional zip lock type bags of sand on top of the re-assembled CD  player or DAC will still help some, as other components other than the crystal can be sensitive to vibration, and additional mass and damping will weaken the vibrations that do get through t to the crystal.  Once you have sand bagged a CD player or DAC crystal, cones or other vibration control devices may react differently than they did before, so experimentation is the order of the day.

It is not as likely that high-end expensive CD players will respond as much as less expensive models, because the top of the line players are likely to have better quality crystals in them than lower end players, but I am sure there are some surprises out there in this regard. My CD player, a JVC top of the line single play unit responded very nicely to this tweak.

Jon Risch

All applicable copyright laws apply, all rights reserved, except transmission by USENET and like facilities granted.  Any use or inclusion in print or other media are specifically prohibited.  The informational content is not warrantied in any way or form, and any use of said content are at the reader's own risk, the author shall not be held responsible in any way for any damages or injuries arising from the content of this web site.  Common safety practices are encouraged at all times.  Do not fold, spindle or mutilate.