CD Rom Oscillator

Hey all, i need some help designing a circuit, in which i can vary the frequency of a crystal inside a cd rom, i have already identified that the specific one i am using has one rated at 12Mhz. my question is, can i vary it within the circuit it has already, or do i have to scrap the existing circuit and create my own? if so, how would i go about it? my aim is to control the frequency, therefore changing the speed at which a song is played back on the CD.

Thanks all

MadMatt

How much variation do you want?
Crystals have a very small range that they can be shifted in frequency.

Why? :grinning:

my question is, can i vary it within the circuit it has already, or do i have to scrap the existing circuit and create my own?

You'll have to create your own circuit. Assuming your new oscillator is built-around a crystal, you'll have to start with a higher frequency and divide-down to the frequency you need. The higher frequency you start with, the more precision/resolution you can get when you divide.

if so, how would i go about it? my aim is to control the frequency, therefore changing the speed at which a song is played back on the CD.

That would ONLY work with an audio CD player or if your CD-ROM drives has analog audio out. An older CD-ROM drive may have audio out, but newer ones don't.

When you play a CD on your modern computer, the speed and pitch are usually determined by the soundcard... If you slow-down disc read-speed too much, you'll just cause glitches & dropouts.

Windows Media Player can [u]change the speed[/u], but I'm not sure if it works with CDs... You might have to rip the CD to WAV or MP3 first.

With any audio editor you can alter speed with or without altering pitch. But, audio editors are designed for making "permanent" changes to audio files (WAV, MP3, etc.) rather than working in real-time during playback. ([u]Audacity[/u] is a FREE audio editor.)

DVDdoug:
([u]Audacity[/u] is a FREE audio editor.)

And a superb project overall.

Thanks for all the replies! Hmm, I found a couple on the internet if you search on Google "crystal frequency control circuit" but they all involve variable capcitors, not resistors. I want maybe +-30% in speed? It's what DJ decks usually have. The final question, why? Well, it's for one of my subjects, we are to design our own system, some people, are making security systems, quad copters, and I wanted to make a sub woofer amp with a speaker box (have done this already) and to control 2 CD Roms (which I did with atapiduino) but I now need to change the speed of them, just like a DJ does on decks? Is it possible to do so without a computer? (Not using Audacity, etc) thanks for that :slight_smile:

MadMatt

OP wants to change the audio playback speed of old computer CD-ROM drives.

I told him that it was only possible if he could adjust the clock of the D/A converter.
Remove the crystal, and replace it with a variable oscillator.

So make a variable oscillator that can be adjusted from e.g. 5-15Mhz, and inject it in the circuit.
Try to Google things like RF test oscillator.

For several weeks I keep telling him to drop it, and use e.g. a free DJ program like Mixxx.
Convert your CDs to Mp3, and be amazed.
Speed control, beat detection/syncing and much more.

Leo..

well, I think we agree with you...

you CAN buy digitally controlled crystal oscillators... in fact, I have one that I removed from some defunct gear that can be adjusted using 3 pins to 8 unique frequencies between 5 and 15 mhz but the chip controls which frequencies...

If it even works... the whole thing sounds very "Rube Goldberg-ish" to me.

Yes, but I need to continue this, I have already committed myself to it, if i can't change the speed of the song from a cd rom, I can't beat mix, therefore, it destroys all I have done so far, it was based around this very concept, to control the speed of the song playing, WITHOUT using software and computers. This is what I need, but I can't do this without some help. Also, did you design that circuit?

Sorry if I sound stubborn, but this is how I can amaze the teacher and get the best marks I can for VCE, I want to be an electrical engineer, and this would be great to do for a starting point.

MadMatt

Try a crystal with a different frequency.
If the speed changes, you know you have the right one.
There might be more than one crystal in a CD-ROM.
Leo..

Wawa:
Try a crystal with a different frequency.
If the speed changes, you know you have the right one.
There might be more than one crystal in a CD-ROM.
Leo..

Will do :slight_smile: thanks for that!

MadMatt

I want maybe +-30% in speed?

You can not pull a crystal by that much, it is much less than 1% before it stops.

but they all involve variable capcitors, not resistors.

Why would you think that a resistor could pull a crystal?

At 30% you don't want a crystal anyway you need a variable frequency oscillator.

Maybe the Si5351 would do the trick (see here). It's easier to calculate and generate stepped integer frequencies instead of smoother fractional ones; though as you have a small frequency range you may be able to pre calculate more frequency steps to smooth out speed changes.

A subset of integer frequencies in the 12MHz +- 30% generated by the Si5351

 8.0078, 8.0729, 8.1055, 8.2031, 8.3008, 8.3333, 8.3984, 8.4635, 
  8.4961, 8.5937, 8.6914, 8.7240, 8.7891, 8.8542, 8.9844, 9.1146, 
  9.1797, 9.2448, 9.3750, 9.5052, 9.5703, 9.6354, 9.7656, 9.8958, 
  9.9609, 10.0260, 10.1562, 10.2865, 10.3516, 10.4167, 10.5469, 10.6771, 
  10.7422, 10.8073, 10.9375, 11.0677, 11.1328, 11.1979, 11.3281, 11.4583, 
  11.5234, 11.5885, 11.7187, 11.9141, 11.9792, 12.1094, 12.2396, 12.3047, 
  12.5000, 12.6953, 12.7604, 12.8906, 13.0208, 13.0859, 13.2812, 13.4766, 
  13.5417, 13.6719, 13.8021, 13.8672, 14.0625, 14.2578, 14.3229, 14.4531, 
  14.5833, 14.6484, 14.8437, 15.0391, 15.1042, 15.2344, 15.3646, 15.4297, 
  15.6250, 15.8203, 15.8854, 16.0156, 16.1458, 16.2109, 16.4062, 16.6016,

Riva:
Maybe the Si5351 would do the trick (see here). It's easier to calculate and generate stepped integer frequencies instead of smoother fractional ones; though as you have a small frequency range you may be able to pre calculate more frequency steps to smooth out speed changes.

A subset of integer frequencies in the 12MHz +- 30% generated by the Si5351

 8.0078, 8.0729, 8.1055, 8.2031, 8.3008, 8.3333, 8.3984, 8.4635, 

8.4961, 8.5937, 8.6914, 8.7240, 8.7891, 8.8542, 8.9844, 9.1146,
  9.1797, 9.2448, 9.3750, 9.5052, 9.5703, 9.6354, 9.7656, 9.8958,
  9.9609, 10.0260, 10.1562, 10.2865, 10.3516, 10.4167, 10.5469, 10.6771,
  10.7422, 10.8073, 10.9375, 11.0677, 11.1328, 11.1979, 11.3281, 11.4583,
  11.5234, 11.5885, 11.7187, 11.9141, 11.9792, 12.1094, 12.2396, 12.3047,
  12.5000, 12.6953, 12.7604, 12.8906, 13.0208, 13.0859, 13.2812, 13.4766,
  13.5417, 13.6719, 13.8021, 13.8672, 14.0625, 14.2578, 14.3229, 14.4531,
  14.5833, 14.6484, 14.8437, 15.0391, 15.1042, 15.2344, 15.3646, 15.4297,
  15.6250, 15.8203, 15.8854, 16.0156, 16.1458, 16.2109, 16.4062, 16.6016,

Awesome!! Thanks so much :smiley: I will try this

MadMatt

Remember the original crystal is in an oscillator circuit, its not a clock input circuit, so you
may have to be lucky - typically one of the two pins will act as a clock injection point, but
not always. There will be a maximum rate at which you can vary the frequency before the
control loops in the drive lose lock, don't expect miracles.

The other thing to note is that the chip produces desecrate frequency values, that is they are in steps and not continuously variable.

"desecrate"? Was that a Cupertino? Should be "discrete"

MadMatt:
Sorry if I sound stubborn, but this is how I can amaze the teacher and get the best marks I can for VCE

Or not as the case may be.

I want to be an electrical engineer, and this would be great to do for a starting point.

If you want to become an engineer you must learn to design things yourself, not rely on help from the internet.

Russell.

Grumpy_Mike:
The other thing to note is that the chip produces desecrate frequency values, that is they are in steps and not continuously variable.

Using fractional maths it will do pretty much any frequency in the range but the difficulty is figuring out the correct fractional maths to get that desired frequency.

Using fractional maths it will do pretty much any frequency in the range

But that chip is not set up for it and you get phase jitter when you do that technique making it useless for audio work.