CD/DVD drive power consumption

[shunker]

Hi
I've been trying to find data on the power consumption of a CD/DVD drive at different rotational speeds of the turntable motor that spins the disk, but no luck.

I'm trying to get this data myself by measurement. However, this requires that I be able to control the speed of the motor (more or less) precisely (in particular, in a Constant Angular Velocity mode). Is there any way I can do this speed control in software?

-shunker

 

[HammerBot]

There isn't any significant change in power consumption at different rotational speeds. Probably only a few watts. This is not going to overload your PSU, unless you have a way too small supply. In that case it is much more likely that your PSU would belly up if you run some CPU demanding software. The increase in power consumption here is much more significant.

<i><b>Engineering is the fine art of making what you want from things you can get</b></i>

 

[shunker]

I was looking at CD/DVD drives on portable devices like laptops, DVD movie players, etc. Here, a few watts could make a difference in the battery lifetime. During a DVD movie playback, the DVD drive would then consume a significant portion of the total power, and it may be worth trying to reduce this.

 

[HammerBot]

I see your point. But then again a few watts compared to the total of the laptop (60W+ ??) is only a few percent.
I have never seen any manufacturer who specifies the power consumption that precicely, so if you want to know it you must measure it yourself.

<i><b>Engineering is the fine art of making what you want from things you can get</b></i>

 

[JP5]

I think you would need to hack your CD firmware, unless the current firmware allows software to control CAV. It would be easier to just test an older, slower drive and see what the power draw was.

 

[shunker]

Thanks for your reply, Hammerbot, and sorry for the late reply. I checked the voltage and current consumption on my laptop while the CD drive was not used and while it was used. The voltage reported by the battery monitor was about 15 V.

i) I played an audio CD (sound output muted), and the difference between the currents during and after playing was about 150 mA.
ii) I wrote a small program to copy sectors 5 to 9000 of the drive, one sector at a time. This data was not saved to a file, so the hard disk was not involved. The current change here was about 600 mA.

In case i, the speed must be 1x. In case ii, I did not set any speed explicitly, so the drive must have chosen an optimal speed itself. So, there was about 7 W difference in powers at different speeds.

I believe either of these speeds must be well within the highest possible speed, as the drive was very quiet in both cases. When I copied a 100 MB file from the same disc from the command line though, the disc made a lot of noise, so higher speeds are possible, and can consume quadratically higher power.

I'm trying to control the speed in software through the SELECT_SPEED command that many drives support. However, my laptop drive does not support this feature (confirmed by using the GET_CAPABILITY command), so I'm trying this on other drives.

 

[shunker]

You're right jp5. The uniform CD driver in linux, and the SCSI Multimedia command specifications only seem to have a command called SELECT_SPEED or SET_CD_SPEED. There is a field called rotational control, however, it says
00 - CLV recording
01 - CAV recording
others -reserved

Burning DVD-RWs in CAV seems to be a popular application of the 01 flag. But I don't know if it will work with CAV reading too.

I guess I'll have to dig up an obsure opcode from somewhere for CAV.

 

[shunker]

You're right jp5. The uniform CD driver in linux, and the SCSI Multimedia command specifications only seem to have a command called SELECT_SPEED or SET_CD_SPEED. There is a field called rotational control, however, it says
00 - CLV recording
01 - CAV recording
others -reserved

Burning DVD-RWs in CAV seems to be a popular application of the 01 flag. But I don't know if it will work with CAV reading too.

I guess I'll have to dig up an obsure opcode from somewhere for CAV.

 

[HammerBot]

Exactly where did you masure this current? On the battery supply? In that case you also measure the increased power consumption of the CPU etc. You have to measure the current drawn only by drive on both 5 and 12 V supplies.

<i><b>Engineering is the fine art of making what you want from things you can get</b></i>

 

[shunker]

Yes, this is the battery current. You're right, ideally, I should've measured this at the CD drive's power supply.

However, the audio playback would be done in hardware on the CD drive (not involving the CPU - I confirmed this by observing the CPU utilization while it was playing), so the number for that may not be wrong.

And for the second case where I copied the CD sectors, the CPU utilization averaged to about 2%, so this would've had some effect. I don't have actual numbers, but I'm guessing that for a 100% CPU utilization, if the average power is 30 W, then, in my case, the additional average power consumed by the CPU would be 0.6 W.

 

[HammerBot]

You should measure the consumption of drive to be sure about it. I find it unlikely that that small motor in the drive has such high power consumption.

<i><b>Engineering is the fine art of making what you want from things you can get</b></i>