Depends on the life expectancy for which you want to design the equipment. 4,5 is what we use here. The lower the ratio, the shorter the average life.Stream said:No, I just have 10pcs of 40V shottky and 8 pcs of 60V shottky diodes... I heard about safety ratio *3-4, but not so high, as you said
Yes, it would be safe, if your transformer has two secondaries. For a center-tapped transformer it makes no sense.Stream said:
Schottkies switch faster. That means lower peaks and less ringing. That is why people expect them to sound better as well. Lower peaks mean, you can use a smaller safety margin.Igla said:
4,5 is a rule of thumb, not a fixed limit, below which the diodes would blow up immediately, and it comes probably with its own safety margin as well. It is a value that is based on average measurements and experience of more learned people than I am. Going below it can]/i] shorten the service life, but must not necessarily do that. Or if it does, it may be of no importance, because the rest of the amplifier may fade away faster anyhow. Audio equipment also has the advantage that it is usually not constantly driven at full throttle. Less strain increases the life expectany again, so it may level the smaller safety margin out.
CMOY
For what its worth, my CMOY puts out about 30mA and its not wanting even with low impedance headphones. See my CMOY web page.
Of course people here like to go overkill, good luck and cannot wait to see your results.
For what its worth, my CMOY puts out about 30mA and its not wanting even with low impedance headphones. See my CMOY web page.
Of course people here like to go overkill, good luck and cannot wait to see your results.
What's wrong with a little headroom? I found my first "cmoy" with OPA2134 to be a bit weak when driving large diaphragm headphones with heavy bass music.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=136852
http://www.diyaudio.com/forums/showthread.php?s=&threadid=136852
30mApk into 8ohm phone is just 3.6mW.
If the reactance of the load is as bad as severe reactance speakers then peak transients can approach three times what is expected into a nominal resistive load.
That reduces the effective maximum output to just 10mApk into 8ohms.
This is equivalent to 0.4mW at maximum volume.
Now apply the 20dB headroom to account for music dynamic range and you have just 65db SPL available on average music level into 90dB/mW phones.
Most listeners would claim they can't hear anything at that low SPL.
An opamp with a limit around 30mApk is just not suitable for driving low impedance headphones.
Listen to the physics.
If the reactance of the load is as bad as severe reactance speakers then peak transients can approach three times what is expected into a nominal resistive load.
That reduces the effective maximum output to just 10mApk into 8ohms.
This is equivalent to 0.4mW at maximum volume.
Now apply the 20dB headroom to account for music dynamic range and you have just 65db SPL available on average music level into 90dB/mW phones.
Most listeners would claim they can't hear anything at that low SPL.
An opamp with a limit around 30mApk is just not suitable for driving low impedance headphones.
Listen to the physics.
AndrewT said:
Huh?
P = I^2 (R)
P = .03*.03 * 8
P = .0072 = 7.2 mW not 3.6 mW?
The lowest impedance phones I have are 16 ohms with 105 db/mW, so even 1 mW will make my ears ring.
The highest impedance are 600 ohms, 88 db/mW. Here its about 540 mW being produced (.03*.03*600)
Anyway, sounds fine to me but I'm not a headphone fanatic. I do see the advantage in being able to supply more current and my next headphone amp will do more. Maybe it will be the amp that comes from this thread?
Let's use your example of 105dB/mW phones and Ipk~=30mA.
Max power = 0.01^2 * 8 = 0.8mW.
listening at -20dB ref max power results in 84dB average level.
No argument, that is ~ 18dB louder than the earlier example.
even 74dB SPL and 30dB of overhead for transients will sound adequate to most listeners.
Max power = 0.01^2 * 8 = 0.8mW.
listening at -20dB ref max power results in 84dB average level.
No argument, that is ~ 18dB louder than the earlier example.
even 74dB SPL and 30dB of overhead for transients will sound adequate to most listeners.
30 mA is Ipeak. You should calculate with Ieff. 0,707 * 0,03 = 0,021, then you get 3,6 mW.lgreen said:
The same principle. When you use Ieff, you get 270 mW, if you have Ueff = 12,6 V at the output. With those op amps you can count yourself lucky, if you have 12,6 V peak.lgreen said:
For 540 mW you would need an output Ueff of 18 V = Upeak 25,4 V. That is rather a Gainclone than a CMoy.[/B][/QUOTE]
thanks
OK, thanks guys. I calc the power and RMS stuff about 1 time every 2 years so I always forget how its done.
Also, for me its not the opamp, its the virtual ground that is limiting my current to about 35mA. So swapping out the opamp is not a solution.
Apologize for going off-topic. Back to the amp under discussion.
OK, thanks guys. I calc the power and RMS stuff about 1 time every 2 years so I always forget how its done.
Also, for me its not the opamp, its the virtual ground that is limiting my current to about 35mA. So swapping out the opamp is not a solution.
Apologize for going off-topic. Back to the amp under discussion.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.