OK, how much is too much?
Jun 15, 2002 at 6:18 AM Thread Starter Post #1 of 9

finleyville

100+ Head-Fier
Joined
Nov 30, 2001
Posts
373
Likes
13
When building the META42 and space is not an issue, how much power capacitance is overkill for 24V from the wall? 1000uF? 2200uF a side?

As for a different question, will a 24V wallwart be too much for the AD8620 opamp? Is that too close to its limitations?
 
Jun 16, 2002 at 3:42 AM Post #2 of 9
Quote:

how much power capacitance is overkill for 24V from the wall? 1000uF? 2200uF a side?


I haven't tested it myself, but there have been posts in the past of people using tens of thousands of mics of power capacitance. Not with the META42, but with other headphone amps. Personally I've never gone beyond 3000 uF, and even then, I did it only because I had the caps for it. I didn't seriously test it against an amp with a smaller amount of power capacitance.

Quote:

will a 24V wallwart be too much for the AD8620 opamp? Is that too close to its limitations?


It depends on the wall wart. If you get an unregulated wall wart, they tend to go up in voltage with light loads. They only get down to 24V when supplying their full rated current level. You might get 27V out of a "24V" unregulated wall wart.

Another problem with unregulated wall warts is that they put out a direct analog of the input voltage -- if you get a 10% surge in input voltage, the output will go up by 10% as well. That would put the voltage over the 26V max of the 8620.

Even if you go with a regulated wall wart, not all will really hit their voltage targets. The Radio Shack "regulated" wall warts (which are switching power supplies, by the way, not _linear_ regulated) tend to behave just like the unregulated ones at low loads.

Now, if you've got a linear regulated 24V supply that's guaranteed out to about 130V, then you're probably safe. Personally, though, I'd probably not risk it. I doubt the chip needs that much voltage to sound its best. I'll bet you'll find that you can get excellent performance at 15V or so.
 
Jun 16, 2002 at 8:12 AM Post #3 of 9
Thanks Tangent,

I'm asking about the voltage because I'm worried about driving difficult headphones like the senn's. Finding a good regulated power source that's not huge or hugely expensive is more difficult than I thought. At least for the power requirements I'm looking for. Thanks for all the help.
smily_headphones1.gif
 
Jun 16, 2002 at 8:31 AM Post #4 of 9
With Senns, you have to have a fair amount of "swing" relative to other headphones. For example, many headphones for portable use will achieve full rockin' volume with about 300 mV. Senns might require up to 2V, depending on the model, and your definition of "rockin'".
smily_headphones1.gif


Now, an op-amp has a maximum voltage swing that depends on its supply voltages. If the input signal is precisely centered around the midpoint between the supply voltages, an OPA132 with a 300 ohm load will start clipping at around 8V p-p supply with a 2V signal. If the load is higher (e.g. with the 64 ohm Senn HD-570) it will clip at even higher voltages.

Now, all this assumes that the input signal is centered around the mid-point of the supply voltages. This does happen in the META42 amp, because of the precision voltage reference (TLE2426) and the power supply buffer. But in a lesser amp that uses a simple resistor divider for its power supply, it's really easy to shift the power supply voltages relative to virtual ground. Since the input signal is centered around the vground, the signal will shift as well, making the amp clip at higher voltages.

My point is that there's a lot of lore going around in these forums about how Senns require lots of voltage to sound good. I've helped spread some of that lore. But, if you use a strong power supply design, you can get by with much lower supply voltages. Given any op-amp, I doubt you can hear the difference between 15V and 24V, with Sennheiser headphones.

(By the way, you need to update your profile. It doesn't say which Senns you have. Inquiring minds want to know, finleyville.
smily_headphones1.gif
)
 
Jun 16, 2002 at 9:26 AM Post #5 of 9
Quote:

With Senns, you have to have a fair amount of "swing" relative to other headphones. For example, many headphones for portable use will achieve full rockin' volume with about 300 mV. Senns might require up to 2V, depending on the model, and your definition of "rockin'". My point is that there's a lot of lore going around in these forums about how Senns require lots of voltage to sound good. But, if you use a strong power supply design, you can get by with much lower supply voltages. Given any op-amp, I doubt you can hear the difference between 15V and 24V, with Sennheiser headphones.


Thanks. That is exactly what I needed to know.
smily_headphones1.gif


Quote:

Now, all this assumes that the input signal is centered around the mid-point of the supply voltages. This does happen in the META42 amp, because of the precision voltage reference (TLE2426) and the power supply buffer. But in a lesser amp that uses a simple resistor divider for its power supply, it's really easy to shift the power supply voltages relative to virtual ground. Since the input signal is centered around the vground, the signal will shift as well, making the amp clip at higher voltages.


Is this what you mean about DC offset?
confused.gif


Quote:

By the way, you need to update your profile. It doesn't say which Senns you have. Inquiring minds want to know, finleyville


Actually, I don't have them yet. I'm waiting until the world tour makes it to the 'Burgh so I can sample all the goodies. I'll probable know my next can purchase after that. I'm just preparing in case my ears choose a more difficult headphone to drive instead of my V6's, Grado's, or the HD 590's.
biggrin.gif
 
Jun 16, 2002 at 9:52 AM Post #6 of 9
Imagine a graph of a sine wave, centered around 0 on the y axis. If the signal goes up to 1 on the y axis and down to -1, you have a 2V peak-to-peak signal. (Which is not the same thing as 2V RMS, but let's ignore that for simplicity's sake.)

Now imagine that the signal is shifted up so the sine wave is centered around +.1 -- its upper peak is 1.1V, and the lower peak is -0.9V. If this were a voltage, it would still be 2VAC p-p, but with a 0.1V DC offset.

In the context of a split power supply like in a CMoy or a META42 amp, shifting the voltage means that the voltage potential from virtual ground to each rail is different. You might get +4V and -5V from a 9V battery, which would mean virtual ground has shifted by 0.5VDC.

The DC offset I think you're talking about, finleyville, is on the op-amp's output. If your virtual ground shifts, then you don't really get a of DC offset on the op-amp's output, because the headphones use the virtual ground, too -- as far as they are concerned, the signal is still centered around virtual ground. When we say there's a DC offset on the op-amp's output, it's relative to the virtual ground, whatever voltage potential it may have.

The problem with DC offsets on the op-amp's output is that your headphones work best when the output signal is precisely centered around the ground potential. The more DC offset there is, the poorer they perform, until you get the extreme case where the signal is entirely positive (or negative) relative to ground, in which case the driver is pushed in one direction constantly -- it doesn't move back and forth, and thus doesn't produce sound. It may even be damaged.

With FET-input op-amps, you have to work pretty hard to get more than about 20mV of DC offset, which is completely ignorable in most circumstances, given that the signal level is hundreds of millivolts at minimum. It's with bipolars, where offsets can be multiple volts, that you have to worry about DC offset.
 
Jun 16, 2002 at 3:50 PM Post #7 of 9
Why not make your own power supply? That's what I'm doing.

Don't let mains voltage scare you away, just use common sense and be careful. You only really deal with that mess getting into the trafo anyhow, once it's stepped down it's *much* more sane.

I'm planning on using 4*5000uf Cerafine caps (total overkill) and push the stepped-down 30V into sevearl stages of regulation so I'll have a +/-24V,18V,15V,12V,9V,4.5V. Each voltage step down will use the previous regulated section to feed it's regulators. Then I'll push the regulated rails into a pair of EL2001s (in parallel) to stabalize the rails for the up-to .5V they could be off from the regulators.

Even as overkill as this is it should be fairly easy, though expensive, to make.
 
Jun 17, 2002 at 12:36 PM Post #8 of 9
Quote:

With FET-input op-amps, you have to work pretty hard to get more than about 20mV of DC offset, which is completely ignorable in most circumstances, given that the signal level is hundreds of millivolts at minimum. It's with bipolars, where offsets can be multiple volts, that you have to worry about DC offset.


Huh? Neglecting bias current for now, bipolars can usually acheive an order of magnitude lower offset voltage than FETs. This is of course excluding FET chopper op-amps.

High-speed bipolars usually exhibit less than 50 uA of bias current, 7 uA of offset current. Precision bipolars usually sit at the tens of nanoAmps while superbeta bipolars demonstrate sub nanoAmp bias current. Unless MegaOhm resistor combinations are being used, only the highest bias bipolars will prove to be troublesome. Of course MegaOhm resistors are trouble one way or another.

Nezer: Cascading so many stages of power regulation strikes me as a thoughtless way of providing power. Precision regulators/references can be cheaply purchased. Coupled with an addtional transistor a high-drive, precision power supply can be obtained.
 
Jun 17, 2002 at 5:06 PM Post #9 of 9
I wasn't talking about input offset voltage. I don't remember which spec it is, but as I understand it, either input bias or input offset current is high enough with the bipolars I've used so that you have to match impedances on the op-amp's input or suffer large DC offsets on the output.
 

Users who are viewing this thread

Back
Top