This is the case in my CD player with unregulated +/-15V lines. Does it affect sound if the positive and negative input voltages are not the same number? How much does it matter?
Last edited:
Thanks, I will probably see if I can put in 9V regulators to stabilize the voltage. It's slightly fluctuating now which I presume does matter?
The op-amp's PSRR may be sufficiently high to render minor power supply voltage fluctuations inconsequential. However, in general, you should always take measures to minimize power supply rail noise.
I'd take care that the peak-to-peak output voltage can be reached on opamps running on the regulated +/- 9V. If output gets within a couple of volts of the power rails, some opamps could clip. A well-filtered (thus any voltage variations are well below 20Hz), but higher and unregulated voltage (closer to the 13 and 14.5V) may be preferable.
Most opamps have large PSRR at audio frequencies, and really lousy PSRR at MHz frequencies and above... so a bit of ripple or sag on the rails is unlikely to matter (as long as the opamps are suffifiently far away from clipping that it doesn't increase THD), but watch out for HF noise.
Add some regulators please. 9 Volt ones are OK, anything better than unregulated voltages with possible incoming ripple or RF garbage. It would not hurt to add coils and caps so that CLC filtering is before the new regs. It would be even better to add a complete new symmetric PSU with a separate very small 2 x 12 V transformer and 12 V regs so that the opamps get their +/- 12 V DC again. Depending on the make and type of the transformer you might even use 15 V regs as most put out way higher voltages than specified.
Please check if the original opamps are bipolar or JFET types and check the things benb stated indeed.
Please check if the original opamps are bipolar or JFET types and check the things benb stated indeed.
Last edited:
Which CD player is it? I strongly doubt that the analog stage is fed by unregulated power supply. As for the assymetry of + and - voltages, it is found very often in op amp supply rails. Probably not intended, but it is beneficial for suppressing higher order distortion.
.
In hopes of clarifying, realize that in any circuit "ground" is an arbitrary. It's just a point we define as zero, and then all voltages are measured with one meter probe placed on that point. Theoretically we could pick any random point and call it "ground."
But we pick some logical point as ground, not a random point. In fact, we logically build a power supply--or connect two batteries--with a midpoint that we can reasonably define as zero volts, or ground.
But in the real world transformer outputs vary, and so does battery voltage. So our supposed midpoint is never precisely that. For instance, two 9 volt batteries, total of 18 volts, supposedly have a midpoint of 9 volts, which we arbitrarily define as ground and call zero volts.
But it could easily happen that one battery is at 9 volts, another is at 8 volts, a total of 17 volts. In that case the midpoint--which is, by definition, ground--would be not at 9 volts, but at 8.5 volts. So then what happens?
For all practical purposes nothing happens. A very small amount of amplifying ability is lost, but in the real world amplifiers are never designed to run at maximum output (and would probably overheat if they did), so this very small loss doesn't matter.
The lost amplifying ability is called "headroom," and this lost ability means it's possible for one side (positive or negative) of an audio signal to suffer distortion before the other side does. But again, in the real world this doesn't happen because the amp doesn't run at maximum.
I've talked about batteries, but it works the same way with transformers. As a matter of fact, you wouldn't be wrong to think of any power supply as just a battery(s) that never runs down.
The upshot is that it makes no difference, you don't have to do anything. If anybody tells you different ask them how a single-supply circuit works.
Sometimes this stuff is hard to catch onto without visual aids. I hope this might be of some help.
.
In hopes of clarifying, realize that in any circuit "ground" is an arbitrary. It's just a point we define as zero, and then all voltages are measured with one meter probe placed on that point. Theoretically we could pick any random point and call it "ground."
But we pick some logical point as ground, not a random point. In fact, we logically build a power supply--or connect two batteries--with a midpoint that we can reasonably define as zero volts, or ground.
But in the real world transformer outputs vary, and so does battery voltage. So our supposed midpoint is never precisely that. For instance, two 9 volt batteries, total of 18 volts, supposedly have a midpoint of 9 volts, which we arbitrarily define as ground and call zero volts.
But it could easily happen that one battery is at 9 volts, another is at 8 volts, a total of 17 volts. In that case the midpoint--which is, by definition, ground--would be not at 9 volts, but at 8.5 volts. So then what happens?
For all practical purposes nothing happens. A very small amount of amplifying ability is lost, but in the real world amplifiers are never designed to run at maximum output (and would probably overheat if they did), so this very small loss doesn't matter.
The lost amplifying ability is called "headroom," and this lost ability means it's possible for one side (positive or negative) of an audio signal to suffer distortion before the other side does. But again, in the real world this doesn't happen because the amp doesn't run at maximum.
I've talked about batteries, but it works the same way with transformers. As a matter of fact, you wouldn't be wrong to think of any power supply as just a battery(s) that never runs down.
The upshot is that it makes no difference, you don't have to do anything. If anybody tells you different ask them how a single-supply circuit works.
Sometimes this stuff is hard to catch onto without visual aids. I hope this might be of some help.
.
Last edited:
- Status
- Not open for further replies.