regulator IC or zener?

[tempus]

Hey all;

Is there any advantage to using a voltage regulator like a 7809 when you could just use a 9v zener diode? I understand that it does a better job of keeping the voltage steady from a rectified AC source, but what if you've already got a solid DC power source (like from phantom power or a laptop power supply)? It just seems that a zener diode would be easier to use and take up less board space, if there were no other considerations.

Thanks

[R.G.]

There are other concerns.

To start with, zeners are shunt regulators, three-terminals are series regulators. Shunt regulators have to have a series current limiter - resistor, solid state, something - to supply a maximum available current to the shunt element. This is not optional. A zener or other shunt regulator connected to a low impedance voltage source will overcurrent and burn up. When the load uses zero current, the shunt regulator has to eat the maximum current available from the source. As the load uses more current, it takes it away from the shunt regulator until there is zero left for the shunt regulator and the regulation ceases.

I did a lot of zener diode regulator designs back when I designed power supplies for a living. I left with the impression that it is much harder to do a good design of a zener shunt regulator than a series regulator, even if you're doing the series regulator design from discretes instead of three-terminal devices, which were new and unpleasantly expensive at the time. IMHO, zeners are unreliable at any significant current level - like over 10-30ma.

Zeners in particular have wider tolerances than three-terminal regulators unless specially selected, and worse drift. Being a non-feedback device, their suppression of noise and ripple on the incoming power supply is limited.

Shunt regulators work best if the incoming DC power is much larger voltage than the output, so the current limiting parts have better control of the currents. Series regulators work best when the incoming DC voltage is only slightly higher than the output voltage, so less power is wasted.

A three terminal can be just that, a three-terminal. If you're using less than 100ma output, it can be a single TO-92 package. A zener will typically be a DO-4 zener, plus a possibly high power resistor and at least one output cap, so it is usually as many or more parts and more board space.

There are a lot of concerns - and reasons that three terminal regulators are so popular.

[tempus]

Thanks for the explanation RG. I knew there must be some reason...

The application I'm considering the zener for is a small electret type mic, which I use to amplify my classical guitar. I'd like to set things up so that it could safely be powered by 48v phantom power, should the situation arise. Since I'd have to dump at least 40v, I think I'll give one a try (as most regulators don\t want an in/out differential greater than 35v). The incoming DC will also already be well regulated, and the mic will draw a small current.

Thanks again\

[R.G.]

The application I'm considering the zener for is a small electret type mic, which I use to amplify my classical guitar. I'd like to set things up so that it could safely be powered by 48v phantom power, should the situation arise. Since I'd have to dump at least 40v, I think I'll give one a try (as most regulators don\t want an in/out differential greater than 35v). The incoming DC will also already be well regulated, and the mic will draw a small current.

Those details make a difference. In low current cases like this, and cases where the input/output differential is big - like this, again - a zener is a workable solution.

I was responding to the more general questions. Yes, use a zener. Put a 0.1uF ceramic and a 10uF to 100uF across it, and a resistor in parallel as well as a resistor in series.

Is the voltage just for the electret mike, or for a buffer circuit that comes with it? If it's just the electret, the current draw will be microscopic. If there's a buffer, it will be a bit larger, but still not much.

Calculate the series resistor to give the zener about half or less of its maximum allowable current with no load. Also, I'm not too familiar with phantom power - are there current limits on it?

[tempus]

Thanks again RG.
Phantom power is supplied through 2 6.81K resistors (one for each of the out of phase XLR inputs), so the current available is pretty low already.
The electret has a buffer circuit built in, and typically they expect to see 3-10v through a 2.2K resistor. I was planning on using the 2.2K as the series resistor, at least as a starting point. What's the parallel resistor for, and how should it be sized?

Thanks