9,1v zener as protection to a charge pump. good idea?

[pietro_moog]

in a few projects involving a charge pump i used a zener 9,1v as protection diode to keep the incoming voltage from frying the pump, in the case i plug something else than +9v (never happened but it might).
does this make the pump safe for real or is it just a way to add noise?

[chromesphere]

Good question!  My meek knowledge of diodes would say that it is protecting the charge pump as if the voltage goes over 9.1v the zener will start conducting in the forward bias (short to ground i would assume) and protect the chip.

[rullywowr]

Yes. The Zener will conduct in reverse (also called avalanche mode) dumping any voltage above its Zener rating from cathode to anode. This is a good level of protection, but if the current is too much, it can destroy the diode (especially if a current limiting resistor is absent). This means you will have to replace the diode if it fails.

[jubal81]

I've found this causes a lot more problems than it solves.

[midwayfair]

At the very least, you must have a smallish current limiting resistor. About 100R is fine. Most of the time. Note there are a bunch of designs -- even famous ones -- that don't have that CLR and they seem to blow a lot of Zener diodes.

Note that it will also provide some reverse polarity protection. Except that the Zener isn't as robust as a rectifier diode like a 1N4001, and you blow the diode anyway, even with the 100R!

Then there are times when that same 100R, in a current-hungry pedal, will sap a bunch of voltage from the final pedal.

The important thing to remember is that Zeners aren't regulators (don't use them to knock down a supply to the right voltage), and they aren't reverse polarity protection. (AND neither is a 4001 without a big ole CLR!) They are, however, useful for protecting against momentary spikes in power. A regulated supply made for powering pedals (even something like the 1Spot or Powerall) is still better than relying on the Zener.

One last thing: I use a 12V Zener as my overvoltage protection for LT1054s. My one spot never gets anywhere near that, but if it were 9.1V, it would constantly be shunting half a volt, which would heat up and potentially eventually fail.

Scruffie has posted that it's a good idea to have a resistor in parallel with the diode as well. But it's complicated calculating what that resistor's value should be.

Like Jason said, they cause a lot of problems.

If you want to read more in-depth on the subject, it's been covered recently in a DIYSB thread and in a couple threads here, so it might be worth a search.

[electrosonic]

A 1 Watt 9.1 volt zener is only rated to pass about 110mA.  A one spot puts out something like 9.5 volts and is capable of sourcing 1700mA. The zener would appear as a low resistance to ground and quickly smoke.

If you put a 100 ohm current limiting resistor in front the zener, the voltage to the circuit will be held at 9.1 volts with no load, but the voltage will drop 1 volt for every 10mA drawn by the circuit - this may or may not be a problem depending on the current draw.

Haven't considered the issue of zener noise, a cap might be needed parallel the zener to shunt the noise it creates.

Andrew.

[pietro_moog]

so the point is: on paper they works, but in reality they blow easily. so it's not worth it. ok. i'll be extra careful with the power supply.

[zilla]

Related to this, I have built a few pedals for others that use a Zener diode to protect the charge pump. I have been explicitly clear 9VDC regulated power only and even put it on the enclosure, but they still manage to blow up the Zener.

Getting a bit frustrated with this. I'm almost tempted to socket the Zeners to make fixing them easier. I moved up to a 12v Zener because the charge pump can handle an input voltage of 12v.