Zeners are strange, when you're a stranger

Started by MoltenVoltage, May 07, 2012, 02:46:26 PM

Previous topic - Next topic

MoltenVoltage

I have a very low power circuit and want to regualte it at ~ 5 volts without any voltage regulator vampire current as discussed in this thread:
http://www.diystompboxes.com/smfforum/index.php?topic=97307.0

Seljer suggested using a series Zener diode with the battery which I tried (0.5W 5.1v) and it worked, but I discovered that the voltage on the circuit side of the zener varies with the amount of current going through it.  I confirmed this by hooking up an LED to a pot and found that the zener voltage varies proportionally to the amount of current to the LED.

I always assumed zeners allow a "fixed" voltage to flow, but that appears not to be the case at all.

Also, when the current is suddenly reduced (like when a switch is no longer engaged), the voltage level recovers, but curves up as if there were a capacitor discharging.  It takes a very noticeable amount of time when viewed on the scope.

Is this normal, or is the ghost of robby krieger haunting my breadboard?
MoltenVoltage.com for PedalSync audio control chips - make programmable and MIDI-controlled analog pedals!

merlinb

A zener will only maintain a constant voltage across itself when the current through it is above a critical value, perhaps a milliamp or so. Below that and the voltage across it quickly drops (so in this case your output voltage will increase at low current).

A possible solution to your problem is to use a special low-current CMOS opamp as a voltage regulator (just like the Vref buffer you see in some pedals).

R.G.

Welcome to Mother Nature's Fun House.

Zener diodes are not simple if you rely on them for precision, regulation of any significant amount of power, or regulation of very small amounts of power. The zener effect causes a relatively (note that term well: it's not tossed in there to increase the word count) constant voltage to drop across it within certain current bounds. They have internal resistance, both real and in-effect. Things get odd around the voltage where the zener starts being semi-constant-voltage.

Seljer mentioned a micropower regulator; that was a good idea, and worthy of your further consideration. Another is to program the uC to self-regulate its own voltage with an output pin. If it's really low power, two coin cells are an even better idea. Two coin cells for keep-alive and having the uC turn on a higher supply voltage when it goes active is another idea worth thinking about.

Quote from: MoltenVoltage on May 07, 2012, 02:46:26 PM
I have a very low power circuit and want to regualte it at ~ 5 volts without any voltage regulator vampire current as discussed in this thread:
http://www.diystompboxes.com/smfforum/index.php?topic=97307.0

Seljer suggested using a series Zener diode with the battery which I tried (0.5W 5.1v) and it worked, but I discovered that the voltage on the circuit side of the zener varies with the amount of current going through it.  I confirmed this by hooking up an LED to a pot and found that the zener voltage varies proportionally to the amount of current to the LED.

I always assumed zeners allow a "fixed" voltage to flow, but that appears not to be the case at all.

Also, when the current is suddenly reduced (like when a switch is no longer engaged), the voltage level recovers, but curves up as if there were a capacitor discharging.  It takes a very noticeable amount of time when viewed on the scope.

Is this normal, or is the ghost of robby krieger haunting my breadboard?
R.G.

In response to the questions in the forum - PCB Layout for Musical Effects is available from The Book Patch. Search "PCB Layout" and it ought to appear.

Gurner


Gurner

#4
duplicate post .......self deleted

mth5044


boogietone

My Zener moment was after buying a batch of 5.1V ones to provide overvoltage protection for a microcontroller input pin. I put one across the pin and ground. In the circuit, it clamped the voltage to about 3V. Useless.
An oxymoron - clean transistor boost.

R.G.

For a bit of perspective, back at Three Initial Corporation there was a two-day internal education class for beginning engineerlings on zeners. Just zeners, and the perversity thereof.

Zeners are really simple - if you stay far from the zener knee and power above half the rated power. Otherwise, there are a few footnotes to be read.
R.G.

In response to the questions in the forum - PCB Layout for Musical Effects is available from The Book Patch. Search "PCB Layout" and it ought to appear.

MoltenVoltage

I tried putting a tank cap on the circuit side of the zener thinking that might make a difference with the voltage drop but it didn't.

As mentioned however, the circuit works fine with the series zener, since the chips can run off 3-5.5 volts, and the drop is about 1 volt.  Apparently the abrupt voltage drop isn't enough to cause any real issues with the program.  I have a hard time imagining this would hold true if the program were constantly running instead of normally in sleep mode but maybe the chips are more forgiving than I think.  On this chip, the brownout detect is 2.1 volts so presumably that's why it still works.




MoltenVoltage.com for PedalSync audio control chips - make programmable and MIDI-controlled analog pedals!

Cliff Schecht

I agree with RG in that zeners are really simple to use if you follow the rules. For voltage regulation, read the datasheet and look at what current is required for the diode you are using. Use some bypass capacitance to ground and know that you cannot source very much current from this sort of circuit. A zener, bjt and a few resistors do make a decent regulator though. I use this style of circuit (ala the RG Mosfet Follies circuit) to reduce too high of voltage on tube amps. I have 5 or 6 amps now that have a zener diode and a power bjt to knock down anywhere from 30V to 150V depending on how inappropriate the PT was for the build I was doing :P.

Also zeners make a great broadband noise source. They are much more consistent than bjt's too because zeners are designed to operate in reverse breakdown constantly. I've used 8.2V and 6.8V ones in the past with great success. In my testing I never put a zener in a noise generating circuit that didn't produce a very loud white noise when biased correctly (just a resistor and a trimpot to limit current and tune in the noise).

Somewhere in the pile of books I have one just on zener diodes and it goes much more in depth than most anybody would need. Any solid state devices textbook will also have enough on zeners to give most a headache. After beating the topic to death in a few classes and hundreds of blown zeners later I know to follow a few simple rules and they work well :).