stability of a zenner diode

what sort of stability does a zenner diode have ? I may need a low voltage reference and some drift is fine as long as it is not too much, the temperature may be varying from -10 to 50 C but and the current may vary by an order of magnitude

The stability depends on the voltage of the zener, 5.6V is best.
Zener diode - Wikipedia, the free encyclopedia

Why not use the TL431 which is even better and you can set the voltage to the desired level?

I'm looking for a voltage reference as low as 1 volt with as low as possible drop out, it is for my dynamo regulator project (just to make a change) so a zener seems the obvious choice but if it varies too much it will kind of wreck havoc

what is there to stop me using a couple of normal diodes in series ? I guess the junction voltage will vary in function of the current ?

That would be worse than a zener.

Why do you need such a low drop-out voltage?

No stable regulator will give you 1V out with 1.2V in.

You could use an LM431 with a potential divider but the minimum voltage setting is 2.5V.

If you're serious about the drift with temperature not being too critical, how about a forward biased silicon diode at ~.8 Vf , or a Schottky diode at ~.3 Vf. I can't find any data sheets that graph Vf vs Temp. I suppose the silicon diode follows VBE= kTJ/q * ln (Ic/Is)

Ken

Hero999 said:

That would be worse than a zener.

Why do you need such a low drop-out voltage?

No stable regulator will give you 1V out with 1.2V in.

You could use an LM431 with a potential divider but the minimum voltage setting is 2.5V.

Click to expand...

It's for my dynamo controller I need a reference voltage that will only become available when the dynamo is turned so it has to be low enough to be produced by the nearly dead dynamo but high enough to signal: hey I'm alive lets get going, if I use a 78L05 (or low drop out version) I may not get it to kick start, I'll have to check up on how many volts a dynamo will make from residual magnetism before the reg kicks in

Well a TL431 should just output the input voltage minus a tiny leakage current when the voltage is below 2.5V, add a potential divider and you have your reference.

But why not just use a regulator IC and connect it to the battery as I showed you previously?

The ignition switch will prevent the battery from discharging and the dynamo should provide enough current to charge the batteries slowly when the engine is running in idle. If this wasn't a good design then millions of cars that use it wouldn't have been produced.

what you forget is that I am trying to make a drop in replacement for a unit made long before most of the users of this forum (including you and me) were even born and it's a totally different technology and I don't want to rock the boat too much, the unit I am currently trying to replicate had nothing to do with the ignition switch, I did do as you suggest with a previous prototype but have changed my design (from a pic one to a descrete one) and want to raise the stakes and get this one to do exactly as the original.

my main issue at the moment is that I do not know how may volts I will have on the dynamo when it first starts to turn and/or some power is supplied by the ignition lamp, it will probably be enough to run a low voltage reference perhaps even up to 3 V, the rest of the circuit will run off the battery, by using the dynamo to power the reference I will be making a way to keep the circuit off when the engine stops (because the ref will become 0 so the feed back voltage will be way higher and fool the "logic" keeping the mosfet off) and telling it to start up when the engine does.

All right I see your problem but I'd like to confine the discussion to the other thread - having multiple threads dealing with the same project just makes it harder.

Hero999 said:

All right I see your problem but I'd like to confine the discussion to the other thread - having multiple threads dealing with the same project just makes it harder.

Click to expand...

agreed I have a habbit of doing it I'm afraid

One problem with zener’s is it's value changes with temperature (around the 5.1V value it has no change) , however temperature compensated zener’s are available. Within the “TL431 family” there is also a 1.24V .5% version available. If you have to go below a volt, Acutechnology has an AQT102 which is a 600mV .1% reference.