Zener regulator circuit, and power ratings.

[sureshot]

Your probably getting bored with all my questions, but thank you very much for the help given in my posting. Much appreciated.

So I'm looking at zener diodes again and regulator circuits..
I've built the basic zener circuit and that worked fine. I'm wandering if there's a way to increase the output power with a series pass transistor ? I did give this a go with a 2N3055 npn transistor, but noticed the zener diode getting very hot. High power zener diodes seem hard to source. Mine are 5 Watt 13 volt diodes.
I'm thinking along the lines of how linear regulators can be used with little current drawn by them, but a power transistor is doing most of the work. Only in the zener circuit the device dissipates a lot of heat, and the zener power rating has to be chosen for the intended output current. By that i mean the more current you draw the higher wattage the zener needs to be.

But with a linear voltage regulator the power that dissipates stays the same, as in more series pass transistors equals more current. But i can't find a way to do this with a zener diode. Meaning the zener diode just contributes to a very small amount of the total output power, and the series pass transistors do the overall majority of the work load. So in a word, keep the zener diode in sensible limits whilst the series pass transistor does most of the work load with out burning up the zener diode as current increases. I know its possible up to about 1 Amp, but is there a way to do it that achieves up to say 3 Amps or maybe even 5 Amps ? I have searched a fare bit for the answer but have found little on this question.
Thanks for reading, and any help or ideas appreciated.

[HB9EVI]

The question, how much current you can drain, depends from the voltage drop over the series pass transistor, you know Udrop*Idrain=Ploss.
If your zener is getting too hot, you're likely running a too high bias current through it. In a series pass with zener, you have to calculate the necessary base current plus the bias current for the zener. Don't use a 2N3055, there are much better transistors available today; of course you still keep seeing those ones in linear psu, but imo it's a no-go.

[sureshot]

Thank you for your reply. So if I lower the bias current base resistor, I can draw more current from the zener regulator circuit ? And would a pnp transistor configuration for a series pass transistor be a better idea in a zener regulator circuit. What would be the theoretical maximum current that could be drawn from a 5 watt 13 volt zener diode and say a TIP2955 transistor ? Or maybe a TIP147 transistor. Thanks again for the help.

[sureshot]

This is the circuit below I built, but I had to drop that 680 ohm resistor to get anything close to 12 volts output. This circuit claim 3 Amps output current, from memory I loaded it with a 20 watt halogen lamp. That was close to 1.6 Amps and the zener diode was quite hot. I can't remember the value resistor I used, I think it might have been a 100 ohm 2 watt resistor.

[HB9EVI]

what transformer are you using (rating, voltage)?

that circuit has several disadvantages regarding the regulation behaviour; the bias through the zener heavily depends on the load on the circuit, thus the zener voltage shifts. You would be better off using an opamp and place the zener on the regulated side - like that you have a constant bias through the zener and the transistor bias also matches better the load.

[Seekonk]

You can add a second NPN transistor and make it a darlington, C to C and E to B. that would increase the gain and less current would be needed to drive the darlington pair.  That does drop the output another .6V so you could add a diode in series with the zener to raise the voltage by the same amount.  You should look into LM/TL431 adjustable zener circuits.

[sureshot]

Thanks for the replys, yes I know it not the best zener diode regulator circuit. I'm at a loss and can't seem to find high current examples with series pass transistors. I'd like to build a psu that uses a zener diode as the regulator. 2 to 3 Amps would be useful, below about 1.8 Amps it's of little use to me.
I will check out the IC pdf sheet you mentioned. I've not used an operational amplifier in any circuits to date, could be a good idea to see if one is incorporated in a zener diode regulator circuit.

[HB9EVI]

imo it doesn't make any sense looking around for high power zener; it's neither the solution nor it is a good design; this simple setup may look intriguing, it has so many flaws that it's not worth bothering with it, specially not, when you need high or heavily fluctuating currents

take a look on https://tangentsoft.net/elec/opamp-linreg.html

you can see several setups with the zener on the regulated side, which is favorable, when using a shunt voltage reference; also you should consider a different reference than a simple zener - yea, it's working, but it's not great; zener are noisy, they are inaccurate, they have considerable temperature drift - all properties you don't really want in a psu

[sureshot]

Thanks for the link, there looks like plenty to go over there. Could take a while to get my head round a few of them in there. I know the zener regulator isn't perfect. Just on a last hunt for circuits I found this below, it looks like the Darlington arrangment,  but I'm unsure on the resistor values, guess it's a case of breadboard it and see what happens.

[Simon123]

Check this book, page 48+
http://mirror.thelifeofkenneth.com/lib/electronics_archive/Ferranti-ELineTransistorApplications_text.pdf
Circuits on page 50 and 51, much better than your, but still simple.
(calculations are for circuit on page 51)

Using similar circuit for my Variable power supply, 6A and it works fine for my needs ... Not the best but whatever.

Also this book with some zener regulator theory:
http://mmiloslavsky.narod.ru/Dimmer2/Files/References/TVS_and_Zener_Theory_and_Design_Consid.pdf

[sureshot]

Thank you for posting those pdf files, looks like a lot of interesting content in them. I will check them out. Although I didn't see anything over 1 Amp output current, although that was at a glance.

[David Hess]

There is nothing wrong with the zener series regulator other than the lack of current limiting.  Add a couple of normal diodes in series with the zener diode to compensate for the Vbe voltage drops.  Slightly more advanced versions are shown here.

[sureshot]

Thank you for the data, that also looks like some interesting circuits and info to go over. So are my 5 watt zener diodes up to the task of a 3 Amp zener regulator circuit ? Or do i need to source some higher wattage zener diodes ? Thanks again with the help.

[Simon123]

Yes they are up to the task, if you will also use transistors.
By the way that PDF i inluded, which you say only has 1A maximum. You can use formulas provided for higher current (3Amps).

[sureshot]

Thank you for the downloads again, yes I will try and work out the formula for higher current.

[Simon123]

Its just ohms law, and transistor hFE and some kirchoffs voltage and current laws, but if you have questions ask .

[sureshot]

Thanks again. 

[sureshot]

So I've been doing the maths, i didn't want to start another thread as this maths is to do with what I'm trying to achieve. If someone could look over it i would be very grateful.
My maths as follows: 13 volts zener doide of 5 watts..
5/13 = 0.385 mA. Current the zener can deliver. Series resistor for the zener doide, supply voltage 16.92 volts - 13 = 3.92 volts / 0.385mA = 10.18 ohms. Adding a series pass transistor with a minimum hfe of 25. So 0.385mA × 25 = 9.62 Amps. Total current avaliable.  I'm not sure that is right. I would appreciate someone to see if I've got that right. Thank you for reading.

These are the formula I've used in these images.
The images show a 0.5 watt zener diode and an example transistor, also the voltage was just an example i followed using my voltage and power rating parameters.

[CJay]

Your figures look right, haven't checked but it's important to note, the zener dissipates power, it doens't source current, that's what the 5W rating is, it's also important to realise that's a maximum dissipation and it's likely based on ideal circumstances, ambient temperature, cooling etc. etc. so it's not a good idea to design to those figures unless you want to fry zeners, build in some headroom.

Note that you'll need to design for maximum and minimum current drawn, the zener will dissipate maximum power when current drawn from your supply is lowest, if the output of your PSU is open circuit then the Zener will need to be able to deal with all the current that can be supplied via that resistor.

While it's not a problem with the simple design you have, zener's aren't guaranteed to be 'in spec' until they're passing a specified minimum amount of current so regulation can suffer, they're also noisy, it's common to use a zener as an RF noise source (remember, I know what you're using these PSUs for ) so they might not be ideal for your needs.

All in, Zener diodes are useful things but if you're trying to design a PSU, there are better ways to do it and a circuit that simple is going to perform worse than your others that use a 7812 with pass transistor.

[Simon123]

Yes, they appear to be correct, but you want to design it from Output (your goal) towards the input. For example you want like 3A and 12.4V

So for example, you decided you need 3A max output.
And you have a transistor with hFE(min)=15. (i suggest you to use darlington for higher hFE).
This means base current needs to be 3A/15=200mA.
Now you said your supply is about 17V (this value needs to be of a loaded supply).
Now the output will be 12.4V, which means you need 12.4V+0.6V=13V zener.
And voltage across resistor is 17V-13V=4V

Now since you know base current, which will flow threw the zener, when there is no load ... which is 200mA
You can calculate resistor for base. 4V/0.2A=20ohms.

Oh yea, and at the end you can check zener maximum power dissipation; 0.2A*13V=2.6W, but as i suggested use darlington pair, because then zener will dissipate less, temperature wont change as much and it will be more stable.

So yea your calculations are correct, but they are more of circuit analysis then designing.

[sureshot]

Thank you for the replys. Yes i did think this is way above nominal. I got the hfe wrong for the transistor i was going to use, its minimum hfe is 15 it was the TIP35C, but i can find npn darlington transistors. And i did think about the power dissipation of the zener diode, and it all looked to much in my working out figures. 3 Amps output current would be plenty, if i use a darlington transistor won't i have two forward voltage drops ? Say 1.4 Volts. So if my zener diode is rated at 5 watts, i shouldn't take it to its maximum power rating ? Plan a bit less so its not maxed out. Is that right ? Thank you for the explanations, it takes a few attemps to get my head around the numbers.

[Simon123]

Yes, darlington means 2 voltage drops.
Yes, you shouldn't take zener to its max, give it little space.

And stop worrying about how you need 5W zener, from start of your design. Calculate as I showed you in the example above, and then you will know what power rating you need.

[sureshot]

I vaguely remember when i put a 14 Volt zener diode with a TIP142 transistor, with no load on the output the zener diode got way hotter than when i put the circuit under load. I guess it was passing the resistors power.

[Simon123]

Yes, it gets hotter, because current instead of flowing into base of transistor flows into the zener. And if it got really hot, i guess you didn't use good resistor value.

[sureshot]

Yes, darlington means 2 voltage drops.
Yes, you shouldn't take zener to its max, give it little space.

And stop worrying about how you need 5W zener, from start of your design. Calculate as I showed you in the example above, and then you will know what power rating you need.

Yes i will, its starting to sink in now. Thanks for the help with this, that's to everyone that's put up with me.

[sureshot]

Yes, darlington means 2 voltage drops.
Yes, you shouldn't take zener to its max, give it little space.

And stop worrying about how you need 5W zener, from start of your design. Calculate as I showed you in the example above, and then you will know what power rating you need.

Yes i will, its starting to sink in now. Thanks for the help with this, that's to everyone that's put up with me.

From memory that might have been the case, i took the resistor to the maximum. No head room i think in that circuit i put together.

[Simon123]

No worries, by the way i also suggest you when you design and stuff ... circuits, print out schematic and write voltage levels(and currents) across stuff you are interested in.
That way i think you will understand it even faster, since its nice visual representation, not just some stupid math equations lol

[CJay]

You don't have to find a Darlington, you can make one, pick yourself a pass  transistor and then add another 'driver' transistor, have a look at the Ham Jumbo PSU, that's pretty much a classic Zener regulated, Darlington pair based PSU.

As mentioned earlier, if there's no load on the output then the current through the Zener will be maximum, limited by the series resistor so that's why your Zener got hot.

[Simon123]

And there is also another configuration Similar to darlington, called sziklai pair, which gives you 0.6V drop, not 1.2V as darlington ... But for now i suggest darlington, since you are beginner.

[sureshot]

Yes i have looked closley into darlinton and Sziklai pairs, the latter had a slower response time i read, but i don't think that's an issue with this circuit. Buy yes i will stick to darlington transistors, or maybe make one with two bipolar transistors. I read the resistors in the darlington pair improve response times, and help with on to off states. Think that was it. Anyway thanks ever so much for the help from everyone, I'm learning it, be it slowly.