Zener Diode Protection

[Fowkc]

I have a system on breadboard which has both 5v and 12v power lines. On two occasions I've managed to accidentally power the 5v parts from 12v, and hence fry the PICAXE and associated LCD display. (And at £7 per display and £5.70 per 28X, that's a lot to destroy)

If I put a 5.1V zener and small resistor (of suitable power rating, say 1Ohm, 10 Watt) between the 5v line and ground, will this be enough protection, or is there a better way? I mean like this:
<code><pre><font size=2>
Accidental
12V ---------/\/\/\-----------5.1v Out?
Power 1Ohm 10W |
Resistor |
|
__| 5.1V
/\ Zener
|
|
Ground
</font></pre></code>
Obviously I don't want to test this with my precious components in, in case it doesn't work.

(Apologies for the dodgy ASCII art)

Edited by - Fowkc on 2/20/2006 1:50:29 PM

 

[hippy]

The easiest solution would seem to be to use a 5V regulator and power everything from 12V only.

 

[Fowkc]

Yeah, I had the &quot;terrific&quot; idea of putting the 5v regulator in a seperate box.

It's for gliding you see. We use 12V SLA batteries for power, with a 3pin XLR connector (for polarity protection and ruggedness). I built a small box to take the XLR in and give 12V and 5V flying leads out so I could breadboard stuf easily. The 12V lead is a different colour and is clearly labelled &quot;12V&quot;, but it didn't stop me from putting it in the wrong place. Twice.

I suppose going back to a pure 12V + regulator on breadboard system is best. But it seemed like such a good idea at the time...

 

[pbunyan]

the schematic you posted should work, but i i would put the resistor in between the zenor diode and GND, as in the current position it would also limit the power when its running form 5 volts. wire it up with a 5 volt buld or something you could bear to lose, and see if it works. it shoiuld with the modification described, but don thold me to it!

Hope this helps!

 

[Fowkc]

Isn't the resistor there to drop the 12-5 = 7 volts? If I put it below the zener, won't the drop occur just the whatever wiring there is between the zener and the battery? Or am I missing something?

Anyway, I've gone back to a regulator on the board. The idea was to reduce the component count on the board, so if I have to have a large resistor and diode for this protection, I may as well have a regulator instead!

 

[BeanieBots]

Yes, if you did as suggested by pbunyan you would loose any protection the zener offered.
The best bet is the one suggested by Hippy. Namely a 5v regulator on board.

 

[pbunyan]

hmmmm. if you apply 5 volts to the (modified) circuit at the start, it will be fine, and supply five volts, but if you supply 12V then the zenor diode will kick in, and sink it all to ground via the resistor, which is used as a load, to stop a short circuit. is that correct? i thgought so. ah well, it doesnt matter if your using a regulator on the breadboard

Edited by - pbunyan on 2/20/2006 5:10:37 PM

 

[bobrayner]

pbunyan
not correct. If wired as you suggest and you applied 12V you would see 5V across the zener and 7V across the resistor, Total 12V at the top of the zener. Your suggestion would do nothing except blow up the zener and resistor if they were not big enough. Try it.
cheers BobR

 

[Michael 2727]

The diagram is the correct way to do it.
But,,not having worked it out, the resistor
looks very small and 10W is overkill.
The resistors job is to limit the current the
idea is to work out the current you need
plus a little extra, because what is left over
is dropped across the zener as heat
The zener should not have to run hot to work.

 

[BeanieBots]

I fully agree with Michael 2727.
That is the key issue when trying to use a zener for supply regulation. When the circuit is pulling maximum current, the resistor must enough &quot;surplus&quot; to keep the zener conducting. When the circuit pulls minimum, the zener must conduct away the excess. Not only is it extremely inefficient, it does not regulate very well either because a zener is not a perfect device and still exhibits a certain amount of voltage which is proportional to current.
They do have their place but IMHO, NOT as a voltage regulator.

pbunyan, I think you should get some resistors and zeners and have a play with them to see what effect they have and disover for yourself what they can and cannot do.

 

[Dippy]

Have I understood this correctly. You're going to use your zener circuit to power your PICAXE (and all things 5V)?

What have you got against a tiddley regulator? Your circuit is very power-inefficient. Its unlikely you'll produce a better design than the 'Big Boys'.

Dig out the old calculator and work out the amount of power being used in the 10ohm+Zener. And then work out the wattage requirement for your zener. Owch!

Edited by - dippy on 2/21/2006 9:01:45 AM

 

[BeanieBots]

Actually, (and I think I'm largely to blame for the confusion), the original question was how to protect against accidental connection to 12v. Not how to generate a 5v rail.
So, to get back on track.
Whether or not the original circuit described would work regarding accidental connection to 12v depends on several issues.
Firstly, can the 5v circuit tollerate the volt loss caused by the 10R when running from the correct 5v line?
Secondly, can the zener dissipate the power indeffinately without overheating?
(dissipation is 3.5W with the components stated which is significantly higher than the average zener can manage).

The problem with a regulator solution is that most regulators require a headroom of about 3v, so the PICAXE would only see about 2v when fed with 5v.
There are however, special low drop-out regulators available but even these will &quot;loose&quot; several hundred millivolts from a 5v line. (might be no better than R+zener).
It is possible to get &quot;fuseable&quot; resistors.
These are special resistors which will &quot;blow&quot; open circuit very quickly when they exceed a certain power rating. If used in combination with a suitably rated (5W min) zener and maybe a 1R fuseable resistor, they would offer better protection.
Alternatively, use a 100mA fuse (or other suitable value) with the zener. However, the resistance of low value fuses is very high (several 10's of ohms) which is why fuseable resistors are more popular for such applications.

 

[Dippy]

Ah, yes, I see what you mean. Ooops. Have gone blind and daft.

Its just to prevent accidental brief dabs from the 12V rail onto the 5V rail?

Just to clarify for me: The 10 ohm is between the 5V rail and PICAXE +V pin and the zener is from PICAXE +V pin and ground?

I TAKE BACK all what I have said - and I agree with comments about fuses.

All I would say is be cautious about zener ratings, as someone said, they WILL vary with current. I'd get a few values of Zener and have a test run with a bench power supply. But the principle is certainly OK.

Temporarily OTSO Fuses: what about those Bourns resettable 'Multifuses'. I just had a quick look. One that trips at 500mA has an 'ON'res of 1 ohm. Any use?
http://www.farnell.com/datasheets/36437.pdf

 

[pbunyan]

sorry, what id been taught about zeners is different to what wikipedia has to say! yes, that would work as a voltage limiter, but very inefficient.

sorry!

Edited by - pbunyan on 2/21/2006 10:29:01 AM

 

[BeanieBots]

Dippy, yes, they look quite good. Reasonably quick.
At a quick glance, I couldn't see how they reset. I'm always very wary of anything that &quot;self&quot; resets because at the end of the day, it has blown for a reason and if it resets on its own, unexpected consequences can happen.
cf. bi-metalic type used as motor cut-out can cause unexpected motor activation while inspecting for the cause of failure ! !

pbunyan, be a little cautious of wikipedia. Although on the whole it is a very valueable source of information, it is non the less, maintained by anyone who wishes to contribute and I have seen a few very wrong entries. At time of writing, the entry under zener is sound.

Edited by - beaniebots on 2/21/2006 12:53:50 PM

 

[Dippy]

BB, yes I know what you mean. Horses for courses etc.

Can't beat Data Sheet plus calculator. Well you can actually - you get someone else to do it first!

 

[triple crown]

BB is right, the resettables can constantly retry to activate and if there is something wrong with the circuit, the circuit could cause major problems as said, e.g. using high current motors. I beleive in the failsafe system which is fuses.

 

[Dippy]

Like I said before; horses for courses.
In your example you give, and many others too, a fuse or mcb is the best. No argument.

But if you're splashing around with a bread-board where you just might accidentally touch something then having to change fuses can be a pain in the arm. It just depends on the application, eh?

The PolySwitch/Multifuse device resets after the fault is cleared and power to the circuit is removed.

Edited by - dippy on 2/25/2006 10:32:37 PM

 

[Fowkc]

Back to 5V regulation now, it's all fine. My nice box will be recycled to hold something else.

 

[bobrayner]

These &quot;multifuses&quot; or &quot;polyswitches&quot; as they are sometimes called heat up very rapidly when rated threshold is reached and thence maintain just enough current to keep them hot, usually 50 to 100 mA. They only reset when the overload condition or supply is removed. A rise in ambient temperature will reduce the activation point as they are temperature dependant. They are quite often used in series with PA speaker leads where the wiring and/or speakers could be subject to damage.
As for fuses. They DO NOT protect equipment. They are there to protect the SUPPLY and WIRING. Faulty equipment is usually the cause of fuse failure, they take so long to &quot;blow&quot; that if the problem is over voltage or similar supply problem almost all of the semiconductor devices will self destruct while the fuse blows. If over voltage protection using a fuse id required it is usual practise to use a &quot;crowbar&quot; circuit to rapidly put a short on the equipment side if the fuse and purposely fail it.
Hope this clarifies the use of a &quot;fuse&quot;
cheers BobR

 

[Snowface]

if the 12v is constant and already regulated then why not simply use a potenial division circuit? (2 resistors at correct values for the amount of current you need.)

 

[BeanieBots]

Even if the 12v is constant, the load is not. Therefore, the supply voltage will be all over the place.
One of those &quot;looks like it might work in theory but does not in practice&quot; scenarios.