Zener reverse current pulse - can the zener take this?

Try putting a capacitor accros the zener to protect it.

Colin =^.^=

It's OK to write 12V, but not 12volts. That should be with a hyphen: 12-volts, or a space, 12 volts. Likewise 12-ohms or simply 12 ohms, but not 12Ohms.

As for your 20mJ zener, I recommend using a P6KE20A. Only 34 cents at DigiKey.

--
 Thanks,
    - Win

That would be useless since it would charge quickly in couple of micro-seconds.

I'm looking for a cheap solution (as compared to the P6KE diode) since the product is to be sold for less than $10.

-Mike

This is an unusual event, with the battery reversed (what does the switch do in that event?), and it's likely one small zener will handle it OK. But if you want to be more comfortable at a low price, parallel two zeners. The fact that their zener voltages might differ doesn't matter, because at high currents these zeners have a substantial resistive voltage drop so both will share the current.

Tell us more about your switch, it may have a bearing on the scene.

--
 Thanks,
    - Win

Well not if you use the right size capacitor, as I dont know the switching rate its only a gues, but asuming worst case the capacitor is initialy at 20v, lets limit the voltage to 25v wich although its off the graph might well corespond to less than .2A wich puts it nicely in the 2ms pulse handling region, total inductive energy is I^2*L/2 = 20mj wich ignoring zener current for now is absorbed by the capacitor from 20v to

25 v = CV^2/2 therfore C = 177uf. (barring mistakes)

Assuming capacitor is discharged completly each time, its only 64uf.

The zener absorbs the energy from the capacitor but at a peak power of 4w.

Of course now you dont even need a zener, just a resistor will do. but this depends on you pulse rate.

Colin =^.^=

Or put two 1N5240 zeners in series.

--
T

If it's not broken, don't fix it.

You've got 12 or 14V to play with, can you just put a second ~1-cent

1N400x in series with the input and lose the zener entirely?

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

Continuous 1A is a little high for a 1n4007. Two in parallel, or better, something bigger, like a five-cent 1n5400.

--
 Thanks,
    - Win

In message , dated Thu, 24 Aug 2006, Winfield Hill writes

The IEC TC25 police will be calling on you, Win. I wouldn't comment, but I hope your book uses the symbols in the official way; I won't say 'correctly' because that's not an appropriate word for a set of arbitrary rules, however logical. Please look at IEC 60027 (multi-part).

12 volts, 12 V (with a space), 12 ohms if you can't use capital omega. Unit names derived from a person's name do not have a capital letter but symbols so derived do. So 10 joules, but 10 J, 1 millisiemens (not millisiemen!) but 1 mS. 1 millisecond is 1 ms, because it's not named after Professor Millicent Second!

--
OOO - Own Opinions Only. Try www.jmwa.demon.co.uk and www.isce.org.uk
2006 is YMMVI- Your mileage may vary immensely.

John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK

wait a sec folks, what ever happened to the one microamp of forward current?

Even if it is one milliamp, you might be able to use a cheap resistor in series to limit the current without messing up the regulation too much.

[snip]

How about using a power resistor, either in series with the zener to take some of the heat off it, or to replace the zener entirely? E.g., a 27-ohm 10w resistor, which would also quickly dispose of the inductor's stored energy.

James Arthur

Yes of course!!!!!!!! - even a 47uF 50v will do. I have spent 98% of my time writing 50 page long proposals to bring in business, and 2% on design - it is now showing up ;) Anyhow, I appreciate your help very much. BTW, this is a non-repetitive switching.

Thanks, Mike

switching

20v to 4w.

this

your welcome. In that case just replace the zener with a 1k resistor. If you look inside an SMPS you might see this circuit used as part of the snuber for the main switching transistor.

Colin =^.^=

Doesn't such a capacitor cost more than the silicon parts we are recommending, e.g., 1n5400 at < 5 cents, cost?

--
 Thanks,
    - Win

Yes, that's better.

--
 Thanks,
    - Win

Hold up there. Some in the physics, scientific and publishing community eschew a space, specifying the 20mJ vs. 20 mJ approach. I have adopted it for my s.e.d. posts these last two years. I take it you don't approve? BTW, how does one read IEC 60027 (multi-part) and which part? Aren't these the same blokes who say 65 % and 20 C, etc., with an awkward space? How do they defend that?

--
 Thanks,
    - Win

Two 1N5241Bs in series would be better again, if you believe the curves.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

In message , dated Thu, 24 Aug 2006, Winfield Hill writes

Well, you know physicists...(;-)

My attitude to rules is 'Don't break them if you don't have to. If you DO have to, go ahead.' It isn't a question of personal approval. And Usenet is highly informal; it would be very pedantic to insist on a particular format here. But we do get confusion when people really DO mean '10 mHz'. I don't think we've had a case of a real '1 MA' for a long while, but didn't you have something in the '1 MV' region a while back? We certainly haven't (yet) had a case of '1 PA, 1 pA and 1 Pa'.

Doesn't your uni library have copies of standards?

See

This is from the public part of the IEC web site.

No, no space before '%' and the degree sign. 65% and 20°C.

For many years, IEC TC25 was mainly peopled by US scientists and engineers. They've retired now, but most of the groundwork, still valid, was established by your compatriots.

--
OOO - Own Opinions Only. Try www.jmwa.demon.co.uk and www.isce.org.uk
2006 is YMMVI- Your mileage may vary immensely.

John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK

Replace that zener with a resistor.