>Rich says:
>
>> The capacitance of a P/N junction is not fixed, but decreases considerably
>>as reverse potential increases.
>
>But most of this change is at low voltages, depending on the gamma of the
>diode. Up at these piv's, there's very little change in capacity for several
>hundred volts change. The change would make the volts increase across the
>diode, too.
? Not when the avalance V is reached.
>
>> The division of the reverse potential is controlled by the avalanche
>>voltages of all the diodes in a series string
>
>It could be - or it could be the capacity, depending.
>
>A bit of pontificating now. I've been in the semiconductor industry for over
>20 years now, and the first thing is - DON'T TRUST THEM! Where else is there
>an industry that says 'Yes, give me $250K, and I'll do a design for you. If
>it doesn't work, I'll have another go for another $150K.'
>
>If the data sheet doesn't say 'controlled avalanche device', don't assume
>that it is. Might be this week, may not be next. Might be from this
>manufacturer, and not from another.
>
? So, put the device on a hv breakdown tester and find out what's what.
. I have been doing this for over twenty-five years and I can tell you
that they don't make 'em the same way they used to.
>Any parameter not specified on a data sheet can't be relied on for
>repeatability from device to device (and even some that are!)
>
? so test it
>Why do you think the military have JEDEC data sheets for devices? they've
>learned!
>
>Diodes of this class will be tested just about as fast as a handler can go -
>probably 50 to 100 per second, since you need only to test forward drop and
>piv. For a 1000 volt piv diode, you'll test with 1kV reverse; is the current
>less than x microamps? if yes, bin 1, if not bin 2.
>
? 'Tis better to test with more volts and limit the current to 2 - 3 uA.
>You might go 'is current more than y microamps - if yes bin 3' to sort out
>the short circuit ones; bin 2 can then get tested for 800 or 600 or whatever
>volt diodes.
>
>But there won't be sophisticated testing. If the diodes come on a bandolier,
>there's a good chance that they'll be from the same diffusion batch, maybe
>even the same wafer, but that's not guaranteed.
>
? So buy 'em from Rich. He gets 1N5408 on a reel of 1200 units. The
price is currently 20 cents. We will measure and mark the avalanche
voltage for 3 cents per diode.
- cheers, Peter
Rich...
R. L. Measures, 805-386-3734, AG6K, www.vcnet.com/measures
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|