I need an explanation of avalanche ratings on diodes...plz / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

I need an explanation of avalanche ratings on diodes...plz

1 2 next

Move Thread

LAN_403

Patrick

Tue Mar 09 2010, 01:54AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

i want to have series strings of diodes to rectify 20kv square wave AC. i have many different diodes from the 1n4007's (have about 3000 of the buggers) to way more exotic ones. but i heard about diodes that use avalanche condidtions to assure even voltage divison along the whole string, what does this mean and what types of diodes are good for this in the 1000-1800 PIV range with ultra fast recovery ? I usually buy from Mouser or digikey. I connot get a satisfying expalnation from other sources. help is apreciated. i presume 500ns or less of trr, is good.

-Patrick

radiotech

Tue Mar 09 2010, 04:10PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

Avalanche multiplication in SCR's is why they can be latched with a a high voltage pulse without gating. Avalanche diodes are multilayer diodes like diacs which depend on non-destructive punchthrough of a blocking layer. For strings of common diodes in series one thing I have seen on sticks rated for output up to 100kV, the only balancing seemed to derive fron the structure of the mounting which was PI's spaced along an insulating rod and immersed in the transformer tank. Some high dielectric goo could act like balancing capacitors. I wonder what the constant for DMPSO is. 1E5 centistoke used to be available cheaply.

For 4007's series strings consider non soldering connections to avoid heating leads too close to body. Since they are cheap, try them to see if they will work,

ref. Sze S.M. PHYSICS OF SEMICONDUCTOR DEVICES 2nd ed. (1981)-- Bell Labs Murray Hill crowd stuff is pretty good.

MinorityCarrier

Tue Mar 09 2010, 11:14PM

Registered Member #2123 Joined: Sat May 16 2009, 03:10AM
Location: Bend, Oregon
Posts: 312

I believe the avalanche balance you refer to is typical for a string of equivalent-die diodes that is reverse-biased with a voltage that does not exceed the sum of the PIV for all the diodes in the string.

The fastest Trr high voltage diodes I know of are SiC Schottky Barrier rectifiers. Kinda pricey right now, but defect reduction improvements to the SiC substrate, and increased demand in the solar power inverter market may drive down the cost.

Patrick

Wed Mar 10 2010, 01:12AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

yeah....mioritycarrier .... dont some diodes not need biasing resistors or balancing caps for voltage division?

i am trying not to have resistors for voltage dividers, yet not have a impulse break down my diode string!?

im going to the Meriam Library on Chico campus, maybe i can find a book, about diodes.

-Patrick

MinorityCarrier

Wed Mar 10 2010, 05:53AM

Registered Member #2123 Joined: Sat May 16 2009, 03:10AM
Location: Bend, Oregon
Posts: 312

And reversed-biased diodes behave differently from capacitors in what way?

Spent many hours in that library. When I was there, the top floor was unfinished, ran out of funding, and the internet was still the arpanet.

ScotchTapeLord

Wed Mar 10 2010, 06:16AM

Registered Member #1875 Joined: Sun Dec 21 2008, 06:36PM
Location:
Posts: 635

Read the part about the high voltage diode!
Avalanching as I understand means it can withstand a very short overvoltage.
Slight differences in recovery times in diodes will cause uneven distributions of voltage across the string, but avalanche-rated diodes can withstand such short times.

You may want to go with the ones Steve, the man himself, used just to be safe, using similar parameters as his.

Hope I saved you some time and eye strain at the library! :)

Dr. Slack

Wed Mar 10 2010, 07:56AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

More to the point, an avalanche-rated diode will conduct non-destructivly in reverse breakdown as long as the total energy and current is kept below a threshold. The other diodes in the string limit the reverse current to their recovery current, and the total energy by having a finite stored charge. When an ordinary diode starts to conduct at reverse breakdown, it tends to self destruct.

Patrick

Thu Mar 11 2010, 03:39AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Yay! ScocthTapeLord!, Stevehv's (steve ward is the name i believe) site was where i read a blurb about this kind of diode operation.

I Quote Steve ward:

"HV DIODE CONSTRUCTION:

The HV diodes are actually constructed from strings of small diodes wired in series. The particular diode I used (purchased from Newark) is the SF1600. It is a 1600V, 1A, 75nS Trr, avalanche rated diode. I went for 1600V as that was the highest available for the "cheap" diodes. The 75nS Trr is important for running at these high frequencies, so that recovery losses are not significant. The avalanche rating is especially important, and means that I don't have to worry about balancing the diodes. I series 40 of these diodes per each string, giving a 64kV diode. The diodes should only see about 40kv under normal conditions. I soldered the diodes on "perf-board" and then potted them with paraffin wax inside of some 1/2" PVC tube. "

i geuss i will look up SF1600 datasheets.

Patrick

Thu Mar 11 2010, 04:05AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

I hope this doesnt count as a double post, but i cant upload it any other way.

here is the datasheet from mouser.
]sf1600_diodes.pdf[/file]

MinorityCarrier

Fri Mar 12 2010, 06:19AM

Registered Member #2123 Joined: Sat May 16 2009, 03:10AM
Location: Bend, Oregon
Posts: 312

The "Avalanche Rating" you are referring pertains to device ruggedness. The diode is manufactured such that uniform reverse breakdown occurs across the entire die area, rather than 'hotspotting', and the diode is rated to withstand a much higher one-time pulse of reverse-bias energy than a cheapo diode like the 1N4007. Mosfets are avalance rated in much the same way, for the same reason. Where I work we subject our mosfets to an Unclamped Inductive Switching (UIS) tests to verify their avalanche rating.

The 'balancing' you refer to takes place with any similar-die diode. I and others have strung many 1N4007 diodes together for HV multpliers without worrying about 'balancing'. The HV diodes of the '60's were spring-loaded stacks of selenium disc diodes without any 'balancing'. The 'balancing' problem would be more of a concern when mixing different size die types, different current ratings etc., something one shouldn't do with strings of diodes.

That said, avalanche-rated (rugged) diodes are a more robust part, your design will be much more fault-tolerant.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob