Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

donate to eham
   Home   Help Search  
Pages: [1]   Go Down
  Print  
Author Topic: How to test high voltage rectifier diode "blocks" or"encapsulated modules"?  (Read 17192 times)
2E0ILY
Member

Posts: 207




Ignore
« on: September 20, 2012, 11:37:10 AM »

Henry (and others) often used an encapsulated HV diode module or block. Three spade type terminals per block, two blocks used in single phase linear amp supplies, three in 3 phase supplies. I have acquired a few used ones, but do not know how to test them other than bang them in a working circuit, which might be nasty if one is SC. A diode tester won't produce the voltage needed. has anyone got a safe means of doing a basic functionality test on these things out of circuit please? Thanks.
Logged

Best regards, Chris Wilson.
WB2WIK
Member

Posts: 21837




Ignore
« Reply #1 on: September 20, 2012, 11:49:30 AM »

A low voltage power supply and milliammeter can determine if all the junctions are intact and not shorted (or open, which would be rare).

The stacks usually use 600V to 1000V rectifier cells in series to build up to whatever voltage is their rating; a "10kV" stack is almost always 10 to 16 such cells.  In the SEMTECH modules, the diodes are always 1 kV each, so a 10kV stack would be ten in series per leg.

Forward bias the leg using an adjustable LV power supply and load it with a 100 Ohm, 2W resistor in series with a milliammeter.  Start cranking up the voltage until you see 100mA current flowing, and record what the voltage is.  That voltage is how much it took to "turn on" the series junctions.  If it reads 6-7V, for example, there's likely about 10 junctions in series.  The SEMTECH ones would read a bit higher, more like 8-10V for 10 junctions in series, because they use a different kind of junction that has slightly higher Vf than a typical double diffused junction.

Of course that doesn't confirm actual reverse characteristics at all.  But it can confirm if the junctions are intact and you can estimate how many junctions there are.

You need a HV tester (Hipot tester or similar) to actually measure reverse characteristics.

A VOM or DMV won't provide enough bias to turn on a string of devices like this, so as you intimated, that test is pretty worthless.

I'm extremely familiar with how SEMTECH made the modules, but others did things differently.
Logged
2E0ILY
Member

Posts: 207




Ignore
« Reply #2 on: September 20, 2012, 12:44:16 PM »

Thanks for that info. Needless to say I do not have a Hipot tester, so it seems my testing would be quite limited, but I'll give it a shot. As you say you are very familiar with the construction of some makes of thes, are the internals discrete "normal" components just soldered together in series, then potted? It seems quite a feat to get so many discrete diodes in the package size. Don't suppose there are any photos around of just what IS inside them, are there? Thanks again.
Logged

Best regards, Chris Wilson.
WB2WIK
Member

Posts: 21837




Ignore
« Reply #3 on: September 20, 2012, 01:47:40 PM »

I don't know of any photos on line, but I have X-rays of them which show how they're constructed.  Mine are all of Semtech assemblies, which were probably the best ones Henry ever used.

The diodes are discrete metal-oxide encapsulated (hermetically sealed) with solid silver leads and they are arranged side-by-side, spaced about 1/4" apart, arranged in opposing polarities so no two side-by-side are "aiming" the same way, in each "leg."  The leads come straight out (axially) from each diode (both sides) and are clipped off to about 1/4" long each.  A cross bar is welded across opposite pairs of leads to place them all in series.  This is a real silver to silver weld, no solder, and the pinch of the weld "rounds" the welded joint so there is no sharp corner, everything has a radius.  The "end leads" of the last diodes at the end of each "leg" are then welded to terminals which are silver plated copper (with a nickel flash between); those are the terminals you see exposed for connections.

The welded assemblies are placed in a form made specifically to accept the geometry of each assembly and then liquid thermally conductive epoxy (Emerson Cumming) is poured into the preforms to create the potted module you see; while still liquid, the assembly is placed in a vacuum chamber and evacuated, so all air bubbles are drawn out and expelled, leaving a monolithic, air-free potted module.  The epoxy is a good HV dielectric so the diodes can be much closer together than if they were in air.

The reason I know how these were built is I've watched it done. Wink  But again, other companies used different processes.  I know many of the "cheaper" modules just used die stacks that were  not hermetically sealed, used solder to bond them, and weren't vacuum potted.  That cuts cost quite a bit, but results in more failure mechanisms.
Logged
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #4 on: September 20, 2012, 02:05:30 PM »

Henry (and others) often used an encapsulated HV diode module or block. Three spade type terminals per block, two blocks used in single phase linear amp supplies, three in 3 phase supplies. I have acquired a few used ones, but do not know how to test them other than bang them in a working circuit, which might be nasty if one is SC. A diode tester won't produce the voltage needed. has anyone got a safe means of doing a basic functionality test on these things out of circuit please? Thanks.

They test perfectly fine as diodes by measuring forward voltage drop. Just use a variable DC supply with enough voltage to overcome forward drop, and be sure they are all the same for forward drop. If you have one with an open or shorted cell, it will have a different drop.

They are about 1000 volts PRV per cell usually, so a 10 kV will have about 7 volts drop at 50-500 mA. I use a 28 volt variable dc supply and a car stop lamp bulb for current limiting.

I have a high pot tester, so I can check breakdown, but I've never found one yet that does not show on forward drop compared to other known good ones.

Logged
2E0ILY
Member

Posts: 207




Ignore
« Reply #5 on: September 23, 2012, 10:11:28 AM »

Thanks for the replies, that worked perfectly and they are are 10kV ones with a circa 7 volts drop. All tested equal and good, so that's a result!
Logged

Best regards, Chris Wilson.
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #6 on: September 23, 2012, 10:22:30 AM »

Thanks for the replies, that worked perfectly and they are are 10kV ones with a circa 7 volts drop. All tested equal and good, so that's a result!

1kV cells with about .7 volts each means ten cells at 1 kV, or a ten kV rectifier.  :-)
Logged
KC2RLY
Member

Posts: 40




Ignore
« Reply #7 on: February 08, 2013, 03:56:02 AM »

mine explode when bad

http://imgur.com/apJXd6W
Logged
2E0ILY
Member

Posts: 207




Ignore
« Reply #8 on: February 08, 2013, 05:27:54 AM »

mine explode when bad

http://imgur.com/apJXd6W

That looks as if it might have been quite loud? Smiley
Logged

Best regards, Chris Wilson.
KC2RLY
Member

Posts: 40




Ignore
« Reply #9 on: February 08, 2013, 05:41:34 AM »

People always say it sounds like a gunshot. To me it really sounds like a black powder muzzle loader. I can detect a quick fizzle then crack-owwww.
The sound is different when its from to much voltage or current. The voltage gets the fizzle and crack. the ones that die from current just go pfft but loud.

Must be more careful spinning the variac....
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!