The AM Forum

way back before we had solid state - when there were only tube diodes - there were selenium recitifiers

does anyone know how to tell the values of a selenium recitifier?
was it the number of plates?

seen many, it did not seem to be the number of plates. They do have reverse leakage. A very low current test might show where it breaks down but I'd research that before trying it. The current depends on the size of the junction and the heat. Maybe some old catalogs would be best.

Best way to tell a good selenium rectifier is to put the thing in the trash and replace it with silicon diode.

Unless you're trying to restore something to original, replace it.

Fred

Boy do they stink when they fry!

Rotten eggs

If the manufacturer and rectifier numbers are known, the tables from the 1962 ARRL Handbook, attached, might help.

I replaced the selenium diodes in the modulation bias supply in my Heathkit Apache with 1KV silicon diodes. The DC output increased by about 10 volts, which is consistent with the literature I reviewed at the time.  Using Ohm's Law, series resistors can be added to reduce the voltage if necessary and to reduce surge current.

Here are some catalog and spec sheets

A lot of voltage drop through those also

Fred

One thing I found out the hard way years ago when I sometimes didn't check thing twice [or more] before applying power:  For HV units, above about 30 V if I remember, don't forget the series limiting resistor.  Causes all the bad things to happen as brought out by others, above.  Yeah, smell was bad.

I think the smell might be more than just annoying. From talking to some folks early in my career, one of my coworkers said it was quite hazardous. I remember hearing that several places, but he had practical experience.

He was on field trials with a 100 kw sonar mounted in a submarine that had selenium rectifiers. They were down a couple of hundred feet when the selenium stack blew and started venting noxious gases. It was mounted in the forward torpedo compartment, and before he could get out, the captain ordered it sealed to protect the rest of the ship. Fortunately he also ordered an emergency ascent which got them on the surface in a couple of long minutes according to Ed. As soon as they got up they vented the torpedo room to get the fumes out and he was ok. He said it was quite a ride to the surface, including coming partway out of the water and smacking down like that clip of a sub making an emergency ascent that I've seen several times.

I know from my experience with bad rectifiers that it smelled pretty bad. Fortunately that's many moons ago.

 

 

Yes they do.  First hand experience

We've had lots of threads on selenium rectifiers.  Here's one:

http://amfone.net/Amforum/index.php?topic=26755.0

I remember reading somewhere that the fumes from some of these early rectifiers is deadly. Perhaps it was the copper oxide types? Anyone else ever hear of this?

back to my original interest in the Selenium Rectifier

it sounds like there is about specific voltage required per "leaf" - and that the larger the "leaf" the more current can be passed.

but you have really answered the intent behind it all

It was the ("No-No") of what can withstand EMP - so I do not really want to use Selenium Rectifiers

Does anyone know if Silicon Diodes can withstand EMP?  They aren't quite the fragility of transistors or ICs

Like I've said - if I'm going to work toward Emergency Preparedness, I dare not ignor a known "worst" form of disaster.

http://www.atsdr.cdc.gov/toxprofiles/tp92-c3.pdf

The CDC reference on Selenium hazards mentions immediate lung damage from acids, long term congestion, emphysema; spleen damage; liver damage, and not to mention general toxicity.

Burning Selenium rectifiers should be avoided by all means.

If memory serves, selenium sections with the same dimensions can have different voltage & current ratings.  Best to check the manufacturer specs - if you can find them. 

Selenium ages through the years, which alters its electrical properties.  This was known long before silicon diodes became available.  After many decades, the likelihood of failure could be high.  That's why I decided to replace the selenium diodes in my TX-1.

Timely topic! I just recapped a S-119 Sky Buddy II. The selenium rectifier makes a B+ of about 130V. A silicon diode makes 150V. This is with a modest 3 miniature tube load including a 6CM8 audio output. I put the silicon diode in series with the selenium as a safety measure and to use the apparently good selenium rectifier voltage drop to keep the B+ where it probably is supposed to be. The selenium device looks as new. No need to worry about it.

What's the point of putting a silicon diode in series with the selenium rectifier??  Why not remove the selenium and put a series resistor to drop the voltage.

Fred

Hardening electronics to withstand EMP is all a matter of shielding, silicon diodes have been used in military hardware for over forty years now, not just radios and the like but every portable generator in use is in fact an AC alternator that depends on silicon diodes for powering the field. Don’t think the old selenium rectifier junk has been used past the early sixties with the advent of silicon diodes and any of the stock you find today will be so old as not to be worth depending on. If you have any concerns about EMP don’t waste your time on vacuum tubes and primitive technology think more along the lines of a big magnetically shielded box.  Don’t think it matters today but a several years back was involved with a project that would have involved placing shielded equipment shelters at broadcast sites that had a self-contained generator, AM broadcast transmitter and satellite receiver that in the event of massive EMP type event you would go out to afterward and connect to the stations antenna system and be able to broadcast with.  Would imagine that somewhere there must have been plans for huge shielded warehouses full of radios to distribute for people to listen with? I was only involved with the origination side.

The spec sheets I appended above list the equivalent series resistance for each selenium diode type. 

A ballpark range of 5-25 Ohms is typical.  It's a small matter to add a resistor with a  replacement silicon diode.

Leaving the selenium in means no resistor required. Cost-down & Time saved. Don't care about EMP or other stuff on that.