You'll still find lots of these in old equipment. Sometimes you'll smell them too, and if one blows up while you're working in the panel the vile smell will make you leave the area fast.
This is a selenium rectifier. It was popular from around the 1930's until the much better silicon diode took over. It consists of a metal plate with a coating of doped selenium and a tin-cadmium electrode. When heated the tin-cadmium diffuses into the surface of the selenium and creates a semiconductor junction with a forward voltage of around 1V under load, and a PIV (Peak Inverse Voltage) of around 20V.
The low reverse voltage means that for higher voltages the diode plates have to be stacked, and each one adds another forward voltage drop to the stack. This makes the selenium rectifiers much less efficient than modern silicone diodes.
You'll find these in various industrial machines like elevator controllers, bowling pinsetters and various old (but rugged) factory machines. If you're lucky they will still be working, but if you're unlucky the rectifier will have failed with a pungent smell. If you're even more unlucky the rectifier will fail next to you with a loud electrical parp and a jet of extremely stinky smoke that will force you to leave the area.
It's claimed that the smoke is harmful to breathe, but to be honest you're not going to stay around anyway, as the smell is sometimes described as rotten eggs. I'd say it's that plus a sickly sweet smell too.
The modern replacement is the humble 35A block rectifier, which will fit most applications and is easy to mount. It has the advantage of having spade terminals so all you need to do is work out which wires are from the transformer, cut and crimp them, and put them on the AC terminals of the rectifier. Then trace the wires going to the smoothing capacitor and do the same, making sure the positive and negative go to the correct terminals of the rectifier. Changing the capacitor may be a good idea too.
Note that for higher voltage applications the new rectifier will have a lower voltage drop resulting in a higher DC voltage. Although unlikely to be an issue, it may push some older components over the edge. In some instances you may be able to compensate with a slightly lower voltage secondary transformer tap, or a slightly higher voltage primary tap which will result in lower secondary voltage.
If you enjoy these videos you can help support the channel with a dollar for coffee, cookies and random gadgets for disassembly at:-
http://www.bigclive.com/coffee.htm
This also keeps the channel independent of YouTube's advertising algorithms allowing it to be a bit more dangerous and naughty.
#ElectronicsCreators

Let's talk about a really interesting vintage component called the selenium rectifier md, who has worked in really old factory equipment or elevators. Uh will have come across stuff like this. Oh and entertainment equipment as well like bowling, pin setters they've come across these things that are next to transformer and they've got loads of square metal fins just stacked end to end. These are selenium rectifiers and they kind of predated silicon rectifiers.

But once the silicon ones came in, they replace these completely for reasons that will become obvious as we explore it. This one is a a guessing in-house component for british oxygen corporation. It's got a label on it that says international rectifier, uk and it's got a date code, i'm guessing 7104, 1971 um and uh component code at 19-027, and this is actually this stack here is actually two rectifiers mounted into one shaft. If i go down a bit closer, this uh shaft goes all the way through, but there is not normally a connection from that shaft, it's usually purely for mounting the rectifier, but i shall test that.

I shall put it on here and just scan along connections. No, nothing keep in mind. There are diodes, but uh. There is continuity from one end to the other, but not through the actual diodes themselves.

Okay, but if i then go on to say the negative here uh to one of the ac connections, you'll see the classic diode junction, it's a it's really close to a silicon diode. Where is it hold on? Let me just grab a silicon diode. I shall hold that thing, but below fingers, it's not really going to skew the results yeah very similar. If anything, it's got a slightly lower forward voltage drop off load than the silicon diode, but under load that changes the official voltage drop of one of these diodes.

In this bridge, rectifier is approximately one volt, so these uh plates each of these plates is a diode and they make them by getting a piece of metal, and then they put a coating of doped selenium on them. Basically, that selenium with a slight impurity just to adjust the characteristics to what they want, then they coat it with a layer of an alloy of metal and they heat them up and when it gets heated up to the kneeling point, you get a slight the two Layers, the metal surface actually fuses into the germ, the not the germanium, the selenium and it creates the semiconducting junction. There is a downside. Each of these plates is a diode, but it's only rated for around about 20 volts in normal use, with a maximum voltage of 25 volts, it's also very low current.

I would rate this rect for a few hundred milliamps. Here is what happens when you want a higher voltage diode. This isn't a bridge rectifier. This is actually a stack of diodes and series selenium diodes, one two, three, four: five: six that would make this roughly a 120 volt diode radio spheres and it says, type rectifier 3.

So it's a rectifier type. 3! That's how far back go back. It goes really old stuff and it's also it's not rs components. It is.
Radio spheres is printed on this made in england. I wonder if it was made at the same place. So let me show you the configuration of this. I shall bring in a notepad and then we'll talk about the stinky smell it makes when it fails, which is horrific.

Lyft engineers and factory mechanics will know exactly what i'm talking about. Here's the conventional bridge, rectifier layout, where all the diodes point from the negative towards a positive in the case of this one here uh this pin that's sort of aligned differently to the others. Uh, it's a positive, that's a negative, a acn acn and, basically speaking, if you imagine in here, there's a diode between each of these pins forming the square rectifier in the case of the slam rectifiers, they can't just basically well, it's not small. It needs support.

It's on this shaft, so what they do is they've got a threaded rod through them and it forms a sandwich of insulators and conductors that just mates all these plates together and the reason there's a black wire at the end that bridges one side. The other is because it's in a straight line, it loops onto itself again so ignore this diode here by the way uh. So here's the bridge rectifier with uh the two connections at the end loop together and that forms a circle, and that is the negative in this instance. So the two ac, which are these connections here, they're actually marked on here as two yellow dots, the ac, the negative is a blue dot and the positive is a red dot.

The reason there's another diode showing the end here is because, while i was looking at this, i thought why is this one good five plates? It's a bridge rectifier with just an extra diode in the end on the negative end, possibly to create a deliberate voltage. Uh difference across part, the circuitry it would have been probably tube-based circuitry. Whatever this was used in the reason it's black, they will have coated this. You can't shine light onto selenium surface.

It is a photosensitive surface, so they tend to be well dipped. The voltage is an issue because this one is rated uh. Each of these diodes is rated about 20 volts. That's it.

If you want the higher one, you have to go for the stack plates, and that is a problem because it means that, because one volt is being dropped across each plate under load, this thing would actually drop six volts for 100 volt rectifier. Compare that to the humble one n4007, which is so small, i'm struggling to pick it up. This is rated 1000 volts to create an equivalent diode to this. You'd need bigger plates and also you'd need 50 plates stacked together and the voltage drop across it would be 50 volts.

So that explains pretty much why the silicon diode print one up one over the selenium ones, and i can remember i'm not that old, but i can remember the marketing companies in the 80s 90s saying: oh, yes, uh. You definitely want to stick to selenium because it's got greater fault, current handling ability and in a way it does, but but their marketing ploy didn't work and it went the way of the uh. Well, it went the way of things that go go out of fashion. It went out of fashion now they're they're.
You can still get them on ebay, which is nice. That's where i got these ones now fill your mods over time. They will develop resistance and when they develop resistance, they develop heat as they get hotter. They break down.

They cause problems that, because they're developing that resistance, it doesn't really help it cause them to break down faster, and you end up the situation that they just fail and when they do omg, it's horrific. These things just basically shoot flames out in all directions, and this thick acrid stinking smoke. The smoke you'll find warnings, saying don't ever breathe the smoke. Oh, it's so dangerous, don't breathe that you'll get cancer off.

This terrible smoke comes off these. In reality, it's not as bad as they make out, but you won't want to hang around the vicinity anyway. The smokers coming off as fine particulate, selenium dioxide and it stinks well. The description comment description online is rotting eggs.

I wouldn't call it rotting eggs. There is a certain eggy component, but it's a nasty sickly, vile smell, so you don't have to worry too much about breathing too much of this because you wouldn't want to. As soon as one of these just goes. Bang i've had one go bang.

While i was working in the panel, i stuck a fuse and that's why the fuse had blew in it had gone, bang previously and stank the whole control room out. But as soon as that happens, the first thing i want to do is get out of the area, not because it's harmful, but because the smell is so bad. It really is awful. So when you go into old, say vintage elevator controllers that may use a selenium rectifier with a transformer for the brake 100 volt brake.

Perhaps so something like this, it would be stacks. It'd be like the full rectifier would be like one two, three four of these stacked together and you often find if that's gone wrong. People say: oh, what can i do to replace this? I don't even know what this is. The answer.

Is you replace it with this? This is more than amply rated. This is a typical uh. I think this one's rated 800 volts or something like that and it's rated 35 amps. It just blows the selenium rectifiers out of the water.

All you do in the panel is you've got the you'll find along that big, huge stack. You've got the four connections going on. Two will be from the transformer. If you look at these pins, it's the square bridge rectifier arrangement, you've got the all the pins going.

One direction, except for one, that's the positive: they always either put a chamfer or a shorter or longer leg. Uh longer legs just say: there's a better description of that or the marker. In this case it's positive. So the positive is this one: the negative will be dragging the opposite, and the two transformer connections will be ac here and then the diagonally opposite one to the ac, the other side.
In the case of the panel, you literally just drill a eighth inch hole. Three millimeter hole put a self-tapper in take the connections off the original, selenium rectifier put crimp terminals the spade receptacles on, and you just press them onto this and that's more or less it. But there is something you should know the voltage drop across. This is going to be one volt per diode.

If it is a high voltage application, you've got a big, huge, selenium rectifier. Everything will have been compensated for the fact that if it's got say this number of plates say the break rectifier. It's got. This number of plates it'll be dropping six volts in the case of say, 100 volt brake coil.

Six volts isn't going to be huge, but you should be aware that if you pop one of these in its place - and it was quite a high voltage - uh cylinder - rectifier - really big long one - then you may find it does pose a risk of slightly overvolting things. It's usually not that critical, it's going to be more critical in this of 1 000 voltage type. You know big stack of 50 diodes, but it can cause problems. Uh and some of those problems were problems that were just about to happen, and it was just on the edge and replacing with this rectifier will either make it perform better, a nice snappier action or it will make coils that were close to edge burn out.

It's just something you should know, but there we go now. When you go into a panel, you will recognize what these are. You'll know that it's not the end of the world. If one has let out the magic and disgusting smoke and you'll know that you can just basically in most instances pop in a new rectifier, but if there's the facility in the transformer for different voltage taps, it might be advantageous to lower it down, say, for instance, If it was a 220 volt input transformer and it was like 100 volt output, but it had other taps.

You might want to move it from 220 up to 240, because that'll lower the voltage from the uh, the secondary down. When you do that, and it just can nudge things in the right direction, if you're replacing this with this but interesting things fascinating things and if you've not smelled one feeling, then that's something to look forward to in the future.

15 thoughts on “Selenium rectifiers – the smelliest components ever”
  1. Avataaar/Circle Created with python_avatars Neil Bain says:

    I think I remember these from valve radios. Usually there was a rectifier valve, but the occasional radio had a selenium rectifier. My friend, who fixed these things, had a few choice words about selenium, few of them good.

  2. Avataaar/Circle Created with python_avatars Chas Large says:

    Ahhh this reminded me of the stinky EHT triplers in 70s and 80s TVs that failed, we had to send the old ones back so clipped them under the rear door of the Escort Estate on the way back to the workshop. Thanks for the memory 😜😜

  3. Avataaar/Circle Created with python_avatars Stewart Thompson says:

    I still have a bunch of Selenium Rectifiers around here somewhere. They were very common when I first got into Electronics in the late 60s early 70s. Generally tougher than the Solid State equivalents available at the time. People found that just changing them with Solid State could cause over-voltage problems as Clive explained. We quite often had to deliberately add some resistance to balance things out.

  4. Avataaar/Circle Created with python_avatars Hola! Leland Clayton says:

    Whenever I come across a selenium rectum-fryer as I call them I replace them. They do have an awful smell when they burn out. Had one in a Guitar Amp I was repairing and sure enough the garage started to smell like rotten eggs. Also keep in mind they attend to have the Anode and Kathode mixed up, I found that out when I was putting in a Diode to take place a Selenium in a old RCA VTVM.

  5. Avataaar/Circle Created with python_avatars richfiles π says:

    I love vintage electronics… I have over 100 vintage calculators, including an HP 9100A, a Friden EC-132, an SCM Cogito 240SR, A Sony Sobax ICC-600W, and a Wang 360SE. I have an old Sylvania Halovision TV, an pair of old electronic organs, a Conn and a Gulbransen (that uses hybrid modules), and much much more. In addition, i have various internals on display too… The relay module from an elevator, the multiple tube oscillators and amplifiers of another electronic organ, and one of the things I have on display are a pair of selenium rectifiers. They were coated in bright orange. Both are 6-fin high voltage types. The smaller one is a full bridge rectifier (total of 24 fins) with 1.25 inch (just over 3cm) square fins. The big one is a half wave rectifier (12 fins), but it has 2 inch (just over 5cm) fins! it's huge! I just love how they look, so they are one of the many odds and ends electronic bits that stay out on display at my place.

  6. Avataaar/Circle Created with python_avatars hattree says:

    Question, I've heard something about the fail open/closed being different between the selenium rectifier vs silicon. Isn't there a difference in the way you need to fuse it? I ask because I have a jukebox with one of these things in it that runs the selection circuit.

  7. Avataaar/Circle Created with python_avatars Ryan Davis says:

    These are also used in locomotives. They do stink when they die. These come in zeners as well. I had a large freewheeling rectifier short out when one of my coworkers started a locomotive up after he failed to properly terminate the auxiliary generator. I had told him how to be sure of the leads but he took the advice of a novice electrician. Basically the rectifier exploded above my head.
    Clearly I was not happy and had some very unkind words for him. Don’t hook up a DC auxiliary generator on a locomotive backwards or bad things happen. There is a process for testing if in doubt about its polarity.

  8. Avataaar/Circle Created with python_avatars Aarash Navabi says:

    In our product we use a bridge rectifier made of four diode‘s. The first diode always heats up and dies. We believe this occurs due to a thermal run away when the power keeps looping around the rectifier. Is it possible that extra diode is there to take the load off of the diodes’ that build the rectifier?

    Please consider this and let me know if that’s possible?
    If so adding a diode to my design might fix our issue.

  9. Avataaar/Circle Created with python_avatars Suzuran says:

    i have one of these in my 1948 tube radio. but i think it may have some long life left as the radio it self still has all of its oem made tubes. its a low ran table set as 0 scratches on the bakelite and face plates its pretty much new though plug wire is all nice and petrified from age and dinged up but likely from being moved a lot or jammed in stuff when stored or whatever causes it.

  10. Avataaar/Circle Created with python_avatars Jay Brooks says:

    Big, can you show how the magnetic rectifiers work? It was kind of like a relay. I’ve seen them on stuff in museums.. They were put on cheap car battery chargers in the 20’s. It was cheaper and more compact than a mercury bulb rectifiers.

  11. Avataaar/Circle Created with python_avatars Hola! Mariner785 says:

    I once walked into the Wheelhouse of a 1950's built ship and immediately went over to one of the chart table cupboards where the echo sounder workings were and sure enough there was a whisp of smoke and the running light was off. The Mate said the Secunny's (Indian helmsman) guts were a bit ripe that morning so I pointed out that the Selenium rectifiers, of which there were quite a few in the cupboard, had burnt out. Profuse apologies as they both thought it was the other !

  12. Avataaar/Circle Created with python_avatars John C says:

    Selenum is a pretty stinky thing.. hot burning selenium is not pleasant.. My dad ran a precious metal recovery firm and they processed most things including selenium. Everything he and my brother touched stank of selenium, a sweet garlicky smell.. their clothes and towels etc were contaminated. Even 22 years after he died there are certain things that still have a faint odour.

  13. Avataaar/Circle Created with python_avatars ⸸BadRomanceDjinn⸸ says:

    Crossing…Industries???…but there is an old joke that "when Selenium Chemists get on the bus or tram, everyone else gets off!" 😂😂😂

    And Tellurium is said to have a garlic odour with the Selenium scent, yeah, no garlicy farts for me thanks lol 😂 Every element down from Oxygen stinks to high heaven! Your choices are: Rotten Egg Sulphur, Farty Selenium or Garlicy-Farty Tellurium! 😂

  14. Avataaar/Circle Created with python_avatars Ruben Proost says:

    I wonder how they compare to tube rectifiers. I have an old tube radio here that uses a tube rectifier but the transformer shorted when someone replaced the 400mA fuse with a 35A car fuse. The replacement transformer was from a radio with a selenium rectifier and thus only had one secundary high voltage winding. Had to turn the tube rectifier in a bridge by adding 2 silicon diodes and the resulting b+ is about 10% higher than specified.

  15. Avataaar/Circle Created with python_avatars Ian William Tait says:

    About 30 years ago I worked for a transformer manufacturer in Sydney. We also made a few power supplies for the SRA. (State Rail Authority). One of the items we made was on a steel tophat section of 2mm steel. The components were, one almost full wound toroidal transformer. The mounting screw had to be plastic sleeved for insolation, 1 20watt adjustable w/w resistor, and a selenium bridge rectifier with an extra plate as a blocking diode. We had to use the selenium bridge because the original units built maybe 40 or 50 years earlier had the selenium bridge. The design had to be the same. the toroidal transformer is more efficient… FROM GOOGLE 1878 -1883 – The Ganz Company (Budapest, Hungary) uses induction coils in their lighting systems with AC incandescent systems. This is the first appearance and use of the toroidal shaped transformer.

Leave a Reply to Chas Large Cancel reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.