Transistor That Changed the World! The 2N1613 Planar Transistor - F-J's Physics - Video 245

Ajouté : 2026-05-11

[00:00:00]Hello and welcome back to the home lab and I've got a rather different kind of video for you today. What we're going to look at is a transistor from the early 1960s that started a revolution in microelectronics manufacture and production and this is the 2N1613 transistor a silicon planar transistor.

[00:00:25]>> [music] [music] >> So today we're going to have a look at this transistor. This is the 2N1613 and it was an absolute revolution in the design of semiconductor devices back in the day.

[00:00:52]What I'm going to do is I'm going to open it up and we're going to have a look inside and I'm going to compare it to an older design of transistor so you can see why it was so revolutionary for its time and stay tuned to the end of the video because at the end of the video I'm going to set you your own little transistor challenge which I think you'll enjoy.

[00:01:13]So before we start cutting transistors open and explaining why this one is so very different from the ones that went before it I just want to say a huge thank you to all of you for supporting my channel and watching the videos that I make and of course again thank you to PCBWay for sponsoring this video and helping encourage me make lots of videos. Of course if you're into electronics and they're clearly the place to go to not only have they got an absolutely fantastic website which you really must go and have a look at and there's a link below to that but of course if you're building your projects they do fantastic printed circuit boards and when you look at the website I'm sure you'll find something there that will help you with the next project that you're making.

[00:01:59]The 2N1613 or CV7440 simple and unassuming TO5 can hide the really interesting construction that lurks within.

[00:02:10]It's an NPN planar bipolar junction transistor manufactured from silicon by a planar process and it was developed by Fairchild semiconductors as a highly reliable component especially for use in the Minuteman missile with its nuclear warhead capability.

[00:02:26]This transistor was one of the very first to use planar processes in its manufacture and more of why that's so important in a moment.

[00:02:37]To make a comparison I brutally opened up a CV7349 or 2N1304 and whilst doing so it was interesting to note that a suspicious white powder came pouring out of it a smuggling method that I'd not yet come across but in all seriousness it's interesting to note that if this has happened to the silicon thermal transfer and protection paste which I assume was once a sort of liquid as it were or at least a sort of pasty material then one needs to be careful about using these old stock transistors. I must admit that that's not something I'd thought of before.

[00:03:14]Towers describes the CV7349 or 2N1304 as an NPN germanium RF medium current switch germanium alloy transistor which was fairly typical of the manufacturing processes at that time. It was developed by TI in the late 1950s for high speed computer switching applications and it had a PNP complement the 2N1305.

[00:03:41]So let's get the microscope out and have a look at this older 2N1304 transistor that I've opened up so you can see its construction and then we can compare it to the 2N1613 and see why that is so radically different and why it's light years ahead of this design.

[00:04:01]So looking under the microscope you can see this older 2N1304 transistor has these junctions with the wires attached to it so you can see right in the middle and I'm not sure I can even poke at this but I'm going to try.

[00:04:17]Okay over there is the collector with its lead attached to it and it's worth looking at the little blob that's on there as well because that's not very well controlled it's just positioned on it and then that's lying on the base and then if we turn it over which might be a bit tricky to do.

[00:04:40]Just going to bend the legs a bit.

[00:04:45]There we go.

[00:04:47]Okay it's quite difficult to hold these still and I'll just focus in on that.

[00:04:55]Hidden [snorts] inside there is the emitter junction so I'm going to try and get an even better angle on that.

[00:05:01]There we go and focus that up.

[00:05:07]And there's our emitter junction and the important thing about that is you'll see that the actual solder or the joint that's made is very uncontrolled it's a sort of blob that's just put on.

[00:05:22]So inside this transistor we've got its junctions which are very poorly protected and the positioning of the leading wires is not very controlled either so all sorts contamination can get in we might not be in quite the right place on the slice of germanium that's in there so you have to test loads and loads of these and you're going to get all sorts of different readings for the transistors. It's not really a very precise manufacture technique. So what we need is something that improves upon this greatly and that's where the 2N1613 comes into play.

[00:06:05]But now let's have a look at the subject of our video the 2N1613 or here the CV7440 opened up so that you can see the difference in how it was manufactured.

[00:06:17]Compared with the older germanium alloy transistor this is something totally different and a sea change in transistor junction manufacture.

[00:06:28]This transistor uses a planar diffusion process to construct its NPN silicon junctions. No evidence here of alloying or the use of beads of metal such as indium to form alloy junctions as in the OC71 if you've seen that video of mine.

[00:06:46]With this transistor a new process was used and one that would revolutionize circuit design in the following years.

[00:06:53]It all began with this single transistor simple design which led to highly integrated and complex groups of transistors being formed on a single silicon substrate and it therefore paved the way for the silicon chip we've all become accustomed to.

[00:07:12]Jean Hoerni patented the process in 1959 a bright guy as he gained two PhDs in physics something us humble scientists can only dream of.

[00:07:23]Whilst this was only one transistor on a tiny piece of silicon reliability was massively improved and the process he developed made mass production much easier.

[00:07:34]Very soon devices were made with more than one transistor on the silicon die particularly the flip-flop circuits developed by Jay Lathrop's team at Fairchild which were packaged in those multi-lead round TO style cans that you may have seen on older circuit boards.

[00:07:51]So let's now look at why the 2N1613 was such a groundbreaking transistor for its day and the technique that was used to manufacture it and when I do those sorts of things it's always interesting to go back and look at the original patents and here's Hoerni's patent from 1962 I think it was put in in 1959 if I remember correctly. So here we go yeah filed 1st of May 1959 I knew there was a reference somewhere.

[00:08:19]So what we'll do is look at the one page which is really good it shows the diagrams of the process that was developed. So you'll recognize this as a silicon wafer but you'll also notice there are lots of sort of little places on here little dots and that's where the transistors are going to be manufactured or diodes for that matter and that shows the construction of a diode and here the construction of what they sometimes refer to as a triode but in fact it's a double diffused transistor. And they're just peppered all over the place which is really quite interesting.

[00:08:57]So the first thing you deal with is silicon dioxide disc that you'll all be familiar with but the clever bit is you take a group five metal and you use that as a dopant to dope the whole of the disc so what you're left with is a layer that is an N-type semiconductor.

[00:09:18]Now once you've got that so there it is 32 your N-type semiconductor in cross-section you then put a layer of silicon dioxide on top and this process is a well-known process and you put put it in an oven just over 1000°C and you can bubble oxygen through water and pass it over it and it will develop an oxide layer. So all we've got here is our first part of the junction the N layer so we need a P layer and then another N layer.

[00:09:54]And this is where it gets clever.

[00:09:56]The next thing you do etch a little hole in this. Now, you can do that with hydrofluoric acid, and if any of you know anything about hydrofluoric acid, I wouldn't go anywhere near it. But, there's another technique which is very commonly used, which is a photoresist technique. So, you cover the whole thing with a paint that's sensitive to light, and then you put on top of that a mask that has a hole in it this sort of size, and you shine light on it, usually ultraviolet light. And where the ultraviolet light goes through the hole in the mask, the photoresist paint can be washed away.

[00:10:33]Then, what you can do is take the item you've got with its um little hole in it here, and you can put it back in the oven uh with a group three dopant um being very, very careful to make sure um that it's not gallium because uh gallium can actually pass into silicon oxide, and silicon oxide here is an insulator.

[00:10:56]So, it's really important to appreciate that this layer of silicon dioxide here is an insulator, but we've got a gap in it um that we've produced, and it's in the oven, and in is going to diffuse the P layer, which is the group three uh metal dopants. So, um if that was the whole story, you'd really have a diode junction now.

[00:11:18]But, what we want to do is create a transistor, so we've got to put another junction not on top of this, but we're going to diffuse it into this middle layer, which is the base.

[00:11:33]So, if you followed me so far, um the rest of it is fairly easy. So, we've now got our N layer, which is the the the sort of plate, the um silicon disc that we started off with. Uh we've then got our P layer, and what we're going to do is we're going to get silicon dioxide again all the way across that. So, we're going to completely seal that in.

[00:11:54]We're then going to put our uh etching process in place, or we're going to do our photoresist method, and we're going to open up a narrower window. Look at the difference in the window widths here. So, there's a bit of an overhang there labeled 38.

[00:12:09]Then, what we do is put it back in the oven, and these arrows represent heat being put into it, so you can consider it to be the oven process. And we're going to put it in there with a group five type metal that's a dopant, and that dopant is going to go into the P layer. So, it's going to diffuse into it. So, we diffused in our P layer here, that's the first stage. Then, we diffused our N layer in, so this is a double diffuse transistor. And what you'll notice is it's not got any thicker. And this central layer here is completely sealed inside, completely protected, and that's really important part of the design.

[00:12:53]So, we've now got our three layers, NPN, and we've also got our two junctions here.

[00:13:01]So, we've got an emitter, a base, and we can really control the size of these junctions to the size that we want. And then at the bottom, we've got our collector. And it's important to note the collector is often much bigger on transistors uh because electrons and the charge carriers uh don't go in a straight line. They sort of go all over the place in all directions. So, you make the collector quite large um to collect them.

[00:13:27]Okay. So, we've done that. All we've got to do now is put some connections on it so we can connect to the outside world.

[00:13:36]So, to the final bit of the process, uh we again cover the whole thing with an insulating layer of uh silicon dioxide, and then using photoresist um or etching methods, uh remove bits of it to make little windows. And you see the windows here? Uh we've got a little block of insulator there. So, we can get into the bottom layer, the collector. We can get into here, the base. And we can get in here to the emitter.

[00:14:04]So, the next stage is to electroplate the top of it um usually with aluminum, and um I did hear that that often causes another layer of silicon dioxide to form, so the whole thing is sort of sealed in.

[00:14:17]And that aluminum can be bonded to little gold or uh wires that interconnect, and then will go on to the leads um leading out of the transistor case. So, if you look very carefully, that um you've got a complete insulator here, and here, and here, and there's going to be an insulating layer all over there.

[00:14:37]So, the transistor is a number of things. It's very small.

[00:14:42]The junctions are controlled in size and made very small, which makes it very fast for switching applications.

[00:14:49]And um all the junctions are sort of covered up with insulator.

[00:14:55]Just the leads coming in are where the connections are. So, there can't be any short circuiting, and the leads can be positioned very, very accurately. So, what you're left with is something that was as thick as it was when it started, which is so different from earlier transistors. And now you'll see what you're looking at here, you're looking at the disc here that's the collector, and they've opened up windows on it to do um the uh impregnation or the um the P type um diffusion into the disc. And then you would cover it again with silicon dioxide and open up a smaller window, etc. So, I think this shows the larger window to begin with, then a smaller window, and smaller again. So, it's sort of concentric. So, when we look at the transistor under the microscope, you will see this sort of design, and you'll understand how it works. And I'm sure you've noticed it.

[00:15:52]The wires that connect to it all come out of the same side with this design.

[00:15:58]It's very different from the older type of transistors.

[00:16:02]So, now let's look at the 2N1613.

[00:16:06]And the first thing you'll notice is it is so much smaller than the older transistor. Secondly, you'll see that it's completely flat.

[00:16:15]Um you can see those large brown rings at the top of the picture. They're the insulators uh for the lead-in wires, and then you can see the tiny little wires coming in and going on to the pieces of at least the piece of doped uh silicon that's there.

[00:16:31]And they're positioned very precisely, so you can get a really good construction of transistor. And then the rest of it is covered with its oxidizing layer, so the junctions can't get damaged or contaminated.

[00:16:46]Uh so, this is an absolute sea change from the old-fashioned transistor design we saw earlier.

[00:16:55]So, zooming in a little bit further, you can really now see the sort of concentric design of this, the way they've used that single piece of doped N type silicon, and then diffused a P layer into it, covered that up, and then diffused another N layer on top. And so, um those of you not seen this before, um if I showed you this as a photograph, would you go, "Oh, that's obviously a transistor." It looks so radically different. But, of course, we could use this now on a larger piece of silicon to produce lots of flat planar transistors, and then begin to join them up. So, this is the birth of the microcircuit or the microchips that we all know today.

[00:17:41]Oh, by the way, zooming in a little bit further, I think you can see some of the sort of aluminizing of it where the connections were made, and I'm not totally certain, but I think what you can see is little crystals of aluminum forming. And there has been some mention of failure modes on these transistors, and it's something called the purple plague, which I won't go into here. Uh but, this is a problem um that happens between aluminum and gold layers, and you get some uh metallic problems there, which can cause damage to the transistor junctions. But, at least you can see there the window that's been opened up, and how the transistor's constructed.

[00:18:26]So, that's the groundbreaking 2N1613 transistor. And when you're building circuits in the future using modern microchips and modern transistors, just bear a little thought for this little fella that started the ball rolling back in the early 1960s.

[00:18:42]Now, if there could be an OBE for transistors for services to the electronic industry, I reckon this one would be a pretty good candidate.

[00:18:55]Just before we finish, it's worth saying that um I know quite a bit about uh early transistors, their history, and their manufacture, and what they were used for. But, it's only when you make a video like this that you realize there's an awful lot you still don't know. And in making this video, I did quite a lot of research, and I now feel much better informed than I was before. So, um there are lots of transistors out there with really interesting stories and histories behind them. So, my challenge to you, pick a transistor that you don't know a great deal about, but goes back in time maybe uh perhaps before you were designing circuits, or pick one that just looks interesting. Occasionally, I look online, and I see transistors, and I think, "Oh, that's an interesting shape and design and uh encapsulation, etc." Get a hold of one of those transistors, and do lots of research on it. Find out about it, its history, what it was used for, how it was manufactured, why it looks like it does. And I think you'll really enjoy the process. And when you've done that, let me know in the comments how you got on and what you found out. And if you think there's another electronic device that I should feature, then do please let me know that as well.

[00:20:08]So, thanks for joining me today and letting me share my interest in this early planar transistor, and I hope you enjoyed its story.

[00:20:17]Anyway, do stay to the end of the video because I do put some links in there to other videos that I've made, and I might include a few extra shots that didn't go in the main body of the video.

[00:20:27]So, your job now is to get to the end of the video and then go transistor hunting and see if you can do the same thing. Find an interesting transistor and explore all about its history and uses, and then do please let me know in the comments, and I'll be seeing you next time.

[00:20:51]>> [music] [music] [music] [music] [music]