According to Pete - July 5th, 2011

Check out the latest installment of According to Pete.

Favorited Favorite 0

Well, we are back at it after a long weekend of BBQs and fireworks. And there's no better way to jump into a new week than with the latest installment of "According to Pete." In this edition, Pete delves into the world of diodes and transistors. Check it out!

We hope this helps shed some light on a sometimes confusing topic! As always, feel free to leave questions in the comments and we'll do our best to answer them. Hope everyone had a fun, safe holiday (or regular) weekend! Cheers!


Comments 66 comments

  • The silicon is doped with Arsenic and Boron.

    • Blacklab1 / about 13 years ago * / 1

      So the Silicon doping list is:
      N-type: Arsenic or Phosphorus.
      P-type: Boron or Gallium.
      If I understand things right there are others, but are not commonly used.
      I cant wait for O-Pete to explain the new Intel 22nm Tri-Gate Transistor (Nuts&Volts July 2011 Pg 10). I guess this new transistor does not leak as much.

    • SomeGuy123 / about 13 years ago / 1

      Tasty.

    • Ah... much obliged, thanks. I knew that once.

  • No-one / about 13 years ago / 2

    Yay! Op-amps & mosfets next month! =) Can't wait!

    • Well......
      Maybe mosfets, but I really want to start digging into the project a bit. Op amps the time after that. I swear!

  • Larkmore / about 13 years ago / 2

    I think Pete switched the convention for direction of current flow midway through. For the diode he had the current flowing from positive voltage to negative (the way I'm used to) and for transistor he had it flowing from negative to positive (the "real" flow). Either can be correct so long as you're consistent...

    • MattQ / about 13 years ago / 1

      Busted! But since he was talking about holes and electron flow at the same time, I'll give him a pass on this one. But we'll be watching. Bum bum bummmmm... Side note: glad someone is taking the time to explain op amps. I've grown tired of reminding coworkers over an over again how op amps work. I think people just mentally freeze up when they see op amps. Now I'll be able to just send them a link. Great work!

    • SomeGuy123 / about 13 years ago / 1

      Arguments like this annoy me almost as much as the argument concerning the naming of bipolar junction transistors due to the fact that they aren't actually bipolar.
      No-one is right. No-one is wrong.

      • wbeaty / about 13 years ago / 2

        Wow, I never met anyone else who realized this. All diodes are PN diodes, right? :) Shottky NnN catswhisker transistors!
        But if the magic isn't all in the PN junction ...we're lost, forever lost.

    • Ach... I really gotta watch that. I switch between in my head fairly regularly, and I generally think in electron flow (learned that first).

    • JWRM22 / about 13 years ago / 1

      For the usual current flow in a transistor, its easy to remember that the current follows the arrow in the schematic. The arrow on a PNP is backwards.
      Now you all know the basics... Now make an transistor amplifier. I've had a 30 hour class about it, and i still don't fully understand.

      • Brodie / about 13 years ago / 1

        Making basic low power transistor amplifiers can be quite easy once everything "clicks" and you get the feel for things (With BJT's anyway, MOSFETS are a whole other world). When you want to go to 2 transistors or matched stereo etc thats when things start to become painfully complicated.

      • For the usual current flow in a transistor, its easy to remember that the current follows the arrow in the schematic. The arrow on a PNP is backwards.
        By "usual" I think you mean conventional. The convention has been to define current by the "flow of positive charges", which don't exactly flow.
        The PNP schematic arrow isn't backwards, it still points in the direction of conventional current flow, it's on the Emitter side pointing inwards.
        ADDITIONAL: BJT's point in the direction of conventional current flow, but MOSFET's point in the direction of electron flow.
        And you're right, transistor amplifiers are challenging, especially considering that there are hundreds, if not thousands, of different topologies for amplifiers. From BJT, MOSFET, Linear, Switching, Class A, B, A/B, C, D, E, yadda yadda yadda. There are textbooks upon textbooks on the details of amplifiers. Not to scare off any n00bs, there are fairly easy amplifier circuits depending on what you want to do.

        • JWRM22 / about 13 years ago / 1

          Isn't that just a matter of opinion? Inwards, outwards or backwards. If you turn the arrow on the emitter 180 deg the picture would be a PNP, yes the arrow is pointing inwards. (The arrow still represents the emitter, and ofc the conventional current follows this arrow too.)
          For most people its confusing to talk about different current flows. I just use the "conventional flow", positive to negative. But do what ever feels right to you.
          There are indeed many types of transistor amps. I was just referring to an amp with one transistor and 4-8 resistors and some caps... I haven't found a use for them yet, but it was fun to experiment with it.
          example: http://electroschematics.com/5294/transistor-amplifier-design/
          Thanks for posting the "proper English translation". Its not always going smooth. Its challenging to find the right words, and specialty when it gets technical.

          • I hope I don't come off rude, I just think it is important to be clear, especially when things can be confusing.

            Isn't that just a matter of opinion? Inwards, outwards or backwards.
            Saying that the arrow is backwards can sound like it's in the wrong direction. Inwards and outwards are pretty absolute. No matter how a BJT is rotated or flipped on a schematic, n-type always points out and p-type always points in.
            One of my favorite things about this site is that there are readers and posters from many different backgrounds, both in knowledge and language. I appreciate the challenges faced by both groups, so I try to add clarity when I think people might get confused. If I have offended anyone with my comments, I apologize.

        • FWIW, likely nothing, the arrow points to the N material. or, at least, that's how I've always thought about it.

          • bTimmeh / about 13 years ago / 1

            I always learned it this way.
            PNP is
            Pointing
            iN
            Please
            NPN is
            Not
            Pointing
            iN
            and the arrow is always on the emitter :P saves me time when looking through other peoples schematics. thanks for the insightful video! i learned a little more about transistors.

          • FWIW, likely nothing, the arrow points to the N material. or, at least, that's how I've always thought about it.
            Ummm... the arrow points away from the transistor in the N-type BJT and the P-type MOSFET, so I don't really understand your comment.
            Besides, you can always just look it up if you need.
            Edit: Okay, so I get what you mean about the arrow pointing to the N-type, but for BJTs both the Collector and Emitter are N-type. I have been unable to find any explanation for why the symbols are the way the are. Anyone?

            • Member #195064 / about 13 years ago / 1

              That's not true, an NPN has an n-type emitter and collector but a PNP BJT has a p-type emitter and collector, hence P-N-P.

              • Absolutely, you are right. I was referring to the same n-type BJT that I mentioned earlier, but I should have been clearer. Thank you!

  • customer207 / about 13 years ago / 1

    Hey, looks like we are fellow NEI grads!
    Any fond memories?

  • dkhawk / about 13 years ago / 1

    Thanks for these videos! Looking forward to more in the future!

  • cjh08052 / about 13 years ago / 1

    Pete;
    Great video! Could you please post a reference for some of the math you were doing? Personally, I tend to forget (or don't know when) to take into account the voltage drops when trying to calculate the value of a limiting resistor. In the example, in the video, you had to take into account the value (or the side effect) of the limiting resistor at the base to determine the value of the resistor for the LED.
    BTW, Thanks for being so clear with these videos; it helps a lot!
    ;-> (cjh)

    • paralab / about 13 years ago * / 1

      Hey cjh,
      If/when Pete gets around to this i'm sure he'll have something better but I found EDIT: scratch that useful.
      EDIT: Looked twice and noticed that the equations there don't go for beta independence so its a no-no.

  • mark1968 / about 13 years ago / 1

    Pete, thanks for that. I am familiar with the concepts of approximations in Diodes and many other devices such as the First Approximation (AKA Ideal Approximation) where we ignore the barrier potential of the device, the Second Approximation which includes the barrier potential, and the Third Approximation concept which includes both the barrier potential and the bulk resistance defined as the resistance of the N-Region material PLUS the resistance of the P-Region material as an idealistic concept at best which is derived from the instructional concept that the P and N materials are first created and then magically combined at an abrupt junction when in actuality they are actually created within the same crystalline substance during the manufacturing process and thus, as a result, forms a gradual change from one type material to the other. Sorry if I am a bit more technical than most, just my nature and a result of my preferred learning style. I was just simply wondering where the 0.6V concept came from rather than the usual concepts covered in textbooks that state to use 0.3V for Germanium and 0.7V for Silicon that generally are used. I do have some ideas for making a video or series of videos but I am still developing some ideas to include along those ideas. Are you offering me a job or are there any available jobs there for me at Sparkfun? :)

  • mark1968 / about 13 years ago / 1

    Sorry, my last sentence of my last post may not have made sense. It was supposed to say: I was wondering if you had any other reasons for why using a Zener diode as a stand alone voltage regulating device in a DC power supply is a bad design choice, if so, please share.

    • MattQ / about 13 years ago / 1

      Also should note that Zener's have poor ripple rejection. If you're looking for a well regulated supply, zener can't provide it. Also, if you're looking at a circuit that will have a widely varying load current, you'll need a really beefy zener, both inefficient at low load current as the power absorbed by the zener goes up as the load goes down, (Pzener = Przener - Pout). Not a big deal if the zener is sourcing an amplifier (like in the video), but if you want to run a super bright LED or motor, you're looking at trouble. Also, can't adjust a zener, so it's much more convenient and precise to use a linear regulator.

    • ... wanna make a video? Good info, and all true, but probably more than most people need to get their hands dirty - though I'm sure it's appreciated, none-the-less, so much thanks. But to answer your questions...
      Diode drop approximations are just that: approximations. If I know I'm using a germanium diode, I'll use 0.3V. If silicon, 0.6V. If I'm feeling particularly saucy, I may just use 0.7V. But I've never been bit by being a tenth of a volt off in an approximation, because I already know that the circuit it going to be slightly off from my approximation. I tend to ballpark a circuit quickly, then prototype and compare to what I think it's going to do rather than spend a lot of time calculating initially. YMMV.
      Other reasons for not using a zener as a regulator? Not really. They're just not the best choice given what else is readily available. Like I said in an earlier response, if you're in a parts crunch (and you're not eating your parts - Paradoxial, I'm looking at YOU)...

      • ...(ran out of space?) go for it. If you've got a choice and need better regulation... well, there you are.

  • mark1968 / about 13 years ago / 1

    Pete, FYI (as a review of what you already seem to know) and for anyone else who wants to know:
    N-Type material is created by doping the molten silicon with Pentavalent atoms (Atoms with 5 valence shell electrons). Pentavalent atoms include: Arsenic, Antimony, and Phosphorous.
    P-Type material is created by doping the molten silicon with Trivalent atoms (Atoms with 3 valence shell electrons). Trivalent atoms include: Aluminum, Boron, and Gallium.
    I wanted to share this as you stated in the video that you had forgotten the elements for doping the material to create the N-Type and P-Type materials.
    By the way I noticed that you stated that the forward voltage drop across the diode (AKA Barrier Potential) and the base-emitter Diode or PN junction in your discussion is about 0.6V but I always learned it as approximately 0.3V for Germanium and approximately 0.7V for Silicon. Are you using another approximation or a roundabout figure for what actually appears across most components? I do know that when measured the measured voltage drops do range in value between 0.6V and 0.7V.
    Also, I noticed that you stated that it was not a good design to use a Zener diode as a device for regulating a power supply. I do know that the Zener does cause the circuit to use more power than perhaps the Series Pass Transistor regulator that appears in most Semiconductor textbooks (but is never given much discussion on its design) or even most dedicated IC voltage regulators that are readily available. I believe also that while Zener diodes provide good regulation, with only minor fluctuations (which are within a few tenths of a volt) during their operation in circuit and that the Zener, when used as a voltage regulator in a power supply, may also generate excessive noise in the circuit. However, they are very good for instructional purposes with beginners who are learning about simple ways of how to create simple regulated DC power supplies, and I am sure that you would agree with that. I was just wondering if you had any other reasons for why using the Zener diode as a stand alone voltage regulation device in power supplies, if so, please share.

  • godzillatron / about 13 years ago / 1

    great video, helped me a lot.
    Where I can find more of those videos?

  • MoriFi / about 13 years ago / 1

    Now mosfets, mosfets are easy.

  • MoriFi / about 13 years ago / 1

    This still confuses me.

  • qzjake / about 13 years ago / 1

    an example of the circuits in action always makes me feel more comfortable.
    Thank you for the presentation, favorite so far. You have a talent for making these things simple.

    • SomeGuy123 / about 13 years ago / 1

      He is a great teacher, but he is a bit under equipped. That whiteboard needs a renovation.

  • Blacklab1 / about 13 years ago * / 1

    Pete- Could you please get a new white board. It's real hard to read your numbers. It acts like you have put one too many Post-its (tape ect) on your white board and it's not letting the ink stick to the board. It's kind of distracting not being able to see what you write. Thanks.

    • Noted. I thought I had it licked this time, but apparently not. I'll at least have a bigger dry erase marker by next time.

      • Member #105131 / about 13 years ago / 2

        I know that it may seem tempting, but licking the Dry Erase board clean pretty much always produces this result. At best.

  • SomeGuy123 / about 13 years ago / 1

    I made a small reference/cheat sheet to discern between PNP and NPN transistor symbols. You can find it here: http://flic.kr/p/9ZZhEE.

  • SpeakEZ / about 13 years ago / 1

    I really hate to date myself, but back when electronic devices (calculators, clocks, copiers, etc.) first appeared they all used zener diodes driving the base of a transistor (or transistor pair) to regulate the voltage. This is especially true before SSI chips even existed.

  • elproducts / about 13 years ago / 1

    I liked your presentation. Well done. I do have to add one comment though about Zeners. Zener's can be effective regulators. I agree that voltage regulator parts are preferred in many designs but I've seen some very clever low cost/very small designs that used a zener as a regulator. Early automotive electronic modules used zeners so it can be done successfully.
    Check out Microchip's Compiled Tips and Tricks Book for a few zener examples. (http://ww1.microchip.com/downloads/en/DeviceDoc/01146B.pdf) Forrest Mimms books can also give you a few good ideas. And isn't a voltage reference just a voltage regulator?

    • Oh... potato, tomato...
      I'm not saying they can't be used as a regulating device, there are just better choices. If you're in a parts crunch, there are a bunch of different ways to achieve regulation. But if I'm designing a product, I'm going to make a better choice.

  • Ekisu / about 13 years ago / 1

    could have the base and collector resistors been replaced by a single emitter resistor?

    • What that effectively does is add negative feedback. The more you try to turn it on, the more voltage is dropped on the emitter resistor. That, in turn, reduces the voltage on the base-emitter junction, turning the transistor off. If you were on a desert island and only had one resistor to work with, go for it. Beyond that, I'd avoid it.

      • SomeGuy123 / about 13 years ago / 2

        If I was stuck on a desert island with only one resistor, I'd probably end up eating it.

      • MattQ / about 13 years ago / 1

        Ahh, you're missing one important exception: current source. One of the simplest ways to make a fixed current source is to put a fixed resistance at the emitter of a transistor, a zener on the gate, and the load at the collector. Current sources are certainly far less popular than voltage sources, but it's still a very useful schematic to remember.

      • Blacklab1 / about 13 years ago * / 1

        Pete- if someone was foolish to try to use just the emiter resistor as a way to control the current through the LED- wouldn't you end up with a over heating problem with the transistor? Isn't that why you normaly use a 2n2222 NPN switching transistors fully on or fully off?

        • Overheating? Not necessarily. I don't mean to say that emitter resistors aren't used at all, just not for a switching application. When run in linear mode, like in an amplifier, a resistor between the emitter and ground can be quite helpful it setting up your DC bias. Something called a "universal bias"...
          ...but that's another story.

  • D_rock / about 13 years ago / 1

    Wow very relevant. I just designed a board for the sparkfun 4 x 7segment displays and used pnp's for the multiplexing.

  • Nomad / about 13 years ago / 1

    First off Pete, Thank you for taking your time to do these videos. This is exactly what I need to kick start me on getting educated on this stuff.
    You mentioned that Beta on transistors vary per unit, so is there a good way to verify Beta on the specific piece you're using, do you just saturate the Base-Emitter connection and then use your multi to see what value you get across Collector-Emitter? Or is it not important since we should all be designing our transistors to be Beta independent?

    • Always shoot for beta independence and don't bother calculating, unless you enjoy the exercise or you're trying to be clever with a one-off. Check the data sheet for the minimum value give your bias conditions. That's the number that you're (mostly) guaranteed.

  • Squintz / about 13 years ago / 1

    Wow!, When I heard you were doing transistors I thought "This guys could never explain a transistor in 20 minutes". I was wrong. I think you did very well at covering the basics. I was getting a little dizzy from all the camera angle changes. But the lesson seemed perfect for the audience.

  • cgapeart / about 13 years ago / 1

    I have taken many electronics classes both at the highschool and in a university for an engineering degree. That has to be one of the most straight forward and useful discussions that outlines component selection, and what you actually need from the datasheet. Thanks.
    p.s. To remember BJT transistors on a schematic: NPN stands for Not Pointing iN.

  • hadesto / about 13 years ago / 1

    purely awesome sir. I def. have to say that if THIS video had been available when I was taking electronics 001 in college I would have been one VERY smiley camper (dunno about the happy part :p ). Hope this really keeps going. REALLY liking the series!!

  • sephers / about 13 years ago / 1

    how long's that cartoon been on your board :D

  • Aha! 1st comment. Sorry...

    • tetsujin / about 13 years ago / 3

      If you need to apologize for your comment before you even hit "submit", maybe you shouldn't post it at all...?

    • sephers / about 13 years ago / 1

      ....really?

      • 1st comment, yes, but SparkFun sorts highest rated comments to the top, so it pretty much defeats the whole "First!" posting.

        • Yea, LOL, but because I live in the UK, by the time I get home i have usually missed the first batch of comments, so mine is usualy 10th. Anyway, it was still first by time: 10:08

Related Posts

Recent Posts

Tags


All Tags