I've been making DIY synths since I was a teenager--not necessarily GREAT synths, but making them nonetheless. I never took an electronics course so everything was sort of self-directed learning, and pretty lackluster at that. Never in the past 10 years have I found such a helpful series of tutorials, and had I had this available to me when I was 18, I can only imagine what I'd have made! You should be very proud of yourself because very rarely does someone with a good understanding of a concept actually know how to explain it in a digestible way. Every single segment of these videos absolutely blows my mind and has re-kindled my drive and inspiration to start building again. Once this bullshit pandemic is over, I may have to go on a road trip back to Berlin and personally give you a high five to thank you for how great these videos are.
hell of a compliment, thanks! real-life high five sounds great - been thinking of doing some form of get-together/builder's jam when the circumstances allow it!
I'm a CS and CE grad with like 3 years of practical lab experience and your video series is the VERY FIRST TIME I've clearly understood so many fundamentals! Thank you!
Seconded. This is the perfect way to fuse concept with practice, and the resulting lesson is not only enjoyable but incredibly informative. After watching these, I feel a little more equipped to go out and read more for myself.
This was probably the most intuitive representation of a transistor that I have ever seen. All these videos explanations are extremely well done, thank you!
Your mechanical representation of a transistor was the most relatable I've ever seen - it just made total sense to me the minute I saw it. Thank you for that - and for the rest of the design concepts here!
I am currently an electrical engineering student (2 years in) and I have learned more from your videos than I have learned in school lol. You explain things amazingly, I am loving your videos, thank you for making these!!
I've been fooling around with DIY guitar pedals/synth DIY for the last two years and this is by far the best explanation of a transistor I've encountered. The way you combine analogies with real-circuit problems is very effective.
I'm doing the preliminary research to start my journey towards building my own Poly Synth from scratch. I have no idea what I'm doing, well that was the case til I started watching this series, now I feel like I actually stand a chance! What an excellent education this has been so far. Thank you so much!
I like your analogies with water; they're well thought out. However, you've explained bjts like they're fets, which is a common problem I find on the internet. BJTs don't have a "linear" region for physics reasons. BJTs have 4 regions: off, forward active, saturation, and reverse active. All bjt have an exponential relation for input base voltage vs output collector current while in the forward active region, this doesn't just exist "below" 0.7V, but is true so long as the voltage at the base is less than the voltage at the collector. Your work is great in general, and I look forward to more of your videos.
hey david, thanks for the reply! as for the linear region (which is also referred to as the forward active region): i was talking about the relation between base-emitter current and collector-emitter current there, not base voltage and collector current. from what i can tell, the relation between those two currents is approximately linear while Vb > 0.7V, and is determined by the transistor's hfe: Ic = hfe x Ib. see here: web.engr.oregonstate.edu/~traylor/ece112/lectures/bjt_reg_of_op.pdf also, from what i've read, BJTs are generally current controlled (save for that
Yeah. Like I said, it's very common to find people who think that BJTs are "current controlled." That just doesn't exist. For any transistor. Anywhere. All transistors are power input, power output devices. Even fets need a current to charge and discharge the gate, and they will leak into the body of the transistor just as surely (though much more slowly) as a bjt. As for hfe: it's not controlled and it's not static (however it will always be the ratio of collector current to base current because that's it's definition). There a lot of data sheets that will give you hfe curves. These will show the hfe varying from 10 to 250 for something like a 2n2222. Controlled in this case means if you take a random 2n2222 from a box and apply 1mA of current at the base you might get 250mA out of that one. But the next might be 200 or 150 or yet another number. Where as if you apply 0.7V from the base to the emitter you will get a very similar current out of the collector with a variance closer (+/-)5-15 mA. Unfortunately these are common misconceptions because semiconductor physics is difficult and manufacturers don't like to hand out their semiconductor technical data willingly for people to look at. I'm also extremely disappointed with the quality of what Organ State is producing. The last graph is just WRONG in so many ways. It states that beta is constant, which it isn't. it states that Ic=B*Ib only for the linear region, which is also false and for two reasons beta is DEFINED as the ratio of input current and output current and thus Ic=B*Ib at all times unless the device is in cut-off. That also failed to mention reverse-active operation of a bjt as a valid option, which leads me to believe that this a VERY basic document for rough guess work in designs and was intended to lead into or be proceed by lectures on fets which only have three regions of operation. www.electronics-notes.com/articles/electronic_components/transistor/current-gain-hfe-beta.php Here's a semi-decent link to how hfe/beta works in bjt's. It's not perfect, but it's more than adequate to embarrass that paper by Organ State. (And yes it too states that BJT are current controlled which I strongly disagree with)
@@MoritzKlein0Now this is a separate reply because it's a separate subject. Why did you not get 12V at the base of the transistor. The answer is simple, you forced the current through a resistor first. You placed a 100K ohm resistor between the transistor and the potentiometer. If the BJT is drawing a static 1mA through the base then that 1mA needs to pass through a 100K ohm resistor first that means the resistor drops (V=I*R) V=0.001A*100000ohm = 100V!!!!! In other words, you're not feeding it the POWER that it needs to reach saturation or hardly even to reach turn on. This means you're operating in what I know as "Sub-threashold" (although that is a term for mosfets not BJT's, let's borrow it anyways until someone else can correct me). Bjt's will draw current into the base (npn) or push current out of the base (pnp) due to semiconductor physics, this means you need to pay attention to how much current it needs which is determined by Beta which comes from collector current which is exponentially proportional to base voltage.
@@FuncleMonster in the link you gave, under the section "Variations of current gain", it's stated that "normally the bipolar transistor is biassed to operate in its linear region for analogue signals and it can be assumed that the current gain is constant. Accordingly, for good linear operation, the transistor should be operated well within its operating range and not running into the rails or drawing excessive current for the particular semiconductor device." is that also wrong in your opinion? if so, how come you can use an NPN to amplify audio signals without distorting them?
@@MoritzKlein0 Yes I do. Because by that same logic, there should be "forward linear" and a "reverse linear" region, yet they only talk about "linear" in that paragraph. The irony being that two sections earlier they talked about hfe as "This is the current gain for a transistor expressed as an h parameter (hybrid parameter). The letter f indicates that it is a forward transfer characteristic, and the letter e indicates it is for a common emitter configuration. The small letter h indicates it is a small signal gain. hfe and small signal Beta are the same" which acknowledges that there are other ways of measuring current gain which are not "forward." Also your quote blatantly states *ASSUMED THE CURRENT GAIN IS LINEAR.* It also goes on to stipulate a long list of conditions to maintain this assumption. That's fine, if you want to be 85% correct or what you're working with doesn't need true linearity. And again, just because something is approximately linear doesn't mean that every device follows the same linear curve. BJT1 could be more of a Y=~200*X where as BJT2 could be more of a Y=~134*X. These are both linear, but they're not the same. Also I found it rather ironic that you pulled that quote from the section of website labeled "Variations of Current Gain" and where the first paragraph under that reads "It is normally expected that the value of current gain β for a bipolar transistor is constant, however there are some variations that occur in the value of β or hFE." which again means that they even acknowledge their approximation and were telling you the non-linear dependencies of current gain. As for why audio signals can be amplified without distortion: They can't. That's why we measure the distortion as both THD (Total Harmonic Distortion) and IMD (Inter-Modulation Distortion). It's also why we apply both local and global negative feedback to linearize circuits in things like op-amps and power amps. There's a lot of hand waving in circuits, and that's cause people don't want to do the math. There are a lot of "approximates" and "it works good enough" and "I did a parameter sweep, this value is best" because circuits theory isn't as easy as people want it to be, but I promise that if you understand the devices and the theory, you might only have to revise your circuit 2 or 3 times as opposed to iterating through 100 spice sims just to figure out the optimum value of one resistor. Spice is a great way to check your answers, not find them. Just like these websites and non-textbooks are a great way of getting started, but not as the manual for an actual circuit using multiple IC's.
You are a true teacher. I have watched so many videos trying to explain these things and just can't understand whats happening at the volt and resistance level. I especially appreciate your water analogy for a transistor. I don't think I've ever seen it applied that way. Nearly every transistor explanation on the internet bewilders people with electrons, holes, doping, and other concepts. I appreciate knowing those things, but they simply aren't helpful in applying it to actual circuits.
In year 1 electrician trades course, and my teacher is great, but hearing what he's explained in a new way is really good to have. I'm so excited to try to put these concepts to the test
I've worked with electronic components for a few years now building drones and synth modules but I never knew what was going on inside these mysterious little items! I feel enlightened. Thank you for these videos.
hands down the best video ive ever seen on you tube. ive dabbled with electronics for years repairsing and modding stuff after watching your first 2 videos ive learned more then i have in 20 years and its building a synth which is something ive wanted to do for years. seriously thank you for such informitive and fun videos.
I fully understand how a transistor works; we received more than one water analogy in college. And yet I can never skip past your simple explanations and wonderful illustrations. I wish I had you as a resource 15 years ago! Thanks!
Reading through the comments before I maybe type what everyone else is....nah dont bother...so I repeat: this is the best method of teaching electronics I've ever seen. Better than any professor I had while earning my EE degree!
This series is really great! The water analogies are helpful, and this is the first tutorial I’ve seen that talks about how to control the oscillator with voltage. Thanks for doing these!
I'm really enjoying this series and look forward to finding out how to get full and accurate CV control of the CD40106-based oscillator. Thanks for your hard work.
These videos are so great man. Following along on breadboard. I'm super excited to learn about how the whole temperature dependence thing works in your next video.
“Super high pitched drone music” - makes me think of Aphex Twin's “Ventolin”, or lots of stuff by Merzbow. Seriously: great tutorial. I’ve never been able to get my head around analog circuits, but I learned some things watching this series!
Hey, it's me again from the comments of part 1. Just making a notes of the bits I didn't get to hopefully help make future videos easier to understand. Again, loving the series so far. For pitch control we need to change the resistance of the drain resistor. I get that, but why not just use a potentiometer? I didn't understand this until you finished the transistor/octaves are exponential bit, but these parts felt out of order. We already know what a resistor and a voltage divider is, and I feel like a potentiometer would've made more sense as a next step. Had you explained how a pot works, hooked it up in place of the static drain resistor and showed that it does alter the pitch, then explained that this isn't ideal for musical purposes because of how the pot is linear, then explain we want an exponential voltage change due to the way the musical scale works, then explained/demonstrated that putting the linear voltage into the transistor results in an exponential voltage output, I feel it would've made more sense and flowed better. (Sorry for the run on sentence) Next, in the scaling/offset part, I understood the scaling part immediately. However, the diagram of the combined offset/scale part threw me off a little. Because you explained scaling first then called back to offset voltage, I assumed that was the order the circuit would do these things. Explaining them in this order was great for understanding the scale/offset problem, but once you got to 21:32 had you pointed out more explicitly that we are modifying the offset first and then scaling I wouldn't have got lost. I did understand after a few rewatches, so it might be a moot point.
keep in mind that this only works under very specific circumstances. (as far as i know, only when the emitter is tied to ground and the collector is always at a positive voltage level.)
Excellent video, Moritz. In fact, all of your videos that I have watched thus far are superb. You should consider designing some vacuum tube based circuits as well as they provide certain audio characteristics that are difficult if not impossible to achieve with solid state devices.
Man this is awesome. Sir thanks so much for sharing as well as for your time and effort. I'm going step-by-step and it's just fun. Your explanatory drawings ... awesome.
You open with the question I had from Part 1! I hadn't noticed the R value - LOL. I hadn't thought of transistors as fluidic devices before. The V/Oct was exactly the CV question I've had for years. "Cats concert" is a wonderfully descriptive term.
Great video! Thanks:) Perhaps attaching a schematic of what's been done at the end of every video or in the description could help those following along!
Great video! I had to think for a moment at 17:08 where you said that the "the voltage at the base is unaffected". Then it hit me that as long as the voltage is less than the "diode drop" at the Vbe junction (let's say it's about .7v and in your case it is because your range is 350mv to 550mv), then all of the voltage is dropped by the diode and that is why the voltage does not change across the resistor. Of course, if you go above .7v, then there WILL BE a voltage drop across the resistor.
As a former computer programmer, I've got "perfect pitch" for powers of two. A ration of 450:1 is 8 1/2 octaves (well, 8.8). In other words, as we observed, from supersonic tones to distinct clicks, rather than a bass tone
so here's the full list, including all the components we need for the next episode. i've never used mouser but i'll see if i can whip up a set/kit/whatever there. however, if you have a local electronics shop, i'd strongly encourage you to buy there! > basics: 1x breadboard jumper wires/bridges set 2x 9V battery plus clips OR 1x dual 9-15V power supply > ICs 1x TL074 quad op amp 1x 40106 hex schmitt trigger inverter > diodes: 2x 1N1418 diode > capacitors: 1x 2.2 nF foil capacitor 2x 1uF foil capacitor > resistors: 1x 1M resistor 5x 100k resistor 2x 10k resistor 1x 4k resistor 1x 1k5 resistor > potentiometers: 3x 100k linear mono potentiometer 1x 1k linear precision trim potentiometer > transistors: 1x BC548C NPN transistor 1x BC558C PNP transistor > jack sockets: 2x 6.35mm OR 2x 3.5mm
@@MoritzKlein0 hi, did you ever create a BOM on mouser, or another electronic components site? I'd love to follow along, but I'm a super early beginner and a bit afraid I'd buy the wrong type of components.
Looks like I was watching this video a year ago! OK I have a few insights into adjustable capacitors (usually air gap capacitors). They are typically used for high voltage. There are some lower-voltage ones. But there are two other issues. They are BIG, for one (much bigger than a potentiometer); and they do not have much of a capacitance range. Also they can be like $60 or more so building them into commercially available designs may be difficult. (Also the reasons stated in the video as well)
Hi Moritz! First of all, let me ust say what great tutorials! I remember my teacher using a similar water analogy in the first year of school, back when I studied electronics. It's really a neat way of explaining how elctricity "works"! I'm really looking forward to your next videos, and so I'm kinda wondering: When is the next one gonna be here? How often do you think you'll be posting new episodes? Sorry for nagging, btw... ;-) Thanks, Thomas
hi thomas, glad to hear you enjoy the videos! the next episode is already done, it'll go up soon. i'm trying to have a bimonthly kinda rhythm, but it always depends on the roadblocks i face writing the scripts! sometimes it takes a while to break things down, especially if there's only rally math-y explanations online..
I’m loving this series (and the VCF); your diagrams and explanations with the water analogy are perfect! I’ve been learning electronics as a hobby for a little while now and understand what different components do, but it’s always been hard to visualize how they interact in practice to make a circuit function. I feel like I’ve learned more from these few videos than from all the other reading/watching I’ve done so far! Really great stuff, can’t wait to see more! One question; can an opamp be used to create the schmitt trigger inverter?
I'm experimenting with this on breadboard, and i get very inconsistent results depending on the capacitor type i use for the oscillator. Electrolitics give a much louder sound, ceramic are very quiet, PET are in between, PP are almost silent... Did i get something wrong?
hahahaa I just put a light dependent resistor in for the one that controls the pitch and I can't stop playing around with it! These things are so much fun, I can leave the lights on and "play" Ode to Joy by covering the resistor wth my hand, or I can turn the lights off and play the oscillator with a flashlight. This almost makes me want to build a Theremin... maybe after I finished all the filter videos.
in an old radio we change the oscillator frequency by changing the capacitor. but it's an extremely low value capacitor and it can only change over a very small range. perfect if you need to to precisely adjust relatively high frequencies (like from 550 to 1720 kHz in an A.M. radio)
Great stuff, looking forward to the next one! I was wondering one thing as a n00b: why not use a FET instead of a transistor? Since these are voltage controlled devices and are also usable as variable resistor? I suppose there is a good reason that I don't know of :)
to be honest, i haven’t really looked into FETs yet so i‘m kinda clueless there! it might be doable, but i haven‘t seen a design using them yet. maybe they don’t have that exponential voltage to current relation? i‘ll look into it!
@@MoritzKlein0 I have a hard time understanding the info about it online, it's all pretty vague and technical. I suppose the use of transistors is documented better because of it's longer history. But considering the exponential curve, that might be the issue yes! But more in general it would be nice to have an alternative for the kinda messy diy vactrol to plug CV in. Anyways, thanks for the great content!
@zvm FETs are transistors too - just fyi! and yeah, i also used to fantasize about some simple, clever component that could to what a vactrol does, but less messy. as far as i can tell, that doesn't exist - you'll need to use different components for different scenarios. i've got a filter series planned next, maybe i'll look into FETs for that!
@@MoritzKlein0 I have simulated some circuits with this paper as source: www.vishay.com/docs/70598/70598.pdf?fbclid=IwAR0QEFo9vZrmlNF0DQf5qAeiMSvMQPql9sfDqieE4mD6RQo1iuvbxD9aahc I used the circuit on the bottom of page 3, and in the simulations it seemed to be a very efficient simple circuit that acts as a voltage controlled resistor. I haven't tested it in real life yet, as I'm still waiting for my ordered FETs. However, I find it hard to believe that it will work perfect, as this study is from 1997 and I haven't seen this simple circuit all over the place. I hope I can test it soon!
I finished building this on the breadboard today! It was a lot of fun, and I learned a lot of new stuff. Really excited to be able to make these things myself, but that'll take some more time. On that note, this week I'm receiving lots of parts, a microcontroller and some midi plugs. I'm going to try and make my own midi-controller, I'll also be sharing this on my channel. Do you also do programming and digital style modules Moritz?
These videos are amazing, I can't thank you enough. I'm following along, and at this step, everything seems to be functioning properly, except it is still extremely quiet. I have it plugged into a guitar amp which I'm sure is not the best, but even at a decently high volume, the VCO is still very quiet. I've gone over the videos a few times to make sure my circuit matches and I can't find any issues. Is this normal? edit: I figured out that my op-amp was bad, causing the issue.
01:07 I am jaded by all these inaccuracies. I love the videos because they give me a chance to cite the errors contained therein. The circuit shown here does not generate a real sawtooth waveform, rather, a mediocre approximation (due to the use of a Schmidt trigger in lieu of a regular op-amp). With an op-amp connected as shown here, the waveform would be even more distorted. When a capacitor discharges through a fixed resistance, you don't get a linear ramp. the timing capacitor will begin to discharge at a reverse exponential rate, multiplied by a constant whose value is determined by the DC offset at the _base_ of the waveform. A real sawtooth waveform has an inverse linear ramp in the first quadrant, preceded by an undefined, but ideally, linear ramp in the first quadrant starting from the origin. The only way to achieve the inverse linear ramp would be to replace the resistor with a constant current drain circuit. I don't know what this circuit has to do with exponential conversion, however, because I cannot seem to find part 1 or 3, or any other subsequent videos. therefore, I am forced to take the risk of interpreting this video out of context.
I don't get the "adding voltages". If I have a 5V source and a 1V source (with common ground) and I put them through the 2 resistors...I suddenly get 6V? What am I missing here?
Ok...I think I figured it out. You're actually adding the CURRENTs from the 2 inputs through the 2 resistors. Assuming the 1V source is constant, it will add a constant portion of current to the output current. Even if the 5V input signal drops to 0V, the 1V source will still provide some current at the output, thus allow for a voltage at that point (depending on the load's resistance/impedance), not allowing the 5V signal to drop to zero and thus effectively 'raising' the signal.
Thanks! As a very green hobbyist, that is hands down the most helpful yt electronics channel I've seen so far. I do have a question though. At 16:30, how come the resistance you put between the power rail and the transistor base doesn't affect the voltage? I mean, I do understand it limits the current, but isn't the resistor a load, why no voltage drop?
it does affect the voltage, but only slightly. since the transistor is barely turned on, only a few nanoamps will actually flow into the base - which is not enough to make the voltage at the base drop significantly!
Do the transistors need to be matched? I managed to find and use a pair with HFE within 2 or 3 (they were some crazy outliers) but they are pretty rare from the batch i got (1 pair out of about 100 transistors). Also I was watching a documentary about ARP and they mentioned that their engineer actually took the transistors to a sander to remove as much of the package as possible before thermally bonding the transistors together. I didn't go that far since I only had one pair, but it sounds fun.
are you talking about the npn/pnp-pair? if so, i’m fairly certain you don’t need to match them. also, generally in synth design, you match transistors for Vbe, not hfe - and, only afaik, you can’t even match npns and pnps for Vbe. take this with a grain of salt though!
![Kumamoto Masters Japan 2024 | Gregoria Mariska Tunjung (INA) [5] vs. Akane Yamaguchi (JPN) [2] | F](http://i.ytimg.com/vi/EIjZybhsWqA/mqdefault.jpg)
Seriously the best electronics tutorial I've encountered
Seconded
I Terded
Third it*
Seriously, I have a degree in electronics and I’m learning a few new tricks from this
Easily
When a Mechanical Engineer suddenly understands electronics, you know you’ve done something special! Thank you for sharing you knowledge!
0000
I've been making DIY synths since I was a teenager--not necessarily GREAT synths, but making them nonetheless.
I never took an electronics course so everything was sort of self-directed learning, and pretty lackluster at that.
Never in the past 10 years have I found such a helpful series of tutorials, and had I had this available to me when I was 18, I can only imagine what I'd have made!
You should be very proud of yourself because very rarely does someone with a good understanding of a concept actually know how to explain it in a digestible way.
Every single segment of these videos absolutely blows my mind and has re-kindled my drive and inspiration to start building again.
Once this bullshit pandemic is over, I may have to go on a road trip back to Berlin and personally give you a high five to thank you for how great these videos are.
hell of a compliment, thanks! real-life high five sounds great - been thinking of doing some form of get-together/builder's jam when the circumstances allow it!
Now I understand that the frequency knob of the VCO is simply adding an offset voltage to the transistors base..Thanks
I'm a CS and CE grad with like 3 years of practical lab experience and your video series is the VERY FIRST TIME I've clearly understood so many fundamentals! Thank you!
So many things in the field are just taught in a vacuum, with no context. It makes it really hard to understand how it all relates in nature...
I like how you're walking through the circuitry as well as the issues faced in getting things to work and scale properly.
Fantastic - Moritz, I have the feeling you are going to be the father of a generation of new synthesisers. That is awesome!🙂
hell of a compliment, thank you!
Seconded. This is the perfect way to fuse concept with practice, and the resulting lesson is not only enjoyable but incredibly informative. After watching these, I feel a little more equipped to go out and read more for myself.
This was probably the most intuitive representation of a transistor that I have ever seen. All these videos explanations are extremely well done, thank you!
Your mechanical representation of a transistor was the most relatable I've ever seen - it just made total sense to me the minute I saw it. Thank you for that - and for the rest of the design concepts here!
I am currently an electrical engineering student (2 years in) and I have learned more from your videos than I have learned in school lol. You explain things amazingly, I am loving your videos, thank you for making these!!
I've been fooling around with DIY guitar pedals/synth DIY for the last two years and this is by far the best explanation of a transistor I've encountered. The way you combine analogies with real-circuit problems is very effective.
You're a modern day Ray Wilson of synth DIY
I'm doing the preliminary research to start my journey towards building my own Poly Synth from scratch.
I have no idea what I'm doing, well that was the case til I started watching this series, now I feel like I actually stand a chance! What an excellent education this has been so far. Thank you so much!
I like your analogies with water; they're well thought out. However, you've explained bjts like they're fets, which is a common problem I find on the internet. BJTs don't have a "linear" region for physics reasons. BJTs have 4 regions: off, forward active, saturation, and reverse active. All bjt have an exponential relation for input base voltage vs output collector current while in the forward active region, this doesn't just exist "below" 0.7V, but is true so long as the voltage at the base is less than the voltage at the collector.
Your work is great in general, and I look forward to more of your videos.
hey david, thanks for the reply! as for the linear region (which is also referred to as the forward active region): i was talking about the relation between base-emitter current and collector-emitter current there, not base voltage and collector current. from what i can tell, the relation between those two currents is approximately linear while Vb > 0.7V, and is determined by the transistor's hfe: Ic = hfe x Ib. see here: web.engr.oregonstate.edu/~traylor/ece112/lectures/bjt_reg_of_op.pdf
also, from what i've read, BJTs are generally current controlled (save for that
Yeah. Like I said, it's very common to find people who think that BJTs are "current controlled." That just doesn't exist. For any transistor. Anywhere. All transistors are power input, power output devices. Even fets need a current to charge and discharge the gate, and they will leak into the body of the transistor just as surely (though much more slowly) as a bjt.
As for hfe: it's not controlled and it's not static (however it will always be the ratio of collector current to base current because that's it's definition). There a lot of data sheets that will give you hfe curves. These will show the hfe varying from 10 to 250 for something like a 2n2222. Controlled in this case means if you take a random 2n2222 from a box and apply 1mA of current at the base you might get 250mA out of that one. But the next might be 200 or 150 or yet another number. Where as if you apply 0.7V from the base to the emitter you will get a very similar current out of the collector with a variance closer (+/-)5-15 mA.
Unfortunately these are common misconceptions because semiconductor physics is difficult and manufacturers don't like to hand out their semiconductor technical data willingly for people to look at. I'm also extremely disappointed with the quality of what Organ State is producing. The last graph is just WRONG in so many ways. It states that beta is constant, which it isn't. it states that Ic=B*Ib only for the linear region, which is also false and for two reasons beta is DEFINED as the ratio of input current and output current and thus Ic=B*Ib at all times unless the device is in cut-off. That also failed to mention reverse-active operation of a bjt as a valid option, which leads me to believe that this a VERY basic document for rough guess work in designs and was intended to lead into or be proceed by lectures on fets which only have three regions of operation.
www.electronics-notes.com/articles/electronic_components/transistor/current-gain-hfe-beta.php
Here's a semi-decent link to how hfe/beta works in bjt's. It's not perfect, but it's more than adequate to embarrass that paper by Organ State. (And yes it too states that BJT are current controlled which I strongly disagree with)
@@MoritzKlein0Now this is a separate reply because it's a separate subject. Why did you not get 12V at the base of the transistor. The answer is simple, you forced the current through a resistor first. You placed a 100K ohm resistor between the transistor and the potentiometer. If the BJT is drawing a static 1mA through the base then that 1mA needs to pass through a 100K ohm resistor first that means the resistor drops (V=I*R) V=0.001A*100000ohm = 100V!!!!! In other words, you're not feeding it the POWER that it needs to reach saturation or hardly even to reach turn on. This means you're operating in what I know as "Sub-threashold" (although that is a term for mosfets not BJT's, let's borrow it anyways until someone else can correct me). Bjt's will draw current into the base (npn) or push current out of the base (pnp) due to semiconductor physics, this means you need to pay attention to how much current it needs which is determined by Beta which comes from collector current which is exponentially proportional to base voltage.
@@FuncleMonster in the link you gave, under the section "Variations of current gain", it's stated that "normally the bipolar transistor is biassed to operate in its linear region for analogue signals and it can be assumed that the current gain is constant. Accordingly, for good linear operation, the transistor should be operated well within its operating range and not running into the rails or drawing excessive current for the particular semiconductor device." is that also wrong in your opinion? if so, how come you can use an NPN to amplify audio signals without distorting them?
@@MoritzKlein0 Yes I do. Because by that same logic, there should be "forward linear" and a "reverse linear" region, yet they only talk about "linear" in that paragraph. The irony being that two sections earlier they talked about hfe as "This is the current gain for a transistor expressed as an h parameter (hybrid parameter). The letter f indicates that it is a forward transfer characteristic, and the letter e indicates it is for a common emitter configuration. The small letter h indicates it is a small signal gain. hfe and small signal Beta are the same" which acknowledges that there are other ways of measuring current gain which are not "forward." Also your quote blatantly states *ASSUMED THE CURRENT GAIN IS LINEAR.* It also goes on to stipulate a long list of conditions to maintain this assumption. That's fine, if you want to be 85% correct or what you're working with doesn't need true linearity. And again, just because something is approximately linear doesn't mean that every device follows the same linear curve. BJT1 could be more of a Y=~200*X where as BJT2 could be more of a Y=~134*X. These are both linear, but they're not the same. Also I found it rather ironic that you pulled that quote from the section of website labeled "Variations of Current Gain" and where the first paragraph under that reads "It is normally expected that the value of current gain β for a bipolar transistor is constant, however there are some variations that occur in the value of β or hFE." which again means that they even acknowledge their approximation and were telling you the non-linear dependencies of current gain.
As for why audio signals can be amplified without distortion: They can't. That's why we measure the distortion as both THD (Total Harmonic Distortion) and IMD (Inter-Modulation Distortion). It's also why we apply both local and global negative feedback to linearize circuits in things like op-amps and power amps.
There's a lot of hand waving in circuits, and that's cause people don't want to do the math. There are a lot of "approximates" and "it works good enough" and "I did a parameter sweep, this value is best" because circuits theory isn't as easy as people want it to be, but I promise that if you understand the devices and the theory, you might only have to revise your circuit 2 or 3 times as opposed to iterating through 100 spice sims just to figure out the optimum value of one resistor. Spice is a great way to check your answers, not find them. Just like these websites and non-textbooks are a great way of getting started, but not as the manual for an actual circuit using multiple IC's.
Your analogy of water pressure and how the NPN transistor operates really clears up the confusion I've had how to use them. Thank you.
You are a true teacher. I have watched so many videos trying to explain these things and just can't understand whats happening at the volt and resistance level. I especially appreciate your water analogy for a transistor. I don't think I've ever seen it applied that way. Nearly every transistor explanation on the internet bewilders people with electrons, holes, doping, and other concepts. I appreciate knowing those things, but they simply aren't helpful in applying it to actual circuits.
thanks, glad to hear - that's exactly what i'm going for, since i had the exact same experience when i was starting out!
In year 1 electrician trades course, and my teacher is great, but hearing what he's explained in a new way is really good to have. I'm so excited to try to put these concepts to the test
I've worked with electronic components for a few years now building drones and synth modules but I never knew what was going on inside these mysterious little items! I feel enlightened. Thank you for these videos.
hands down the best video ive ever seen on you tube. ive dabbled with electronics for years repairsing and modding stuff after watching your first 2 videos ive learned more then i have in 20 years and its building a synth which is something ive wanted to do for years. seriously thank you for such informitive and fun videos.
I fully understand how a transistor works; we received more than one water analogy in college. And yet I can never skip past your simple explanations and wonderful illustrations. I wish I had you as a resource 15 years ago! Thanks!
Reading through the comments before I maybe type what everyone else is....nah dont bother...so I repeat: this is the best method of teaching electronics I've ever seen. Better than any professor I had while earning my EE degree!
Finally, after first part my VCO didn't have sound, but after replacing 100K resistor to 1M i makes sound!
This series is really great! The water analogies are helpful, and this is the first tutorial I’ve seen that talks about how to control the oscillator with voltage. Thanks for doing these!
I'm really enjoying this series and look forward to finding out how to get full and accurate CV control of the CD40106-based oscillator. Thanks for your hard work.
Just wanted to tell you that your videos are amazing. Everything is expertly explained and with perfect pacing.
These videos are so great man. Following along on breadboard. I'm super excited to learn about how the whole temperature dependence thing works in your next video.
same
i have builded quite a few 40106 oscillators already, now i can make them VCOs. thank you very much for the best synth diy tutorial i have ever seen
“Super high pitched drone music” - makes me think of Aphex Twin's “Ventolin”, or lots of stuff by Merzbow.
Seriously: great tutorial. I’ve never been able to get my head around analog circuits, but I learned some things watching this series!
+1 for aphex twin & merzbow!
Hey, it's me again from the comments of part 1. Just making a notes of the bits I didn't get to hopefully help make future videos easier to understand. Again, loving the series so far.
For pitch control we need to change the resistance of the drain resistor. I get that, but why not just use a potentiometer? I didn't understand this until you finished the transistor/octaves are exponential bit, but these parts felt out of order. We already know what a resistor and a voltage divider is, and I feel like a potentiometer would've made more sense as a next step.
Had you explained how a pot works, hooked it up in place of the static drain resistor and showed that it does alter the pitch, then explained that this isn't ideal for musical purposes because of how the pot is linear, then explain we want an exponential voltage change due to the way the musical scale works, then explained/demonstrated that putting the linear voltage into the transistor results in an exponential voltage output, I feel it would've made more sense and flowed better. (Sorry for the run on sentence)
Next, in the scaling/offset part, I understood the scaling part immediately. However, the diagram of the combined offset/scale part threw me off a little. Because you explained scaling first then called back to offset voltage, I assumed that was the order the circuit would do these things. Explaining them in this order was great for understanding the scale/offset problem, but once you got to 21:32 had you pointed out more explicitly that we are modifying the offset first and then scaling I wouldn't have got lost. I did understand after a few rewatches, so it might be a moot point.
Awesome video and great explanation with clear formulation of possible problems, looking forward to the next video!
Explained better than electrical engineering professors.
I haven't seen a video (and the others in the series of course) with this much value in a long time, great stuff!
I just made notes and tomorrow I will start ordering components. the best tutorial so far. bravo!
perfect, have fun!
I never realized that each octave is double that of the previous. I just assumed it was linear.
mind blown
This is the best style of education for me.
Your analogies are the best I've ever seen
Super explanation, best video series I found so far on Synth DIY. Thank you!
Thats the best explanation of the way an analog VCO works I've ever found. Thanks.
Ahh right, I had no idea that these things were so much a sequence of problems to be resolved. Very interesting!
not to get too philosophical, but i'm starting to believe that everything's a series of small, easy-to-solve problems - even the scary, complex stuff!
@@MoritzKlein0 you could be on to something there! 😉😁
I didn't know I could use a transistor as a voltage controlled resistor. Thanks for the tip!
keep in mind that this only works under very specific circumstances. (as far as i know, only when the emitter is tied to ground and the collector is always at a positive voltage level.)
Excellent video, Moritz. In fact, all of your videos that I have watched thus far are superb. You should consider designing some vacuum tube based circuits as well as they provide certain audio characteristics that are difficult if not impossible to achieve with solid state devices.
your explanations help me understand electronics much better. thank you so much!
Thanks so much! Your videos are incredibly detailed and really help me arrange scattered pieces of information into logical structures in the brain
Fantastic series! I've been wanting to learn to make musical circuits for a long time and this is really helpful. You rock!
these videos are insanely useful, amazing presentation and very purpose driven, love em
Wow this series is so awesome already 🙌 looking forward to the next episode. Thanks!
Great explanation about the basics of VCO's!
this is the best lesson I've ever received. thank you
I love this series and eagerly await each new video!
Man this is awesome. Sir thanks so much for sharing as well as for your time and effort. I'm going step-by-step and it's just fun. Your explanatory drawings ... awesome.
Great video series! It's really good and you clearly explain all the concepts
You open with the question I had from Part 1! I hadn't noticed the R value - LOL. I hadn't thought of transistors as fluidic devices before. The V/Oct was exactly the CV question I've had for years. "Cats concert" is a wonderfully descriptive term.
yeah - i only realized after i had already written the script for the first episode and was too lazy to correct it. glad you enjoy the videos!
Thought my tinnitus was acting up at the start of the video.
Oh man!!! Just dont stop what you are doing!!! Its fantastic!!! God bless you!!!
Great video! Thanks:) Perhaps attaching a schematic of what's been done at the end of every video or in the description could help those following along!
These are brilliant. You certainly have the teaching gene!
Might give this a go use up some of my Schmitt triggers bought a ton of them.
thank you... finally i start understanding something of the things happening below the surface :)
Simply amazingly well thought through and performed. And congrats to 1k subs.
Gruß in die Hauptstadt aus der Provinz
danke dir! welche provinz genau?
@@MoritzKlein0 Sachsen-Anhalt, B ist die alte Heimat. Freue mich auf den nächsten Teil.
Thank you for making this teaching easy to understand.
I love the content and the way it is explained. Could you in future show us the DIY power supply you use to power the breadboard?
yes, i'm planning to do a video on that after i'm done with the VCO!
@@MoritzKlein0 Awesome!
It will be super useful 😃
you got pumped for the next video, well done.
Excellent! Really enjoying these tutorials.
Ur tutorials are really on point! Thanks for sharing ur knowledge, and hope to learn more from u in the future! ❤️
Great video! I had to think for a moment at 17:08 where you said that the "the voltage at the base is unaffected". Then it hit me that as long as the voltage is less than the "diode drop" at the Vbe junction (let's say it's about .7v and in your case it is because your range is 350mv to 550mv), then all of the voltage is dropped by the diode and that is why the voltage does not change across the resistor. Of course, if you go above .7v, then there WILL BE a voltage drop across the resistor.
yes and no - the way i talk about transistors here is quite imprecise. check my VCA video, i think the explanation there is much more accurate!
I'm learning a lot from your videos, thank you very much!
Thank you soooo much! It is realy inspiring, can't waint to see the next steps!
As a former computer programmer, I've got "perfect pitch" for powers of two. A ration of 450:1 is 8 1/2 octaves (well, 8.8). In other words, as we observed, from supersonic tones to distinct clicks, rather than a bass tone
better than almost all EE courses I've taken @_@...
Perfect tutorials man. Thanks a lot for this series!
Great info and brilliant explanations
This is amazing! Looking forward to the next episodes!
Remind me to build a synthesizer out of valves, pipes and water
This is a great series, I'd like to order the parts to follow along, would you be able to publish a full BOM and perhaps a link to Mouser or Tayda?
yep, i'll see what i can do!
so here's the full list, including all the components we need for the next episode. i've never used mouser but i'll see if i can whip up a set/kit/whatever there. however, if you have a local electronics shop, i'd strongly encourage you to buy there!
> basics:
1x breadboard
jumper wires/bridges set
2x 9V battery plus clips OR 1x dual 9-15V power supply
> ICs
1x TL074 quad op amp
1x 40106 hex schmitt trigger inverter
> diodes:
2x 1N1418 diode
> capacitors:
1x 2.2 nF foil capacitor
2x 1uF foil capacitor
> resistors:
1x 1M resistor
5x 100k resistor
2x 10k resistor
1x 4k resistor
1x 1k5 resistor
> potentiometers:
3x 100k linear mono potentiometer
1x 1k linear precision trim potentiometer
> transistors:
1x BC548C NPN transistor
1x BC558C PNP transistor
> jack sockets:
2x 6.35mm OR 2x 3.5mm
Moritz Klein fantastic, thank-you
@@MoritzKlein0 hi, did you ever create a BOM on mouser, or another electronic components site? I'd love to follow along, but I'm a super early beginner and a bit afraid I'd buy the wrong type of components.
Looks like I was watching this video a year ago! OK I have a few insights into adjustable capacitors (usually air gap capacitors). They are typically used for high voltage. There are some lower-voltage ones. But there are two other issues. They are BIG, for one (much bigger than a potentiometer); and they do not have much of a capacitance range. Also they can be like $60 or more so building them into commercially available designs may be difficult.
(Also the reasons stated in the video as well)
just when you thought they couldn't get any less appealing :(
Hi Moritz! First of all, let me ust say what great tutorials! I remember my teacher using a similar water analogy in the first year of school, back when I studied electronics. It's really a neat way of explaining how elctricity "works"! I'm really looking forward to your next videos, and so I'm kinda wondering: When is the next one gonna be here? How often do you think you'll be posting new episodes? Sorry for nagging, btw... ;-) Thanks, Thomas
hi thomas, glad to hear you enjoy the videos! the next episode is already done, it'll go up soon. i'm trying to have a bimonthly kinda rhythm, but it always depends on the roadblocks i face writing the scripts! sometimes it takes a while to break things down, especially if there's only rally math-y explanations online..
This is an amazing series
genuinely can't like these videos enough
one correction at 3:16, a varicap (or varactor) excists, which is a voltage controlled capacitor, and it's also a diode
Thank you so much, Moritz !
I’m loving this series (and the VCF); your diagrams and explanations with the water analogy are perfect! I’ve been learning electronics as a hobby for a little while now and understand what different components do, but it’s always been hard to visualize how they interact in practice to make a circuit function.
I feel like I’ve learned more from these few videos than from all the other reading/watching I’ve done so far! Really great stuff, can’t wait to see more!
One question; can an opamp be used to create the schmitt trigger inverter?
hi matt! yes, very easily. check this: tinyurl.com/yxq6qgfb
@@MoritzKlein0 Wonderful! Thanks! I keep rewatching the videos to try to grasp everything, gets pretty interesting by part 4 🤩
I'm experimenting with this on breadboard, and i get very inconsistent results depending on the capacitor type i use for the oscillator. Electrolitics give a much louder sound, ceramic are very quiet, PET are in between, PP are almost silent... Did i get something wrong?
Thank you for these amazing videos!
Hi bro! This is the best chanel! (Seriously good)
hahahaa I just put a light dependent resistor in for the one that controls the pitch and I can't stop playing around with it! These things are so much fun, I can leave the lights on and "play" Ode to Joy by covering the resistor wth my hand, or I can turn the lights off and play the oscillator with a flashlight. This almost makes me want to build a Theremin... maybe after I finished all the filter videos.
Thanks for your efforts, this is really instructive and clear.
in an old radio we change the oscillator frequency by changing the capacitor. but it's an extremely low value capacitor and it can only change over a very small range. perfect if you need to to precisely adjust relatively high frequencies (like from 550 to 1720 kHz in an A.M. radio)
Itching for the next one...
on it!
Great stuff, looking forward to the next one!
I was wondering one thing as a n00b: why not use a FET instead of a transistor? Since these are voltage controlled devices and are also usable as variable resistor? I suppose there is a good reason that I don't know of :)
to be honest, i haven’t really looked into FETs yet so i‘m kinda clueless there! it might be doable, but i haven‘t seen a design using them yet. maybe they don’t have that exponential voltage to current relation? i‘ll look into it!
@@MoritzKlein0 I have a hard time understanding the info about it online, it's all pretty vague and technical. I suppose the use of transistors is documented better because of it's longer history.
But considering the exponential curve, that might be the issue yes! But more in general it would be nice to have an alternative for the kinda messy diy vactrol to plug CV in.
Anyways, thanks for the great content!
@zvm FETs are transistors too - just fyi! and yeah, i also used to fantasize about some simple, clever component that could to what a vactrol does, but less messy. as far as i can tell, that doesn't exist - you'll need to use different components for different scenarios.
i've got a filter series planned next, maybe i'll look into FETs for that!
@@MoritzKlein0 I have simulated some circuits with this paper as source: www.vishay.com/docs/70598/70598.pdf?fbclid=IwAR0QEFo9vZrmlNF0DQf5qAeiMSvMQPql9sfDqieE4mD6RQo1iuvbxD9aahc
I used the circuit on the bottom of page 3, and in the simulations it seemed to be a very efficient simple circuit that acts as a voltage controlled resistor. I haven't tested it in real life yet, as I'm still waiting for my ordered FETs.
However, I find it hard to believe that it will work perfect, as this study is from 1997 and I haven't seen this simple circuit all over the place. I hope I can test it soon!
@@KeurslagerKurt looks promising! let me know how your tests go!
I finished building this on the breadboard today!
It was a lot of fun, and I learned a lot of new stuff.
Really excited to be able to make these things myself, but that'll take some more time.
On that note, this week I'm receiving lots of parts, a microcontroller and some midi plugs.
I'm going to try and make my own midi-controller, I'll also be sharing this on my channel.
Do you also do programming and digital style modules Moritz?
These videos are amazing, I can't thank you enough. I'm following along, and at this step, everything seems to be functioning properly, except it is still extremely quiet. I have it plugged into a guitar amp which I'm sure is not the best, but even at a decently high volume, the VCO is still very quiet. I've gone over the videos a few times to make sure my circuit matches and I can't find any issues. Is this normal?
edit: I figured out that my op-amp was bad, causing the issue.
Sehr schön, bin auf den nächsten Teil gespannt.
thanks so much! you make it so easy to understand
Great explanation and very easy ..we love it ...thanks
A ring oscillator will be tunable over some range by changing the supply voltage, good enough for a PLL.
Thanks for making these videos!
01:07 I am jaded by all these inaccuracies. I love the videos because they give me a chance to cite the errors contained therein. The circuit shown here does not generate a real sawtooth waveform, rather, a mediocre approximation (due to the use of a Schmidt trigger in lieu of a regular op-amp). With an op-amp connected as shown here, the waveform would be even more distorted. When a capacitor discharges through a fixed resistance, you don't get a linear ramp. the timing capacitor will begin to discharge at a reverse exponential rate, multiplied by a constant whose value is determined by the DC offset at the _base_ of the waveform. A real sawtooth waveform has an inverse linear ramp in the first quadrant, preceded by an undefined, but ideally, linear ramp in the first quadrant starting from the origin. The only way to achieve the inverse linear ramp would be to replace the resistor with a constant current drain circuit. I don't know what this circuit has to do with exponential conversion, however, because I cannot seem to find part 1 or 3, or any other subsequent videos. therefore, I am forced to take the risk of interpreting this video out of context.
I don't get the "adding voltages". If I have a 5V source and a 1V source (with common ground) and I put them through the 2 resistors...I suddenly get 6V? What am I missing here?
it’s more mixing than adding - didn’t make this clear enough. 5V and 1V with two equal resistors will give you the average between the two - so 3V!
Ok...I think I figured it out. You're actually adding the CURRENTs from the 2 inputs through the 2 resistors.
Assuming the 1V source is constant, it will add a constant portion of current to the output current. Even if the 5V input signal drops to 0V, the 1V source will still provide some current at the output, thus allow for a voltage at that point (depending on the load's resistance/impedance), not allowing the 5V signal to drop to zero and thus effectively 'raising' the signal.
Thanks! As a very green hobbyist, that is hands down the most helpful yt electronics channel I've seen so far. I do have a question though. At 16:30, how come the resistance you put between the power rail and the transistor base doesn't affect the voltage? I mean, I do understand it limits the current, but isn't the resistor a load, why no voltage drop?
it does affect the voltage, but only slightly. since the transistor is barely turned on, only a few nanoamps will actually flow into the base - which is not enough to make the voltage at the base drop significantly!
thanks a bunch!
Do the transistors need to be matched? I managed to find and use a pair with HFE within 2 or 3 (they were some crazy outliers) but they are pretty rare from the batch i got (1 pair out of about 100 transistors). Also I was watching a documentary about ARP and they mentioned that their engineer actually took the transistors to a sander to remove as much of the package as possible before thermally bonding the transistors together. I didn't go that far since I only had one pair, but it sounds fun.
are you talking about the npn/pnp-pair? if so, i’m fairly certain you don’t need to match them. also, generally in synth design, you match transistors for Vbe, not hfe - and, only afaik, you can’t even match npns and pnps for Vbe. take this with a grain of salt though!