Thanks for watching. For more Op Amp Circuits please see: Temperature-Independent Current Circuit Design with Op Amp, BJT, Zener and Schottky Diodes ruclips.net/video/hFbnjbddUvs/видео.html On-Chip Current Source with BJT Transistors ruclips.net/video/Rs7gEMk03dw/видео.html Thermometer Circuit Design with Op Amp & BJT transistor ruclips.net/video/55YsraFE0rg/видео.html Push-Pull Power Amplifier Design with Op Amp, Sziklai Darlington Transistors ruclips.net/video/8BFzsi7-Vbs/видео.html Instrumentation Amplifier with Electronic Gain Control ruclips.net/video/C4tghZ-q6Zs/видео.html Push-Pull Power Amplifier with Darlington Transistors ruclips.net/video/866MYibo8yE/видео.html Analog Logarithm Computer ruclips.net/video/RpKEq5WyoLg/видео.html Op Amp Analog Computer Differential Equation Solver ruclips.net/video/ENq39EesfPw/видео.html Sawtooth Oscillator Design ruclips.net/video/2eUsGPfqbW4/видео.html Triangle Oscillator Op Amp circuit ruclips.net/video/JF5Up_cuL9k/видео.html Sallen-Key Analog Filter Design Tutorial ruclips.net/video/KwUnQXbk7gM/видео.html How to find Bode Plot, Freq Response, Transfer Function of Analog Filters ruclips.net/video/vZFkPeDa1H8/видео.html Universal Analog Filter Design ruclips.net/video/2J-0msXZE2o/видео.html Laplace Transform Example and S-domain circuit analysis: ruclips.net/video/ps8N5TPM_qU/видео.html Op Amp circuit Bode Frequency plot ruclips.net/video/BLVzuuqAlZs/видео.html Lowpass Butterworth Filter: ruclips.net/video/UzCjkwqy-9w/видео.html Analog Computer to Raise Signal to power n ruclips.net/video/IUTlBH1UraE/видео.html Differential Equation Solver Analog Circuit ruclips.net/video/R3X5AYNZGEI/видео.html Complex Sinusoid Oscillator ruclips.net/video/GXRhmwmS5Zk/видео.html Full-Wave Rectifier circuit example ruclips.net/video/DJJMNU-CYcg/видео.html Sawtooth Waveform Generator design with OpAmp, JFET, BJT ruclips.net/video/5zHXTx-Vl20/видео.html op amps Circuit with feedback loops to design an analog computer that solves a second order differential equation ruclips.net/video/HeZRtnRXpEI/видео.html For more analog circuits and signal processing examples see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt I hope these Circuit design and analysis videos are helpful. 🙂
@@ЁбрагимИпатенкоибнАдхарма Thank you! Glad that you liked this Wilson Current Source Circuit video. Alternative design techniques are discussed in ruclips.net/p/PLrwXF7N522y48AAPxFaQlowim4-8gYoWz Hope this circuit playlist is interesting as well.
Hello Engineering Prof, Thanks for your clear and detailed explanation of analog electronic circuits. Look forward to more internal building blocks and circuitries of analog IC from you.
You are welcome. Thanks for watching, your interest & encouraging comment. Glad that you like this Wilson Current Mirror Tutorial video. For more analog circuit examples please see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. I hope these Circuit videos are useful as well. 🙋♂️
Thanks for watching and sharing your thoughts and suggestions. Whether to have Zener Diode positioned above the resistor or below the resistor you will end up with its own set of advantages and disadvantages. But I agree that one of the advantages of Zener diode connected between Vsupply and Resistor is better supply noise rejection for the reference current Iref. For more regulator circuits please see ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt . Thanks again.
@@STEMprof Actually, swapping the Zener and R1 will massively improve supply noise rejection. In the circuit you show, the current Iref is given by (Vs - Vz - Vbe) / R2, which depends linearly on Vs. If you swap the Zener and R1, Iref is given by (Vz - Vbe) / R2, which is independent of Vs. You then use a lower value Zener (3.3V to 3.9V), which has a temperature coefficient around -2.3mV/K which would cancel out most of the temp coefficient of the PNP's Vbe.
Agree, Zener diode as shown amplifies impact of Vsupply on Ireference (5% change of Vsupply causes ~10% change of Iref). It is better to swap R1 and Zener diode. P.S. The video is great
@@JumpingJack-w2l Thank you for watching & sharing your thoughts. Glad that you like this Wilson Current Mirror Tutorial video. Given that reference current Iref = (Vs-0.7-Vzener)/R2 then sensitivity = delta_Iref/Iref = delta_Vs/(Vs-0.7-Vzener) which means sensitivity increases if Vs is reduced or if Vz is increased. In this example with Vs=10v , Vzener=4.7v , if 10 volt supply is increased 5% to 10.5V, then Iref=(10.5-0.7-4.7)/4.6k = 1.109 mA indicating 10.9% increase in Iref as you noted. For more analog circuit examples pls see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. Thanks again 🙋♂️
You're welcome. Thanks for watching. Glad that this Wilson Current Mirror Tutorial video is useful. For more analog circuit examples pls see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. Thanks again 🙂
Thanks for sharing your thoughts. I agree that one of the advantages of Zener diode connected between Vsupply and Resistor is better supply noise rejection for the reference current Iref. Given that reference current Iref = (Vs-0.7-Vzener)/R2 then sensitivity = delta_Iref/Iref = delta_Vs/(Vs-0.7-Vzener) which means sensitivity increases if Vs is reduced or if Vz is increased. In this example with Vs=10v , Vzener=4.7v , if 10 volt supply is increased 5% to 10.5V, then Iref=(10.5-0.7-4.7)/4.6k = 1.109 mA reflecting 10+% increase in current. Alternative design techniques are discussed in the circuit videos ruclips.net/p/PLrwXF7N522y48AAPxFaQlowim4-8gYoWz Hope this circuit playlist is interesting.
Thanks for watching & your follow-up question. Substituting Ic in equation 1 using equation 2 results in the equation shown at 17:40 , and then a brief algebraic operation results in finding output current Io as a function of the reference current Iref in the form of Io = Iref * (B^2+2B)/(B^2+2B+2) . I hope this is helpful.
You're welcome. Thank you for the comment. Glad that my explanation was helpful. 🙂 For more analog circuit examples please see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. I hope these Circuit design/analysis videos are helpful as well.
Thanks for watching and sharing your thoughts. I agree that one of the advantages of Zener diode connected between Vsupply and Resistor is better supply noise rejection for the reference current Iref. For more examples please see: Voltage Regulator Design with Op Amp and BJT Transistor ruclips.net/video/rI9f6-DyXxQ/видео.html Voltage Regulator Op Amp Circuit with Foldback current limiting ruclips.net/video/VN4_qF9DvBM/видео.html Regulator Design with BJT Darlington Transistors & Zener Diodes ruclips.net/video/ArisQp7V0Ac/видео.html For more analog circuit examples pls see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. I hope these circuit design & analysis videos are helpful. Thanks for watching 🙏
Prof: thought I might open a conversation with you regarding circuit design and the way I was taught. (did not say I learned ;-}) In school we young aspirational student were taught circuit analysis via KVA: Loop currents:etc. Ok well and good. Then we graduated, and went to work. Now on occasion we faced the problem of generating new circuits. But for this we were not well prepared, yes analysis tools we had, but not real 'generation of new circuits" methods. --- I think what was missing and could have been included was an extensive discussion about how to think of transistors, Rs,Cs,Ls,... in a way that maps the components into behavioral models such as used in Spice. In this way using models like current sources, voltage sources and the controlled versions of each, one could "construct" a new circuit functionally and then start by including real world components with the specific constraints such as temperature variance, input currents,... and all the rest of the "real" component selection problems. Personally I had an extensive technical training prior to going to engineering school - so I had some basis or understanding of many circuits and was able to use that to develop new ideas. Sometimes good ideas, sometimes not. You may at some time feel that you have analyzed a sufficient number of typologies and might be interested in demonstrating your own method of generating new circuits. I for one would be interested and I encourage you to consider doing so. cheers d
Thank you for your interest in this channel and for sharing your thoughts. I appreciate. To your point, a good number of circuits in this channel are actually my own designs. I will post more new design examples in my Analog and Op Amp Circuits playlist ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt I hope you find them interesting. 🙂
A zenner diode is a noisy device, as well as a LED. So, better results will be with an additional simple transistor that will give 0.6v for R2. That's enough. Or we can use two sequential diodes in a same sot-23-4/sot-23-3 package. The second thing is that it's better to have a Vsupply independent current (or in other words to increase a PSRR), and so the zenner must go first, and then R1. If we still need a higher reference voltage, tl431 is a good alternative to a zenner. FYI x4 matched(!) transistors chips like MMPQ, TAS, MAT.. are no longer manufactured. For a long time we purchase x5-x10 quantities and select them manually (not matched transistor arrays are a different thing. here we have only a same temperature for them all, but as we are not able to match, usually transistor arrays are used as switches) PS MAT14 that still remains on sale is for 11 euros, bc850 goes for 10 cents. So... x25.
Thank you for watching and sharing your insights and practical suggestions. You have good points regarding Zener diode and potential workaround using Texas Instrument's TL431 precision shunt regulator. Also, thanks for reminding that many of multi-transistor matched BJT chips are no longer manufactured. I agree that there are better ways to reduce noise and improve power supply rejection ratio (PSRR). To your point, one of the advantages of Zener diode connected between Vsupply and Resistor is better supply noise rejection for the reference current Iref. In the existing configuration in this example with Zener Diode connected below the resistor, given that reference current Iref = (Vs-0.7-Vzener)/R2 then sensitivity = delta_Iref/Iref = delta_Vs/(Vs-0.7-Vzener) which means sensitivity increases if Vs is reduced or if Vz is increased. In this example with Vs=10v, Vzener=4.7v, if 10volt supply is increased 5% to 10.5V, then Iref=(10.5-0.7-4.7)/4.6k = 1.109 mA which means 10.9% increase in reference current for 5% increase in supply voltage. For more examples please see: Voltage Regulator Design with Op Amp and BJT Transistor ruclips.net/video/rI9f6-DyXxQ/видео.html Voltage Regulator Op Amp Circuit with Foldback current limiting ruclips.net/video/VN4_qF9DvBM/видео.html Regulator Design with BJT Darlington Transistors & Zener Diodes ruclips.net/video/ArisQp7V0Ac/видео.html For more analog circuit examples pls see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. I hope these videos are interesting as well. Thanks for watching 🙏
Thanks for sharing your observations. Analog Devices offer the following matched transistor packs www.analog.com/en/parametricsearch/10988 ,, there is also a list when searching in Digikey and also ebay.
Prof: I thought I would provide you a request for a video. Show modeling of springs, dampers, etc in the S-domain as in solving mechanical systems with LTSpice.
Thanks for your interest and for your suggestions. Please see the following videos: Using Laplace Transform to solve spring mass mechanical system transient response ruclips.net/video/I4mqQQ9hSKc/видео.html Double spring mass system analysis using Laplace Transform ruclips.net/video/lCyJGtUK9zs/видео.html S-Domain Double spring-mass mechanical system analysis ruclips.net/video/I5qTyPIo-xg/видео.html Oscillation frequency of Tuning Fork computed using Laplace Transform ruclips.net/video/CLY4f9RZo_U/видео.html I hope these Mechanical system Analysis videos are helpful and interesting 🙂🙏
I'd be curious to know how the output impedance varies over frequency with typical devices. For example, if you're using this as a current sink for a pair of matched transistors that are forming the tail for the two emitters of a differential amplifier, where the differential amplifier's transistors are similar to the transistors in the current mirror. Are you going to hit the upper frequency limits of the differential amplifier before the current mirror becomes a worry, or will the current mirror loses its "stiffness" as a good current sink at a much lower frequency?
Thanks for watching and your good question. Operation Bandwidth of Wilson Current Mirror depend on the choice of Transistor and also whether additional passive or active frequency compensation techniques are used or not. For instance to increase bandwidth we can insert a resistor between Base of T1 and the T2 Diode (BJT transistor in Diode formation). A suggested value for this Resistor is R = 1/gm1. This passive bandwidth compensation technique in addition to choosing proper matched BJT transistors should support high frequency operation. I will post more circuit examples in the Analog Circuits Collection: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. I hope this is helpful. 🙂
It seems to me that the design mirrors the current through R2, which is going to vary directly as a function of supply voltage. If your supply goes up by 4.6V to 14.6V, R2 current would double and your output current would also double. Seems to me this defeats the whole point of aconstant current source, and swapping the locations of R1 and the Zener would make the R2 current a function of only the Zener voltage and not the supply (ignoring beta, temperature, etc). Or am I missing something?
Thanks for watching and sharing your observations and practical considerations. You have valid points. I agree that one of the advantages of Zener diode connected between Vsupply and Resistor is better supply noise rejection for the reference current Iref. In the existing configuration in this example, given that reference current Iref = (Vs-0.7-Vzener)/R2 then sensitivity = delta_Iref/Iref = delta_Vs/(Vs-0.7-Vzener) which means sensitivity increases if Vs is reduced or if Vz is increased. In this example with Vs=10v , Vzener=4.7v , if 10 volt supply is increased 5% to 10.5V, then Iref=(10.5-0.7-4.7)/4.6k = 1.109 mA which means 10.9% increase in reference current for 5% increase in supply voltage. Swapping Diode and R1 would resolve this to a good extent. For more examples please see: Voltage Regulator Design with Op Amp and BJT Transistor ruclips.net/video/rI9f6-DyXxQ/видео.html Voltage Regulator Op Amp Circuit with Foldback current limiting ruclips.net/video/VN4_qF9DvBM/видео.html Regulator Design with BJT Darlington Transistors & Zener Diodes ruclips.net/video/ArisQp7V0Ac/видео.html For more analog circuit examples pls see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. Thanks for watching 🙂
@@STEMprof Thanks for including those links. I'll take a look at them. In my experience, the supply voltage of a circuit tends to be much more variable (due to supply tolerance, loading, transients, and ohmic distribution drop) than any 0.03 percent themal coefficient of the Zener. So it's a bit misleading to talk about how stable and high performance the Zener is, when there are orders-of-magnitude larger sensitivities on unstable inputs that will dominate the ultimate performance. Anyways, thanks for this video, always enjoy watching some circuit analysis!
@@zxborg9681 Thanks again for watching and sharing your insights. I also hope that you like the video Push-Pull Power Amplifier Design with Op Amp, Sziklai Darlington Transistors ruclips.net/video/8BFzsi7-Vbs/видео.html . 🙋♂
why not put the Zener diode in the place of R1 ?? Would not do that better? At least for say half cases. Part the Zener family in positive drift and negative drift, and use one schematic for each.... My proposal, even better than 50% of cases, because it is meant to also cover for the drift of the supply.
Thanks for watching and sharing your thoughts. I agree that one of the advantages of Zener diode connected between Vsupply and Resistor is better supply noise rejection for the reference current Iref. Given that reference current Iref = (Vs-0.7-Vzener)/R2 then sensitivity = delta_Iref/Iref = delta_Vs/(Vs-0.7-Vzener) which means sensitivity increases if Vs is reduced or if Vz is increased. In this example with Vs=10v , Vzener=4.7v , if 10 volt supply is increased 5% to 10.5V, then Iref=(10.5-0.7-4.7)/4.6k = 1.109 mA . For more analog circuit examples pls see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. I hope this is helpful.
@@STEMprof Actually using zener diode like this propagates all voltage supply instabilities (ripple, bad stabilization) to the current source in addition to temperature drift. Usually zener diode is connected in parallel to base junction and emitter resistor. In your circuit it works well only with one supply voltage. Because changing supply voltage cause change voltage on on R2 (Ur2 ~= Ur1 - 0.7) and therefore changes in output current. If you swap R1 and zener you can change supply voltage in much bigger range with much smaller output current change (changin supply voltage with the same R1 cases change of current through the zener and therefore little change voltage across it)
@@dgo42 Thank you for watching and sharing your observations and practical considerations. You have good points. I agree that one of the advantages of Zener diode connected between Vsupply and Resistor is better supply noise rejection for the reference current Iref. In the existing configuration in this example, given that reference current Iref = (Vs-0.7-Vzener)/R2 then sensitivity = delta_Iref/Iref = delta_Vs/(Vs-0.7-Vzener) which means sensitivity increases if Vs is reduced or if Vz is increased. In this example with Vs=10v , Vzener=4.7v , if 10 volt supply is increased 5% to 10.5V, then Iref=(10.5-0.7-4.7)/4.6k = 1.109 mA which means 10.9% increase in reference current for 5% increase in supply voltage. Regarding Thermal sensitivity of emitter voltage: with the selected 4.7v Zener Diode placed as shown in this example, it is currently a decent -0.6mV/C . If we swap the R1 and Zener Diode as you suggested, then I agree that we need negative temperature coefficient for Zener Diode and hopefully close to nominal -2 mV/C of BJT emitter-base voltage. The closest Zener Diode in the 1N52xx Zener Diode family from Vishay is the 1N5228 3.9V Zener with -0.06%V/C temperature coefficient that results in ~ -2.35 mV/C change in Zener Diode nominal 3.9V voltage. This translates to a mere ~ +0.35 mV/C increase in emitter voltage which is slightly better than the -0.6 mV/C of the existing configuration. For more examples please see: Voltage Regulator Design with Op Amp and BJT Transistor ruclips.net/video/rI9f6-DyXxQ/видео.html Voltage Regulator Op Amp Circuit with Foldback current limiting ruclips.net/video/VN4_qF9DvBM/видео.html Regulator Design with BJT Darlington Transistors & Zener Diodes ruclips.net/video/ArisQp7V0Ac/видео.html For more analog circuit examples pls see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. I hope this is helpful. Thanks for watching 🙏
You're welcome. Thanks for watching & your interest. Glad that you like this video. For more Analog Circuits pls see: On-Chip Current Source with BJT ruclips.net/video/Rs7gEMk03dw/видео.html Thermometer Circuit Design with Op Amp & BJT transistor ruclips.net/video/55YsraFE0rg/видео.html Instrumentation Amplifier with Electronic Gain Control ruclips.net/video/C4tghZ-q6Zs/видео.html I am writing my textbook. Examples are not from a specific book. I either design my own examples or they are variants of interesting problems or topics that I have seen in different resources. I use multiple applications. For additional analog circuits and signal processing examples see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt I hope these Circuit videos are interesting. Thanks for watching 🙋♂️
Thanks for watching and your interest in this Wilson Current Source circuit. This can potentially be used in any application that requires a practically constant current sink with very high output impedance. An example application is in ruclips.net/video/5zHXTx-Vl20/видео.html Sawtooth Waveform Generator circuit that requires a current sink which was implemented using an Op Amp. We can replace the current sink used in that Sawtooth Oscillator Circuit with this Wilson Current mirror. I hope this explanation is helpful. 🙋♂
Prof: this circuit is of course very important to understand, so a good topic. However, IMHO your explanation became more clumsy than your other videos. Also note IMHO Vcompliance of 0.2+0.6 is optimistic. I would tend to require at least Vce = 2v on T3. Also my memory seems to suggest that a signal diode in series with the zener may provide better thermal regulation. Do I recall that correctly?
Thanks for watching and sharing your feedback and suggestions. Please let me know which part of this explanation seemed more clumsy than before? How do you think I could have done better? Regarding Voltage compliance, as mentioned in the video the absolute lowest output voltage is 0.8-0.9 volt but it doesn't mean it is recommended to get down to this absolute lowest voltage. My practical recommendation for min outout voltage of this circuit is 1.4 volt. Regarding Zener Series with signal Diode, depends on whether temperature coefficient signs are counteracting or not. It only helps if one has positive and the other negative temp coefs in such a way that resulting delta temp coef is less than the Zener Diode temperature coefficient (abs value). For more analog circuits examples pls see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt I hope this is helpful.
@@STEMprof All forward-biased silicon pn junctions have a temperature coefficient of around -2mV/K, this includes both signal diodes and the emitter-base junction of a transistor. The Zener diode, on the other hand, has two distinct mechanisms of breakdown when reverse biased. At low voltages, the Zener effect predominates, which is quantum tunnelling allowing electrons to tunnel from the valence band to the conduction band. This has a negative temperature coefficient. At higher voltages, avalanche breakdown occurs, which is the acceleration of minority carriers in the depletion region by the high electric field, which then collide with bound electrons to create electron-hole pairs. This has a positive temperature coefficient. In a constant current source/sink, you connect the Zener between the transistor base and the same rail as the emitter resistor is connected to. Then selecting a Zener with a negative temperature coefficient close to -2mV/K will cancel out much of the temperature drift of the transistor Vbe. Alternatively, you may be able to find a Zener whose temperature coefficient is close to zero, in which case adding a silicon diode in series with it will again give a degree of thermal cancellation with the transistor Vbe, keeping the voltage across the emitter resistor relatively independent of temperature.
Thanks for watching and sharing your thoughts. The example Zener Diode in this video has ~ + 0.03%V/C temperature coefficient. So adding silicon signal Diode in series should help. Thanks again.
@@STEMprof Let's do the analysis. In the circuit you presented, the voltage across R2 is given by Vs - Vz - Vbe, and that is the voltage we want to keep constant. So ideally, we want Vz and Vbe to have equal and opposite temperature coefficients. The value for dVz/dT works out to be +1.4mV/K, but dVbe/dT is typically -2mV/K, giving a combined rate of change of the voltage at the emitter of T4 (the pnp transistor) of around 0.6mV/K, which is not bad, but could be improved by changing the Zener to 5.6V, as that has dVz/dT = 2.1mv/K, for a net rate of change of around 0.1mV/K. Of course, sample variations can upset that result, but it's best to aim for that. In this case, adding a small signal diode in series with the Zener makes things worse. However, if we swap the Zener and R1, then the voltage across R2 is now Vz - Vbe, so we want the temperature changes to be equal and in the same direction, i.e. we want the Zener to have a temperature coefficient close to -2mV/K and that's achievable with a low voltage Zener between about 3V and 3.9V, but they would have to be run at 5mA or more to avoid the poor "knee" where the dynamic resistance goes up rapidly at lower currents.
well prof it is like most tasks, going into a task cold results in mistakes along the way that need to be recognized and corrected. if on the other hand you were to practice the lesson first and take notes, then you would be able to present the material in a more linear way without the corrections etc. it would also help if your equations were laid out one after the other without being stuffed into blank spaces in the circuit diagram. basically a more orderly lesson for your students would be powerful. cheers. @@STEMprof
I am a person involved in practical electronics, I am NOT a very good mathematician becauuse I am a visual thinker and as such can’t follow the msthematics to the n th dimension, my ability stops at three dimensions, those that I can visualize in my head. Your channel is very good in the respects that in brings the theory closer to practice and there is a vast unaddressed gulf between these two. I have looked back at your older videos and see it is a mixture of mathematics and electronics. I see older electronics videos with circuits containing op-amps and 3D networks of resistors in cubes or dodecahedrons. This type of thing is only any use as a mathematical excercise, that being to mathematically reduce the complex 3D resistor arrays to a single resistor and as such is only an excercise for students but has no real practical value in the real world. If I wrer presented with such a problem, I’d go to my bench, build the array from real resistors and measure between the assigned nodes with my multimeter on the resistance range…effectively “short-circuiting” all the mathematics that I’m not really good st anyway! What I do see in your overall channel is a trend away from these puerly theoretical maths excercise circuits toward more practical, real world circuits like linear power supplies, class A-B audio amps and the like. I think a good development would be to design a circuit then Actually Build it. Aaron Lanterman does exactly this on his channel. This circuit here, a sutudent deminstration of a Wilson Current Mirror….the circuit that removed many passives from circuit designs allowing them to integrated onto silicon and ushering in the “analog revolution”, actually is far more practically useful than might be apparent from what is dhown here. I will explain by example. Imsgine you have a very wide bandwith signal, an old analog video signal for exanple, a signal with digital, analog and RF components and extending “from DC to Daylight”. Imagine this signal has a DC offset thst one seeks to alter or remove entierly. One could couple the video signal through a capacitor to block the DC component, but since it contains RF, it has to be terminsted in a low impedance do stray reacrances don’t alter its upper frequency components. So a HUGE capacitor is required, more than 1000uF to reduce time-constant effexts on the digital components (sync pulses) of the signal….so almost whatever you choose to do, you will be trading off one problem (DC offset) for another, (integrated sync pulses)! This circuit actually is the “answer” to the issue described above. Firstly it will sink a constant current from the breakdown voltage of Q3, so c45v, right down to about two vbe drops anove ground, (or a negative rail of you choose it as the circuit reference). This circuit slso has the very high output impedance as so elegantly demonstrated, 50Meg! So if you reference this circuit to a Quiet -5v rail and use it to pull current from a known low source impedance video signal, it will pull a precisely defined current, 1mA via the source impedance, (usually 75 Ohms foe video) and this it will move the video signal down by 75mv with no descernable distortions or alterations bar small juction capacitances of the collector-base junction of Q3 attenuating the high RF components of the video signal slightly. If this circuit were to be configured as a VOLTAGE CONTROLLED current source/sink, then the DC component of the video signal could be altered st will and without need for use of capacitors. In conjunction with low pass filters, precision rectifiers and the like, complex circuits like black level clamps could be realized. Admittedly analog video is now dead, but what excellent circuits for demonstration. This circuit could slso be used in sawtooth oscillators, voltage adjustable one-shots and a myriad of practical uses. One question I would always ask thoriticians, “But what is it for, (besides demonstrating the mathematics.)! I would LOVE to see detailled videos that bridge the theoretical and the practical going all the way from Laplace Transforms and how they are used to design servo feedback loops in things like the boost converter in my LCD screen TV that drives the backlight LEDs! There are loads of channels of people fixing TVs, there are loads if channels of prople doing interesting and abstract mathematics, but almost nothing to bridge the gap between them. Here, on this channel I see a trend from the abstract to the practical and would love to see it extrapolated all the way! Cheers, “The Globe Collector”.
Thank you for watching, your interest in this channel and for sharing your thoughts and observations. I also appreciate your detailed and encouraging comment. I will try to post a reasonable balance of interesting practical and theoretical topics and videos. Please see following further examples given your mentioned topics of interest: Sawtooth Oscillator Design ruclips.net/video/2eUsGPfqbW4/видео.html Triangle Oscillator Op Amp circuit ruclips.net/video/JF5Up_cuL9k/видео.html Sallen-Key Analog Filter Design Tutorial ruclips.net/video/KwUnQXbk7gM/видео.html How to find Bode Plot, Freq Response, Transfer Function of Analog Filters ruclips.net/video/vZFkPeDa1H8/видео.html Universal Analog Filter Design ruclips.net/video/2J-0msXZE2o/видео.html Laplace Transform Example and S-domain circuit analysis: ruclips.net/video/ps8N5TPM_qU/видео.html I hope you find these practical Analog Circuit Design and Analysis videos useful and interesting. Thanks again 🙏
Thanks for watching. Glad that you like this Wilson Current Mirror Tutorial video. For more analog circuit examples see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. I hope these Circuit design videos are helpful. 🙂
Thanks for watching. For more Op Amp Circuits please see:
Temperature-Independent Current Circuit Design with Op Amp, BJT, Zener and Schottky Diodes ruclips.net/video/hFbnjbddUvs/видео.html
On-Chip Current Source with BJT Transistors ruclips.net/video/Rs7gEMk03dw/видео.html
Thermometer Circuit Design with Op Amp & BJT transistor ruclips.net/video/55YsraFE0rg/видео.html
Push-Pull Power Amplifier Design with Op Amp, Sziklai Darlington Transistors ruclips.net/video/8BFzsi7-Vbs/видео.html
Instrumentation Amplifier with Electronic Gain Control ruclips.net/video/C4tghZ-q6Zs/видео.html
Push-Pull Power Amplifier with Darlington Transistors ruclips.net/video/866MYibo8yE/видео.html
Analog Logarithm Computer ruclips.net/video/RpKEq5WyoLg/видео.html
Op Amp Analog Computer Differential Equation Solver ruclips.net/video/ENq39EesfPw/видео.html
Sawtooth Oscillator Design ruclips.net/video/2eUsGPfqbW4/видео.html
Triangle Oscillator Op Amp circuit ruclips.net/video/JF5Up_cuL9k/видео.html
Sallen-Key Analog Filter Design Tutorial ruclips.net/video/KwUnQXbk7gM/видео.html
How to find Bode Plot, Freq Response, Transfer Function of Analog Filters ruclips.net/video/vZFkPeDa1H8/видео.html
Universal Analog Filter Design ruclips.net/video/2J-0msXZE2o/видео.html
Laplace Transform Example and S-domain circuit analysis: ruclips.net/video/ps8N5TPM_qU/видео.html
Op Amp circuit Bode Frequency plot ruclips.net/video/BLVzuuqAlZs/видео.html
Lowpass Butterworth Filter: ruclips.net/video/UzCjkwqy-9w/видео.html
Analog Computer to Raise Signal to power n ruclips.net/video/IUTlBH1UraE/видео.html
Differential Equation Solver Analog Circuit ruclips.net/video/R3X5AYNZGEI/видео.html
Complex Sinusoid Oscillator ruclips.net/video/GXRhmwmS5Zk/видео.html
Full-Wave Rectifier circuit example ruclips.net/video/DJJMNU-CYcg/видео.html
Sawtooth Waveform Generator design with OpAmp, JFET, BJT ruclips.net/video/5zHXTx-Vl20/видео.html
op amps Circuit with feedback loops to design an analog computer that solves a second order differential equation ruclips.net/video/HeZRtnRXpEI/видео.html
For more analog circuits and signal processing examples see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt
I hope these Circuit design and analysis videos are helpful. 🙂
Nice educational current mirror analysis and explanation video!
@@ЁбрагимИпатенкоибнАдхарма Thank you! Glad that you liked this Wilson Current Source Circuit video. Alternative design techniques are discussed in ruclips.net/p/PLrwXF7N522y48AAPxFaQlowim4-8gYoWz
Hope this circuit playlist is interesting as well.
Hello Engineering Prof, Thanks for your clear and detailed explanation of analog electronic circuits. Look forward to more internal building blocks and circuitries of analog IC from you.
You are welcome. Thanks for watching, your interest & encouraging comment. Glad that you like this Wilson Current Mirror Tutorial video. For more analog circuit examples please see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. I hope these Circuit videos are useful as well. 🙋♂️
Swap the position of R1 and Zener is more reasonable to avoid the Vs disturbance
Thanks for watching and sharing your thoughts and suggestions. Whether to have Zener Diode positioned above the resistor or below the resistor you will end up with its own set of advantages and disadvantages. But I agree that one of the advantages of Zener diode connected between Vsupply and Resistor is better supply noise rejection for the reference current Iref. For more regulator circuits please see ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt . Thanks again.
@@STEMprof Actually, swapping the Zener and R1 will massively improve supply noise rejection.
In the circuit you show, the current Iref is given by (Vs - Vz - Vbe) / R2, which depends linearly on Vs.
If you swap the Zener and R1, Iref is given by (Vz - Vbe) / R2, which is independent of Vs.
You then use a lower value Zener (3.3V to 3.9V), which has a temperature coefficient around -2.3mV/K which would cancel out most of the temp coefficient of the PNP's Vbe.
@@RexxSchneider Thanks again for watching & sharing your thoughts and suggestions.
Agree, Zener diode as shown amplifies impact of Vsupply on Ireference (5% change of Vsupply causes ~10% change of Iref). It is better to swap R1 and Zener diode. P.S. The video is great
@@JumpingJack-w2l Thank you for watching & sharing your thoughts. Glad that you like this Wilson Current Mirror Tutorial video. Given that reference current Iref = (Vs-0.7-Vzener)/R2 then sensitivity = delta_Iref/Iref = delta_Vs/(Vs-0.7-Vzener) which means sensitivity increases if Vs is reduced or if Vz is increased. In this example with Vs=10v , Vzener=4.7v , if 10 volt supply is increased 5% to 10.5V, then Iref=(10.5-0.7-4.7)/4.6k = 1.109 mA indicating 10.9% increase in Iref as you noted. For more analog circuit examples pls see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. Thanks again 🙋♂️
Thank you for another excellent video
You're welcome. Thanks for watching. Glad that this Wilson Current Mirror Tutorial video is useful. For more analog circuit examples pls see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. Thanks again 🙂
I would exchange the Zener and R1 so the set current is not so dependent on the supply voltage.
Thanks for sharing your thoughts. I agree that one of the advantages of Zener diode connected between Vsupply and Resistor is better supply noise rejection for the reference current Iref. Given that reference current Iref = (Vs-0.7-Vzener)/R2 then sensitivity = delta_Iref/Iref = delta_Vs/(Vs-0.7-Vzener) which means sensitivity increases if Vs is reduced or if Vz is increased. In this example with Vs=10v , Vzener=4.7v , if 10 volt supply is increased 5% to 10.5V, then Iref=(10.5-0.7-4.7)/4.6k = 1.109 mA reflecting 10+% increase in current. Alternative design techniques are discussed in the circuit videos ruclips.net/p/PLrwXF7N522y48AAPxFaQlowim4-8gYoWz
Hope this circuit playlist is interesting.
How did u get equation Ir(B^2+2B/B^2+2B+2)? 19:00
Thanks for watching & your follow-up question. Substituting Ic in equation 1 using equation 2 results in the equation shown at 17:40 , and then a brief algebraic operation results in finding output current Io as a function of the reference current Iref in the form of Io = Iref * (B^2+2B)/(B^2+2B+2) . I hope this is helpful.
@@STEMprof oh okay thank you, electronics is so difficult its crazy how smart you are that you know how this stuff works
You're welcome. Thank you for the comment. Glad that my explanation was helpful. 🙂 For more analog circuit examples please see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. I hope these Circuit design/analysis videos are helpful as well.
Swap R1 and the Zenerdiode to get a constant current source that don't change the reference crurrent if the Power Supply voltage is changing..
Thanks for watching and sharing your thoughts. I agree that one of the advantages of Zener diode connected between Vsupply and Resistor is better supply noise rejection for the reference current Iref. For more examples please see:
Voltage Regulator Design with Op Amp and BJT Transistor ruclips.net/video/rI9f6-DyXxQ/видео.html
Voltage Regulator Op Amp Circuit with Foldback current limiting ruclips.net/video/VN4_qF9DvBM/видео.html
Regulator Design with BJT Darlington Transistors & Zener Diodes ruclips.net/video/ArisQp7V0Ac/видео.html
For more analog circuit examples pls see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. I hope these circuit design & analysis videos are helpful. Thanks for watching 🙏
Prof: thought I might open a conversation with you regarding circuit design and the way I was taught. (did not say I learned ;-})
In school we young aspirational student were taught circuit analysis via KVA: Loop currents:etc.
Ok well and good.
Then we graduated, and went to work.
Now on occasion we faced the problem of generating new circuits.
But for this we were not well prepared, yes analysis tools we had,
but not real 'generation of new circuits" methods.
---
I think what was missing and could have been included was an extensive discussion
about how to think of transistors, Rs,Cs,Ls,... in a way that maps the components into
behavioral models such as used in Spice. In this way using models like current sources, voltage sources and the controlled versions of each, one could "construct" a new circuit functionally and then start by including real world components with the specific constraints such as temperature variance, input currents,... and all the rest of the "real" component selection problems.
Personally I had an extensive technical training prior to going to engineering school - so I had some basis or understanding of many circuits and was able to use that to develop new ideas. Sometimes good ideas, sometimes not.
You may at some time feel that you have analyzed a sufficient number of typologies and might be interested in demonstrating your own method of generating new circuits.
I for one would be interested and I encourage you to consider doing so.
cheers d
Thank you for your interest in this channel and for sharing your thoughts. I appreciate. To your point, a good number of circuits in this channel are actually my own designs. I will post more new design examples in my Analog and Op Amp Circuits playlist ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt
I hope you find them interesting. 🙂
A zenner diode is a noisy device, as well as a LED. So, better results will be with an additional simple transistor that will give 0.6v for R2. That's enough. Or we can use two sequential diodes in a same sot-23-4/sot-23-3 package. The second thing is that it's better to have a Vsupply independent current (or in other words to increase a PSRR), and so the zenner must go first, and then R1. If we still need a higher reference voltage, tl431 is a good alternative to a zenner.
FYI x4 matched(!) transistors chips like MMPQ, TAS, MAT.. are no longer manufactured. For a long time we purchase x5-x10 quantities and select them manually (not matched transistor arrays are a different thing. here we have only a same temperature for them all, but as we are not able to match, usually transistor arrays are used as switches)
PS MAT14 that still remains on sale is for 11 euros, bc850 goes for 10 cents. So... x25.
Thank you for watching and sharing your insights and practical suggestions. You have good points regarding Zener diode and potential workaround using Texas Instrument's TL431 precision shunt regulator. Also, thanks for reminding that many of multi-transistor matched BJT chips are no longer manufactured. I agree that there are better ways to reduce noise and improve power supply rejection ratio (PSRR). To your point, one of the advantages of Zener diode connected between Vsupply and Resistor is better supply noise rejection for the reference current Iref. In the existing configuration in this example with Zener Diode connected below the resistor, given that reference current Iref = (Vs-0.7-Vzener)/R2 then sensitivity = delta_Iref/Iref = delta_Vs/(Vs-0.7-Vzener) which means sensitivity increases if Vs is reduced or if Vz is increased. In this example with Vs=10v, Vzener=4.7v, if 10volt supply is increased 5% to 10.5V, then Iref=(10.5-0.7-4.7)/4.6k = 1.109 mA which means 10.9% increase in reference current for 5% increase in supply voltage. For more examples please see: Voltage Regulator Design with Op Amp and BJT Transistor ruclips.net/video/rI9f6-DyXxQ/видео.html
Voltage Regulator Op Amp Circuit with Foldback current limiting ruclips.net/video/VN4_qF9DvBM/видео.html
Regulator Design with BJT Darlington Transistors & Zener Diodes ruclips.net/video/ArisQp7V0Ac/видео.html
For more analog circuit examples pls see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. I hope these videos are interesting as well. Thanks for watching 🙏
Quad matched packages are almost non-existent, I agree. But it's possible to find some dual ones around.
Thanks for sharing your observations. Analog Devices offer the following matched transistor packs www.analog.com/en/parametricsearch/10988 ,, there is also a list when searching in Digikey and also ebay.
Prof: I thought I would provide you a request for a video.
Show modeling of springs, dampers, etc in the S-domain as in solving mechanical systems
with LTSpice.
Thanks for your interest and for your suggestions. Please see the following videos: Using Laplace Transform to solve spring mass mechanical system transient response ruclips.net/video/I4mqQQ9hSKc/видео.html
Double spring mass system analysis using Laplace Transform ruclips.net/video/lCyJGtUK9zs/видео.html
S-Domain Double spring-mass mechanical system analysis ruclips.net/video/I5qTyPIo-xg/видео.html
Oscillation frequency of Tuning Fork computed using Laplace Transform ruclips.net/video/CLY4f9RZo_U/видео.html
I hope these Mechanical system Analysis videos are helpful and interesting 🙂🙏
I'd be curious to know how the output impedance varies over frequency with typical devices. For example, if you're using this as a current sink for a pair of matched transistors that are forming the tail for the two emitters of a differential amplifier, where the differential amplifier's transistors are similar to the transistors in the current mirror. Are you going to hit the upper frequency limits of the differential amplifier before the current mirror becomes a worry, or will the current mirror loses its "stiffness" as a good current sink at a much lower frequency?
Thanks for watching and your good question. Operation Bandwidth of Wilson Current Mirror depend on the choice of Transistor and also whether additional passive or active frequency compensation techniques are used or not. For instance to increase bandwidth we can insert a resistor between Base of T1 and the T2 Diode (BJT transistor in Diode formation). A suggested value for this Resistor is R = 1/gm1. This passive bandwidth compensation technique in addition to choosing proper matched BJT transistors should support high frequency operation. I will post more circuit examples in the Analog Circuits Collection: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt.
I hope this is helpful. 🙂
It seems to me that the design mirrors the current through R2, which is going to vary directly as a function of supply voltage. If your supply goes up by 4.6V to 14.6V, R2 current would double and your output current would also double. Seems to me this defeats the whole point of aconstant current source, and swapping the locations of R1 and the Zener would make the R2 current a function of only the Zener voltage and not the supply (ignoring beta, temperature, etc). Or am I missing something?
Thanks for watching and sharing your observations and practical considerations. You have valid points. I agree that one of the advantages of Zener diode connected between Vsupply and Resistor is better supply noise rejection for the reference current Iref. In the existing configuration in this example, given that reference current Iref = (Vs-0.7-Vzener)/R2 then sensitivity = delta_Iref/Iref = delta_Vs/(Vs-0.7-Vzener) which means sensitivity increases if Vs is reduced or if Vz is increased. In this example with Vs=10v , Vzener=4.7v , if 10 volt supply is increased 5% to 10.5V, then Iref=(10.5-0.7-4.7)/4.6k = 1.109 mA which means 10.9% increase in reference current for 5% increase in supply voltage. Swapping Diode and R1 would resolve this to a good extent. For more examples please see:
Voltage Regulator Design with Op Amp and BJT Transistor ruclips.net/video/rI9f6-DyXxQ/видео.html
Voltage Regulator Op Amp Circuit with Foldback current limiting ruclips.net/video/VN4_qF9DvBM/видео.html
Regulator Design with BJT Darlington Transistors & Zener Diodes ruclips.net/video/ArisQp7V0Ac/видео.html
For more analog circuit examples pls see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. Thanks for watching 🙂
@@STEMprof Thanks for including those links. I'll take a look at them. In my experience, the supply voltage of a circuit tends to be much more variable (due to supply tolerance, loading, transients, and ohmic distribution drop) than any 0.03 percent themal coefficient of the Zener. So it's a bit misleading to talk about how stable and high performance the Zener is, when there are orders-of-magnitude larger sensitivities on unstable inputs that will dominate the ultimate performance. Anyways, thanks for this video, always enjoy watching some circuit analysis!
@@zxborg9681 Thanks again for watching and sharing your insights. I also hope that you like the video Push-Pull Power Amplifier Design with Op Amp, Sziklai Darlington Transistors ruclips.net/video/8BFzsi7-Vbs/видео.html . 🙋♂
why not put the Zener diode in the place of R1 ?? Would not do that better? At least for say half cases. Part the Zener family in positive drift and negative drift, and use one schematic for each....
My proposal, even better than 50% of cases, because it is meant to also cover for the drift of the supply.
Thanks for watching and sharing your thoughts. I agree that one of the advantages of Zener diode connected between Vsupply and Resistor is better supply noise rejection for the reference current Iref. Given that reference current Iref = (Vs-0.7-Vzener)/R2 then sensitivity = delta_Iref/Iref = delta_Vs/(Vs-0.7-Vzener) which means sensitivity increases if Vs is reduced or if Vz is increased. In this example with Vs=10v , Vzener=4.7v , if 10 volt supply is increased 5% to 10.5V, then Iref=(10.5-0.7-4.7)/4.6k = 1.109 mA . For more analog circuit examples pls see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. I hope this is helpful.
@@STEMprof Actually using zener diode like this propagates all voltage supply instabilities (ripple, bad stabilization) to the current source in addition to temperature drift. Usually zener diode is connected in parallel to base junction and emitter resistor. In your circuit it works well only with one supply voltage. Because changing supply voltage cause change voltage on on R2 (Ur2 ~= Ur1 - 0.7) and therefore changes in output current. If you swap R1 and zener you can change supply voltage in much bigger range with much smaller output current change (changin supply voltage with the same R1 cases change of current through the zener and therefore little change voltage across it)
@@dgo42 Thank you for watching and sharing your observations and practical considerations. You have good points. I agree that one of the advantages of Zener diode connected between Vsupply and Resistor is better supply noise rejection for the reference current Iref. In the existing configuration in this example, given that reference current Iref = (Vs-0.7-Vzener)/R2 then sensitivity = delta_Iref/Iref = delta_Vs/(Vs-0.7-Vzener) which means sensitivity increases if Vs is reduced or if Vz is increased. In this example with Vs=10v , Vzener=4.7v , if 10 volt supply is increased 5% to 10.5V, then Iref=(10.5-0.7-4.7)/4.6k = 1.109 mA which means 10.9% increase in reference current for 5% increase in supply voltage. Regarding Thermal sensitivity of emitter voltage: with the selected 4.7v Zener Diode placed as shown in this example, it is currently a decent -0.6mV/C . If we swap the R1 and Zener Diode as you suggested, then I agree that we need negative temperature coefficient for Zener Diode and hopefully close to nominal -2 mV/C of BJT emitter-base voltage. The closest Zener Diode in the 1N52xx Zener Diode family from Vishay is the 1N5228 3.9V Zener with -0.06%V/C temperature coefficient that results in ~ -2.35 mV/C change in Zener Diode nominal 3.9V voltage. This translates to a mere ~ +0.35 mV/C increase in emitter voltage which is slightly better than the -0.6 mV/C of the existing configuration. For more examples please see:
Voltage Regulator Design with Op Amp and BJT Transistor ruclips.net/video/rI9f6-DyXxQ/видео.html
Voltage Regulator Op Amp Circuit with Foldback current limiting ruclips.net/video/VN4_qF9DvBM/видео.html
Regulator Design with BJT Darlington Transistors & Zener Diodes ruclips.net/video/ArisQp7V0Ac/видео.html
For more analog circuit examples pls see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. I hope this is helpful. Thanks for watching 🙏
Excellent explanation as always, thank you!! My friend what app do you use for writing?
Are those examples are from Sedra's book??
You're welcome. Thanks for watching & your interest. Glad that you like this video. For more Analog Circuits pls see: On-Chip Current Source with BJT ruclips.net/video/Rs7gEMk03dw/видео.html
Thermometer Circuit Design with Op Amp & BJT transistor ruclips.net/video/55YsraFE0rg/видео.html
Instrumentation Amplifier with Electronic Gain Control ruclips.net/video/C4tghZ-q6Zs/видео.html
I am writing my textbook. Examples are not from a specific book. I either design my own examples or they are variants of interesting problems or topics that I have seen in different resources. I use multiple applications. For additional analog circuits and signal processing examples see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt
I hope these Circuit videos are interesting. Thanks for watching 🙋♂️
@@STEMprof I mean what application do you use for writing in your device?
Thanks for your interest. Yes, I replied in my earlier response. Please see my earlier comment.
What is usually connected to a circuit like this and where in the circuit is it connected to?
Thanks for watching and your interest in this Wilson Current Source circuit. This can potentially be used in any application that requires a practically constant current sink with very high output impedance. An example application is in ruclips.net/video/5zHXTx-Vl20/видео.html Sawtooth Waveform Generator circuit that requires a current sink which was implemented using an Op Amp. We can replace the current sink used in that Sawtooth Oscillator Circuit with this Wilson Current mirror. I hope this explanation is helpful. 🙋♂
@@STEMprof thank you very much, yes this was helpful 🙂
Prof: this circuit is of course very important to understand, so a good topic.
However, IMHO your explanation became more clumsy than your other videos.
Also note IMHO Vcompliance of 0.2+0.6 is optimistic.
I would tend to require at least Vce = 2v on T3.
Also my memory seems to suggest that a signal diode in series with the zener may
provide better thermal regulation.
Do I recall that correctly?
Thanks for watching and sharing your feedback and suggestions. Please let me know which part of this explanation seemed more clumsy than before? How do you think I could have done better? Regarding Voltage compliance, as mentioned in the video the absolute lowest output voltage is 0.8-0.9 volt but it doesn't mean it is recommended to get down to this absolute lowest voltage. My practical recommendation for min outout voltage of this circuit is 1.4 volt. Regarding Zener Series with signal Diode, depends on whether temperature coefficient signs are counteracting or not. It only helps if one has positive and the other negative temp coefs in such a way that resulting delta temp coef is less than the Zener Diode temperature coefficient (abs value). For more analog circuits examples pls see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt
I hope this is helpful.
@@STEMprof All forward-biased silicon pn junctions have a temperature coefficient of around -2mV/K, this includes both signal diodes and the emitter-base junction of a transistor. The Zener diode, on the other hand, has two distinct mechanisms of breakdown when reverse biased. At low voltages, the Zener effect predominates, which is quantum tunnelling allowing electrons to tunnel from the valence band to the conduction band. This has a negative temperature coefficient. At higher voltages, avalanche breakdown occurs, which is the acceleration of minority carriers in the depletion region by the high electric field, which then collide with bound electrons to create electron-hole pairs. This has a positive temperature coefficient.
In a constant current source/sink, you connect the Zener between the transistor base and the same rail as the emitter resistor is connected to. Then selecting a Zener with a negative temperature coefficient close to -2mV/K will cancel out much of the temperature drift of the transistor Vbe.
Alternatively, you may be able to find a Zener whose temperature coefficient is close to zero, in which case adding a silicon diode in series with it will again give a degree of thermal cancellation with the transistor Vbe, keeping the voltage across the emitter resistor relatively independent of temperature.
Thanks for watching and sharing your thoughts. The example Zener Diode in this video has ~ + 0.03%V/C temperature coefficient. So adding silicon signal Diode in series should help. Thanks again.
@@STEMprof Let's do the analysis. In the circuit you presented, the voltage across R2 is given by Vs - Vz - Vbe, and that is the voltage we want to keep constant. So ideally, we want Vz and Vbe to have equal and opposite temperature coefficients. The value for dVz/dT works out to be +1.4mV/K, but dVbe/dT is typically -2mV/K, giving a combined rate of change of the voltage at the emitter of T4 (the pnp transistor) of around 0.6mV/K, which is not bad, but could be improved by changing the Zener to 5.6V, as that has dVz/dT = 2.1mv/K, for a net rate of change of around 0.1mV/K. Of course, sample variations can upset that result, but it's best to aim for that. In this case, adding a small signal diode in series with the Zener makes things worse.
However, if we swap the Zener and R1, then the voltage across R2 is now Vz - Vbe, so we want the temperature changes to be equal and in the same direction, i.e. we want the Zener to have a temperature coefficient close to -2mV/K and that's achievable with a low voltage Zener between about 3V and 3.9V, but they would have to be run at 5mA or more to avoid the poor "knee" where the dynamic resistance goes up rapidly at lower currents.
well prof it is like most tasks, going into a task cold results in mistakes along the way that need to be recognized and corrected. if on the other hand you were to practice the lesson first and take notes, then you would be able to present the material in a more linear way without the corrections etc. it would also help if your equations were laid out one after the other without being stuffed into blank spaces in the circuit diagram. basically a more orderly lesson for your students would be powerful. cheers. @@STEMprof
I am a person involved in practical electronics, I am NOT a very good mathematician becauuse I am a visual thinker and as such can’t follow the msthematics to the n th dimension, my ability stops at three dimensions, those that I can visualize in my head.
Your channel is very good in the respects that in brings the theory closer to practice and there is a vast unaddressed gulf between these two. I have looked back at your older videos and see it is a mixture of mathematics and electronics. I see older electronics videos with circuits containing op-amps and 3D networks of resistors in cubes or dodecahedrons. This type of thing is only any use as a mathematical excercise, that being to mathematically reduce the complex 3D resistor arrays to a single resistor and as such is only an excercise for students but has no real practical value in the real world. If I wrer presented with such a problem, I’d go to my bench, build the array from real resistors and measure between the assigned nodes with my multimeter on the resistance range…effectively “short-circuiting” all the mathematics that I’m not really good st anyway!
What I do see in your overall channel is a trend away from these puerly theoretical maths excercise circuits toward more practical, real world circuits like linear power supplies, class A-B audio amps and the like. I think a good development would be to design a circuit then Actually Build it. Aaron Lanterman does exactly this on his channel.
This circuit here, a sutudent deminstration of a Wilson Current Mirror….the circuit that removed many passives from circuit designs allowing them to integrated onto silicon and ushering in the “analog revolution”, actually is far more practically useful than might be apparent from what is dhown here. I will explain by example.
Imsgine you have a very wide bandwith signal, an old analog video signal for exanple, a signal with digital, analog and RF components and extending “from DC to Daylight”. Imagine this signal has a DC offset thst one seeks to alter or remove entierly. One could couple the video signal through a capacitor to block the DC component, but since it contains RF, it has to be terminsted in a low impedance do stray reacrances don’t alter its upper frequency components. So a HUGE capacitor is required, more than 1000uF to reduce time-constant effexts on the digital components (sync pulses) of the signal….so almost whatever you choose to do, you will be trading off one problem (DC offset) for another, (integrated sync pulses)!
This circuit actually is the “answer” to the issue described above. Firstly it will sink a constant current from the breakdown voltage of Q3, so c45v, right down to about two vbe drops anove ground, (or a negative rail of you choose it as the circuit reference). This circuit slso has the very high output impedance as so elegantly demonstrated, 50Meg! So if you reference this circuit to a Quiet -5v rail and use it to pull current from a known low source impedance video signal, it will pull a precisely defined current, 1mA via the source impedance, (usually 75 Ohms foe video) and this it will move the video signal down by 75mv with no descernable distortions or alterations bar small juction capacitances of the collector-base junction of Q3 attenuating the high RF components of the video signal slightly. If this circuit were to be configured as a VOLTAGE CONTROLLED current source/sink, then the DC component of the video signal could be altered st will and without need for use of capacitors. In conjunction with low pass filters, precision rectifiers and the like, complex circuits like black level clamps could be realized. Admittedly analog video is now dead, but what excellent circuits for demonstration.
This circuit could slso be used in sawtooth oscillators, voltage adjustable one-shots and a myriad of practical uses.
One question I would always ask thoriticians, “But what is it for, (besides demonstrating the mathematics.)!
I would LOVE to see detailled videos that bridge the theoretical and the practical going all the way from Laplace Transforms and how they are used to design servo feedback loops in things like the boost converter in my LCD screen TV that drives the backlight LEDs!
There are loads of channels of people fixing TVs, there are loads if channels of prople doing interesting and abstract mathematics, but almost nothing to bridge the gap between them. Here, on this channel I see a trend from the abstract to the practical and would love to see it extrapolated all the way!
Cheers, “The Globe Collector”.
Thank you for watching, your interest in this channel and for sharing your thoughts and observations. I also appreciate your detailed and encouraging comment. I will try to post a reasonable balance of interesting practical and theoretical topics and videos. Please see following further examples given your mentioned topics of interest: Sawtooth Oscillator Design ruclips.net/video/2eUsGPfqbW4/видео.html
Triangle Oscillator Op Amp circuit ruclips.net/video/JF5Up_cuL9k/видео.html
Sallen-Key Analog Filter Design Tutorial ruclips.net/video/KwUnQXbk7gM/видео.html
How to find Bode Plot, Freq Response, Transfer Function of Analog Filters ruclips.net/video/vZFkPeDa1H8/видео.html
Universal Analog Filter Design ruclips.net/video/2J-0msXZE2o/видео.html
Laplace Transform Example and S-domain circuit analysis: ruclips.net/video/ps8N5TPM_qU/видео.html
I hope you find these practical Analog Circuit Design and Analysis videos useful and interesting.
Thanks again 🙏
❤❤❤❤
Thanks for watching. Glad that you like this Wilson Current Mirror Tutorial video. For more analog circuit examples see: ruclips.net/p/PLrwXF7N522y4c7c-8KBjrwd7IyaZfWxyt. I hope these Circuit design videos are helpful. 🙂