Also, what I love about your teachings and schematics is that the different sections are broken up into pieces. I'm a software dev, and of course that's what we do, but never really thought about it for analog circuits. I'm starting to recognize different common patterns (even said to myself "current mirror" when first looking at it).
This is so cool! As an undergraduate EE student i often get tired of such an intense course but your videos inspire me to carry on! Amazing stuff as usual. Thank you again!
Curve tracers seem to have gone the way of the dodo, and any that you can find are the same price as a decent car, despite it being such a useful thing for people tinkering with basic electronics. I, for one, would love to see you design one from scratch, that would maybe be an advanced project, but doable for the average mortal who already has a scope. I find your tutorials so easy to follow that I would definitely try to build one.
I'm going through an EE program right now, and your practical videos (with all the little tips and tricks from experience) are a boon! Keep up the great vids!
+jason cater Glad to hear it. Good luck with your studies, and be sure to let me know if there is a topic or two for future videos that would help you and your fellow students.
I'm doing some T.A. work with the EE 101 class currently and your LM555 and Op Amp Vids have been invaluable! However, for my regular classes we are building amplifiers (audio range and then moving into RF) and one thing that hasn't been addressed is causes of noise and distortion. I am aware that this is a HUGE subject that covers volumes of books, but there isn't much talk about the practical "gotchas" (improper loading between stages, stray RF, building on a bread board VS PCB, etc.) and simple mistakes that can be made while designing and building BJT amplifiers. A video that addresses the most common types of interference/noise in BJT amplifiers (and how to avoid them) would be awesome (IMHO). Thanks again for all your work!
+jason cater Lots of good topics. Noise is a BIG topic. Sources of distortion in BJT amps is usually a matter of signal level and bias conditions. Feedback almost always helps with linearity. Try not to make too much gain in any one stage - that will usually help with linearity / distortion. Anything besides Class A add distortion by design, which is often reduced by filtering.
Great!! it works for me, I'm not EE of any degree I just like electronics and study one year at high school some technical basics of electronics, back at my younger days, thanks for sharing your knowledge
Thanks Alan for your like, I'm an old guy that likes electronics wish is my pasión and I don't leave these opportunities without trading what I anjoy most
These videos are alot of fun. You make my scope more and more valuable as I learn from these videos. I only wish I had your experience to design these circuits. Thanks!
Many thanks Allen for this very instructive video. You would do me a great favor in continuing on with the ability to test pnp and p-mosfet transistors. Regards, RJM
I'm making this circuit at the moment, I have the stairstep circuit working I need to add some potentiometers so I have the option to set some variations. I want to be able to set the frequency higher and lower, now it oscillates at 185Hz exactly. I also want to be able to set the amplitude and if I can add the ability to input a DC offset that would be great. I'm also adding a switch with the resistor for driving the gate of a MOSFET and the resistor to ground so I can do both and don't have to change the circuit every time. First I'm making the circuit on a breadboard so I know for sure that it works with the components that I'm using and then I'll make it more permanent on a perf board and solder it all together. I already made a different function generator using two 555 timers that have the output synchronized but the second chip outputs the signal inverted. On that function generator I can set the duty cycle from about 20% to 80% at the moment so that could use some improvement, I can set the frequency from 1Hz up to about 450KHz, I can also change the amplitude of both outputs and all the settings work independent so changing one setting doesn't change the other ones. So when I have this circuit working I'm going to try and put all the different function generators together on one single PCB, it would be nice if I can use the timer chips and the op amps for the different purposes to keep the component count a bit lower but if I can't get it to work it doesn't really matter because I already have all the parts and those are pretty cheap but I don't want to use up everything I have for this function generator, if I can save some components I don't have to buy new ones when I want to start tinkering around and use it for something else. w2aew, your circuits took a long time for me to really understand what is happening but I've learned a LOT just from watching your videos and making the circuits myself, because there are some differences in the components that I'm using it was a challenge for me to get my circuits to work properly but as you already said it is the best way to learn this stuff. I wish you have a great day w2aew.
I was expecting a more exponential rise of the C-charge, but in this way (early cut-off) it it quite linear. Otherwise a simple current source would do a linear work. But maybe it doesn't matter at all, as we are plotting the resulting collector current on the y-axis, it really doesn't matter in what fashion the Vce is changing, as long as it is incremental, I suppose. We just get more or less points in some regions.... Once again interesting circuit 👍
Este circuito era exatamente o que estava procurando. Encontrei aqui por acaso..... kkkk...vou fazer este experimento. Obrigado por disponibilizar o circuito e colocar nele a referencia que o inspirou em fazer o experimento. (Brasil Português ) This circuit was exactly what I was looking for. I found here by chance ..... kkkk ... I'm going to do this experiment. Thank you for making the circuit available and putting a reference on it that inspires you to do the experiment. (Brazil language Portuguese)
nice video, there are certainly a number of ways to generate a starecase. I like the analog aproach since it is easy to alter the number of steps. Thanks and thumbs up!
Very nice Alan. Enjoyed the video. I have a old restored Eico 460 scope sitting around that I have been thinking about adding a curve tracer circuit to.
Excellent! For a hobbyist like me, that's gonna be more than enough. Certainly better than just picking a pair of transistors with a similar gain at a fixed current.
10:15 "...if I was going to set out to build a curve tracer I probably wouldn't start with this..." Why? Is it unnecessarily complicated, inaccurate, or ineffective? What is the ideal design (or design elements to seek out) for someone with an entry level digital scope? I've seen #197 and #49, your two other videos on curve tracers. I probably don't "need" a curve tracer, but I feel like I learn a lot more about practical electronics by building circuits like this. Plus, no one complains about having too many tools to work on projects ;) -Jake
I probably wouldn't start with this mainly because it is a bit inflexible in setting up step sizes, etc. I'd probably start with some kind of microcontroller and a few DACs, etc. I would bet there are a few kits out there...
As always I learn things from you. I do electronic repair for a living I can think of a few occasions this could have been helpful. Wish I had taken college when I had the time and funds to do it. You should pursue this and design and build one for multiple voltages and types. A short video on how someone could modify it to be usable over a broad spectrum on transistors, NPN/PNP/MOSFETs etc. would be interesting for sure. Thanks for the Video. Best Wishes and Blessings Keith
Enjoyed your video Allen! Is there a curve tracer that sweeps from 60~ to about 10MHZ? When checking caps I need a higher sweep rate to read small caps, 10MHZ preferred for junctions of various transistors/diodes. I've been using 60~ over the years and its not giving me the coverage I want! Your circuit of "#232" is great but its an over kill vs the one I have at present. Thank you, 73's , Gary Grove CET
Nice circuit Alan. Like the presentation too. I like your jigsaw approach with the drawings, piecing them together as overlays. I have a question about the capacitor that is introduced with the second drawing, the one being discharged periodically with a transistor. You said the charge doesn't go near Vcc so the trace is the bottom end of the cap's charge cycle and is therefore rather linear, no problem with that but your trace has the capacitor charge flattening off just before the reset pulse arrives. If the cap doesn't reach Vcc and the Transistor isn't being driven then why would the cap flatten off nearly 40uS before the pulse. (Yes you can see this in HD, Visible at around 6:23)
+Michael Hawthorne - Good question. While Alan said the yellow trace was "essentially the same thing" but after the op amp, it is actually the X connection showing Vce. You can see that he doesn't change the probe connection point from 7:30 (where you can still see the ramp being clamped) through the end of the video where it is in XY mode. Perhaps it would have been a bit better if he connected the 3rd scope channel to that capacitor to make it a bit clearer. 20/20 3rd person hindsight! ;-)
+Michael Hawthorne +ElmerFuddGun is right. I adjusted the gain of the op amp stage so to swing to the positive rail just before the end of the step. This allowed me to use less than 1 RC time constant on the R-C circuit so that it was somewhat linear - then use the gain of the op amp circuit to bring it to full rail swing. Good observation! Yes, I should have moved the probe to the cap, or added another probe to show that.
+Michael Hawthorne w2aew ElmerFuddGun sporadic -Z Thanks for your replies guys. I saw it again and this time I caught Alan say 'After the op amp' at 6:28. I suppose having a tighter control on component tolerances, matched transistors in the current mirror and a constant current source for the charging cap, you could make a pretty accurate curve tracer.
+sporadic -Z Nice one, I forgot all about Depletion types, the circuit I'm currently designing may have benefited from these. Too late though the circuit is in the breadboard stage. Saved the PDF though.
Awesome project! I'm trying to replicate this circuit to trace the Id curve for MOSFETs. Theoretically, there shouldn't be any issues with the circuit, but I seem to be getting some errors that I'm trying to understand. Have you used this with any specific MOSFET before?
Can this circuit be modified to switch between NPN and PNP transistors? I'd also like to be able to test "power" transistors.Thanks for providing us with clear, concise, and understandable tutorials.
I love these general-purpose circuits-in-action videos, thanks! What's that "hole" in the plot on the lower gain transistor (9:34). Looks like the first few base currents aren't conducting till a much higher Vce than the higher base currents.
+CodeKujo You're just seeing the distortion due to the non-linearity of my little PNP current mirror - not acting as a mirror until it has sufficient bias.
So, I made the curve tracer circuit and I can see that it wants to work but my signals are so small that my scope can't make the curve tracer visible, and I've set both channels at the lowest of 10mV per division and I can see the right kind of waves but very, very vague, mostly some dots. You didn't mention what kind of transistors you used for the add-on circuit and you also didn't mention what the op amp is so I grabbed a few transistors and an op amp that I had. The NPN transistor I used is the "LC945P" which has a beta of 343, the two PNP transistors for the current mirror I used the "A949 Y" for, with a beta of 180. And the op amp I used is the "KA393" which is a dual operation comparator and when I check the input signal coming from the 555 timer, that's okay and the NPN transistor changes the signal and amplifies it and I can see that the comparator is doing something with the signal so on the output I see something that resembles the signal that you've drawn on the schematic at the output of the op amp, in my case the comparator and all that seems to work fine. Now when it comes to the PNP transistors that's a different story and because the collectors are tied together and the bases too without any resistors between the two there's not really much I can do I think, so it would be nice if you can share with me what the beta was for the PNP transistors that you used, I don't think that I have the same parts here but I may have a few that are better matching with the ones that you used for the current mirror. I know that it is a long time ago that you've made this circuit and this video but can you recall which transistors you used? I really want to have this circuit working so I can show it off to my friends together with the other function generators that I made, and I find it pretty nifty that I managed to do it all with analog circuits, even the more complicated waves look very nice and sharp and I haven't even soldered it all together so most of it is running on the breadboards. It's so satisfying when I finally got it working. I hope you get back to me soon so I can finish the last bit and make it work and show it off to my friends. Best regards, Ricardo Penders.
A KA393 is a dual COMPARATOR, it is NOT and OP AMP! They are not interchangable. You'll need to replace it with an op amp. The PNP transistors I used were just general purpose low-power devices. Actual part number isn't very important, you just want them to be generally fairly matched with beta (within 20% or so). I probably used 2N3906 transistors in mine. Note that the collectors on the PNPs are NOT tied together! Please check the schematic and your circuit. The bases are tied together, and the emitters are tied together, but the collectors are definitely not.
@@w2aew OOPS, I switched the collector and the emitter connections I think, I know I have the bases correct for sure, the top side of the two PNP's which is coming from the op amp those aren't the collectors? Are those the emitters, and the collectors go to the oscilloscope probes for the X and Y, right?
@@w2aew I made some changes to the circuit, I'm using the UA741 single op amp now and now I have the bases and the emitters tied together as it should be. So, now I'm getting in the right direction and I see completely different signals as before, much stronger and at the right kind of voltage levels that my scope is able to pick up. However, my curve tracer on my scope looks nothing like yours, it's drawing a curve alright and I do see the different levels whenever the stairstep goes one step higher the output of the curve tracer follows with the stairstep wave. I'll have to make a video of this because without you seeing the actual curve tracer I can't explain anything of it since I have no idea how it works, I'll try to make the video today and let you know when I have it uploaded here on UA-cam.
@@w2aew Persistence pays off, I've got it working correctly now, only my curve tracer is the other way around so from right to left. I had to change the timing capacitor that gets the signal from the 555 timer through the NPN transistor so where you used a 10nF capacitor I'm using a 1nF ceramic cap (102) instead of the (103) and now that I changed that one cap the curve tracer looks much better, maybe I can improve it a bit more using a bit smaller cap even I have to try that.
Could you use voltage from NE555 capacitor to drive X axis? Just add offset to op amp, and use it as inverting op amp. I think that shape of ramp voltage to drive collector doesn't affect shape collector curves. If it is exponential instead linear it just mean that it will sweep faster to start of curve than at end. But speed are low enough that it shouldn't affect curve shape. You still get collector current for applied voltage in every dot. EDIT: Great video, as always! Thanks
+Pokojni Tozo Yes, that would be another way to go - nice clever re-use of the 555 timing capacitor voltage. Would probably work out to be slightly fewer components.
Alan,Great video, I kinda saw this one coming while watching video 231. None-the-less, I have a Tek 576 that is awaiting some of my free time to start heading up the learning curve (pun intended). This video is a great jump start. I'll bet drawing a load line should be made quite easy with the 576. Any hint/tips/kinks/suggestions are welcome. 73, W1SEX
Thank you for so many excellent videos. At 2:35 you mentioned that you make the base resistor large so that it acts as a current source. How does the particular voltage level act as a current source when Rb is very high? Is that just because the resistor translates a locally constant voltage to a locally constant current?
+Eric Bauer Voltage sources have a very low source impedance (so that the voltage doesn't change vs. load changes), and Current sources have a very high source impedance (so that the current doesn't change vs. load changes). Making this resistor fairly large makes the source connected to the base of the test transistor looks more like a stepped current source (due to the low input impedance of the transistor).
Hi Alan. Could the linearity of the ramp generator be improved by using a current control transistor instead of the 100 kohm resistor? Of course, the linearity may be good enough as is for the job in hand. Another nice video with good illustrations and description; as well a good learning project for beginners.
What kind of circuit would you use, if you wanted to match 2 transistors. Is there a wheatstone bridge version of a curve tracer, where you can put both transistors in circuit against each other?
It really comes down to what you need to match (Vbe, Beta, leakage, etc. or threshold/cutoff, Idss, etc.) - it is possible to design circuits that would indicate how well matched some parameters are (but not all with the same circuit). Most curve tracers have a way to measure/compare two devices, often by being able to switch between one and the other quickly.
Pardon the dumb question, but I'm wondering about the current mirror you added to the circuit. Did you use general PNP transistors? I'm thinking 2N3906's. Though I am a real beginner, I find that the clarity of your explanations, drawings and demos are very understandable (though some require multiple views). Thank you for all your work. Finally, for your "Basics" videos, I would love to see something on how to chose between the many types of oscilloscopes available today. Also, a video on all the capabilities of your current scope (the MDO4104C and associated probes) would be very interesting.
What if you used the stairstepped output, ran it through a comparator to generate the pulse to reset the 555 and discharge the capacitor at the stairstep opamp? Wouldn't that make that the pulse generated by the leftmost opamp unnecessary? Reason I am asking is that in the application I am interested using this in, I want to control/adjust the max stairstep voltage, as well as the size of each step.
@@w2aew I worked at Radio Shack in the late 80s and recall those "mims" books. Wish I still had them. Might have a fighting chance to understand what was going on!
+davecc0000 I've already dismantled a portion of this circuit. Changing it to test enhancement mode n-channel mosfets is simple - simply short the resistor that is in series with the "base" for the BJT version. Changes to test PNP, Pchannel or depletion mode FETs would be a bit more involved, but still not difficult.
Wouldnt it be nicer using a simple current source (i.e transistor, 2 diodes and 2 resistors) instead of the 100k ohm resistor to charge the capacitor? Or was this just a quick n dirty 'good enough' approach?
+tHaH4x0r Yes, this was just a quick n dirty approach - using just the first part of the RC charging curve followed by a gain block - it's "linear enough". Actually, no real need for it to be linear, since the XY display mode is being used, you wouldn't notice even it was more exponential. Sure, a current source driving the cap would make a fine linear ramp generator, it that was really required.
QUESTION: Is there not a fairly easy way to make this work with a PNP transistor? Made this stairstep circuit a while back and was going to build a transistor curve tracer and thought of this. Would really like it to work with PNP transistors also. And power transistors. Wasn't sure if this would work with them either. Love your videos! You explain things very well. Thanks!!
There is a fairly easy way to make it sorta work for PNP transistors. Get rid of the current mirror transistors. Connect the Emitter of the test-PNP directly to the output of the upper right op amp (where the emitters of the current mirror was connected). The base connects to the same place. The collector goes to the 270ohm resistor. X and Y connections are to the emitter and collector of the test-PNP. Note that the curves will be flipped upside-down, and the maximum current curve will happen first.
Minimalist = cleverist. :) You mentioned matching toward the end; it occurred to me that it might not be that hard to add drive and "output capture" for a second transistor, subtract the currents, and... Bob's your uncle! Just a thought; I don't know if people do that sort of thing when they're trying to match transistors. Hope things are well with you!
+Jegadeesan Ramasamy The current mirror is made with general purpose PNP devices like 2N3906. I used the same rail-rail op amp that is used in the stairstep generator.
hi Alan. just did this circuit. im using a MCP602(ebay). i know its fake(?) but it works on supply of 5.8Vdc(max 6v as per datasheet). now upon probing around, I get a jittery pulses from 555. the other opamp schmitt osc, is stable. im not using 'good capacitors' for the timing ckt as it is only low freq. each chip has its own decoupling 100nF. im using MLCC caps 100nF 10nF. would this cause the pulse jittering on the 555? ceramic caps seem to be microphonic. I wonder if this is a bad thing for timing circuit? maybe I just should use mylars instead?
What is jittery about the pulses from the 555? Pulse width? Pulse repetition interval? Pulse amplitude? When measuring them, what are you triggering the scope on?
hi. just forget the 'jitter'. my mistake. its because its being 'switched' by the schmitt. removing the op-amp. I get jitter-free osc on 555. heres a quick video on mine. maybe in your future videos, youd consider a pnp mod for this. and maybe jfets as well? 'matching' for phaser use(guitar fx).. I even tried checking J201, I could see a trace but apears to be 'amplified curve'.. with a few tweaks on scope amplitude, could be possibly usable ua-cam.com/video/qpyMLIrIuXk/v-deo.html BIG thanks 👍
Great video! I just built the whole circuit. It works great for BC547C and BC547B. But only if I plot collector currents of 1-2 mA. If I plot curves up to Ic=10 mA or so, I get a linear slope in the left side of each curve (longer, for larger collector currents), before the curve gets almost flat. For Ic=10 mA, the curve is not almost flat until about 3V collector-emitter. This is not "real". A current sensing collector resistor shows that current gets "flat" at little about saturation voltage. Any idea why this happends?
This circuit is far from being an ideal curve tracer, so there certainly will be issues like this. It is quite possible that the op amp used to drive the current mirror in the collector isn't able to supply the current (which would be 2x the collector current).
True. I'm using LM324 which output current is about 1.5 - 3 mA. I placed a simple emitter follower between op amp output and current mirror emitters (no resistors, just a bare BC547C), but nothing changed.
Also, what I love about your teachings and schematics is that the different sections are broken up into pieces. I'm a software dev, and of course that's what we do, but never really thought about it for analog circuits. I'm starting to recognize different common patterns (even said to myself "current mirror" when first looking at it).
This is so cool! As an undergraduate EE student i often get tired of such an intense course but your videos inspire me to carry on! Amazing stuff as usual. Thank you again!
I had no idea about curve traces until I watched a Mr Carlson's Lab, and once I saw that chart the operation of a transistor really solidified.
Curve tracers seem to have gone the way of the dodo, and any that you can find are the same price as a decent car, despite it being such a useful thing for people tinkering with basic electronics. I, for one, would love to see you design one from scratch, that would maybe be an advanced project, but doable for the average mortal who already has a scope. I find your tutorials so easy to follow that I would definitely try to build one.
It would be a fun project, if I could ever find the time...
Wow, such a complex circuit in such a simple way that even a non EE person can understand it. Kudos my friend!
That's cool! You could also use the stairstep generator to drive the lines on a CRT to make the scan lines !
Super educational circuit! It's great that it has a purpose too, but there's a massive value just in the fundamentals it demonstrates.
the way you educate others is really appreciable Mr.W2AEW , good going
Fantastic series. Really interesting and made easy to watch with your clear and effective explanation style. Thanks for making them.
You are awesome, and I really appreciate you making the valuable videos that you do! ☺️
I'm going through an EE program right now, and your practical videos (with all the little tips and tricks from experience) are a boon! Keep up the great vids!
+jason cater Glad to hear it. Good luck with your studies, and be sure to let me know if there is a topic or two for future videos that would help you and your fellow students.
I'm doing some T.A. work with the EE 101 class currently and your LM555 and Op Amp Vids have been invaluable!
However, for my regular classes we are building amplifiers (audio range and then moving into RF) and one thing that hasn't been addressed is causes of noise and distortion. I am aware that this is a HUGE subject that covers volumes of books, but there isn't much talk about the practical "gotchas" (improper loading between stages, stray RF, building on a bread board VS PCB, etc.) and simple mistakes that can be made while designing and building BJT amplifiers. A video that addresses the most common types of interference/noise in BJT amplifiers (and how to avoid them) would be awesome (IMHO).
Thanks again for all your work!
+jason cater Lots of good topics. Noise is a BIG topic. Sources of distortion in BJT amps is usually a matter of signal level and bias conditions. Feedback almost always helps with linearity. Try not to make too much gain in any one stage - that will usually help with linearity / distortion. Anything besides Class A add distortion by design, which is often reduced by filtering.
this is my favorite electronics channel.
very well explained
damn we need a REAL practical electronics instructor like this at my school. good job on all your videos. i learn a lot from them.
Great!! it works for me, I'm not EE of any degree I just like electronics and study one year at high school some technical basics of electronics, back at my younger days, thanks for sharing your knowledge
Thanks Alan for your like, I'm an old guy that likes electronics wish is my pasión and I don't leave these opportunities without trading what I anjoy most
Circuits like this are a great way to teach electronics. Nice way to introduce transistor characteristics. Big thumbs up from me.
Brilliant video and always fantastic engineering content on this channel. One of my faves!
These videos are alot of fun. You make my scope more and more valuable as I learn from these videos. I only wish I had your experience to design these circuits. Thanks!
Many thanks Allen for this very instructive video. You would do me a great favor in continuing on with the ability to test pnp and p-mosfet transistors. Regards, RJM
I'm making this circuit at the moment, I have the stairstep circuit working I need to add some potentiometers so I have the option to set some variations.
I want to be able to set the frequency higher and lower, now it oscillates at 185Hz exactly.
I also want to be able to set the amplitude and if I can add the ability to input a DC offset that would be great.
I'm also adding a switch with the resistor for driving the gate of a MOSFET and the resistor to ground so I can do both and don't have to change the circuit every time.
First I'm making the circuit on a breadboard so I know for sure that it works with the components that I'm using and then I'll make it more permanent on a perf board and solder it all together.
I already made a different function generator using two 555 timers that have the output synchronized but the second chip outputs the signal inverted.
On that function generator I can set the duty cycle from about 20% to 80% at the moment so that could use some improvement, I can set the frequency from 1Hz up to about 450KHz, I can also change the amplitude of both outputs and all the settings work independent so changing one setting doesn't change the other ones.
So when I have this circuit working I'm going to try and put all the different function generators together on one single PCB, it would be nice if I can use the timer chips and the op amps for the different purposes to keep the component count a bit lower but if I can't get it to work it doesn't really matter because I already have all the parts and those are pretty cheap but I don't want to use up everything I have for this function generator, if I can save some components I don't have to buy new ones when I want to start tinkering around and use it for something else.
w2aew, your circuits took a long time for me to really understand what is happening but I've learned a LOT just from watching your videos and making the circuits myself, because there are some differences in the components that I'm using it was a challenge for me to get my circuits to work properly but as you already said it is the best way to learn this stuff.
I wish you have a great day w2aew.
thanks it was an excellent presentation
Great video. A very neat circuit. Thank you for taking the time, I learn a lot from you.
That cut-out schematic "trick" is superb!
This channel helped me learn a lot about electronics.Thank you sir
I was expecting a more exponential rise of the C-charge, but in this way (early cut-off) it it quite linear. Otherwise a simple current source would do a linear work. But maybe it doesn't matter at all, as we are plotting the resulting collector current on the y-axis, it really doesn't matter in what fashion the Vce is changing, as long as it is incremental, I suppose. We just get more or less points in some regions.... Once again interesting circuit 👍
I'm always happy to see your videos! They are very interesting and are explained very effectively! Thank you for your work.
Este circuito era exatamente o que estava procurando. Encontrei aqui por acaso..... kkkk...vou fazer este experimento. Obrigado por disponibilizar o circuito e colocar nele a referencia que o inspirou em fazer o experimento. (Brasil Português )
This circuit was exactly what I was looking for. I found here by chance ..... kkkk ... I'm going to do this experiment. Thank you for making the circuit available and putting a reference on it that inspires you to do the experiment. (Brazil language Portuguese)
nice video, there are certainly a number of ways to generate a starecase. I like the analog aproach since it is easy to alter the number of steps. Thanks and thumbs up!
Very nice Alan. Enjoyed the video. I have a old restored Eico 460 scope sitting around that I have been thinking about adding a curve tracer circuit to.
Excellent! For a hobbyist like me, that's gonna be more than enough. Certainly better than just picking a pair of transistors with a similar gain at a fixed current.
Excellent Explanation as always by Alan, Thanks a lot for your time and efforts.
Well explained and simple circuit. Nice job.
Thank you! Wonderful presentation.
Fantastic. So well explained and demonstrated. Thank you.
Really awsome video (as always). Please do much more of these.
Nicely done and explained. Thanks.
Thank you Alan as always enjoyed and had my full attention. Great circuit to build and learn
10:15 "...if I was going to set out to build a curve tracer I probably wouldn't start with this..."
Why? Is it unnecessarily complicated, inaccurate, or ineffective? What is the ideal design (or design elements to seek out) for someone with an entry level digital scope?
I've seen #197 and #49, your two other videos on curve tracers. I probably don't "need" a curve tracer, but I feel like I learn a lot more about practical electronics by building circuits like this. Plus, no one complains about having too many tools to work on projects ;)
-Jake
I probably wouldn't start with this mainly because it is a bit inflexible in setting up step sizes, etc. I'd probably start with some kind of microcontroller and a few DACs, etc. I would bet there are a few kits out there...
Thanks again for another great video, Alan!
That's very informative and neat, many thanks for sharing.
very satisfying to see the A4 punched holes lining up
+Newbie1991 Yes, those were my registration marks!
Enjoyed Watching the video. Thank you for sharing.
Great video. Keep them coming!
Thanks for this good educative video.
As always I learn things from you. I do electronic repair for a living I can think of a few occasions this could have been helpful. Wish I had taken college when I had the time and funds to do it. You should pursue this and design and build one for multiple voltages and types. A short video on how someone could modify it to be usable over a broad spectrum on transistors, NPN/PNP/MOSFETs etc. would be interesting for sure. Thanks for the Video. Best Wishes and Blessings Keith
Very good presentation. Is the 270 Ohm resistor essentially the collector resistor that would be the load line of the transistor?
I have the simulation in Proteus and it works for my, thanks. I am going to upload the video in my UA-cam chaanel
Your schematics are really neat, it's like you're a human CAD program :) If i would draw that circuit you wouldn't recognise it
Enjoyed your video Allen! Is there a curve tracer that sweeps from 60~ to about 10MHZ? When checking caps I need a higher sweep rate to read small caps, 10MHZ preferred for junctions of various transistors/diodes. I've been using 60~ over the years and its not giving me the coverage I want! Your circuit of "#232" is great but its an over kill vs the one I have at present. Thank you, 73's , Gary Grove CET
Nice circuit Alan.
Like the presentation too. I like your jigsaw approach with the drawings, piecing them together as overlays.
I have a question about the capacitor that is introduced with the second drawing, the one being discharged periodically with a transistor. You said the charge doesn't go near Vcc so the trace is the bottom end of the cap's charge cycle and is therefore rather linear, no problem with that but your trace has the capacitor charge flattening off just before the reset pulse arrives.
If the cap doesn't reach Vcc and the Transistor isn't being driven then why would the cap flatten off nearly 40uS before the pulse. (Yes you can see this in HD, Visible at around 6:23)
+Michael Hawthorne - Good question. While Alan said the yellow trace was "essentially the same thing" but after the op amp, it is actually the X connection showing Vce. You can see that he doesn't change the probe connection point from 7:30 (where you can still see the ramp being clamped) through the end of the video where it is in XY mode. Perhaps it would have been a bit better if he connected the 3rd scope channel to that capacitor to make it a bit clearer. 20/20 3rd person hindsight! ;-)
+Michael Hawthorne +ElmerFuddGun is right. I adjusted the gain of the op amp stage so to swing to the positive rail just before the end of the step. This allowed me to use less than 1 RC time constant on the R-C circuit so that it was somewhat linear - then use the gain of the op amp circuit to bring it to full rail swing. Good observation! Yes, I should have moved the probe to the cap, or added another probe to show that.
+sporadic -Z Yes, exactly, and on purpose (gain on op amp stage purposely set to make it hit the rail at the end of the step).
+Michael Hawthorne
w2aew
ElmerFuddGun
sporadic -Z
Thanks for your replies guys.
I saw it again and this time I caught Alan say 'After the op amp' at 6:28.
I suppose having a tighter control on component tolerances, matched transistors in the current mirror and a constant current source for the charging cap, you could make a pretty accurate curve tracer.
+sporadic -Z
Nice one, I forgot all about Depletion types, the circuit I'm currently designing may have benefited from these. Too late though the circuit is in the breadboard stage. Saved the PDF though.
Excellent as usual.
Awesome project! I'm trying to replicate this circuit to trace the Id curve for MOSFETs. Theoretically, there shouldn't be any issues with the circuit, but I seem to be getting some errors that I'm trying to understand. Have you used this with any specific MOSFET before?
Can this circuit be modified to switch between NPN and PNP transistors? I'd also like to be able to test "power" transistors.Thanks for providing us with clear, concise, and understandable tutorials.
Great explanation. Now what is going on inside the transistor under test ... what happens to the electrons as the base current change with each step??
As the base current increases, more carriers are drawn from the emitter and are swept to the collector.
Big thanks for this nice impressive example.
Cool! Very useful and thanks for waveform-explanation. Make more videos like this :)
Great video! Thank you very much! You are my "electronic" mentor :)
I love these general-purpose circuits-in-action videos, thanks!
What's that "hole" in the plot on the lower gain transistor (9:34). Looks like the first few base currents aren't conducting till a much higher Vce than the higher base currents.
+CodeKujo You're just seeing the distortion due to the non-linearity of my little PNP current mirror - not acting as a mirror until it has sufficient bias.
Great Presentation
Always learning-- Great presentation..
Thanks...
So, I made the curve tracer circuit and I can see that it wants to work but my signals are so small that my scope can't make the curve tracer visible, and I've set both channels at the lowest of 10mV per division and I can see the right kind of waves but very, very vague, mostly some dots.
You didn't mention what kind of transistors you used for the add-on circuit and you also didn't mention what the op amp is so I grabbed a few transistors and an op amp that I had.
The NPN transistor I used is the "LC945P" which has a beta of 343, the two PNP transistors for the current mirror I used the "A949 Y" for, with a beta of 180.
And the op amp I used is the "KA393" which is a dual operation comparator and when I check the input signal coming from the 555 timer, that's okay and the NPN transistor changes the signal and amplifies it and I can see that the comparator is doing something with the signal so on the output I see something that resembles the signal that you've drawn on the schematic at the output of the op amp, in my case the comparator and all that seems to work fine.
Now when it comes to the PNP transistors that's a different story and because the collectors are tied together and the bases too without any resistors between the two there's not really much I can do I think, so it would be nice if you can share with me what the beta was for the PNP transistors that you used, I don't think that I have the same parts here but I may have a few that are better matching with the ones that you used for the current mirror.
I know that it is a long time ago that you've made this circuit and this video but can you recall which transistors you used?
I really want to have this circuit working so I can show it off to my friends together with the other function generators that I made, and I find it pretty nifty that I managed to do it all with analog circuits, even the more complicated waves look very nice and sharp and I haven't even soldered it all together so most of it is running on the breadboards.
It's so satisfying when I finally got it working.
I hope you get back to me soon so I can finish the last bit and make it work and show it off to my friends.
Best regards,
Ricardo Penders.
A KA393 is a dual COMPARATOR, it is NOT and OP AMP! They are not interchangable. You'll need to replace it with an op amp.
The PNP transistors I used were just general purpose low-power devices. Actual part number isn't very important, you just want them to be generally fairly matched with beta (within 20% or so). I probably used 2N3906 transistors in mine. Note that the collectors on the PNPs are NOT tied together! Please check the schematic and your circuit. The bases are tied together, and the emitters are tied together, but the collectors are definitely not.
@@w2aew OOPS, I switched the collector and the emitter connections I think, I know I have the bases correct for sure, the top side of the two PNP's which is coming from the op amp those aren't the collectors?
Are those the emitters, and the collectors go to the oscilloscope probes for the X and Y, right?
@@RicardoPenders Emitters have the arrows on them.
@@w2aew I made some changes to the circuit, I'm using the UA741 single op amp now and now I have the bases and the emitters tied together as it should be.
So, now I'm getting in the right direction and I see completely different signals as before, much stronger and at the right kind of voltage levels that my scope is able to pick up.
However, my curve tracer on my scope looks nothing like yours, it's drawing a curve alright and I do see the different levels whenever the stairstep goes one step higher the output of the curve tracer follows with the stairstep wave.
I'll have to make a video of this because without you seeing the actual curve tracer I can't explain anything of it since I have no idea how it works, I'll try to make the video today and let you know when I have it uploaded here on UA-cam.
@@w2aew Persistence pays off, I've got it working correctly now, only my curve tracer is the other way around so from right to left.
I had to change the timing capacitor that gets the signal from the 555 timer through the NPN transistor so where you used a 10nF capacitor I'm using a 1nF ceramic cap (102) instead of the (103) and now that I changed that one cap the curve tracer looks much better, maybe I can improve it a bit more using a bit smaller cap even I have to try that.
Could you use voltage from NE555 capacitor to drive X axis? Just add offset to op amp, and use it as inverting op amp.
I think that shape of ramp voltage to drive collector doesn't affect shape collector curves. If it is exponential instead linear it just mean that it will sweep faster to start of curve than at end. But speed are low enough that it shouldn't affect curve shape.
You still get collector current for applied voltage in every dot.
EDIT:
Great video, as always! Thanks
+Pokojni Tozo Yes, that would be another way to go - nice clever re-use of the 555 timing capacitor voltage. Would probably work out to be slightly fewer components.
Maybe you could draw up or create a UA-cam on your modified design?
Alan,Great video, I kinda saw this one coming while watching video 231. None-the-less, I have a Tek 576 that is awaiting some of my free time to start heading up the learning curve (pun intended). This video is a great jump start. I'll bet drawing a load line should be made quite easy with the 576. Any hint/tips/kinks/suggestions are welcome. 73, W1SEX
+Paul Topolski Lucky you, with the 576! I haven't used one in over 30 years, so can't really offer any tips on that!
Thank you for so many excellent videos. At 2:35 you mentioned that you make the base resistor large so that it acts as a current source. How does the particular voltage level act as a current source when Rb is very high? Is that just because the resistor translates a locally constant voltage to a locally constant current?
+Eric Bauer Voltage sources have a very low source impedance (so that the voltage doesn't change vs. load changes), and Current sources have a very high source impedance (so that the current doesn't change vs. load changes). Making this resistor fairly large makes the source connected to the base of the test transistor looks more like a stepped current source (due to the low input impedance of the transistor).
very good circuit, very useful
Hello, Excellent explanation.
You used MSO. Is there any problem using DSO?
Thanks
no problem at all
one of the applications of this circuit is Stepped FMCW Radar (sweep voltage to VCO)
i am referring to that ramp with steps :)
Hi Alan. Could the linearity of the ramp generator be improved by using a current control transistor instead of the 100 kohm resistor? Of course, the linearity may be good enough as is for the job in hand. Another nice video with good illustrations and description; as well a good learning project for beginners.
Nice work Alan, thanks. Man, I wish I could afford a scope like yours...
+Robert Calk Jr. Me too! It wouldn't be possible for me if I didn't work for Tek.
What kind of circuit would you use, if you wanted to match 2 transistors.
Is there a wheatstone bridge version of a curve tracer, where you can put both transistors in circuit against each other?
It really comes down to what you need to match (Vbe, Beta, leakage, etc. or threshold/cutoff, Idss, etc.) - it is possible to design circuits that would indicate how well matched some parameters are (but not all with the same circuit). Most curve tracers have a way to measure/compare two devices, often by being able to switch between one and the other quickly.
In contrast to most of us, he knows what he's doing.
Pardon the dumb question, but I'm wondering about the current mirror you added to the circuit. Did you use general PNP transistors? I'm thinking 2N3906's.
Though I am a real beginner, I find that the clarity of your explanations, drawings and demos are very understandable (though some require multiple views). Thank you for all your work.
Finally, for your "Basics" videos, I would love to see something on how to chose between the many types of oscilloscopes available today. Also, a video on all the capabilities of your current scope (the MDO4104C and associated probes) would be very interesting.
+Gary Krouth Yes, I used a pair of 2N3906 transistors for the current mirror. I'll add your video suggestions to my list.
What if you used the stairstepped output, ran it through a comparator to generate the pulse to reset the 555 and discharge the capacitor at the stairstep opamp?
Wouldn't that make that the pulse generated by the leftmost opamp unnecessary?
Reason I am asking is that in the application I am interested using this in, I want to control/adjust the max stairstep voltage, as well as the size of each step.
Yup, that would work.
Great video!
I have a some questions:
Is it possible to make it work as curve tracer for NPN and PNP transistors?
What improvments can be done?
...and I thought your teaching skills were superb!
Your schematic drawing skills are second to none.
Forrest Mims has nothing on you :)
Thank you - I learned clean schematic drawing methods from the Forrest Mims Engineers Notebook series.
@@w2aew I have the collection.
A prime example of "Student surpassing the master".
@@w2aew I worked at Radio Shack in the late 80s and recall those "mims" books. Wish I still had them. Might have a fighting chance to understand what was going on!
TNX for another great video !
Cool stuff. How can we increase the Vce. 100V should be fine.
beautiful ! and all analog. great stuff!
That’s beautiful!
Thanks Alan.
I'd like to build a simple CT for bp's and fet's of all flavors. Could be an interesting video to modify this circuit a little bit more?
+davecc0000 I've already dismantled a portion of this circuit. Changing it to test enhancement mode n-channel mosfets is simple - simply short the resistor that is in series with the "base" for the BJT version. Changes to test PNP, Pchannel or depletion mode FETs would be a bit more involved, but still not difficult.
+w2aew
Yes, saw N mod comments. Look forward to P mods. Thanks.
Simply wonderful.
Nice circuit! Does it work for pnp as well as npn?
It would need some modification for PNP devices.
Wouldnt it be nicer using a simple current source (i.e transistor, 2 diodes and 2 resistors) instead of the 100k ohm resistor to charge the capacitor? Or was this just a quick n dirty 'good enough' approach?
+tHaH4x0r Yes, this was just a quick n dirty approach - using just the first part of the RC charging curve followed by a gain block - it's "linear enough". Actually, no real need for it to be linear, since the XY display mode is being used, you wouldn't notice even it was more exponential. Sure, a current source driving the cap would make a fine linear ramp generator, it that was really required.
QUESTION: Is there not a fairly easy way to make this work with a PNP transistor? Made this stairstep circuit a while back and was going to build a transistor curve tracer and thought of this. Would really like it to work with PNP transistors also. And power transistors. Wasn't sure if this would work with them either. Love your videos! You explain things very well. Thanks!!
There is a fairly easy way to make it sorta work for PNP transistors. Get rid of the current mirror transistors. Connect the Emitter of the test-PNP directly to the output of the upper right op amp (where the emitters of the current mirror was connected). The base connects to the same place. The collector goes to the 270ohm resistor. X and Y connections are to the emitter and collector of the test-PNP. Note that the curves will be flipped upside-down, and the maximum current curve will happen first.
Minimalist = cleverist. :)
You mentioned matching toward the end; it occurred to me that it might not be that hard to add drive and "output capture" for a second transistor, subtract the currents, and... Bob's your uncle! Just a thought; I don't know if people do that sort of thing when they're trying to match transistors.
Hope things are well with you!
Great video
Mr Wolke, how should I adapt the schematic for non rail to rail op amps?
Fascinating...cheers.
How do you modify this circuit for PNP transistors?
Is it only for transistors curve tracing?, or can be mosfet and IGBT?
It could be made to work with them
Could you do a video on how to match JFET's. Could this circuit be used for JFET's? Thanks.
fantastic, congratulations and THANKS!!!
Does it work with PNP transistors as well ? (newbie question)
+Iacobescu Alin Unfortunately no.
+w2aew, can you please specify which opamp and PNP transistors you used? Thanks!
+Mauro Sedrani This information is in the show notes page - link at the bottom of the video description.
www.qsl.net/w2aew//youtube/curvetracer.pdf
+w2aew Thanks Alan, but I can't see the type of components for the integration circuit (current mirror PNP transistors and amplifier).
+Mauro Sedrani Same op amp will work as shown in the parts list. PNP devices are any general purpose small signal PNP like 2N3906, etc.
+w2aew Thanks a lot for your patience :)
brilliant video.
Sir, nice circuit .what is transistor used for current mirror and the last op amp no is not given .simple and usefull test for transistor
+Jegadeesan Ramasamy The current mirror is made with general purpose PNP devices like 2N3906. I used the same rail-rail op amp that is used in the stairstep generator.
Could you please give another link to the schematics used in this video? The current link won't open
I fixed the link, try it now.
www.qsl.net/w2aew/youtube/curvetracer.pdf
Hi can you make it to view the Negative region
or negative resistances
really great channel
Thanks Allen
hi Alan. just did this circuit. im using a MCP602(ebay). i know its fake(?) but it works on supply of 5.8Vdc(max 6v as per datasheet). now upon probing around, I get a jittery pulses from 555. the other opamp schmitt osc, is stable. im not using 'good capacitors' for the timing ckt as it is only low freq. each chip has its own decoupling 100nF. im using MLCC caps 100nF 10nF. would this cause the pulse jittering on the 555? ceramic caps seem to be microphonic. I wonder if this is a bad thing for timing circuit? maybe I just should use mylars instead?
What is jittery about the pulses from the 555? Pulse width? Pulse repetition interval? Pulse amplitude? When measuring them, what are you triggering the scope on?
hi. just forget the 'jitter'. my mistake. its because its being 'switched' by the schmitt. removing the op-amp. I get jitter-free osc on 555.
heres a quick video on mine. maybe in your future videos, youd consider a pnp mod for this. and maybe jfets as well? 'matching' for phaser use(guitar fx).. I even tried checking J201, I could see a trace but apears to be 'amplified curve'.. with a few tweaks on scope amplitude, could be possibly usable
ua-cam.com/video/qpyMLIrIuXk/v-deo.html
BIG thanks 👍
Great video!
I just built the whole circuit. It works great for BC547C and BC547B. But only if I plot collector currents of 1-2 mA.
If I plot curves up to Ic=10 mA or so, I get a linear slope in the left side of each curve (longer, for larger collector currents), before the curve gets almost flat.
For Ic=10 mA, the curve is not almost flat until about 3V collector-emitter.
This is not "real". A current sensing collector resistor shows that current gets "flat" at little about saturation voltage.
Any idea why this happends?
This circuit is far from being an ideal curve tracer, so there certainly will be issues like this. It is quite possible that the op amp used to drive the current mirror in the collector isn't able to supply the current (which would be 2x the collector current).
True. I'm using LM324 which output current is about 1.5 - 3 mA.
I placed a simple emitter follower between op amp output and current mirror emitters (no resistors, just a bare BC547C), but nothing changed.
You are a wizard!