Alan Wolke, you are amazing person! ... Iam now 45yo ... but watching your videos is fulfilling my dreams about electronics from my childhood. Finally I found your lectures!
This video is still relevant in oct '23. I found some regular opamps in one of my bins, so I ordered the required r2r opamps and I will be building this circuit soon, just for fun and education.
If the school teachers are as motivated and interesting as you are, we would have an army of young radio/electronics enthusiasts. Well done Alan! Good on ya mate!!!
What a refreshing video, to see an oscillator in this day and age that's not some super integrated proprietary PLL kind of deal or a microcontroller running one line of code is just very pleasant.
Great little tutorial on this interesting circuit. It's a very visual circuit on the scope as well; so that makes it more educational to visualize what is happening. Thanks for all your great vids.
Impatient to try this but waiting for rail to rail op amps to arrive...tried LM324. Got the relaxation oscillator running and adjusted pulse timings. Then the 555 pulse generator and added the reset. More adjustment to get about 12 steps between resets. Then added the pulse integrator but results were not as expected. Reset was not taking the integrator capacitor to ground. Eventually figured out that the 8k2 base resistor needed to be reduced to turn on a 2N3904 sufficiently to get down to ground during a reset. Guess this is a consequence of not using a rail to rail op amp. More adjustment of the base drive and finally an output close to yours. Not quite as square steps but close. Spent a few hours working on this circuit because it looked fascinating and have learnt lots about basic electronics. There is no substitute for doing - great that your videos are, there is even more to be gained by building and modifying. Been great fun and learning.
I will give that a go ! This has got to be one of those few times when i wanted a circuit that does more with simple parts and better than i could of expected ...all the links very very helpful thank you !
+w2aew please man make a video on analog sinusoidal generators something simple like you made in the video Simple 3 transistor sawtooth generator / oscillator no DDS i need something from ,,0``-6 Mhz or 0-1 Mhz and explain it like you done it on that video
A little something for the first oscillator is you can use a linear pot and 2 diodes in the feedback network so you can have a near constant frequency ( roughly, you will have to vary C to get different frequencies) but where the duty cycle is variable over a very wide range, from around 5% to 95% as you turn the knob. Can be used with the whole circuit to sample the output to get a specific step using a differentiator driving a sample gate ( 4066 switch). But I can see a use for an analogue staircase generator where you can select the number of steps easily just by using RC delays.
+The Current Source I do as well! I have 2 night rider circuits running right now just for my pleasure; an analog circuit with the venerable 555, 4017, and a bunch of diodes, and one using an Arduino. I prefer the analog circuit because it just seems to me to be warmer and more fluid than the digital Arduino circuit. Is it just me?
Hi Alan, It reminds me of the basis of a transistor curse tracer on an oscilloscope of the 70's, when at school. It could be a "fun circuit" to see you build :-). Your videos are greatly instructive. Thank you for share so much knowledge with us. Regards, RJM.
Neat circuit! BTW that output stage is called a Deboo integrator, one of an interesting class based on Howland current sources. It's great for single-supply as it's non-inverting, unlike conventional Miller integrators.
Very interesting as usual and a little over my head at times - lol Well presented valuable training. Thank you for providing it and Keep up the great work!
Very nice little circuit. If you put about a 47K resistor on the base of Q2/Q1 to ground, it'll work better with 2N2222 transistors - they tend not to turn off very fast and hold the 555 in reset too long so you only get 4 or 5 steps out of it. Just a thought :) Love your videos.
+RF Burns Playing around on the bench, I managed to turn it into a basic transistor curve tracer by adding one more op amp, 3 more transistors and a handful of passives... Maybe I'll post another video...
Thank you! Your videos always inspire me to experiment with LTspice. I made one with transistors only containing just a single oscillator. Works in LTspice, but I don't know if it works in real life. :D
the reseting op amp may be another 555 timer .. I have too much ot this buggers and that would i do.. Everyone can adapt the circuit with his own view .. The final op am is neat ! Great job
I use a LM358 as OPAM ( i have no "rail to rail" one) and I need to put a diode between 8k2 resistor and transistor bases in orden to achive transistors off state. So I get a nice stairstep.
Thanks for your awesome training techniques they are very clear and understanding. Have you ever experience any three phase circuits to parallel AC generators, or do you have any ideas on what chip should be use for phasing sequence two or more AC generators
You are a great teacher! I'm curious: what applications would you use a staircase generator or signal? I seem to recall part of the old NTSC color TV signal having one?
+agentsmithisalive Let me give a big old "/me too"' to this comment. From the previous comments I can guess it would be used in a curve tracer but just a couple lines in the video about what they are used for would have been really helpful for people like me who admire the beauty and cleverness of the circuit but wonder why one would build it.
+agentsmithisalive Playing around on the bench, I managed to turn it into a basic transistor curve tracer by adding one more op amp, 3 more transistors and a handful of passives... Maybe I'll post another video...
+llhand Playing around on the bench, I managed to turn it into a basic transistor curve tracer by adding one more op amp, 3 more transistors and a handful of passives... Maybe I'll post another video...
+w2aew That would be great -- would love to see a curve tracer! BTW, I was partially correct: it's the NTSC *monochrome* signal that uses a stair-step pattern (for the luminance signal). See: www.eetimes.com/document.asp?doc_id=1272387 Thanks again and 73!
What is the purpose of the 6.8 kOm resistor, used in the Schmitt trigger relaxation oscillator circuit, connected between opamp output and ground? From threshold voltage point of view, this resistor is unnecessary - it does not change Vth+ nor Vth-. It only loads opamp output with more current.
The circuit was originally designed using an op amp whose output stage would add crossover distortion when swinging to the rail due to the design of its push-pull output stage. The resistor to ground kept the output stage in Class A. You're right, not needed with this particular op amp.
There is a bar on top of the reset pin on the diagram, which means it is an active low reset, which means the reset is default high. When there is a low, the circuit resets. But your scope is showing an active high reset - whenever the pulse is high, it resets.
This reminds me of a AGC circuit I stumbled into building on breadboard about 30 years ago... The accumulator drove attenuators, a detector drove the accumulator, and a timer reset it to detector level after the hang time. Fast attack, no pumping, fixed hang time; for something so crude it worked great. Sadly I've long lost the schematic I drew.
I got it going but got some extra lines. Is 50mhz Rigol scope fast enough?? On another note I held a match to the DUT and watched the curves rise and fall, that was cool.
As always great explanation. I'm assuming this circuit must have a lower frequency boundary? on the basis that the step will decay and there will be a point where the sag makes this too nonlinear, is this right?
I thing you can automate it by letting the relaxation oscillator making the high frequency pwm and let the 555 in monoflop configuration which gets triggered by a high at out to make the discharging step at the output.
Does the output stage need to be as complex? Could we just have a diode from 555 output -> capacitor positive side then put a buffer/voltage-follower there?
Your tektronix mdo4104c oscilloscope - where would be a good place to find such a scope? I have seen some for $17k to $23 which is too high for a hobbyist like my self. Could you recommend a similar but less expensive scope?
Hi Alan, As always,your content and presentation skills are absolutely impeccable !! Hope I did not miss it,but could you clarify why you have the 6k8 resistor to ground on the output of U1. TIA.
+EmbSysDev You didn't miss anything. I was experimenting with a couple of different op amps, and some of them worked better with a pull-down resistor. This op amp doesn't need it. So, it really could be eliminated without any peril.
In my case, the circuit didn't work until I added this resistor. In fact, the stairstep looked all wrong at first until I brought this resistor down to 1K which made it much better. It is still a very noisy plot, it seems like the output of the 555 gets all noisy as soon as I connect it to the op amp. Still investigating how to get it clean as Alan's graphs on the scope.
Really nice video. IMHO had the right amount of content and some nicely paced descriptions. Just the right level for me. I was wondering... does it need an output buffer... any load will disrupt the stairstep? Also, it's a nice old school design but isn't it actually cheaper nowadays to use a small pin-count controller?
+NivagSwerdna The op amp used in the pulse integrator has a low impedance output, so no need for buffer. Of course, it would likely be cheaper to use a cheap uC to do this, but - it would need to be programmed, and you wouldn't learn about the cool analog circuitry. What fun is that? ;-)
+Vale Sarbu Both transistors are turned on "hard", meaning both are going into saturation. No special matching needed in this case - just use the same type of transistor. Of course, you could simply use one more resistor to drive each base independently from the reset pulse generator.
Youre great and your final output stairstep* wave/signal is worth goldif it is possible to make this work in the 400V-1200V range, it is because the wave form is exactly right for a Quasi continues wave tesla coil!! So my question is how to amplify it even more? * :)
Probably a stupid question here, but as a newbie, can I use my op amp oscillator in place of the 555? If so what modifications would you need to make? I have limited access to materials at the moment.
The staircase waveform looks linear on the oscilloscope (before the "steps" are carved into it). However, the main capacitor is charged by a 10k resistor on the output of the 555, rather than by a current source, so I would expect it to be exponential. Maybe it is actually exponential, but just doesn't appear to be so, or have I misunderstood something?
I have a Heathkit IG-1275 Log/Lin Sweep Generator. I would like to display markers, say 10 marks both Log and Linear, across the X axis on an Oscilloscope using the scopes Z axis input to show the marks. Is this possible to display marks both linear and logarithmic across the X axis using the Sweep Gate Output, Sweep Output, or Analog Output of the IG-1275 using the Step Generator circuit #231?
+Gregory Burns You probably only need the 555 pulses. Reset the 555 at the start of the sweep, and then adjust the period of the 555 pulses to get the timing that you need. Of course, this will only work with the Linear sweep. For the log sweep - maybe multiple window comparators to detect the sweep voltage at various thresholds...
So if you make the 555 pulses wider, that means more time to increase the capacitor voltage so the steps would have a bigger voltage difference between?
Am i right if I say the 6.8K resistor forms a voltage divider with the 8.2K connected to Q1? Otherwise the 6.8K blows up my calculation of the trigger points forming the voltage hysteresis of the Schmitt-trigger circuitry.
No, it doesn't make a voltage divider with the 8.2k resistor. It is simply a load on the output of the opamp and can be ignored in the calculation of the voltage hysteresis.
@@w2aew Thank your answer. It comforts me knowing that I wasn't completely wrong. I had some thoughts of this resistor being a load my self. By replacing the 6.8k resistor with an infinite high one in my recalculation the trigger points and hysteresis finally made sense.
Dear Sir, Yes, absolutely. I am trying that but I am learning and use capture, compare of PWM, but at the age of 55, it is little too much to learn, so I felt if you have any solution I would implement it. In fact I do not need the DAC, as I just want to generate thyristor firing pulses for Phase angle control of thyristor from simple PID time-proportional output of the temperature controller. Anay ways thank you very much for your attention.
Hi, After time domain, let's go to frequency domain.. So could you make a video about making a "frequency staistep" for a spectrum analyzer ? Thank's ! May be Mr Carlson's could do this too.. Who will be the first to challenge this ?
Hi I like these kind of videos just wondring if i can give you a veroboard ver that i have done for this circuit To help others that might want to make this. I have had fun making this on veroboard over the weekend. Thanks for reading
Sorry to bother you again but I have another question. In the reset circuit, what is the function of the 6.8k resistor between the op-amp output and ground? I breadboarded just this part and played around with the resistor. The only thing I noticed was a slight level increase of the pulse when I disconnect the resistor. Thank you in advance.
+Alwin Leerling Not really needed. I was playing with a couple of different op amps, and a few needed a pull down to completely shut off the transistors. With many rail-rail op amps, this resistor isn't needed anymore.
There is no "general" way to do it. Each resistor has a function. It may be to create a certain voltage drop, to become a load, to establish a certain bias voltage or operating current, to form some specific timing by setting a current or voltage. Sometimes a value is critical to circuit operation, many times it isn't super critical - it all depends on the circuit.
+byszovd Actually - not really needed. I was playing with different op amps in the circuit, and some needed a pull-down to fully shut off the transistors. Using a good rail-rail op amp, this resistor could be eliminated.
@@w2aew I just tried the link again but the only thing I see is a new open tab doing nothing, I don't know why it's not working for me I'm not blocking anything.
+w2aew Ah, I see. As the voltage across the capacitor increases, the decreasing current through the resistor from the 555 is compensated by the opamp's output, so the current through the capacitor is consistent.
Hi w2aew. When transistor is used to discharge capacitor (like Q2 on the schematics in the movie), what are the rules for not to blow up the transistor? I mean, Q2 acts as kind of short circuit of the capacitor which then gives out its charge to ground through transistor momentarily. There is no particular "current limiter" which protects the transistor from overcurrent/overload. What parameters of transistor should be concerned respective to capacitance and actual voltage on capacitor to consider such solution safe?
In general, if you turn on the transistor relatively slowly, you'll limit the peak current that the transistor conducts with. This can be done with the drive circuit, or often time with just a resistor in series with the base.
Ok, I am one step further to understand. Specialized drive circuit is somehow self explanatory, so let's omit it. Let's consider step voltage which drives the transistor. I can think of a fet transistor where gate-source capacitance together with some resistance in series with the gate create small rc constans which turns the transistor "relatively slowly". What I don't fully understand is the example of resistor in series with the base of junction transistor. I know that this resistor will limit current in the base to keep the transistor saturated. But what mechanism will be involved that assures this transistor turns on "relatively slowly" when step voltage is applied? Should I simply assume max collector current 'Icollmax' and choose the value of resistor in base using 'Ibase = Icollmax / beta' equation?
A resistor in series with the base of a BJT can do two things for you in this situation. It can limit the base current, which will limit the collector current. It can limit it a lot before saturation is reached. You can use the equation you suggested when the transistor is in the active region, but it doesn't apply in saturation. In saturation, the "beta" reduces to about 10-20. It can also form and RC circuit with the transistor's base-collector capacitance which can slow the switching speed slightly (or additional capacitance can be added).
Can it be used for making an analog to digital converter? For example, you compare the analog input voltage with the height of the stairstep and then output a digital signal when the analog input equals the stairstep's height.
It isn't critical. In some rail-rail op amps, you can get a little cross-over distortion when the op amp output switches between sourcing and sinking current. A little pulldown resistor like this ensures that he op amp output is always sourcing current, so it helps to prevent any crossover distortion in the output.
Alan Wolke, you are amazing person! ... Iam now 45yo ... but watching your videos is fulfilling my dreams about electronics from my childhood. Finally I found your lectures!
I love the style of your videos - it's almost a cliché, but it really is like having a great teacher sit next to you and explain stuff. Thank you.
This video is still relevant in oct '23. I found some regular opamps in one of my bins, so I ordered the required r2r opamps and I will be building this circuit soon, just for fun and education.
This is a great reason to get the breadboard out...cheers
Bravo! In all the time I've been following YT, I've only found two such excellent educators. You and DiodeGoneWild.
Couldn't agree more !
The way you explain these circuits is engaging even for a novice like me. Using the scope as a demonsration is icing on the cake. Thanks!
If the school teachers are as motivated and interesting as you are, we would have an army of young radio/electronics enthusiasts. Well done Alan! Good on ya mate!!!
What a refreshing video, to see an oscillator in this day and age that's not some super integrated proprietary PLL kind of deal or a microcontroller running one line of code is just very pleasant.
+FFcossag I have a lot of oscillators that are uC and PLL free on this channel ;-)
***** Crap, I just got busted as a really bad subscriber for not knowing that, didn't I?
Not at all, with over 230 videos, you can certainly be excused for not knowing everything that's here. Thanks for being a subscriber.
Thanks so much. It's my new years resolution to build and describe 20 circuits. I may give the stairstep generator a shot!
Great little tutorial on this interesting circuit. It's a very visual circuit on the scope as well; so that makes it more educational to visualize what is happening. Thanks for all your great vids.
Impatient to try this but waiting for rail to rail op amps to arrive...tried LM324. Got the relaxation oscillator running and adjusted pulse timings. Then the 555 pulse generator and added the reset. More adjustment to get about 12 steps between resets. Then added the pulse integrator but results were not as expected. Reset was not taking the integrator capacitor to ground. Eventually figured out that the 8k2 base resistor needed to be reduced to turn on a 2N3904 sufficiently to get down to ground during a reset. Guess this is a consequence of not using a rail to rail op amp. More adjustment of the base drive and finally an output close to yours. Not quite as square steps but close.
Spent a few hours working on this circuit because it looked fascinating and have learnt lots about basic electronics. There is no substitute for doing - great that your videos are, there is even more to be gained by building and modifying. Been great fun and learning.
I *LOVE* to hear when my videos inspire folks to build, experiment, test, and learn!!! Thank you!
That's a pretty elegant circuit, thanks.
Nice circuit analysis Alan.
Thanks for your time to build it on breadboard...
I will give that a go ! This has got to be one of those few times when i wanted a circuit that does more with simple parts and better than i could of expected ...all the links very very helpful thank you !
My new favorite YT channel. This is amazing!!! Thank you so much for taking the time!
73 N0BPS
+Brian Streufert Thank you - I hope you enjoy my other 230 videos just as much!
+w2aew please man make a video on analog sinusoidal generators something simple like you made in the video Simple 3 transistor sawtooth generator / oscillator no DDS i need something from ,,0``-6 Mhz or 0-1 Mhz and explain it like you done it on that video
A little something for the first oscillator is you can use a linear pot and 2 diodes in the feedback network so you can have a near constant frequency ( roughly, you will have to vary C to get different frequencies) but where the duty cycle is variable over a very wide range, from around 5% to 95% as you turn the knob.
Can be used with the whole circuit to sample the output to get a specific step using a differentiator driving a sample gate ( 4066 switch). But I can see a use for an analogue staircase generator where you can select the number of steps easily just by using RC delays.
+SeanBZA Thaks - that's the whole idea here - just get folks thinking about the possibilities, and other circuit ideas.
This was just plain fun. Well done as always. Thanks a lot Alan.
Such an amazing circuit.. ties in a lot of concepts together.. great for learning
Beautifully explained as usual. Love the analog stuff
+The Current Source I do as well! I have 2 night rider circuits running right now just for my pleasure; an analog circuit with the venerable 555, 4017, and a bunch of diodes, and one using an Arduino. I prefer the analog circuit because it just seems to me to be warmer and more fluid than the digital Arduino circuit. Is it just me?
Always nice to Watch and have the schematic besides. Many could learn from the way you do.
Great video. This works really well with swapping out the 555s and op amp with 15ms trigger clocks for CV generation for electronic music gear.
Oh.... that's what the RESET pin is for! Still learning. Great video as always and thanks for the education!
Hi Alan, It reminds me of the basis of a transistor curse tracer on an oscilloscope of the 70's, when at school. It could be a "fun circuit" to see you build :-). Your videos are greatly instructive. Thank you for share so much knowledge with us. Regards, RJM.
Did you check the next video in series? I did make a basic transistor curve tracer using this:
ruclips.net/video/ZOLLoa2fH24/видео.html
Neat circuit! BTW that output stage is called a Deboo integrator, one of an interesting class based on Howland current sources. It's great for single-supply as it's non-inverting, unlike conventional Miller integrators.
Oh, excellent! I didn't know there was a name for it. I see now that Maxim has some application notes on it. Nice! Thanks for sharing!
جزاك الله خيرا وزادك علما والله اكبر
Thank you for providing the schematic. I'm going to print it out and put it on my wall.
Very interesting as usual and a little over my head at times - lol
Well presented valuable training. Thank you for providing it and Keep up the great work!
Allen thanks for doing these videos, great job
+joel hall ment Alan,sorry
Beautiful circuit, and crystalclear explanation. Thanks for sharing!
Very clearly explained as usual- I am playing with a similar stair-step circuit for use as an analog frequency divider
Very nice little circuit. If you put about a 47K resistor on the base of Q2/Q1 to ground, it'll work better with 2N2222 transistors - they tend not to turn off very fast and hold the 555 in reset too long so you only get 4 or 5 steps out of it. Just a thought :) Love your videos.
Again a great lesson..happy and healthy 2020
*Vielen Dank* für Deine wunderbaren Videos!
Bitte schön
Amazing explanation!
As always ... Great job. Wonderfull explanation. Thank you.
I really do enjoy your circuit analysis videos. Are there future plans for this circuit? - A video linearity tester? - A transistor curve tracer?
+RF Burns No plans yet - but using as a start on a curve tracer did cross my mind...
*****
I would love to see that! And perhaps build for myself.
+RF Burns Playing around on the bench, I managed to turn it into a basic transistor curve tracer by adding one more op amp, 3 more transistors and a handful of passives... Maybe I'll post another video...
*****
Cool! Can't wait!
Thank you! Your videos always inspire me to experiment with LTspice. I made one with transistors only containing just a single oscillator. Works in LTspice, but I don't know if it works in real life. :D
Sounds like you'll have to go build it and see!
Seems like a 556'd be a good fit too :) fun stuff
Nice lesson as always Alan! It will be interesting to see what I can do with it controlling the reset pin with resistance rather than voltage.
the reseting op amp may be another 555 timer .. I have too much ot this buggers and that would i do.. Everyone can adapt the circuit with his own view .. The final op am is neat ! Great job
Another enjoyable video. Thank you! Would it be possible to ramp down instead of up? or even more daring, ramp both up and down like a triangle wave?
Awesome video Alan. thanks for posting
Very interesting circuit, and Great Video. Thanks for sharing.
Really awesome circuit analysis! Thanks.
A very great explanation - perfect job!
thanks - I have to add one of these to the system I am currently designing and this video really made the analysis simple
Great video content as usual. Thank you.
I use a LM358 as OPAM ( i have no "rail to rail" one) and I need to put a diode between 8k2 resistor and transistor bases in orden to achive transistors off state. So I get a nice stairstep.
simply beautiful.
Thanks for easy to understand review, as usual :-)
Your schematics are very neat !
What a master piece!!
Thank you, it's very interesting video. I wanna watch your videos on analog circuits from basic to the one more complicated.
very very interesting circuit. thank you
Thanks for your awesome training techniques they are very clear and understanding. Have you ever experience any three phase circuits to parallel AC generators, or do you have any ideas on what chip should be use for phasing sequence two or more AC generators
I'm sorry, I really don't have any experience with 3-phase power circuits.
Still a great watch!
You are a great teacher! I'm curious: what applications would you use a staircase generator or signal? I seem to recall part of the old NTSC color TV signal having one?
+agentsmithisalive Let me give a big old "/me too"' to this comment. From the previous comments I can guess it would be used in a curve tracer but just a couple lines in the video about what they are used for would have been really helpful for people like me who admire the beauty and cleverness of the circuit but wonder why one would build it.
+agentsmithisalive Playing around on the bench, I managed to turn it into a basic transistor curve tracer by adding one more op amp, 3 more transistors and a handful of passives... Maybe I'll post another video...
+llhand Playing around on the bench, I managed to turn it into a basic transistor curve tracer by adding one more op amp, 3 more transistors and a handful of passives... Maybe I'll post another video...
+w2aew That would be great -- would love to see a curve tracer! BTW, I was partially correct: it's the NTSC *monochrome* signal that uses a stair-step pattern (for the luminance signal). See: www.eetimes.com/document.asp?doc_id=1272387 Thanks again and 73!
Love this! Please do more videos like this if you can!
What is the purpose of the 6.8 kOm resistor, used in the Schmitt trigger relaxation oscillator circuit, connected between opamp output and ground? From threshold voltage point of view, this resistor is unnecessary - it does not change Vth+ nor Vth-. It only loads opamp output with more current.
The circuit was originally designed using an op amp whose output stage would add crossover distortion when swinging to the rail due to the design of its push-pull output stage. The resistor to ground kept the output stage in Class A. You're right, not needed with this particular op amp.
@@w2aew Thanks for explanation.
Could the circuit at 2:19 be modified to produce PWM signal by replacing those 2 timing resistors with a pot?
Yes, of course.
I figured, though I wonder, would it work from 0% to 100% and would it be linear?
+Power Max It would be reasonably linear, but there would be some non-linearities due to the diode behavior.
There is a bar on top of the reset pin on the diagram, which means it is an active low reset, which means the reset is default high. When there is a low, the circuit resets.
But your scope is showing an active high reset - whenever the pulse is high, it resets.
When the pulse goes high, it turns on Q1 which pulls the reset pin low. Q1 inverts the signal.
This reminds me of a AGC circuit I stumbled into building on breadboard about 30 years ago... The accumulator drove attenuators, a detector drove the accumulator, and a timer reset it to detector level after the hang time. Fast attack, no pumping, fixed hang time; for something so crude it worked great. Sadly I've long lost the schematic I drew.
+ZlayaCo6aka Say it ain't so!
Wow, this is amazing - got to go build one!
Nice video once again! Fun circuit, clear explanation, as usuaL, Thanks! I see many potential uses, gives me ideas!.
+jeromekerngarcia Perfect - glad to be an inspiration!
I got it going but got some extra lines. Is 50mhz Rigol scope fast enough?? On another note I held a match to the DUT and watched the curves rise and fall, that was cool.
Sure - this is a pretty slow circuit.
As always great explanation. I'm assuming this circuit must have a lower frequency boundary? on the basis that the step will decay and there will be a point where the sag makes this too nonlinear, is this right?
Yes, that is correct.
I thing you can automate it by letting the relaxation oscillator making the high frequency pwm and let the 555 in monoflop configuration which gets triggered by a high at out to make the discharging step at the output.
Ooo, I like this a lot! Simple components and design is sexy! I wonder how it sounds...
Does the output stage need to be as complex? Could we just have a diode from 555 output -> capacitor positive side then put a buffer/voltage-follower there?
As always great! Awesome!
Thanks a lot!
Your tektronix mdo4104c oscilloscope - where would be a good place to find such a scope?
I have seen some for $17k to $23 which is too high for a hobbyist like my self. Could you recommend a similar but less expensive scope?
I like your excellent explanations.....well done...!!!!
Like you voice over circuit analyzing technique.
That's very look like a kind of flip-flop circuit! Very awesome how you do without memory circuit or microcontroller! THK!
So cool!!
Hi Alan,
As always,your content and presentation skills are absolutely impeccable !!
Hope I did not miss it,but could you clarify why you have the 6k8 resistor to ground on the output of U1.
TIA.
+EmbSysDev You didn't miss anything. I was experimenting with a couple of different op amps, and some of them worked better with a pull-down resistor. This op amp doesn't need it. So, it really could be eliminated without any peril.
In my case, the circuit didn't work until I added this resistor. In fact, the stairstep looked all wrong at first until I brought this resistor down to 1K which made it much better. It is still a very noisy plot, it seems like the output of the 555 gets all noisy as soon as I connect it to the op amp. Still investigating how to get it clean as Alan's graphs on the scope.
Really nice video. IMHO had the right amount of content and some nicely paced descriptions. Just the right level for me.
I was wondering... does it need an output buffer... any load will disrupt the stairstep? Also, it's a nice old school design but isn't it actually cheaper nowadays to use a small pin-count controller?
+NivagSwerdna The op amp used in the pulse integrator has a low impedance output, so no need for buffer. Of course, it would likely be cheaper to use a cheap uC to do this, but - it would need to be programmed, and you wouldn't learn about the cool analog circuitry. What fun is that? ;-)
How can you predict that both Q1 and Q2 will turn on having the bases in parallel and a single resistor? Extremely well matched?
+Vale Sarbu Both transistors are turned on "hard", meaning both are going into saturation. No special matching needed in this case - just use the same type of transistor. Of course, you could simply use one more resistor to drive each base independently from the reset pulse generator.
very informative and very interesting video. I like your channel.
Youre great and your final output stairstep* wave/signal is worth goldif it is possible to make this work in the 400V-1200V range, it is because the wave form is exactly right for a Quasi continues wave tesla coil!! So my question is how to amplify it even more? * :)
Would be nice to see this circuit actually doing something
Probably a stupid question here, but as a newbie, can I use my op amp oscillator in place of the 555? If so what modifications would you need to make? I have limited access to materials at the moment.
The staircase waveform looks linear on the oscilloscope (before the "steps" are carved into it). However, the main capacitor is charged by a 10k resistor on the output of the 555, rather than by a current source, so I would expect it to be exponential. Maybe it is actually exponential, but just doesn't appear to be so, or have I misunderstood something?
This iş What I also getting 1 volt straight line don't know what is wrong
Did you used to make synthesyzers?
I have a Heathkit IG-1275 Log/Lin Sweep Generator. I would like to display markers, say 10 marks both Log and Linear, across the X axis on an Oscilloscope using the scopes Z axis input to show the marks. Is this possible to display marks both linear and logarithmic across the X axis using the Sweep Gate Output, Sweep Output, or Analog Output of the IG-1275 using the Step Generator circuit #231?
+Gregory Burns You probably only need the 555 pulses. Reset the 555 at the start of the sweep, and then adjust the period of the 555 pulses to get the timing that you need. Of course, this will only work with the Linear sweep. For the log sweep - maybe multiple window comparators to detect the sweep voltage at various thresholds...
So if you make the 555 pulses wider, that means more time to increase the capacitor voltage so the steps would have a bigger voltage difference between?
Yes, exactly!
Am i right if I say the 6.8K resistor forms a voltage divider with the 8.2K connected to Q1? Otherwise the 6.8K blows up my calculation of the trigger points forming the voltage hysteresis of the Schmitt-trigger circuitry.
No, it doesn't make a voltage divider with the 8.2k resistor. It is simply a load on the output of the opamp and can be ignored in the calculation of the voltage hysteresis.
@@w2aew Thank your answer. It comforts me knowing that I wasn't completely wrong. I had some thoughts of this resistor being a load my self. By replacing the 6.8k resistor with an infinite high one in my recalculation the trigger points and hysteresis finally made sense.
Dear Sir,
Yes, absolutely. I am trying that but I am learning and use capture, compare of PWM, but at the age of 55, it is little too much to learn, so I felt if you have any solution I would implement it. In fact I do not need the DAC, as I just want to generate thyristor firing pulses for Phase angle control of thyristor from simple PID time-proportional output of the temperature controller.
Anay ways thank you very much for your attention.
Are you using tantalum or ceramic caps?? Thanks, having a go myself.
ceramic
Almost have it going. Only have one curve in XY mode. Stair step looks good though. Using LMC6492. Thanks.
Hi,
After time domain, let's go to frequency domain..
So could you make a video about making a "frequency staistep" for a spectrum analyzer ?
Thank's !
May be Mr Carlson's could do this too.. Who will be the first to challenge this ?
Hi I like these kind of videos just wondring if i can give you a veroboard ver that i have done for this circuit
To help others that might want to make this.
I have had fun making this on veroboard over the weekend.
Thanks for reading
Very cool!
Sorry to bother you again but I have another question.
In the reset circuit, what is the function of the 6.8k resistor between the op-amp output and ground?
I breadboarded just this part and played around with the resistor.
The only thing I noticed was a slight level increase of the pulse when I disconnect the resistor.
Thank you in advance.
+Alwin Leerling Not really needed. I was playing with a couple of different op amps, and a few needed a pull down to completely shut off the transistors. With many rail-rail op amps, this resistor isn't needed anymore.
In general, how do you come up with the resistors value?
There is no "general" way to do it. Each resistor has a function. It may be to create a certain voltage drop, to become a load, to establish a certain bias voltage or operating current, to form some specific timing by setting a current or voltage. Sometimes a value is critical to circuit operation, many times it isn't super critical - it all depends on the circuit.
What is the purpose of that 6.8k resistor in relaxation oscillator section?
+byszovd Actually - not really needed. I was playing with different op amps in the circuit, and some needed a pull-down to fully shut off the transistors. Using a good rail-rail op amp, this resistor could be eliminated.
The link to the schematic isn't working anymore.
I just tried it, seems to be working for me
@@w2aew I just tried the link again but the only thing I see is a new open tab doing nothing, I don't know why it's not working for me I'm not blocking anything.
@@RicardoPenders This link works for me: www.qsl.net/w2aew/youtube/stairstep.pdf
Are the heights of each stair step the same or do the steps get smaller as the capacitor charges?
The steps are the same height. The first step may be taller because of the different charge time after reset.
+w2aew Ah, I see. As the voltage across the capacitor increases, the decreasing current through the resistor from the 555 is compensated by the opamp's output, so the current through the capacitor is consistent.
Hi w2aew. When transistor is used to discharge capacitor (like Q2 on the schematics in the movie), what are the rules for not to blow up the transistor? I mean, Q2 acts as kind of short circuit of the capacitor which then gives out its charge to ground through transistor momentarily. There is no particular "current limiter" which protects the transistor from overcurrent/overload. What parameters of transistor should be concerned respective to capacitance and actual voltage on capacitor to consider such solution safe?
In general, if you turn on the transistor relatively slowly, you'll limit the peak current that the transistor conducts with. This can be done with the drive circuit, or often time with just a resistor in series with the base.
Ok, I am one step further to understand. Specialized drive circuit is somehow self explanatory, so let's omit it. Let's consider step voltage which drives the transistor. I can think of a fet transistor where gate-source capacitance together with some resistance in series with the gate create small rc constans which turns the transistor "relatively slowly". What I don't fully understand is the example of resistor in series with the base of junction transistor. I know that this resistor will limit current in the base to keep the transistor saturated. But what mechanism will be involved that assures this transistor turns on "relatively slowly" when step voltage is applied? Should I simply assume max collector current 'Icollmax' and choose the value of resistor in base using 'Ibase = Icollmax / beta' equation?
A resistor in series with the base of a BJT can do two things for you in this situation. It can limit the base current, which will limit the collector current. It can limit it a lot before saturation is reached. You can use the equation you suggested when the transistor is in the active region, but it doesn't apply in saturation. In saturation, the "beta" reduces to about 10-20. It can also form and RC circuit with the transistor's base-collector capacitance which can slow the switching speed slightly (or additional capacitance can be added).
Thanks for all explanation!
Can it be used for making an analog to digital converter? For example, you compare the analog input voltage with the height of the stairstep and then output a digital signal when the analog input equals the stairstep's height.
Interesting approach, but would be pretty limited to low resolution due to finite step sizes.
I think I understood everything in the first part except what the 6.8K resisistor is doing (s. 4:29).
It isn't critical. In some rail-rail op amps, you can get a little cross-over distortion when the op amp output switches between sourcing and sinking current. A little pulldown resistor like this ensures that he op amp output is always sourcing current, so it helps to prevent any crossover distortion in the output.
@@w2aew Wow- thanks so much for the quick response. Love your videos!