This was another deep concept master piece. I design all analog, digital, power electronics circuits and all works. But I never gone in this much depth. The more I watch your video, the more I realise that I am beginner and there is long way to go. Thanks for sharing your understanding as well as a way to understand things with us.
I like this teaching approach. It feels more practical and accessible. I watched despite my current interests residing mostly in digital electronics. Thanks for the upload. -Jake
Totally agree! A real joy to follow this video because on this level there is so much interesting new knowledge for components we use (and have used) so much! This is joyfull nostalgic but also superusefull for our new work!
Great to learn to see trough all that interesting how&why from the IC-disign on the level of the individual transistors! These videolessons get better and better, and joyfuller and joyfuller to follow this kind of most interesting ins&outs!(forgive me my unexperienced use of english, but for technical terms that's luckilly never a problem) Especially interesting was the explaining of the currentsouce to get the output lower than 0.6v for low sinking currents. Never saw trough that "design trick" till now, and it is such an usefull knowledge!
Sometimes a circuit needs to track near VEE. I use LM358 or CA3130 or CA3140 for this purpose. Other circuit needs to track near VCC. I use LM301 or TL071 for this purpose. Thank you professor Sam Ben-Yaakov, sir.
Thank you professor. Such an amazing video. I have a doubt not in the video though. Yesterday I was using a Chinese clone of the LM741 in unity gain mode (output shorted to inverting terminal and input fed to the non-inverting terminal). I was using a single supply of 12V. I knew the 741 was not a rail-to-rail op-amp but still tried. The first observation was, as I change the input (using 5k potentiometer), the output was swinging all the way from 0 to 12V. Just to confirm again, I varied the input voltage within 0V and 0.4V range and the output followed seamlessly (the frequency of adjustment was
Interesting. Perhaps they changed the input stage design, Can you send me the schematics of the circuit you used and the catalog number of component? Was the op amp capable of delivering current at output over full range of input? Please respond to sby@bgu.ac.il
Wowow, this video is amazing, thank you professor! I got a little question:at the input bjt, we use VCM=VB=VC=0, but in the real world VB should be bigger than VC. How much VB should be bigger than VC approximately, and what is the scale? I want to get intuation about the low rail closeness. Thank you very much
"but in the real world VB should be bigger than VC.". No, VB cab be = to VC and the bjt is still in the linear ose since the junction C-B is not conducting.
Hi professor, is Q11/Q12 truly a Darlington without the collectors tied? Q11 looks like a CC voltage buffer to me. What is the function of the active load on Q10 CC stage, I'm only familiar with active load on CE. Thanks for another excellent video!
Darlington function does not rely on connected collectors. A resistor in Q10 will draw current proportional to power supply voltage (this unit has a constant current consumption) and will reduce the gain.
Dear Dr. Sam, thank you for your detailed explanation. 1. Can you clarify on slide 8-8, the outer bigger triangle represents the actual op-amp but the inner small triangles are gain stages made of BJTs, but not opamps. Am I right?
How did they make a constant current source? I guess the CCS for current sinking at the output can't be a bipolar transistor as it has a CE voltage drop itself. I think a bare 0.5k resistor there is more than enough to let it drive microcontrollers, mosfet gates and the low power logic-operational components you would use opamp for
The OnSemi LM324 dataheet shows a little more detail and states that it is in fact a bipolar transistor based current mirror Q8/Q10 (and the current is derived from a single JFET current source). There also exist die photographs of an LM358 (half of an LM324) at vintageteardown[dot]com; die name is GO158B6.
Dear Dr.Sam, this op-amp is based on BJT technology. What betterment in characterstics an op-amp can offer if it is made with mosfets? Any popular opamp made with mosfet is available in market like LM324? If it does, I would like to see a video in contrast to BJTs. Please reply. Thank you.
MOSFETs pro: low power, low input bias current. Cons: high output impedance low operating voltage. There are no FETS op amp that match the 2.5V -30V operation range. LMV324 is sort of a hybrid limited to 5V. LM324.
OH Oh! Let me guess! It was internal compansation and reletively novel low power output stage (?) just a complete guess haha, I use this opamp for hobby electronics way too much.😜 0:44 Oh yeah I forgot the common mode input range does include zero as well. That does make it very convenient and always screws up first year EE's 😂
Hi Power Max, please pardon my bluntness: I sense that your way of looking at electronics is somewhat superficial? Here is a quick test: what is the reason for the slew rate limits of op amps?
@@sambenyaakov Stability would be my answer. I have several very high speed opamps (I can't remember the part number) but they are not stable at unity gain without an external compensation capacitor. As I understand they have a gain higher than 1 at high frequency, where the phase reaches and exceeds 360 degrees total. The end result is negative phase and gain margin. Such opamps can perform better when some effort goes into the control system design. For some applications not requiring tough performance requirements a simple LM324, UA741, TL074, etc. can perform just fine, and is simpler to get up and running. Hence the popularity of these parts. Great video as always, and I always learn something new!
@@sambenyaakov Sorry for butting in, but let me try. Seeing the 3-stage design as shown in the vid, the slew rate limit appears inherent in the purpose of the stages. The input stage emits a current that depends on the voltage input difference, and the subsequent stage must convert it to output voltage, so it's a current integrator. But the current must have limits either way, which will necessarily limit the slew rate. The speed would also depend on the apparent capacitor value across the second stage, and the choice would present a compromise between stability and speed. Have i gotten it right or am i talking out of my behind? Well it's definitely the latter, since i don't actually recall seeing opamp structure being analysed before.
This was another deep concept master piece. I design all analog, digital, power electronics circuits and all works. But I never gone in this much depth. The more I watch your video, the more I realise that I am beginner and there is long way to go. Thanks for sharing your understanding as well as a way to understand things with us.
Thanks
I like this teaching approach. It feels more practical and accessible. I watched despite my current interests residing mostly in digital electronics.
Thanks for the upload.
-Jake
Thanks Jake for note.
Great video again. Good to understand.
Thank you, Prof. Ben-Yaakov!
Thanks
Nice video, I liked the concept, perhaps you could make another featuring other iconic IC's.
Thanks. Will see.
Totally agree! A real joy to follow this video because on this level there is so much interesting new knowledge for components we use (and have used) so much! This is joyfull nostalgic but also superusefull for our new work!
I can't even imagine the hard work to do 25 minutes of such a great class with all these slides
👍Thanks
Its really the joy of learning for all who watch these videolessons
Thank you Prof.Ben-Yaakov for a very interesting and enjoyable explanation.
Thank for note.
Very good indeed. Thanks Professor for Sharing. Can you also give a similar lecture on LM10? Appreciate in advance.
Will see
Great to learn to see trough all that interesting how&why from the IC-disign on the level of the individual transistors! These videolessons get better and better, and joyfuller and joyfuller to follow this kind of most interesting ins&outs!(forgive me my unexperienced use of english, but for technical terms that's luckilly never a problem)
Especially interesting was the explaining of the currentsouce to get the output lower than 0.6v for low sinking currents. Never saw trough that "design trick" till now, and it is such an usefull knowledge!
Thanks Rob for sharing your thoughts.
Sometimes a circuit needs to track near VEE. I use LM358 or CA3130 or CA3140 for this purpose.
Other circuit needs to track near VCC. I use LM301 or TL071 for this purpose.
Thank you professor Sam Ben-Yaakov, sir.
HI Ibles, Thanks for sharing relevant information.
Thanks professor for this new vidéo ! Very interesting to understand the compensation, rail and last output mirror
Happy New year 2021 !
Thank and happy new year to you and yours.
Thank you for the video, I use this and lm258 in my products! Regards from Brazil.
👍
Thanks your video is very useful
Thanks.
Thank you professor. Such an amazing video. I have a doubt not in the video though. Yesterday I was using a Chinese clone of the LM741 in unity gain mode (output shorted to inverting terminal and input fed to the non-inverting terminal). I was using a single supply of 12V. I knew the 741 was not a rail-to-rail op-amp but still tried. The first observation was, as I change the input (using 5k potentiometer), the output was swinging all the way from 0 to 12V. Just to confirm again, I varied the input voltage within 0V and 0.4V range and the output followed seamlessly (the frequency of adjustment was
Interesting. Perhaps they changed the input stage design, Can you send me the schematics of the circuit you used and the catalog number of component? Was the op amp capable of delivering current at output over full range of input? Please respond to sby@bgu.ac.il
Wowow, this video is amazing, thank you professor!
I got a little question:at the input bjt, we use VCM=VB=VC=0, but in the real world VB should be bigger than VC.
How much VB should be bigger than VC approximately, and what is the scale?
I want to get intuation about the low rail closeness.
Thank you very much
"but in the real world VB should be bigger than VC.". No, VB cab be = to VC and the bjt is still in the linear ose since the junction C-B is not conducting.
Hi professor, is Q11/Q12 truly a Darlington without the collectors tied? Q11 looks like a CC voltage buffer to me.
What is the function of the active load on Q10 CC stage, I'm only familiar with active load on CE. Thanks for another excellent video!
Darlington function does not rely on connected collectors.
A resistor in Q10 will draw current proportional to power supply voltage (this unit has a constant current consumption) and will reduce the gain.
@@sambenyaakov Thanks! As always I have much to learn.
Dear Dr. Sam, thank you for your detailed explanation.
1. Can you clarify on slide 8-8, the outer bigger triangle represents the actual op-amp but the inner small triangles are gain stages made of BJTs, but not opamps. Am I right?
Indeed.
How did they make a constant current source? I guess the CCS for current sinking at the output can't be a bipolar transistor as it has a CE voltage drop itself. I think a bare 0.5k resistor there is more than enough to let it drive microcontrollers, mosfet gates and the low power logic-operational components you would use opamp for
Good point. I am not sure. Today, current sources are based on band gap reference which is then distributed throughout the chip.
The OnSemi LM324 dataheet shows a little more detail and states that it is in fact a bipolar transistor based current mirror Q8/Q10 (and the current is derived from a single JFET current source). There also exist die photographs of an LM358 (half of an LM324) at vintageteardown[dot]com; die name is GO158B6.
Dear Dr.Sam, this op-amp is based on BJT technology. What betterment in characterstics an op-amp can offer if it is made with mosfets? Any popular opamp made with mosfet is available in market like LM324?
If it does, I would like to see a video in contrast to BJTs.
Please reply. Thank you.
MOSFETs pro: low power, low input bias current. Cons: high output impedance low operating voltage. There are no FETS op amp that match the 2.5V -30V operation range. LMV324 is sort of a hybrid limited to 5V. LM324.
OH Oh! Let me guess!
It was internal compansation and reletively novel low power output stage (?) just a complete guess haha, I use this opamp for hobby electronics way too much.😜
0:44 Oh yeah I forgot the common mode input range does include zero as well. That does make it very convenient and always screws up first year EE's 😂
Hi Power Max, please pardon my bluntness: I sense that your way of looking at electronics is somewhat superficial? Here is a quick test: what is the reason for the slew rate limits of op amps?
@@sambenyaakov Stability would be my answer. I have several very high speed opamps (I can't remember the part number) but they are not stable at unity gain without an external compensation capacitor. As I understand they have a gain higher than 1 at high frequency, where the phase reaches and exceeds 360 degrees total. The end result is negative phase and gain margin. Such opamps can perform better when some effort goes into the control system design. For some applications not requiring tough performance requirements a simple LM324, UA741, TL074, etc. can perform just fine, and is simpler to get up and running. Hence the popularity of these parts.
Great video as always, and I always learn something new!
@@sambenyaakov Sorry for butting in, but let me try. Seeing the 3-stage design as shown in the vid, the slew rate limit appears inherent in the purpose of the stages. The input stage emits a current that depends on the voltage input difference, and the subsequent stage must convert it to output voltage, so it's a current integrator. But the current must have limits either way, which will necessarily limit the slew rate.
The speed would also depend on the apparent capacitor value across the second stage, and the choice would present a compromise between stability and speed.
Have i gotten it right or am i talking out of my behind? Well it's definitely the latter, since i don't actually recall seeing opamp structure being analysed before.