This was fantastic, I million thank yous! I understand the functionality of the classes better, im not fully grasping exactly how I need to apply this to my current amp project yet. Some things take me some time to fully absorb! Great stuff here either way!!
Looking at everything, heat, distortion, class AB is the answer, with a little on the warm side for the bias. Class B was used a lot in old 60/80s portable radio's were battery life was more important than fidelity. Probably where the early, so called transistor sound came from.
Thanks, nice explanation ....I am still asthonished by this subject be'cause it is so missunderstood by many audio fans ..............in fact class AB is an aberration as it is only a switching from class A into class B ...............in every bias current mode between A and commuting to B..................many so called class A amplifiers are in reality hot biased class AB amps.........and most class B amps have a slall amount of bias current .. Tubes don't suffer as much of temperature issues as transistors do........ I like class A amplifiers , as I must admit some good class AB come close .........
Nice video on amplifier biasing and operational classes. I'm learning a lot and filling in my knowledge gaps while watching your videos. Thank you for doing what you're doing! I'm new to circuit simulation, is it possible to look at frequency domain to estimate THD with different bias points?
Hi, nice explanation. I was just wondering, if the positive transistors are running much hotter than the negative transistors, would it be possible to design the circuit so that you have separate bias's for the negative and positive transistors.
That is a great question! But I think placing them on the same heatsink makes them track each other, and as they are a complimentary pair (closely matched) I think that helps them track. Great question...
Thanks for the lesson Eddie. I built on a simple Class A and a simple Class AB amp on Micro-Cap, but I can not get a decent amount of voltage at RL. The biasing seems correct to me. I've tried driving the voltage higher, changing resistor values and transistors. Also, I don't know what all the transistor info is telling me. Do you have a video explaining the data sheet acronyms from Micro-Cap? Even when I find a design on the internet and build it out in Micro-cap, it doesn't seem to work for me. Can I send your both of these files for your review?
oke first of al a real good video, thx for that easy to understand explanation, it's simple this way to understand, you are a good teacher, now for the amp, i like to see the distortion u get from setting the amp in lets say a hot A/B, so let's say the first 10 watt are in class A and the next say from 10 watt to 20 watt its A/B, what's the distortion on that amp bias like that, i think its 0,6 amp until 0,8 amp bias, and what's the distortion if u go the hole way to class A, so the bias at 1,2 amps, with my Hood 1969 amp the bias need to be 1,3 amps to get the hole 20 watts in class A, and maybe not only simulate it but do it to the real amp so that u can see if the simulation is correct or not, is the real amp close to the simulation, that's what i like to see, so again very good video thx for that
The harmonic distribution changes when biased fully class A and can be described as predominantly even, thus the 2nd Harmonic would be highest. When you run class B then the harmonics are predominantly odd order and the 3rd harmonic would be the highest. That is the main sound signature difference between running class A versus class B amps, the class A sounds rounded and musical while the class B sounds harsh or sometimes referred to as clinical sharp or crisp. Cross over distortion on the other hand is, when you hear it absolutely awful and not harmonically related that is why designers choose class AB as explained in the video. John Lindsey Hood 1969 amp simulates exactly as the real world provided you choose the same transistors. Its harmonic distribution and sound depends entirely on the transistor you choose. John's design is very unusual and performs more like a single ended class A than a push-pull class A because the top transistor is almost entirely a current source. The bootstrap is necessary and increases voltage swing on the base of the output transistors.
@@nicoras8803 thx Nico for the explanation, i do learn every day in this hobby, i did build the hood 1969 with the Toshiba 2SC5200 BJT transistors, they sound more musical than the 2N3055 normal in use, and they have more power and deliver a few more watt, but i like the sound of the Toshiba 2SC5200, and i did change the input and output caps, input i did put in a wima mkp and the output capacitor i use a 3300 uF of Japan Chemical and parallel over the electrolytic a Jantzen audio Z-cap mkp cap i only do not now what the capacity of that cap is i mean 6,8 uF or maybe 10 uF or in between
Thanks Bert! Yes, on the board the LED is shorted. It looks like they did that on purpose to take it out of the circuit. I'll have to investigate using it to see what difference it makes.
@@KissAnalog You don't need the led ,, it is for showing the op-amps are working. My first impression was it is a kind of a current source ,, but it isn't...
i have a diy (not made by me) Krell ksa-50 , it is 50watts classA. could i tranform it into a class A/AB mix so i have like 100watts on 8 ohm? just by changing the value of the 'bias-resistors' ?
Thanks for the great question. The bias transistor Q10 is protected by the resistors in the circuit and is used to drop the voltage in the drive circuit to protect the drive circuit. It works really great!
In my Marantz HD DAC1 there is a headphone amp. And they recommend biasing to 50mV. How do I know how high I can go? It was 95mV when I got the unit but I bought it used. Maybe someone biased it hotter. But how do I know where to set it so I don’t get crossover distortion? Thanks
Thanks for the great question! Marantz is one of the few companies that actually had a design team that was bigger than one or two engineers, so I would trust them. Running the bias higher could provide a cleaner - more Class A sound, but it could also be pushing the drivers which would result in higher distortion. So, I'd try the recommended bias - and have someone else you trust tell you which sounded better - with no idea of what you are doing. I fooled myself by making adjustments - letting the bias in my mind make me think I made improvements (placebo effect;).
@@KissAnalog yeah I might take it to a tech to check out the bias but also make some fixes to the power section and replace some regulators with lower noise ones. It’s not bad at all now as I did some measurements ( FFT) but has room for improvement. Plus I’m bored and modding is sometimes fun.
Thanks for asking. It is now free. I've done a number of tutorial videos with this software. Here's the introduction: ua-cam.com/video/bjn-252Oo7A/v-deo.html
What are the limits of your circuit? How weak of an input signal could you amplify with your circuit & get zero distortion on the output? How much output amplification would this circuit provide with zero distortion?
Firstly you should define zero distortion. Is it the capability of the human ear or technically, there is always distortion of some kind. Distortion is directly proportional to the linearity of the device. You will have the least harmonic distortion if you operate the device in exactly the center of their curve where it resembles a straight line. This is not an easy task as you would need a curve tracer to characterize the transistor you are using and not just a data sheet with typical figures for that particular transistor to design your minimal distortion amplifier. At that point it would be possible to have distortion figures in the noise floor of the particular design, provided you remain under the 2nd and 3rd compression point. Typically, an amplifier sensitivity would be that of 3dB above its noise floor. I am of the opinion that -140 dBm is possible, although some claim that -115 dBm is audible. However even the noise in your listening environment will mask much distortion below -80 dBm. (0dBm is 1 mW) Analyzing the 2nd and 3rd order compression points (yes Microcap12 does that) you can predict at what point the 2nd or 3rd harmonic will appear in a particular stage gain. Regarding harmonic distortion, I as a designer do not like 5th, 7th and 9th order harmonics since that sounds very objectionable because it is odd and totally awkward sounding and contribute much to sibilants in voices as well as harshness in cymbals, snares and small percussion instruments.
Thanks for the great question @A B. And thanks @Nico Ras for the great in depth answer! Let me know if there's anything else that you would like to know?
Hello Eddie! Thanks for nice and informative videos. I write to you for some time mail, but unfortunately no respond. Maybe in spam?. Please if you can check it. Thanks in advance!
Thanks Aleksandar! I did just find it, and I must explain that I think I missed it as I am having trouble keeping up with the comments, so I am first going to people who are subscribed to give them priority, and then sometimes I don't get to everyone else. I'll put a video out that will help explain what the problem is. Thank you for sending this, it will be a great video to help others!
This was fantastic, I million thank yous! I understand the functionality of the classes better, im not fully grasping exactly how I need to apply this to my current amp project yet. Some things take me some time to fully absorb! Great stuff here either way!!
Thank you and I appreciate you! Always feel free to ask any questions....but subscribe:)
Looking at everything, heat, distortion, class AB is the answer, with a little on the warm side for the bias. Class B was used a lot in old 60/80s portable radio's were battery life was more important than fidelity. Probably where the early, so called transistor sound came from.
Thanks Michael!
Really nice overview. Gives me courage to start simulating myself.
That's great! Thank you!
Thank you
Good that going inside amplifiers
You bet! And thank you!!
Thanks for the explanation. Learned a lot.
You are welcome! I'm glad to hear it - thanks for the feedback!
Thanks, nice explanation ....I am still asthonished by this subject be'cause it is so missunderstood by many audio fans ..............in fact class AB is an aberration as it is only a switching from class A into class B ...............in every bias current mode between A and commuting to B..................many so called class A amplifiers are in reality hot biased class AB amps.........and most class B amps have a slall amount of bias current ..
Tubes don't suffer as much of temperature issues as transistors do........
I like class A amplifiers , as I must admit some good class AB come close .........
Nice video on amplifier biasing and operational classes. I'm learning a lot and filling in my knowledge gaps while watching your videos. Thank you for doing what you're doing! I'm new to circuit simulation, is it possible to look at frequency domain to estimate THD with different bias points?
Thank you! Yes it is. Please see my Micro-Cap playlist, I'll show how to do this soon;)
Nicely done
Thank you! Cheers!
Nicely explained
Thanks for liking! I appreciate you. Consider subscribing - I like your comments:)
@@KissAnalog Haha I am already and more ;-)
Hi, nice explanation. I was just wondering, if the positive transistors are running much hotter than the negative transistors, would it be possible to design the circuit so that you have separate bias's for the negative and positive transistors.
That is a great question! But I think placing them on the same heatsink makes them track each other, and as they are a complimentary pair (closely matched) I think that helps them track. Great question...
Very nice Information, Thank You
Thanks so much!! I appreciate you!
Thanks for the lesson Eddie. I built on a simple Class A and a simple Class AB amp on Micro-Cap, but I can not get a decent amount of voltage at RL. The biasing seems correct to me. I've tried driving the voltage higher, changing resistor values and transistors. Also, I don't know what all the transistor info is telling me. Do you have a video explaining the data sheet acronyms from Micro-Cap? Even when I find a design on the internet and build it out in Micro-cap, it doesn't seem to work for me. Can I send your both of these files for your review?
Thanks for this feedback! I’d be happy to help. Just email KissAnalog@gmail.com
@@KissAnalog Any luck looking over those two circuits I emailed you Eddie?
Thanks 🙏,Sir,Can you suggest a good sub filter circuit......20-150 hz/120
oke first of al a real good video, thx for that easy to understand explanation, it's simple this way to understand, you are a good teacher, now for the amp, i like to see the distortion u get from setting the amp in lets say a hot A/B, so let's say the first 10 watt are in class A and the next say from 10 watt to 20 watt its A/B, what's the distortion on that amp bias like that, i think its 0,6 amp until 0,8 amp bias, and what's the distortion if u go the hole way to class A, so the bias at 1,2 amps, with my Hood 1969 amp the bias need to be 1,3 amps to get the hole 20 watts in class A, and maybe not only simulate it but do it to the real amp so that u can see if the simulation is correct or not, is the real amp close to the simulation, that's what i like to see, so again very good video thx for that
The harmonic distribution changes when biased fully class A and can be described as predominantly even, thus the 2nd Harmonic would be highest. When you run class B then the harmonics are predominantly odd order and the 3rd harmonic would be the highest.
That is the main sound signature difference between running class A versus class B amps, the class A sounds rounded and musical while the class B sounds harsh or sometimes referred to as clinical sharp or crisp.
Cross over distortion on the other hand is, when you hear it absolutely awful and not harmonically related that is why designers choose class AB as explained in the video.
John Lindsey Hood 1969 amp simulates exactly as the real world provided you choose the same transistors. Its harmonic distribution and sound depends entirely on the transistor you choose. John's design is very unusual and performs more like a single ended class A than a push-pull class A because the top transistor is almost entirely a current source. The bootstrap is necessary and increases voltage swing on the base of the output transistors.
@@nicoras8803 thx Nico for the explanation, i do learn every day in this hobby, i did build the hood 1969 with the Toshiba 2SC5200 BJT transistors, they sound more musical than the 2N3055 normal in use, and they have more power and deliver a few more watt, but i like the sound of the Toshiba 2SC5200, and i did change the input and output caps, input i did put in a wima mkp and the output capacitor i use a 3300 uF of Japan Chemical and parallel over the electrolytic a Jantzen audio Z-cap mkp cap i only do not now what the capacity of that cap is i mean 6,8 uF or maybe 10 uF or in between
Thanks Nico for this great explanation!!
Does the bias currents have any impact on the output noise levels? In terms of white noise etc. Not distortion.
Great question Taylor! I think if there wasn’t enough bias current then that could be a problem.
Good video ,,,,,, but in your schematic the LED is shorted..
Thanks Bert! Yes, on the board the LED is shorted. It looks like they did that on purpose to take it out of the circuit. I'll have to investigate using it to see what difference it makes.
@@KissAnalog You don't need the led ,, it is for showing the op-amps are working.
My first impression was it is a kind of a current source ,, but it isn't...
i have a diy (not made by me) Krell ksa-50 , it is 50watts classA. could i tranform it into a class A/AB mix so i have like 100watts on 8 ohm? just by changing the value of the 'bias-resistors' ?
That’s a great question! It could be that easy, but I’d have to see the schematic to be sure.
what software did you use as simulator? tnx
Micro-Cap which is free now. I have a playlist with a number of videos for MicroCap.
Couldn't you use the resistor diode combination for thermal protection for Q10?
Thanks for the great question. The bias transistor Q10 is protected by the resistors in the circuit and is used to drop the voltage in the drive circuit to protect the drive circuit. It works really great!
Sorry I wasn't clear I meant replace the transistor with a diode and a resistor
In my Marantz HD DAC1 there is a headphone amp. And they recommend biasing to 50mV.
How do I know how high I can go?
It was 95mV when I got the unit but I bought it used. Maybe someone biased it hotter.
But how do I know where to set it so I don’t get crossover distortion?
Thanks
Thanks for the great question! Marantz is one of the few companies that actually had a design team that was bigger than one or two engineers, so I would trust them. Running the bias higher could provide a cleaner - more Class A sound, but it could also be pushing the drivers which would result in higher distortion. So, I'd try the recommended bias - and have someone else you trust tell you which sounded better - with no idea of what you are doing. I fooled myself by making adjustments - letting the bias in my mind make me think I made improvements (placebo effect;).
@@KissAnalog yeah I might take it to a tech to check out the bias but also make some fixes to the power section and replace some regulators with lower noise ones. It’s not bad at all now as I did some measurements ( FFT) but has room for improvement.
Plus I’m bored and modding is sometimes fun.
I have a question. If it's biased with 2 amp, how many watts will be in class a in 8 ohm? Can you show the math? Thank you.
Thank you for the great question! I think this will take a video to do justice. I'd love to give a short answer but it might not be great.
Name of This simulation software
Thanks for asking. It is now free. I've done a number of tutorial videos with this software. Here's the introduction: ua-cam.com/video/bjn-252Oo7A/v-deo.html
What are the limits of your circuit?
How weak of an input signal could you amplify with your circuit & get zero distortion on the output?
How much output amplification would this circuit provide with zero distortion?
Firstly you should define zero distortion. Is it the capability of the human ear or technically, there is always distortion of some kind. Distortion is directly proportional to the linearity of the device. You will have the least harmonic distortion if you operate the device in exactly the center of their curve where it resembles a straight line.
This is not an easy task as you would need a curve tracer to characterize the transistor you are using and not just a data sheet with typical figures for that particular transistor to design your minimal distortion amplifier. At that point it would be possible to have distortion figures in the noise floor of the particular design, provided you remain under the 2nd and 3rd compression point.
Typically, an amplifier sensitivity would be that of 3dB above its noise floor. I am of the opinion that -140 dBm is possible, although some claim that -115 dBm is audible. However even the noise in your listening environment will mask much distortion below -80 dBm. (0dBm is 1 mW)
Analyzing the 2nd and 3rd order compression points (yes Microcap12 does that) you can predict at what point the 2nd or 3rd harmonic will appear in a particular stage gain.
Regarding harmonic distortion, I as a designer do not like 5th, 7th and 9th order harmonics since that sounds very objectionable because it is odd and totally awkward sounding and contribute much to sibilants in voices as well as harshness in cymbals, snares and small percussion instruments.
@@nicoras8803 What is the dynamic range of such a circuit?
Thanks for the great question @A B. And thanks @Nico Ras for the great in depth answer! Let me know if there's anything else that you would like to know?
helmet opening eddie
LOL Thanks Ozy!
Hello Eddie!
Thanks for nice and informative videos. I write to you for some time mail, but unfortunately no respond. Maybe in spam?. Please if you can check it. Thanks in advance!
Thanks Aleksandar! I did just find it, and I must explain that I think I missed it as I am having trouble keeping up with the comments, so I am first going to people who are subscribed to give them priority, and then sometimes I don't get to everyone else. I'll put a video out that will help explain what the problem is. Thank you for sending this, it will be a great video to help others!