Class B Class AB and Class A Amplifier bias explained

Подписаться 31 тыс.

Просмотров 5 тыс.

50% 1

In this video 'Class B Class AB and Class A Amplifier bias explained', I will go over the reasons to bias an amplifier output stage, and what is considered to set the bias correctly. #ClassABias #ClassAAmplifierbiassetting #AmplifierBiassetting #ClassAamplifierbias
20 Watt Class A Amplifier Troubleshooting video: • 20 Watt Class A Amplif...
20W Class A Amplifier Boards; ebay.us/7VvhSQ
Wago Connectors: amzn.to/3W81slJ
_________________________________________________________________________________________
You can support this channel with this link to Patreon - I really do appreciate your support? This support helps me purchase the equipment I use for testing a reviews.
Patreon.com/KissAnalog
Or you can support this channel with a one time donation on PayPal at: KissAnalog@gmail.com
Link to T-Shirts and more: www.redbubble.com/people/kissa...
__________________________________________________________________________________________
Link to Topdon Thermal camera at Amazon: amzn.to/3pA7M99
Link to Pergear Infrared T2S+ camera: www.pergear.com?ref=mGrqM
Link to Amazon Infrared T2S+ IOS camera: amzn.to/42c6hwc
Amazon Infrared T2S+ Android camera: amzn.to/3LkmIQ3
Amazon HT-A2 thermal camera: amzn.to/3NlDRLW
Picoscope 4444: amzn.to/3AYJzfu
Picoscope 3604D: amzn.to/3LDZAfp
OWON XDS3064AE 14 bit Oscilloscope: amzn.to/42junoK
OWON XDS3202A: 14bits 100MHz 2GSs 2Ch Oscilloscopeamzn.to/429O42s
Picoscope 5444 box openings: • PicoScope 5444 Box Ope...
OWON - a first look: • Owon XDS3204AE Oscillo...
Link to lower cost PicoScope: amzn.to/3mVFTaT
Link to Uni-T spectrum analyzer: amzn.to/3V3yJOs
instruments.uni-trend.com/US-... to low cost Power Cable with 10AWG shielded cable: amzn.to/377C8qU
Audiophile Power Cable Rant: • Audiophile Power Cable...
Please use links that cost nothing extra but will help support the channel:
TPA3255 Power Amplifier board: amzn.to/3LgRZTl
GW Instek 4 channel MDO Oscilloscope: amzn.to/3rvY7PK
MicSig Oscilloscope: amzn.to/3B4DcXe
MicSig Differential Scope Probe: amzn.to/3HvRG4A
Pico Oscilloscope: amzn.to/35UytvC
IR Thermal Camera: amzn.to/3J92LJu
Silicone Blue work mat: amzn.to/37a9txL
Kaiweets Store (using this link will help support this channel): kaiweets.com/collections/hot-...
Kaiweets: kaiweets.com?sca_ref=1269738.glN6S8aePH
Other videos that might be of interest:
EEVblog Comments on Fluke Rant: • Capacitor Types in SPI...
Fluke Rant - Are they over priced? • Fluke Rant - are they ...
Fluke Rant - Better Test Leads Please; • Fluke Rant - Better Te...
Fluke TPAK Hanging meter Kit $42 or $10; magnetic hanger comparison: • Fluke TPAK Hanging met...
EEVblog is Santa Claus - Thanks Dave!!: • EEVblog is Santa Claus...
EEVblog comments;) Fluke Rant - are they Over Priced?: • EEVblog comments;) Fl...
The Iconic Fluke 87V: • The Iconic FLUKE 87 wi...
Fluke 27 Exposed and giveaway: • Fluke 27 Exposed and G...
Please support this channel that will not cost anything extra by using my link to Borbede BD-99, this meter is around $47 which is low cost for a high count meter with high accuracy:
amzn.to/3DRgUbZ
Fluke 87V link: amzn.to/3l4QUCq
Hioki 4256 multimeter: amzn.to/3idmexS
Hioki 4282 multimeter (my favorite meter): amzn.to/3xjwGYN
Testo 760-2 Multimeter (on Sale): amzn.to/38PQE4H
Amazon Commercial 90DM610 Multimeter: amzn.to/2XOO4pf
Redfish Clamp-on Power meter: redfishinstruments.com/produc...
TDA8954 Class D amplifier: amzn.to/2LBqvxF
GW Instek MDO-2204EX (has Generator, Power Supply, and Multimeter):amzn.to/3xmCQYa
Micsig Differential Probe: amzn.to/3oUDtVu
Micsig Current Probe: amzn.to/3nRso6v
Hantek Current Probe (low cost): amzn.to/3qrbVHH
PinTek Differential Probe: amzn.to/39SWqlt
Uni-T 2-Channel ARB Generator: amzn.to/3iibBs5'. I'll show how I did the final repair on this

Наука

Опубликовано:

2 июн 2023

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 83

@adrianvonino2574 Год назад

Good quality videos require more time for you to explain, so don't worry about the video being "too long". Those of us who want this will appreciate it !

@KissAnalog Год назад

Thanks so much! I appreciate you!

@DW11111 Год назад

I'll keep watching till I finish the video (I'm only at the 32 minute mark) but I had to pause to make a couple of comments. First, thanks for the push-pull explanation and the need for the 0.22 resistors. I have a 3 stage push pull amplifier I got from a certain chinese website that starts with "Ali" and I absolutely love it. I found some tweaks for it on the internet but didn't quite understand why. This helped a lot on that front. Second, I'm glad I wasn't the only one who commented on the shirt (you're worn quite a few cool ones, the Rush ones being my favourites). I don't know why, but the combination of the shirt, the presentation style, and the fun way you fix stuff made me think of some fun times in the 80's. :) That and you're the Bob Ross of electronics, minus the fro, of course. :)

@KissAnalog Год назад

LOL - thank you! I appreciate you! Send me the info on that amp and the tweaks that you found and I'll make a video commenting on them;)

@DW11111 Год назад

Hi there! I tried to reply but for some reason my comment was not allowed so I'll resend it. If you search for "MX50 Amplifier" you will find it sold by multiple people. They all seem to post the same schematic. The mods I saw were to adjust the quiescent current and to eliminate any DC offset that might happen. The methods I saw were to replace the R22 and R9 resistors with variable ones to adjust the quiescent current and to null DC offset, respectively. One person I saw for R22 chose to use a 680 ohm resistor and a 1k pot to be able to adjust quiescent, for example. In any event, thanks for taking time to look into this amplifier. Thanks to your videos I've learned a lot. I still don't know quite what everything in it does, but I understand a lot more thanks to this video and others you've done in the past.

@johnm4962 2 месяца назад

Thank you I learn so much from you. I appreciate everything you share.

@KissAnalog 2 месяца назад

You are so welcome! I appreciate you!

@fernandogalloso359 Год назад

Excellent video. Thank you for explaining in detail the differences between class B, AB and A. You are the best. Thank you so much.

@KissAnalog Год назад

Thanks so much for your great feedback! I appreciate you!

@davet3804 Год назад

Great stuff , thanks . The legendary JL Hood was asked if his design was true class A or AB and he couldn’t really answer . Audiophiles bang on about class A , but some non linearity is introduced with rising temp . The best amp is the most linear in your environment (speaker impedance Etc) at your power use 🤔

@KissAnalog Год назад

Thanks for that great story! I could see him getting caught off guard. I like that because I imagine his thought was in making an amplifier sound good and biasing it to that end - not meeting some definition;) I agree - the amplifier should be linear.

@fredflintstone8048 Год назад

Sorry about the double post. I appreciated that you went into why there's a resistor in the emitter section of the bpj transistor to handle the case of using bpjs in parallel to prevent runaway. As we know with fets that's not necessary.

@KissAnalog Год назад

Thank you! Post all you want - it has to help the YT analytics;) Yes FETs do have some advantages - like paralleling;)

@poormanselectronicsbench2021 Год назад

It prevents "runaway" by helping the outputs "load share" better, as due to gain and internal resistance differences, if there are multiple BJT's paralleled, yes, one may be taking too much of a load and heating more than the others, so it will be under more stress and more prone to fail earlier than the others.

@KissAnalog Год назад

Yes you are correct of course. The emitter resistors are necessary to force parallel current share - as well as the bias transistor sharing the heatsink for thermal tracking. FETs have that positive Temp Coefficient so they naturally share. I think that was a huge reason that they took over, but I also understand that their harmonics matched a tube better as well.

@paulb4661 Год назад

Negative thermal coefficient and tight matching between devices is what led me to dive into complementary output stage running lateral mosfets. Exicon 10P20S/10N20S selected pairs, based on legendary Renesas J162/K1058, can be paralleled without source resistors. 120-150mA of bias per pair makes the drain current independent of temperature and eliminates their tendency to oscillate, resulting in very stable, non-compensated and simple design. They sound damn good too!

@poormanselectronicsbench2021 Год назад

The funny part is, due to the differences in gain between the N and P channel FET's, some designers, such as Hafler did in the DH-220, used 2 different value GATE resistors for each (470 and 220 Ohms) to help balance that difference. Those Exicon FET's are getting higher in price as I type too, they are some of the few viable replacements for the original DH-220 outputs.

@paulb4661 Год назад

@@poormanselectronicsbench2021 Australian Engineer, late Patrick Turner, whose forte was tube amps, has become enamoured with Renesas fets and found them surprisingly palatable. That alone is quite telling, I like to think. Turneraudio archive is again available online thanks to Ben Tindale. Rds on could be lower- anyway you cut it, you end up with a heater with audio signals amplification as an added bonus and the gate bias difference you mention is a bit of a let down. Other than that, the curves fall on top of one another, in a fashion seldom seen.

@poormanselectronicsbench2021 Год назад

@@paulb4661 There is also a disparity with the internal gate capacitance between P and N channel FET's that has to be compensated for as well, but that is straying well off of the BJT transistor Path that Ed is focused on here.

@KissAnalog Год назад

Thanks for the great info! I'll have to look into these FETs.

@KissAnalog Год назад

@poormanselectronicsbench2021 Thanks for this great info! I'm not sure Rdson is a parameter of concern with an Audio Amp - unless it actually works into the dampening factor. I agree - are the heaters or amps;)

@johnshaw359 Год назад

Thanks for the tutorial. Bias it for the classic first 8 watts in class A then slides gracefully into A/B. Can you build a circuit to show when it's in either class, like an LED peak indicator for kicks, and to visually explain some of what you are explaining? Be nice to build a crazy steampunk version with multiple dials and indicators showing various operational currents and voltages throughout.

@KissAnalog Год назад

Thanks for the great feedback - and great ideas!

@poormanselectronicsbench2021 Год назад

Unless you are making a totally "home brew" amp, it's definitely best to pay attention to service manual info, and go with that first. When Harmon Kardon designed the Citation Sixteen/Sixteen A, which is a Quasi-comp Symmetry AB amp, they chose to bias it at one may think is a high level, that makes the heat sinks get up to 60C / 140F in temp at idle, but, if you read all of the SM notes, they do it because they did studies where the lower distortion of keeping the amp in Class A was found to be less perceptible if an amp is biased to class A up to 17db of it's full output, or 2 1/2 watts PC in a C16/A amp. Considering that most listening is done at that output or below, it eliminates any crossover distortion totally until it is driven harder. Some freak out when the heat sinks get so hot at idle, but for it's best audible performance, the service manual bias value of 50mV across one of the 0.5 ohm resistors, which is 100mA per output transistor, or, what is REALLY high compared to most amps!

@KissAnalog Год назад

Thanks for bring up this amp! No offense to Harmon Kardon, but that type of design topology is not used today. Back in the day the matched NPN/PNP transistors were not that great, so they used NPN in both sides of the push-pull and that created some design challenges. I should do a video just going over that design... With this design topology, we do have a more symmetrical design with matching NPN/PNP so the drive and bias is totally different and more ideal.

@poormanselectronicsbench2021 Год назад

@@KissAnalog It's true, yes, that was a mid 70's design, but at the time it seemed to be one of the unique ones that seemed to reduce the crossover distortion issue in a practical manner, at the cost of making your room warmer and running up your electric bill.

@KissAnalog Год назад

Yes heavy bias was a big tool;) Thanks!

@shean4 Год назад

Thanks for the video. I'd like to see the output stage built on the bench and tested to determine the lowest biasing point that gives acceptable output results (no crossover step and no "wiggle"). I can't see that's there's adequate information from those data sheets alone to determine that point (??)

@KissAnalog Год назад

Great point! From what I see is that they start around 200 mA, so I think that's the min where they were flat, it would have been nice to see the current start from zero. I'll have to do that at the bench to actually show it in action;)

@fredflintstone8048 Год назад

I suppose I think of a push pull circuit that has a distortion in the crossover between the pos and neg amplification devices due to no forward bias as class C. Of course that's the most efficient. I think of a push pull circuit where there is some forward bias to attempt to eliminate any distortion in crossing over as class AB, and then a circuit with only a single amplifying device, or more than one in parallel where the bias is held somewhere in the center of the linear region as class A.

@KissAnalog Год назад

Thanks for your feedback! Yes, you are correct, but heavy biased push-pull is probably the most common for Class A as well - even though single ended is still considered better.

@triodehexode 4 месяца назад

Greetings from Scotland nice work good revision and learning new stuff didn't know that about the manufacturer of PNP transistors. This is helping me a lot have late 60s sweedish Sonab Tuner amp to fix have diagram but no info to set bias its a dual rail + - type. NPN Tip 33B DC coupled to the Speaker. Had to replace them Tip 35s along with one of the driver Transistors and electrolytics. Still working on it . Thanks again 👍.

@KissAnalog 4 месяца назад

Thanks for your great feedback! Are you watching from Sweden?

@triodehexode 4 месяца назад

@@KissAnalog Scotland, the country of bagpipes Haggis and Kilts. I have done that before looking at a word and seeing another that looks similar, Stockholm to me looks similar. Scotland used to have a large electronics industry mostly gone now like shipbuilding before it. Still got whisky.

@KissAnalog 4 месяца назад

Thanks! Please keep the Whisky;)

@marvinmartian8746 Год назад

Love that shirt! You should sell some merch in that style ;-) Thanks for the new video. Very informative and you are a great teacher as always.

@KissAnalog Год назад

Thanks so much! I'll have to see if we can get some shirts in this style. As it is, the designs my wife came up with just don't sell. Not sure if it is the design or people just don't buy merch??

@hoobsgroove Год назад

Why don't we see inductor's instead of a small value resistors for biasing it's a great video very informative

@KissAnalog Год назад

Great question! Resistors are great as a constant impedance/resistance across the frequency of interest. Inductors would be great if they were actually HF switching to smooth the edges. Make sense?

@hoobsgroove Год назад

@@KissAnalog right would it make sense to use both in series would that be a better approach?

@KissAnalog Год назад

I think the inductor placed at the output works for the frequencies of interest.

@andymouse Год назад

Nice work Eddie !.....cheers.

@KissAnalog Год назад

Thank you! Cheers!

@waynegram8907 4 месяца назад

KISS ANALOG, try to make a video lesson about the differences between grid biasing amplifiers and grid "leakage" biasing amplifiers. They are two totally different biasing types but I'm not sure why an amplifier designer would choose grid leakage biasing over grid biasing?

@KissAnalog 4 месяца назад

Thanks for the fantastic topic! I'll do this video;)

@waynegram8907 4 месяца назад

@@KissAnalog thanks can't wait to see it

@crimadellaphone9374 Год назад

For the push/pull parallel Transistors diagram(the 2nd diagram) 30:00 - to use a transistor for the bias network I imagine, you would, in the example of an NPN for the bias transistor, Collector would be connected to the Base of top NPN, Emitter connected to the Base of the bottom PNP, and your bias and audio signal would flow into the Base of the Bias Transistor(3rd NPN replacing the diode network)? Is how I described it how you would use an NPN on this bias network?

@KissAnalog Год назад

Thanks for asking... If I understand your description, that would essentially make the transistor a diode. But, in the schematic I show you can see the bias transistor is actually set up to amplify the current to drive the output stage. It is a fixed amplification set by the bias resistors for a DC value. Does that make sense?

@crimadellaphone9374 Год назад

@@KissAnalog There is confusion somewhere so I'll just ask this way. On the push-pull schematic, you showed and explained, there were only two Transistors and for the bias network there were two diodes but as you went to describe it you were talking about how it would be better if those diodes were replaced by a 3rd Transistor to regulate the bias for the two amplifying Transistors. If I were to use a Transistor to bias the push-pull amplifying Transistors how would I wire up the transistor for controlling the bias. Or did I just completely not understand what you were saying in the video, though I'm certain that you said to replace those two diodes on the bias/base network with a Transistor to handle the biasing. I'm aut!stic thus it could possibly be a misinterpretation, on my part, about what you were saying. Where you say something very similar in the video is right after you bring up the second schematic, as you got to describe the lay out you said something about replacing those diodes with a transistor. Thanks for responding.

@ronchinoy 10 месяцев назад

Im trying to understand how to bias the Rod Elliot PA3 to work as a class A @150 Mamp. Vs 50 Mah for class AB mode.

@fabiotrevisan8922 Год назад

Hey Ed, great video again! I may have missed when you covered the design of this amp itself sometime ago... I'll catch up... But as I look the schematics I see there's no global feedback loop for AC... Only DC stabilization feedback loop (DC output bias servo). So it has no linearization of the output stage... I wonder what distortion ballpark it has overall... I know there's some hype about audiophiles (or audiophiles) against GFB, but not having any?

@KissAnalog Год назад

Thank you! Yes there is only the servo loop for DC bias, so there is no AC feedback. This amp sounds really good, but the distortion is a bit higher - still hard to say if you can actually hear it. I'll do a gambit of testing on it before I give it back to my friend. BTW there is a playlist for this amp which I call the 20 Watt Class A: ru-vid.com/group/PLZ0_iMoMBSsmIHJsHZ3oWuLNbZEqzGz_u

@gkdresden Год назад

22:42 This circuit doesn't work because there is no path for the base current of the pnp transistor to go to. I use a transistor Vbe multiplier (BJT + trimmer potentiometer wiper to base end contacts to emitter and collector) instead of 2 diodes and a trimmer potentiometer. The BJT must be thermally coupled to the heat sink in the immediate vicinity of one of the output transistors. I normally use the opposite side of the heat sink to the output transistor. So it is just a few millimeters away from the hot spot of the device to be cooled.

@KissAnalog Год назад

Thanks for this feedback! Yes, you are right - is your bias then close to what I showed in the schematic for this Class A amp? It is also placed on the heatsink right in the middle of the NPN and PNP transistors.

@rabarebra Год назад

Great video. I have an active studio monitor pair where one has a bit ripple. I tried to find the fault by swapping the thermistors from the OK speaker to the BAD speaker. That didn't seem to be the fault. I found a bad capacitor and it fixed the ripple coming from the drivers. So all thumbs up. BUT, now both the speaker's AC transformers gives a hum. Do you happen to know if it could be bad thermistors? Both speakers always had a inrush that I never liked. The BAD speaker always made me nervous each time I power it up. Any suggestions why the transfomer's sudden hum?

@KissAnalog Год назад

Thank you! You can power off and disconnect from power - then you can measure the ohms of each thermistor. Chances are - it is another capacitor. You might try to place a poly cap in parallel - just be sure you match the voltage rating of the parallel capacitor. Let us know what you find;)

@rabarebra Год назад

@@KissAnalog Thanks for your answer. It is very kind. I've replaced 6 capacitors. 4 big ones (two parallell for each driver) - since these wasn't the problem, there was one blown smaller capacitor in the mains circuit (also in parallell) - so swapped both. The ripple I had buzzing in the drivers themselves is gone, but it is the AC transformer itself that gives a physical hum, and it is mounted tight and nicely. Never hear it hum that much before. I have to say I had the AC transformer on the table for a couple of weeks without using it. Temperature outside has risen, so I wonder if it is humidity or whatever. When I start up the speaker, the hum comes in after say 5-8 seconds. I guess this is the thermistor's getting warmed up, and the resistance going down? I've measured the thermistor, it reads around 8,8 ohms, sometimes 9,5 ohms. It is labeled NTC 10, so I guess it is a 10 ohms thermistor? I'd be so grateful if you had any further tips. I own quite a lot of these speakers, but this pair is the oldest model. About 25 years old. It feels nice with the new caps and they play good, it is just this sudden AC buzzing coming from the unit itself. It is actually hearable from 1 meter away. So I left the back panel open to be sure it is the hum coming directly from the AC unit itself. 1) Can a AC unit be affected by a bad capacitor in the other circuits? 2) Can I test it without any other connection than the mains, say I disconnect the connector going to the print and bridge rectifier? In the schematics it is the main power in, the N going directly to the thermistor, then to the AC's 4-pin connector - so was thinking of disconnecting the output from it. I am very careful and haven't dared to measure it with my multimeter, yet. I am not trained in this field. 3) The odd thing is that it started simultaneously on both speakers. Not sure if it was that I desoldered and cross-swapped their thermistor - maybe bad re-soldering - can that affect it? 4) Manufacturer informed me that the AC transformer can start to hum more by age. To me, they still look pretty sturdy. Great channel!

@rabarebra Год назад

@@KissAnalog Guess I'll just unsubscribe then. Nothing to learn here.

@rabarebra Год назад

@@KissAnalog The speaker's are using a toroid AC transformer. I've now testet by just feeding power to through the thermistor and to the AC transformer without any other connections. The transformer is making a hum. Is this just due to age or can it be fixed?

@jasonmcintosh8212 Год назад

Great video. And like always respect your time you put in. I'm not an expert, but studied amplifier topology since a kid. I personally like to think, the class A is exactly that of a class A. ie idle of and engine would be sitting at 3000 rmp. not very efficient. But class B Amplifiers are extremely efficient, but not so effective on sound reproduction unless Biased slightly on. I like to refer to it as a Class B Hybrid . But that is how i like to refer to the voltage mirror amplifier style of driving a woofer.. what you think?

@KissAnalog Год назад

Thanks for this feedback! I agree - this is a good way to think about them. I think the topology does have a big impact on what works better for bias. The push-pull is great if the transistors are a good complimentary match.

@jasonmcintosh8212 Год назад

@@KissAnalog Thank-you for reply it means a lot. I personally look up to experts like yourself, and you knowledge is gold to a cheap hifi guy like me.

@KissAnalog Год назад

Thanks so much! Thanks for the comment!

@northox Год назад

Great video. Any news about the DIY SMPS?

@KissAnalog Год назад

Thanks and thanks for asking about the SMPS. I will get back to that next week. I think I'm finally catching up and wanted to get this amp back to my friend;)

@northox Год назад

@@KissAnalog awesome news! Will spread the word.

@KissAnalog Год назад

Thank you!

@rubenrinforzi9089 Месяц назад

que slew rate tiene?

@KissAnalog Месяц назад

Great Question! I'll start doing this test;)

@michaelpelumivictor5711 Год назад

Thanks sir for this information pls sir I have question to ask pls how can I calculate amplifier ohm's load.

@KissAnalog Год назад

Thanks for asking, but can you give me a little more information? I'm not sure what you are asking. The Amplifier's load will be a speaker rated at some Ohm level, e.g. 8 ohm or 4 ohm. Or are you asking what is the output impedance of the amplifier?

@michaelpelumivictor5711 Год назад

@@KissAnalog i mean how to calculate amplifier ohm's for speaker

@michaelpelumivictor5711 Год назад

Hello sir , dis what's I mean for instance I build an Power amplifier How can i calculate total loading speaker for each channel, like 4ohms, 8,2

@KissAnalog Год назад

Here is a video on how to measure speaker impedance: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-g1zda0WMDGc.html

@KissAnalog Год назад

Do you mean how to calculate if you have more than one speaker attached to the same channel? If so, you can take the inverse of each and then add them and take the inverse of the sum, e.g. 1/(1/8 + 1/8 + 1/4) = 2 ohms. Or, you take 2 speakers of the same resistance - let's say two 8 ohm speakers and the resistance of these would be half, so 4 ohms. Then take this 4 ohm and add it to the 3rd speaker which is 4 ohms and two 4 ohms would be half - so 2 ohms.