20W Class A Amp Schematic with MicroCap Tutorial

Подписаться 31 тыс.

Просмотров 5 тыс.

50% 1

I have built a schematic for this Class A amplifier, so I will present that in Micro-Cap Spice Simulation tool. It is free, so get Micro-Cap from Spectrum software downloaded. Also, I have a playlist of tutorials for Micro-Cap. It is titled 20W Class A Amp Schematic with MicroCap Tutorial .
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
Class A Power Amplifier Playlist: • Class A Power Amplifiers
Yaung Jing is the brand, and the following is the specifications for it:
Basic Parameters
Supply voltage: DC +/-25V ~ +/-35V
Output power: 20W @ 8 Ohm, 40W @ 4 Ohm, 80W @ 2 Ohm
Speaker impedance: ohm 2 ~ 8 Ohm
Zener Diodes: amzn.to/3qlKoYH
Wire clamps: Wago assortment: amzn.to/3qkY4mQ
Keadic assortment: amzn.to/2N80vKT
Link to 20W Power Amplifier Board: amzn.to/3oB0VWW
Link to silicone lead set: amzn.to/3j23ksq
Link to Blue silicone mat: amzn.to/36qpIqD
Please use this link for the Hiraga Power supply kit:
rover.ebay.com/rover/1/711-53... Power Supply
Please use this link for Aleph 5, Class A Amplifier kit used in this project: rover.ebay.com/rover/1/711-532...
Another Aleph 5 Class A amplifier link:
rover.ebay.com/rover/1/711-53... 5 Amplifier

Наука

Опубликовано:

5 мар 2021

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 54

@stevebailey1682 3 года назад

Eddie...my hat is off to you reverse engineering that board and turning it into a functioning schematic/simulation. Nice Job!

@KissAnalog 3 года назад

Thanks Steve - I appreciate you! It did take some time and eye strain;)

@wafiullah-shafia 3 года назад

Nice Good thank you for showing the schematic It make me understand more

@KissAnalog 3 года назад

You are welcome;) Thanks for the feedback! Let me know if you have questions or comments.

@raindropsrising7662 3 года назад

Awesome. Beyond my basic level but happy to learn

@KissAnalog 3 года назад

You can do it! ;) Thanks for the feedback!

@7469626572696F 2 года назад

This is very instructive and helpful for people that are new to this field and to MicroCap. Thank you for helping all of us. Very generous of you.

@KissAnalog 2 года назад

Thank you Tibia! I appreciate you!

@ubergeeknz 3 года назад

Surprised seeing the distortion performance that there is no global negative feedback aside from the servo! Quite a neat and simple design. Thanks for reverse engineering it!

@KissAnalog 3 года назад

Thank you! I think it works without the global feedback because of the fairly simplistic design. The drive and output transistors are a slave to the drive signal, and don't really care to react on their own, so nothing to control with feedback;)

@gkdresden Год назад

I don't think it is really a simple design. I've seen an open schematic of Linsley-Hoods Class A amplifier from 1969. I would consider his design as relatively simple, but therefore even harder to optimize. If I would like to design a Class A amplifier operating with an op amp I would use the same configuration as in the Class AB. You just need to increase the idle bias current to half of the maximum output current and this is it. In fact I would use an op amp with a higher output current (NJM4562, 70 mA output current) in non-inverting configuration and feed its output current through the bias spreader into a current sink (let's say 20 mA). Then I would use a complementary output stage using power BJTs with a high current gain (D44VH10 / D45VH10, beta = 200 @ 2A). A power supply separation for the op amp is not necessary. It has a very high power supply ripple reduction. I use LTSpice for circuit simulation allready 15 years and I am happy with it.

@nicoras8803 3 года назад

Excellent simulator. Did you know that you can input a WAV file and you can listen to your design. You can sync and compare the original with the schematic output using Audacity visually see if there are some differences in characteristics if you cannot hear it. I owned and MC since conception and still do. Nothing can touch it.

@KissAnalog 3 года назад

Nico, I think I know that, but have never tried it. I'd like to try;) I'm a fan of MC too, I've purchased copies in the past when I did consulting. MicroCap has always been one of the top 3 I think and still is;)

@richardgray8593 3 года назад

@Nico Ras I wonder what is was like running Micro-Cap I on an Apple II.

@versace885 3 года назад

Hi Eddie, great video. Thank you for sharing.

@KissAnalog 3 года назад

Thank you so much!!

@nicoras8803 3 года назад

Q10 is the Vbe multiplier, and it is what keeps the amp in thermal equilibrium. As it heats up, it passes more current and the bases of the drivers are biased less, etc. Keep in mind that unless you replace the real parameters of the transistors as measured, MicroCap will use the same parameters or model statement for the same components i.e. the output devices, and it is essentially perfectly matched, not quite real world but good enough for the simulation. That is why you found some outputs got hotter than others, simply because they are not absolutely matched in real life, but please carry on you are doing a great job. One thing I have noticed and not tried myself to see if there is a difference isthat it seems like there is a space between 470u, it looks as if it is 47 0u. (just a comment)

@KissAnalog 3 года назад

I could run a worst case analysis to see how sensitive the circuit is. I feel like it is pretty solid as I've put a number of hours on it now;) The screen did look like the space between the 470 uF digits. Not sure why. I don't know if I noticed that before this schematic. The placement/design of Q10 in the circuit does seem to do a fantastic job at dialing back the bias with increased heat. I'm thinking about changing out the LED with an actual diode to help balance the 2 transistor Base-Emitter junctions, but the Q10 is doing a great job...

@TRONMAGNUM2099 3 года назад

Good video. Nice to learn a little more about how to use Micro-Cap. Nice shirt BTW. Royal Bliss came up in my recommended viewing. I listened and liked what I heard.

@KissAnalog 3 года назад

Thanks Tron! They are a local band that owns a bar here in Utah. They have amazing groups (well at least before this C thing).

@microknigh7 3 года назад

Cheers Eddie, another interesting vid

@KissAnalog 3 года назад

Thank you!

@johnwright8814 3 года назад

I used an OP2604 op-amp for my Class-A amp because it will work with +/-24v rails.

@KissAnalog 3 года назад

Thanks John - great option! I thought about ordering some higher voltage OP OpAmps. That sounds like a great option. How does the noise figures stack up?

@johnwright8814 3 года назад

@@KissAnalog Not as good as the NE5534, but not of great concern for a power amp.

@arfer1470 3 года назад

Love these videos Eddie👍I would try increasing the integrator cap in the dc servo, say 4.7u or more, get the bottom end roll off as far away from the audio band as possible, reducing phase shift, when I have used servos it makes the bottom end more focused, even if LF roll off was already lower than your speakers could go in the first place. I have seen these exact boards on Aliexpress very tempted to get a pair to play with. Got too many projects in the house at the moment though. Need bigger house 😃

@KissAnalog 3 года назад

LOL Thanks Ian, this is great feedback!

@stevebailey1682 3 года назад

Also Eddie...thanks for introducing me to MicroCap. I have been using LTSpice for a while and although it seems to work fine for my purposes, it appears MC has more options and flexibility.

@KissAnalog 3 года назад

That's awesome! I used LT too like so many of us have. MicroCap is a league ahead. They have to be as they actually sold their software for years - and it wasn't inexpensive. Huge library of parts and circuits.

@nicoras8803 3 года назад

Before choosing your final current study the SOA, Safe operating area of the transistor and make sure that you do not pass more current than is safe. Also keep in mind that you are driving two transistors in parallel and the current in each could be reduced significantly to still operate in class A, but in my opinion as mentioned last night. Bias the amp as much class A as necessary, say 4 or even 5 watt and allow it to run class B for transient signals. Making a space heater for the bragging rights to claim a 20 watt pure class A only lives in the minds of audiophiles. (Just another 2 pennies worth why waste power if you will hardly use it or notice the difference with your ears)

@KissAnalog 3 года назад

Thanks Nico! I agree with you - I'll find a balance of heat vs Class A setpoint;)

@rhirsch81 3 года назад

Q10 is what's known as a Vbe multiplier. Setting R21, R22, R23 as you like will give you a precise voltage drop between the collector and emitter of Q10 that is reasonably current invariant.

@KissAnalog 3 года назад

Thank you!

@MichaelBeeny 3 года назад

Interesting circuit. I wonder if the zener diode was the wrong voltage fitted on assembly? Shows the boards are not tested. I suspect a few people probably notice the LED un lit, see it's shorted out, thinking it's poor assembly and remove the short.

@KissAnalog 3 года назад

Great feedback Michael! You might be right, but I have an idea that they limited the power this way as they didn't know if people would over heat them using just the heat spreader. Probably protecting people from themselves;) That's OK if they had provided instructions on how to implement this with options.

@argcargv 3 года назад

I would think that the LED would be a noise source as well. Probably best to leave it out. The weird thing is, why keep it on the BOM if you don't plan on using it.

@KissAnalog 3 года назад

Thanks Ed, great questions! With the LED shorted as it is now, it isn't a noise source, but with it in the design - I'd agree with you. I think it was left in because it was found after the design/PCB design was finished and it was just cheaper than changing the BOM and it might have been thought as an option to use if wanted. It will work;) I'm thinking of replacing it with an actual diode.

@johnsweda2999 3 года назад

Would it be much effort to change them for FET transistors that would be sweet

@JasonLeaman 3 года назад

MJE devices would be even better than fets too !

@KissAnalog 3 года назад

Biasing FETs would completely change the drive circuit. I do like FET amp designs and love the JFET/MOSFET combos, but I think a BJT output does benefit from the BJT bias/drive circuit as it balances the silicone over temp.

@JasonLeaman 3 года назад

@@KissAnalog BJT's are really good for audio Fet's are not really.. I sent you a pair of Nice MJE matched sets for the JeanHiraga 60watt amp :) You will be impressed with it.

@richardgray8593 3 года назад

I'm asking before watching the video, but do you have your Micro-Cap file available for download?

@KissAnalog 3 года назад

Great question! I am offering a copy to any subscriber who sends me the request via email at KissAnalog@gmail.com

@DAVIDGREGORYKERR 9 месяцев назад

You know the 5 Watt Class A amplifier could it be upgraded to something like 20W using 2X 2N3055 metal cased power transistors and a BD132 power transistor in place of the original transistors on a large heatsink just wondering.

@KissAnalog 9 месяцев назад

Thanks for asking! Which 5 Watt Class A amp are you referring to?

@DAVIDGREGORYKERR 9 месяцев назад

Google for 5 watt Class A amplifier

@jstro-hobbytech 2 года назад

What do people who seem to come off as stalkers get bahahaha

@KissAnalog 2 года назад

I don't know any of those;)

@jstro-hobbytech 2 года назад

@@KissAnalog well said. Thanks man.

@lucianoc7745 3 года назад

The most notorious things about the design that i see so far are mostly PSU related: * It would be good to have some ceramic capacitors at the input of the power supply (parallel to C20 and C23) to filter HF, those electrolitycs will not be able to reject any HF noise that comes from the input power rails. * It´s a shame the way X4 and X8 was connected. Conecting the collector of X4 to the base of Q26 and the emitter to the zener cathode, then X4 base could have been connected to a resistor divider between B+ and GND and have a regulated power supply with adjustable output voltage wit the cost of two extra resistors. * The filter capacitance doesn't seems "well distributed" to me 1. C18 and C21, they are way oversized. Could have been selected as low as 47uF. 2. C19 and C22 are also oversized, i would have chosen 1000uF for C20,23 and reduced C19 and C22 to maybe 100uF. 3. Same history for C26 and C2. * Would have been nice to have a low-value damping resistor across the output inductor, to reduce overshoot and ringing of the LC combination on capacitive loading. Appart from that, Have you tested the amplifier in the enclosure to make sure the opamp doesnt turn into magic smoke?. Those ~20V zener will have a big TC.

@KissAnalog 3 года назад

Thanks for the great feedback Luciano! I agree with your idea of capacitor sizing. I had the same thoughts. I will add HF cap filtering and extra bulk off the bridge. I am considering using inductors too so there is an CLC input off the bridge. Hope to hear more from you!

@lucianoc7745 3 года назад

@@KissAnalog CLC filtering might too overkill for a 20W amp, i would rather improve the power supply filtering. A better option could be add RC filtering to the VAS supply, in order to "separate" from the output stage. One thing i forget to mention, R58 is too low compared to R59 that increases the gain a little (R59 is in parallel with R57). A far greater improvement will be to take the feedback voltage just before L1, it will reduce nonlinearities greatly and it may even help with the assymetrical clipping.

@TAMIL-MUSIC 3 года назад

Is it class a / or class ab

@KissAnalog 3 года назад

It is Class A, but can be set up to be Class A/B too. I will have to see how I set it up. It might be Class A to A/B;) Maybe Class A to 10 Watts then A/B to 20 Watts...or if it doesn't get too hot I'll set it to Class A for the full 20 Watts;)