Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

full details on creating good sound

Status
Not open for further replies.
It's not about power handling, it's about lowering distortion and maintain high bias current. The main distortion of power amp is crossover distortion where the drive change hands from one to the other transistor in the pair( class B). This distortion does NOT depend on the signal amplitude, it is constant amount. This means the lower the signal level, the more obvious the distortion. This crossover distortion affect the sound quality at lower listening level rather than at loud volume. In another words, at high volume, the crossover distortion becomes so small compare to the signal that it's not important.

The ways to lower crossover distortion is to have more pairs of output transistors. The distortion lower by almost half when you double the number of pairs of output transistors. Using more pairs IS the key of lowering crossover distortion.....that make the amp sounds good at lower volume.....which is the key of an audiophile amp.

The other way to avoid crossover distortion is to have higher bias to create a bigger class A region, so in small signal, the amp is operating in class A. I run 0.9A ( 100mA per output pair) to get about 8W of class A with 4ohm or 16W with 8ohm load. With home listening level, you literally have a class A amp. The rest of the power is for head room during fast transient. That's why if you look at the high end amps like Krell, Threshold, they use a lot of pairs of output transistors. It's not for show. I use the Nakamichi PA-7 Stasis amp designed by Nelson Pass for comparing with my amp. PA-7 is almost the same circuit as the Threshold S300. It is very obvious that at lower volume, my amp has more separation, clarity, sound stage than the PA-7. I believe it's because my amp has over double the bias current of the PA-7.

For DJ, you need none of these. Design is a lot easier, you don't need to have that many pairs of transistors. More importantly, you don't want to carry my amp around. It's over 50lbs!!!! You want something that is lighter. A lot of DJ uses class D amp as sound quality for DJ is not that important.
 

You know from the time I have been building circuits I rarely ever build anything for myself, I don't even have an amp at home that I can say is stationary, I have been doing alot of reading the last month but I kinda won't even know what to read up on.

Having a book dedicated to a specific thing would be good but for now I'm without, do you know of any books online about amplifiers that's free or they are all for purchase.
 

I never have a degree in EE, my degree is Chemistry. I am total self studied, I studies to graduate level on my own including all the advance math and electromagnetics. This is the first transistor book I studied, I used the version from the 70s, this is a newer one. It is about the best book I've seen after studying rackful of books on electronics and physics.https://www.amazon.com/gp/offer-listing/0028028333/ref=dp_olp_used?ie=UTF8&condition=used

I don't know of free books, but you can get it for less than $10 used. You cannot just rely on articles, you want to learn and appreciate designing and building amps, you have to invest time. I studied this book, it carried me very far, even got a job as analog IC designer with Exar in 1984 arming mostly by this book.

After this book, you can buy the book suggested to you before, the Bob Cordell book.

You have to know the transistor basic and Malvino will give you the necessary tools. It covers opamp also. Make sure you study this before Cordell's book. I am sure you can ask questions in this forum(not this particular power electronics).

Power amp is just a discrete high power opamp in a lot of sense, you study the internal circuit of opamp, you will understand power amp a lot easier.
 

I'll buy them.

Well one of my aim here was to build a proper preamp and then a proper amplifier. I tend to normally stop at the preamp since that's what I get help with I guess buying Chinese amps is what I'll have to use for my projects for now.
 

One thing I notice, you mentioned a few times about don't want to spend the money. I don't know your career or your financial situation. In case if you think you want to build your own amp to save money.......DON'T. It always cost more...... a whole lot more building your own amp. I spent way over $1000 to build my amp, that is not even including the test equipments. Building amp is for hobby and to learn, not for saving money. Just my set of pcb I layout is over $300 fab in China. My chassis is $325, then a lot of money in filter caps, transistors, wires, connectors...........

If you want fast result, the best way to get into it cheap is to buy used amps and restore them. Audiokarma is a forum emphasize in restoring, not building. You need a lot less knowledge and time to do restore. You can find a lot of ok quality amps and high power amps in reasonable price, get one of them and go to Audiokarma and seek help to restore it so you can use it.

For high power DJ amp cheap, look at Carva

Also look at Crown professional amp:https://www.ebay.com/sch/i.html?_odkw=DJ+amp&_osacat=0&_from=R40&_trksid=p2045573.m570.l2632.R2.TR12.TRC2.A0.H0.Xcrown+amp.TRS0&_nkw=crown+amplifier&_sacat=180014


As I said, these kind of DJ professional amp are a lot cheaper. I don't know anything about these amps, I just know you can get huge amount of watts for cheap.

- - - Updated - - -

I'll buy them.

Well one of my aim here was to build a proper preamp and then a proper amplifier. I tend to normally stop at the preamp since that's what I get help with I guess buying Chinese amps is what I'll have to use for my projects for now.

First all, you have to go to the correct forum to seek help. I don't want to offend people here, but this is not an audio forum. Only two of us replying to you, and I won't be around here much. I just happen to see this thread. You need to join the two forums I mentioned a few times. Tons of people there willing and eager to help. I am new, not an expert by any stretch. I only know one very narrow facet of audiophile amp, I can't even answer or suggest you on DJ amp other than what I saw on ebay. You need to go to the right place.
 
  • Like
Reactions: FvM

    FvM

    Points: 2
    Helpful Answer Positive Rating
I would like you to reply with quote atleast one time when I said I don't want to spend money....

I spend money every single day on electronics for years now, not as long as you I'm sure since I'm young but I build circuits daily and I ready everyday and night

- - - Updated - - -

In my country it definitely costs more to buy amplifiers even buying it used they over price everything, and importing electronics is a no no... It's costs more to clear things that the cost to purchase them at our customs.

So building amplifiers is the best bet for me. Plus the amplifiers I have built so far tend to give better results at times than the ones I buy and im refering to amplifier boards

- - - Updated - - -

Yes I'm aware its not an audio forum but those forum tend to be for experts only, when they explain things it's not for beginners and alot of persons are giving their facts/opinion/experience while that can help a person to learn it can also be confusing. I happen to not have a prob with just audio guru replying.

When I make posts like this I actually only expect him to reply at times others do which is ok also.

- - - Updated - - -

Also, I want to learn to make them other wise I'll just buy Chinese boards and build my own preamplifier stage I'll not challenge repairing old power amplifiers.

Making cases is much cheaper for me than for you and pcb making is also cheaper for me, might not be as high quality as yours though nor we'll designed but seems to be good enough for now I learn as I go

- - - Updated - - -

Diy audio is a forum I check from time to time, probably once or twice I have ever gotten a reply their so I don't even try to make posts since I don't get replies their.

I assume it's because my questions are so basic which in mind should be the easiest things to answer for professionals but they tend not to reply to them so I make posts here since audioguru pays attention to them

- - - Updated - - -

As I have said more than once.... I'll purchase the books you Recommended but just now now since I'm not able to at the moment.
 

Up to you. I just make suggestion to go to the people that been in it for a long time. Might not feel very good that you don't quite understanding them, that you have to study more. But in the long run, you learn faster.

I think I said enough here already. Good luck.
 

Enzy,
There is an DIY audio site in Australia that has many excellent projects. It has good articles too. They have a 1500W amplifier circuit as a joke but it will probably work fine.
**broken link removed**
 

When Radio Shack carried the LM383 IC, it appeared so easy to use that it was irresistible. Only five terminals. 8W (16W if you bridge two of them). I built a simple PA system with one. I put two in a stereo amp for driving desktop speakers. Minimum fuss. Minimum added components. Although it is not as powerful as you want, it could make a handy reliable building block to test with music sources, pre-amps, volume controls, equalizers, etc.
 

Also another thing, I am going to be honest and might be blunt.

Looking at this thread, it's pretty much Audioguru and me designing the circuit for you. You just draw into your schematic and post back. I don't know whether you really get what we are doing. There is a lot of very basic opamp circuit that you missed and we have to correct for you. This might work for simple preamp circuit like this so far. You cannot do this with power amp. You should really know more instead of posting some circuit and have others correct it for you. Then you just build and hope that it will work. Maybe that's the reason people don't reply back or post back something that you don't understand.

Yes, you do have to know more to talk in the other forums, this mean you need to prepare yourself first.

Even if you buy pcb or even stuffed power amp pcb and put it into the chassis, you still need to know more as you have to build the power supply and know how to do grounding. You mentioned about the hum, that's mostly grounding problem. This is another totally different subject. Then comes the worst part, making the amp stable. Not only it takes more knowledge of transistor, opamps, there is a lot of hands on things you need to know when you bring up the amp. That's when you need more people to help you. You cannot just show people picture, drawing and hoping people can make it work for you.

Just to give you some idea, I spent a few months studying Cordell's book and Doug Self's book before I even start the design. Remember I have been designing transistor circuits and analog IC for the last 30 years. Took me over a year to finish my first amp. This is very very intense, I retired and I spent a few hours a day working on my amp to complete it. The last part is to design the poles and zeros to make the amp stable with 0.1uF capacitor right on the output connector of the amp to ensure it's still stable. I had designed closed loop feedback for a while, it still took me a week to make the amp stable with 3300pF, 6600pF, 10000pF, 0.22uF, 0.1uF right at the output connector of the amp with no resistive load. This stage is very hands on, nobody can even help me on this.

It's almost a must to know how to do simulation also. I use LTSpice IV, but still it is not that accurate, only get you to the ball park.

Bottom line, if you think you can build an amp like how you post in this thread, it's not going to happen.

Sorry.
 
  • Like
Reactions: FvM

    FvM

    Points: 2
    Helpful Answer Positive Rating
When Radio Shack carried the LM383 IC, it appeared so easy to use that it was irresistible. Only five terminals. 8W (16W if you bridge two of them). I built a simple PA system with one. I put two in a stereo amp for driving desktop speakers. Minimum fuss. Minimum added components. Although it is not as powerful as you want, it could make a handy reliable building block to test with music sources, pre-amps, volume controls, equalizers, etc.

these are the practices I want to build since im more of a hobbyist.

I think designing circuits would be for someone who has a couple hours per day to read and practice and not all the time I have that but ill gain knowledge as est i can,

thanks

- - - Updated - - -

Also another thing, I am going to be honest and might be blunt.

Bottom line, if you think you can build an amp like how you post in this thread, it's not going to happen.

Sorry.

I think you tend to ignore me saying more than once or twice that this isnt my field im new to it... I build circuits that I see for a while now thats all I asked for help to build an amp and u cant help so its fine you dont need to repeat ways you cant help.

I posted about 2 times in diyaudio and there was no reply and I didnt even ask for help to build anything and I have seen alot of people ask things without reply which I think are good questions but again people like you tend to be too smart to answer simple questions (how ironic)

anyway you said you posted too much already so ill do my own practices and learn what I can when I can.

I read everyday and night I just finished working and its now past midnight in my country and I have work at 6am in the morning so I dont exactly have time hence me asking for so much help anyway as I said to audioguru just now im more of a hobbyist I love to build circuits im not really a designer but I love the idea of it and want to try a few designs of my own also.
 

Maybe it's your attitude in Diyaudio, I am not too smart to help you, look at this thread how much effort I put in correcting your circuit. Ungrateful is what comes to mind. If I were as green as you, I would be a lot more tactful and humble. I am definitely out now.
 
Last edited:

Your words bring forth words so I don't even know why you decide to even reply in that manner.

Audioguru said ntn bad so I say ntn bad to him I'm only replying to what you said as you can see.

Anyway this is a pointless discussion.

You seem to assume alot and say things that weren't said I'm not here for that.
 

I found this schematic what are your thoughts.

 

The high power class-D amplifier schematic uses the IRS2092 driver IC that has good audio spec's. You might need to buy many Mosfets to match three pairs of them.
The circuit is missing short-circuit and thermal overload protection so a failure will be a spectacular show.
 

I see where it has a 15+ and -15v rail and a 12v rail can I use voltage dividers to step down the rail voltages and then use voltage regulators to supply those voltages or would I have to use a seperate transformer?

where it says B+ and B- those would be the DC supply rails? if so using the FEts in the schematic would allow probably about 90v+/-

I have some Irfp150n Fets which would probably allow 50v+/-
 

Suggest to study the IRS2092 data sheet to better understand the power supply requirements and possible circuit options. Most of the driver power is supplied by the +12V on top of negative rail. Using an efficient source for this node matters for overall power dissipation.

The analog low voltage supplies are generated by series resistors and IRS2092 inbuilt zener diodes in the datasheet reference circuit.

You might need to buy many Mosfets to match three pairs of them.
The circuit is missing short-circuit and thermal overload protection so a failure will be a spectacular show.

MOSFETs of switched mode amplifiers can be usually parallel connected without selecting them. Positive Rdson t..c. causes self-balancing in on-state. Separate gate resistors also level the switching losses.

IRS2092 utilizes a protection scheme known as "desaturation" detection with IGBTs. If the voltage drop in on-state becomes too high, the driver is locked. The protection scheme doesn't work very well for low and medium voltage MOSFET, not sure if it can reliably switch-off output shorts. Detecting overload is even more difficult.
 

I'll be reading up on this chip I'm trying to get some to purchase also, I haven't found any as yet though.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top