Actions
Forum Thread Details
  • State Suggested Answer
  • Replies 9 replies
  • Answers 2 answers
  • Subscribers 181 subscribers
  • Views 1642 views
  • Users 0 members are here
Related

High power class D Amp Build (5 channel, spectrum analyzer, pre&post-amp signal leveling)

Former Member

Well I hope this is the right place to start this discussion. I don't know much about element 14.

 

Anyways, to the point.

     The goal here is to make a high power (ie 5000w) 5 channel class d amp with signal leveling before and after the amplification process as well as having a spectrum analyzer. To get even more in depth when I explain this lets say I want it to take an audio signal in, analyze and level that signal, amplify that signal, analyze and level that signal again, send that signal to a spectrum analyzer, and cross that signal over to 5 channels (high-fq-l/r, mid-fq-l/r, and then 1 low-fq) all while maintaining great efficiency and audio quality.

     I need help making the power supply, the eq and spectrum analyzer (thinking about Arduino), the amp it self, and the cross-overs (everything pretty much). And It has to be from scratch (as much as possible).

 

     If I can get a list of all the components I need to have knowledge about I think I can figure out how to make this project come together. That's basically what I'm asking of you. Can Someone get me a list of components to research such as MOSFETS, Microprocessors, etc...?

Parents
  • 0

    This isn't the right place so perhaps someone will move the thread to the audio group where it might be a better fit (it doesn't really have anything to do with Eagle).

     

    Having said that, I can't understand why you would want signal processing (of which leveling is just a special case) after the amplifier - you can't do complex digital signal processing on 5kW audio.

    Arduinos are no use here - you are into DSP territory.

     

    Building a kW level class D audio amplifier is quite an ambitious task - have you any experience of doing this sort of thing ?

     

    First you need to decide if you want to start on the signal processing side or the power amplifier side - then explain (to us and yourself) why you need 5kW which seems like far too much power (where are you going  to install this monster ?) . Once we know the power source and type of load it starts to be possible to define the design of the amplifier.

     

    MK

  • 0 in reply to michaelkellett

         Well, the main goal is keeping the quality of the audio as good as possible. With that said, from what I understand there will be amounts of distortion after the amplification process. With another EQ, one would be able to get rid of some of that distortion... to my understanding at least.

         No I do not have any experience with this particular sort of thing, I am however in college for automation controls and SCADA, so I do have an understanding of what I might be getting into and I can get help along the way. I would have consulted my instructor but he does not get into this sort of thing.

         Anyways, this setup is going to be in a car, and eventually I want to be able to sell them. You see, I'm working on a prototype. I have a very large group of people I know who would love something like this and will pay big money for it. Besides that, my own passion and being an audiophile myself makes this all the more possible.

         The power source is going to be a bunch of battery's and hi-output- alternators much like what you would see in a professional sound competition set up or something like that. I'm not really worried to much about the power source entirely at the moment. Once I figure out what I need from the source to power all the components I want in this thing is when I'll work on it. to be honest I may decide to go up to 10kw. I'm dreaming big here I know but that's not going to stop me. haha.

  • 0 in reply to Former Member

    EQ can't get rid of distortion at the output of a power amplifier - you would need another power amplifier after the EQ - we solve this problem (in a great many control systems not just audio) by using feedback so we compare the actual output with the output we wanted and adjust the signal going to the amplifier or features of the amplifier accordingly.

     

    I'm going to bug you by saying this but you need to walk before you run - my first audio power amplifier was based on a design from the Mullard Audio Handbook and produced 25W. I went on and built quite  a few high power audio amps from scratch - I still have a roughly 200W per channel amp that I built about 40 years ago  - it still works. A 10kW amp is a big and serious project - at his power level you will almost certainly run into serious safety issues (and not just your ears) as well as very difficult engineering problems. You are years of study and effort away from this.

     

    BTW -  5kW is completely un-necessary to achieve a high sound level in a car - I know some people mess with such systems but they have about the same relationship to audio for listening as drag racers do to family saloons.

     

    I can't really help you in a general sense - if you come back with specific questions I'll do my best. if you decide to try your hand at a less ambitious amplifier for a start I can make some suggestions.

     

    MK

  • 0 in reply to michaelkellett

    I appreciate your input. I do understand where you're coming from. I do think I will ask for your advice on where to start with this. Instead of going for such an extreme output right away lets say I want 1000w I don't think that's too extreme. With that I do have questions. How can I make an equalizer from scratch? I know I want it to be 10 band.

  • 0 in reply to Former Member

    When you say you would like to build an equaliser from scratch do you mean design and build at chip level and write your own software  - that's a big task. You can buy a Behringer Ultragraph DEQ102 for £139 - if I were to quote for designing a similar instrument from scratch for a customer to build in low volume I think I would estimate in the region of 800 - 1200 hours work - and I've done this kind of thing before.

    Building some switching power amps is a much more feasible task:

    Google the TI TAS5630 - this chip can manage 300W, TI's support documents are quite good.

     

    MK

     

Reply
  • 0 in reply to Former Member

    When you say you would like to build an equaliser from scratch do you mean design and build at chip level and write your own software  - that's a big task. You can buy a Behringer Ultragraph DEQ102 for £139 - if I were to quote for designing a similar instrument from scratch for a customer to build in low volume I think I would estimate in the region of 800 - 1200 hours work - and I've done this kind of thing before.

    Building some switching power amps is a much more feasible task:

    Google the TI TAS5630 - this chip can manage 300W, TI's support documents are quite good.

     

    MK

     

Children