Help with class AB amp

Discussion in 'General Electronics Chat' started by CRY&PRAY, Jan 6, 2014.

  1. CRY&PRAY

    Thread Starter Member

    Jan 6, 2014
    84
    0
    btw at the end how can i improve it? i can put more transistors.. i have plenty of them :3 i just want to get the maximum efficency and stability at higher voltages like 90 v (45-45).
    Thx :)
     
  2. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    511
    Buy one of these:

    http://www.ti.com/product/lm3886

    There's a reason stereo makers no longer use junky discrete designs like this any more. They have numerous problems. The bias current doesn't track because the output transistor temp is different. There is no current limit or over temp protection. An integrated device such as above eliminates this and gives excellent reliability.
     
  3. bertus

    Administrator

    Apr 5, 2008
    19,521
    3,977
    CRY&PRAY likes this.
  4. CRY&PRAY

    Thread Starter Member

    Jan 6, 2014
    84
    0
    So, pratically build an amp from scratch... it's a disaster :) ho well, ICs sucks badly, they push out a max of 100 watt when respecting their condiction and there's no fun!. I have that omnitronic e900 board. A lot of stuffs burn out on that pcb but I knew all the pieces burnt out. I found that at the trash just in front of the home, so i got it, it's been repaired more times and that makes me think... that's why they trashed it, i'll post some pictures tomorrow to see if it can be repaired, cuz i need a powerfull amp (600w or more) for an event and i'm the guy who will sit 8-10 hours in front of a mixer changing cds and enabling microphones.

    I've repaired that board once and worked... hem not well, it was amplifing but like 100 w from a 1000 w amp, then while playing stuffs exploded and burned out like fireworks :D:D , infact there is a op amp that's burned out, maybe that guy fried out the rest, it's a TL084CN. Also i need to know what these transistors do on the pcb, cuz if they exlode again i knew what to check :)

    All the stuffs written is bruned or to be replaced because it looks bad, all the rest has been tested and working fine
    the pain are those d669a, i think these guys are biasing the big brothers (one for channel), but it's unfindable (don't know if this word exsist)

    [​IMG]

    or go class D, just find a IC that has an analog input for the audio and 4 output to go directly in the base of some mosfets, then just put 4 mosfet, that ic powered with some voltage regulators from the main power supply, and push power in the mosfet.. and putting the filter at the end. The problem is that are no ICs like that :(

    and as usual thx a lot
     
    Last edited: Jan 8, 2014
  5. Veracohr

    Distinguished Member

    Jan 3, 2011
    676
    103
    If you're more interested in designing the whole thing for your amusement, I understand that, but if you're more interested in the end result, you might be able to take an IC like the one linked and use some external transistors to boost the current.
     
  6. atferrari

    AAC Fanatic!

    Jan 6, 2004
    3,254
    1,269
    Sad news for a hobbyist... :( Been there...
     
  7. Dr.killjoy

    Well-Known Member

    Apr 28, 2013
    1,199
    178
    At least you tried ... Now you know how not to do it ..
     
  8. CRY&PRAY

    Thread Starter Member

    Jan 6, 2014
    84
    0
    what if i build a lm3886 amp and replacing the speaker with a final power push pull thing.... pratically the scheme that i did without the 3rd transistor. will it blow? will the 1k resistor limit the epicness?.

    i have a Pyle PTA1000 amp (300w max rms, yeah, hoped that were 1000 rms.. :( ) if i take the output of that amp (the bases of the final transistor) and use a optocopuler and drive a transistor with the PSU of (95+-) of the Omnitronic e900? so pratically, taking a signal well done and building just the last few bit. Or wiring like that: using only the (95v+-) and using some voltage regulators to give 30 v +- at the cirquit and then putting all the power on the final transistors?

    [​IMG]
     
  9. Veracohr

    Distinguished Member

    Jan 3, 2011
    676
    103
    I haven't built such a circuit, so I can't advise much, but I was thinking of something like this circuit:

    http://ampcircuitschematic.blogspot.com/2012/03/3000w-stereo-power-amplifier-circuit.html

    If everything to the left of Q13 & Q14 is an IC (with its own supply as necessary), then you could connect the output to multiple push-pull stages in parallel to boost the current. Maybe that's kind of like what you said, but you confused me. Further advice will be necessary, I'm sure, to do this correctly.
     
  10. CRY&PRAY

    Thread Starter Member

    Jan 6, 2014
    84
    0
    I'll post some photos of the amp (inside) and, like you sayd, i'd like to attach a different push pull stage with higher voltages to the output of that amp, pratically that schematic, but with a double power supply, one for the "epic stuff" in the beginning, and one only for the push pull stage. i just want to attach 2 wires on the bases of the amp (working) and using those 2 wires to drive an other push pull stage with higher voltages. I'm trying to make an hybrid of 2 amps. the PylePro PTA1000 and the remanings of the Omnitronic e900 (power supply and push pull stages). is that possible?
     
  11. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    511
    Anything is possible if you want to throw enough time and money at it.

    As I said before, when the power stage is not thermally contacting the bias circuitry and drivers, the bias currents won't track and you will have thermal runaway failures. And every power stage needs a very fast current limit circuit or else it will just be a fuse waiting to blow up.

    You also need some kind of soft-start and instant speaker disconnect to prevent turn on/off thumps.
     
  12. tubeguy

    Well-Known Member

    Nov 3, 2012
    1,157
    198
    Yep
    Usually (almost always!) the output transistors blow up before any actual fuses blow.
     
    Last edited: Jan 10, 2014
  13. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    511
    Yeah, that was the joke..... power stages without some kind of very fast current limit ARE the fuses....:eek:
     
  14. CRY&PRAY

    Thread Starter Member

    Jan 6, 2014
    84
    0
    Yep... i've experienced that, that amp is a small block of metal with a pcb on a side with 5 transistor touching it, 3 on a side, 2 on the other. 2 of these transistors are the final, what if i put 2 more of them (so totaly of 4) just like the omnitronic and i put 95+- vs on that pcb? (that pcb rns at +-30v) or the thing that i'm says, running the pcb at +- 30v with it's own psu and only the 4 collectors attached to the +-95 v, both psu negative grounded to each other.

    The omnitronic has a antibump thing :3 and i know what my speakers do without a antibump, 've burned a old CIARE 12" dual cone speaker :''''''(
     
  15. CRY&PRAY

    Thread Starter Member

    Jan 6, 2014
    84
    0
    I need a powerfull amp for the 15th February. I'm gonna use the Pyle Pro PTA 1000 if i don't make something better. So i want to be sure at 100% that if i add 2 more transistors to the PCB (there are holes for the PTA1400, they uses the same pcb) and attach ONLY to the collectors an other PSU (+- 95v) and all the rest to it's PSU, will work, or will explode?

    (the transistors are the ones that i've got to repair the Omnitronic but didn't worked with them, other transistors were burn, so i've used them in that amp.)

    and the PTA1000 is chinese powered 100%, i've cried when i saw all those bad soldering jobs :3 but it was cheap so i got it.
     
  16. MrChips

    Moderator

    Oct 2, 2009
    18,530
    5,883

    Here is my advice.

    You're the sound guy who has to make everything run smoothly on 15th February without smoking it otherwise you get more than pie in the face.

    Building an audio amp from scratch can be fun for the hobbyist. There are lots of circuits for up to 10W audio amps.

    There are more circuits that can take you to 100W and beyond.

    As a hobbyist you are going to smoke it every time with your kind of expertise and construction practice.

    You are going to be hard pressed to meet your goal for 15th February.

    Go out and rent a 600W-1000W amp from a sound equipment rental. Cost about $50 for the weekend for mixer, amp, two speakers and a monitor, including all speakers and mic cables.
     
  17. CRY&PRAY

    Thread Starter Member

    Jan 6, 2014
    84
    0
    yeah, you're right :) just need an amp, the speakers are already mounted, we have a plan B, i have 2 PA speakers (FBT verve 12) and a bad ass mixer, but those speakers sounds awfull in a closed zone, there are a lot of reflection over the walls, so i want to use the speakers that are mounted there (without an amp :( ) because are 8 speakers mounted 4 for side spreading the sound better than 2 powerfull one on an angle (not even a side)

    i'll use the Pyle Pro PTA1000 (or hope to repair the Omnitronic in time) just wanted to have some headroom to not run at the max power before clipping.

    Other cool thing about the pyle pro, is that they use the same pcb for the pta1000, 1200, 1400. The 1400 has 2 more power transistors, all the rest is the same, so i thought what if i double the power (like the pta1400) and add the 2 transistors with their resistors? i think it will work, there are no other components, the only thing that i don't know is: the PTA1400 PSU! does it give double the voltage or the current? the transistor that were there before had a max voltage of 140v 10 A, the amp (pta1000) runs at +-45 v, but the pta1400 with 2 more transistors?? maybe at +-60v with more amperes?

    thx :3
     
  18. MrChips

    Moderator

    Oct 2, 2009
    18,530
    5,883
    You cannot take a power amp and just add a couple of transistors and resistors and hope to increase the power output.

    Power amps do not work like that.
     
  19. CRY&PRAY

    Thread Starter Member

    Jan 6, 2014
    84
    0
    I know :3 i'll add those power transistor and then increase the PSU, i know that i can't generate power form nothing... There are the holes for two more transistors and their resistors, just need to know how much i can increase the PSU without fuc***g everything..
     
  20. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    511
    You keep asking the same question but it's one that can't be answered until you actually build it, test it, and dial in the design.

    Exactly.


    I worked in customer service. When customers used to ask me questions like that, I would carefully explain the facts. If they asked again, I would explain it again. If they kept asking, I would just make up a number for an answer and that would satisfy them or at least make them stop calling my phone.

    Pick any number that gives the answer you want.....
     
    Last edited: Jan 11, 2014
Loading...