1. Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.
    Dismiss Notice

About make an audio power amplifier.

Discussion in 'General Electronics Chat' started by Nikolai Petrenko, Jan 30, 2016.

  1. spec

    spec Well-Known Member Most Helpful Member

    Joined:
    Oct 27, 2015
    Messages:
    5,828
    Likes:
    710
    Location:
    Somerset UK GMT
    OBSOLETE. PLEASE SEE POST #87

    Here is an outline circuit of an opamp PSU, that I mentioned previously:

    ETO_2016_02_16_Iss01_PREAMP_PSU_OPAMP.png
     
    Last edited: Feb 16, 2016
  2. granddad

    granddad Active Member

    Joined:
    Jan 18, 2015
    Messages:
    762
    Likes:
    76
    Location:
    Worcestershire UK
    dont know noise spec of L200 , but made in China....
    adjustable-regulator-circuit-using-l200.jpg
     
  3. spec

    spec Well-Known Member Most Helpful Member

    Joined:
    Oct 27, 2015
    Messages:
    5,828
    Likes:
    710
    Location:
    Somerset UK GMT
    The 35V max voltage is between the input terminal of the 78xxx and the 0V terminal. As the 0V terminal is up at 10V in my PSU circuit, the safe input voltage to the PSU is 35V + 10V = 45V
     
  4. dave

    Dave New Member

    Joined:
    Jan 12, 1997
    Messages:
    -
    Likes:
    0


     
  5. spec

    spec Well-Known Member Most Helpful Member

    Joined:
    Oct 27, 2015
    Messages:
    5,828
    Likes:
    710
    Location:
    Somerset UK GMT

    Nikolai,

    It would be much better if you could power the preamp from a separate winding on the main PSU transformer or, better still, from a separate small transformer.
     
    • Like Like x 1
  6. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,573
    Likes:
    950
    Location:
    Canada, of course!
    I do not see how the zener diode-opamp-transistor power supply in post #81 starts. When power is applied then the (+) input and output are at 0V so they might stay like that. I know that the LM358 uses PNP input transistors that try to pull the inputs positive a little but both inputs will go positive a little. But the (+) input has a higher resistor value but again if the input offset voltage is the wrong polarity then it will not work.
     
  7. spec

    spec Well-Known Member Most Helpful Member

    Joined:
    Oct 27, 2015
    Messages:
    5,828
    Likes:
    710
    Location:
    Somerset UK GMT
    :happy: Excellent AG, you have spotted the obvious shortcoming- it is only an outline design and I couldn't be bothered to sort the starting
    spec
     
  8. spec

    spec Well-Known Member Most Helpful Member

    Joined:
    Oct 27, 2015
    Messages:
    5,828
    Likes:
    710
    Location:
    Somerset UK GMT
    Opamp reg with start-up:
    (1) Output voltage = 5V6 * (1+ R8/R11)= 15.68V
    (2) One of the strong features of this topology is that a constant current is maintained into the Zener diode. Ik= (15.68V- 5.6V)/R5 = 4.58 mA

    ETO_2016_02_16_Iss02_PREAMP_PSU_OPAMP.png
     
    Last edited: Feb 17, 2016
  9. Little Ghostman

    Little Ghostman Well-Known Member Most Helpful Member

    Joined:
    Jan 7, 2011
    Messages:
    3,737
    Likes:
    433
    Location:
    SCOTLAND
    Sorry I am WAY to busy to go around deleting your nonsense :D:p, say pretty please LOL
     
    • Funny Funny x 1
  10. Little Ghostman

    Little Ghostman Well-Known Member Most Helpful Member

    Joined:
    Jan 7, 2011
    Messages:
    3,737
    Likes:
    433
    Location:
    SCOTLAND
    Sorry I thought you said ADAM! I know its not my real name but my mum always calls me that! She is forever saying And erm (adam) another thing.........................:D
     
    • Funny Funny x 1
  11. Nikolai Petrenko

    Nikolai Petrenko Member

    Joined:
    Oct 8, 2015
    Messages:
    399
    Likes:
    16
    Location:
    Vietnam-motherland
    Your circuit look pretty like basic of what inside a LM431 and the circuit in my friendy's guitar preamp use 8V2 zenner and LM741 opamp.
     
  12. Nikolai Petrenko

    Nikolai Petrenko Member

    Joined:
    Oct 8, 2015
    Messages:
    399
    Likes:
    16
    Location:
    Vietnam-motherland
    Yes, like my think, I have many EI small transformers, usually dual 5-18V, very well.
     
  13. spec

    spec Well-Known Member Most Helpful Member

    Joined:
    Oct 27, 2015
    Messages:
    5,828
    Likes:
    710
    Location:
    Somerset UK GMT
    Very good move. With a separate preamp transformer you can have an optimum OV signal routing so that distortion artifacts from the massive current and voltage signals of the power amp don't mix with the low-level preamp signals.

    Also you can then easily avoid the excessively large input voltage to the preamp PSU regulators. A high input voltage causes heating and noise in the regulator electronics. Incidentally a series regulators, unlike the Zener shunt regulators that you mentioned, have a big advantage that they do not inject so much noise, ripple, and hash down the 0V line.

    I would also be inclined to build the preamp in a separate, small case too, but that is just my preference, because it provides a more modular system.
     
    Last edited: Feb 17, 2016
  14. Nikolai Petrenko

    Nikolai Petrenko Member

    Joined:
    Oct 8, 2015
    Messages:
    399
    Likes:
    16
    Location:
    Vietnam-motherland
    Put the preamp into a galvanized steel box will keep signal clear from 50Hz hum and RF noise.
    The transformers will be covered by a galvanized steel box too, but I will not make any chance for it to cause shorted turn.
    Main AC 220V wires will be twisted together andwrapped by thin galvanized steel sheet ( but it is not easy wrap fine because of sheet's springy). I think all other wires should be wrap by aluminium foils. Then all those shields will be soldered into main ground point.
    The amp case make from hard work plates as reinforce. Then large thick steel sheets will cover all, sheets also be soldered to main ground point. Only leave series of holes under and above the heatsink to allow air flows.
    No fan because it is noisy and quickly become a dust sucker.
     
  15. spec

    spec Well-Known Member Most Helpful Member

    Joined:
    Oct 27, 2015
    Messages:
    5,828
    Likes:
    710
    Location:
    Somerset UK GMT
    Rolls Royce preamp :cool:

    Although it will do no harm, it will not be necessary to shield any wires except those carrying high voltage AC.
    No, a fan would not be necessary for the low power dissipated in a pre amp.
    As a general point, always design and build equipment with development/servicing/maintenance in mind so that no item is inaccessible. There is an unwritten law that says that, the more inaccessible an item is, the more likely it is to give trouble. For example, with the case design that I proposed for the power amp in the Transistor Equivalent thread, you will notice that all items are accessible, including the underside of the power amp PCBs.

    spec
     
    Last edited: Feb 17, 2016
  16. Mikebits

    Mikebits Well-Known Member

    Joined:
    May 24, 2008
    Messages:
    6,217
    Likes:
    175
    Location:
    San Diego, Ca
    And connectorize your I/O for easy board removal. :D
     
    • Like Like x 1
  17. spec

    spec Well-Known Member Most Helpful Member

    Joined:
    Oct 27, 2015
    Messages:
    5,828
    Likes:
    710
    Location:
    Somerset UK GMT
    It is amazing the number of times you see servicing aspects completely ignored, or worse still made impossible for no good reason. I don't know about American cars (autos) anymore but European cars are now especially bad, leading to unnecessarily high servicing costs. On the first Ford Escorts you could change the gearbox in an hour with no hassle. Just try that on the modern equivalent, the Ford Focus. Mind you, the Austin/Morris Mini was always a pig to work on.

    I came across a blatant case during a design review: this bit of the system comprised a worksurface with a processing unit under it and a mouse and display on the top. The mouse cable passed through a hole near the edge of the worksurface. This meant that to assemble the sub-system the connector on the mouse (26 pin D type in those days) had to be removed, the cable passed through the hole in the worksurface and the connector re soldered. The odd thing was that the mouse failed quite often and necessitated this desoldering/soldering procedure to fit a new mouse. Turning the hole in the worksurface into a slot sorted the situation, but the worrying thing was that the engineer responsible for the design just could not see it and was most indignant that his design had been criticized. Incidentally, after changing to a slot the mouse problems ceased, why I will never know. :wideyed:

    And the engineer. Well he was a newbee, and developed into a solid designer, especially focused on modularity and serviceability. :happy:
     
    Last edited: Feb 17, 2016
  18. Nikolai Petrenko

    Nikolai Petrenko Member

    Joined:
    Oct 8, 2015
    Messages:
    399
    Likes:
    16
    Location:
    Vietnam-motherland
    No, I mean fan for power amp sections.
     
  19. Nikolai Petrenko

    Nikolai Petrenko Member

    Joined:
    Oct 8, 2015
    Messages:
    399
    Likes:
    16
    Location:
    Vietnam-motherland
    My construct right?
     
  20. spec

    spec Well-Known Member Most Helpful Member

    Joined:
    Oct 27, 2015
    Messages:
    5,828
    Likes:
    710
    Location:
    Somerset UK GMT
    Got it.

    With class AB amps, unless you have a high quiescent current or an output power capability of over 50W, power dissipation in the output transistors and drivers is seldom a problem, and certainly would not require a fan. Class A amps are a different matter. In fact, by good heat sink design you can dissipate surprisingly large amounts of power without a fan, but the downside is that the heat sinks tends to be large. In general many thin fins are better than a few thick fins. If you need an illustration take a look at the cylinder head and barrel of a typical air cooled aero engine:

    [​IMG]

    With the heat sink running along the entire width of the case and vertical fins, as in my suggested case design on the Transistor Equivalent thread, heat sinking will be good, especially with the NP aluminum casting technique. :happy:

    Sounds good, but I really would need to see an outline design to comment. One thing that I did not mention before is the question of OV points: you need two OV points: one for signal and one for screening. The way that they should interconnect is difficult to describe, but the screening OV is much less critical than the signal OV.

    I have built amps and preamps with both steel and aluminum cases and the most difficult, from an induced hum point of view, has been steel. This is because steel is a magnetic material and it can induce unwanted currents. I would advise that the preamp be built exactly the same as the power amp, except much smaller and, of course the back panel can be plain sheet, rather than finned. Even if you have a steel back panel you can still bolt the preamp stabilised PSU dissipating elements to it to get adequate heat sinking.

    You mention soldering, and I assume you mean silver soldering rather than soft soldering. Well, that is a good construction technique so long as you look out for distortion due to heat. Another good approach is to use fibre glass or epoxy resin to form joints, making sure that you have positive bonding between panels of course.

    I once made a multi-meter case from fibre glass sheet bonded together with fibre glass resin and mat. Once it had been sanded down and given a coat of epoxy paint, it really looked the part- shame I didn't finish the electronics. :banghead:
     
    Last edited: Feb 18, 2016
  21. spec

    spec Well-Known Member Most Helpful Member

    Joined:
    Oct 27, 2015
    Messages:
    5,828
    Likes:
    710
    Location:
    Somerset UK GMT
    I know you are only having some fun, but Nikolai is 14 years old and is currently a student. His family, culture, and ties are in Vietnam so that is where he is likely to want to live. Yes, he does chop and change and sometimes gets things out of balance, but he is keen to learn and is way ahead of me at that age.

    In those days I too relied on my parents for dosh and went to all sorts of shenanigans to do things on the cheap while still trying to get good quality reproduction from valve amps (no transistors in those days). It is a great shame that I did not know then what I know now because I could have built a fabulous system with the junk I had accumulated. For example a hugely expensive Parmeko ultralinear valve output transformer that was simply being chucked out and a massive mains transformer rescued from some military test equipment.

    I also had access to any number of valves, including 6V6GT, KT66, ECC83 etc. But I simply did not know what to do with this stuff, especially as there were no freely available data sheets in those days and the valves were all identified with military CVxxx nomenclature. The only data I had was the Mullard application report for their 5/10 amp. One example is that there was multiple 6.3V winding on my mains transformer and the 5/10 amp only had one and non had the same current rating of the 5/10 data sheet. They were all a lot more. I thought that this extra current would simply burn out the valve heaters or cause some other problems. It was a similar thing with the transformer HT winding too. And I had no idea that CV type valves were actually superior to their commercial equivalents. The 5/10 circuit was labeled with various voltages and currents. I thought that these were absolute and if not spot on the amp would not perform properly and so on.

    I even had a nice Stentorian base driver that was also being dumped, and rescued a massive 18 inch speaker from a cinema that was being refitted. But I had no idea how to design a suitable enclosure, although I could have done the woodwork. The Stentorian had an impedance of 16 Ohms, but the output transformer had a 15 Ohm secondary so I thought they were incompatible and would generate gross distortion. If someone could have put me on the right track here and there I would have progressed much better.

    So, I say good luck to Nikolai and others in the same position and I for one am only to pleased to help in his endeavors where possible. :happy:
     
    Last edited: Feb 19, 2016
    • Thanks Thanks x 1

Share This Page