+ Reply to Thread
Page 4 of 4 FirstFirst ... 234
Results 31 to 32 of 32

Thread: Class A power amplifiers

  1. #31
    Join Date: Jul 2010

    Location: North Cambs UK, Earth, Sol, Orion - Cygnus arm of galaxy

    Posts: 11,166
    I'm MadeOfDeadGiantStarsThatExplodedEonsAgo.

    Default Who is LJM?

    Well this post is more for anyone that purchases an L25D IRS2012 powered class D amplifier PCB For my own records as well if my PC goes t*ts up

    For some reason I can't fathom there are metal film resistors feeding the gates of the power Mosfets? Metal film resistors are inductive & shouldn't be in there at all. These will be replaced with carbon resistors (on the way now), also as the IRS2092 can supply up to 1Amp peak source & 1.2Amp peak sinking current I have dropped the gate resistor value to 15 ohms. Along with the IRFB4020 internal gate resistance of approximately 2.5 ohms with a 15V supply there will be no problems.

    There are also two snubbers on the PCB, 10 ohms & 150pf, again the resistors are metal films Once again carbon resistors will be fitted.

    Finally (for now) there is zero sign of a ground plane, or should I say three of them as per the IRS2092 application notes. What's even worse is that all the grounds are connected together at the ground/zero volt input. so this will need to be connected to the ground of a +/- supply & all the interference from output filtering, power supply decoupling etc will be superimposed on to the ground feeding the audio input

    I have purchased a load of 0.1mm thick copper sheet & will make the required ground planes, cut tracks where needed & connect them back to the proper grounding point at the capacitor bank.

    If a job is worth doing, it's worth doing right first time.


    On another note the designer appears to have done something rather weird in yet another area. High side current limiting is done by measuring the voltage accross the high side Mosfet, in the application notes there is a resistor from the high side gate that forward biases a diode when the high side is on (thus enabling measurement of current minus the diode voltage). In this circuit the diode is biased on via the high side power supply I can't think of any advantages to this as there is a 450nS delay from the high side switching on to measurement of current. I can't see any reason for this in all honesty. Might even cause problems. The diode was also as slow as hell in reverse recovery, as I'm using a lower voltage a very fast 1N4148 should be fine.

    Anything else to pick to bits... Not right now


    Current limiting done right on the IRS2092 application notes. R1 is connected to the gate of Q1, on this PCB it is connected to the high side power supply or VB


    IRS2092 application note recommended ground planes... Makes sense as 500KHz + can make a lot of noise


    What no ground planes Can anyone spot copper? Oh you will given time... Stuff in red is new value resistors etc to do the job I want it to I'll need to keep a copy of this as most of that PCB & component selection is about to be covered with 0.1mm of copper foil glued on..


    Bye for now..
    Bests, Mark



    "We must believe in free will. We have no choice" Isaac Bashevis Singer

  2. #32
    Join Date: Jul 2010

    Location: North Cambs UK, Earth, Sol, Orion - Cygnus arm of galaxy

    Posts: 11,166
    I'm MadeOfDeadGiantStarsThatExplodedEonsAgo.

    Default

    Well I'd like to once again thank Steve (activexp) for this post he made, I took a look over at DIYAudio & what was on there..

    Loads of information on both the L15D, L20D & some on the L25D which these amp modules happen to be. It looks like these are some of the best implementations of the IRS2092 control chip out there It also looks like I fessed up the chip number in my previous post

    So in accordance with the old saying, "if it aint broke then don't fix it" I'm going to drop the ground plane idea & get on & build the things but with certain new components to increase the current limit etc.

    As it happens the output inductors supplied as standard will be spot on for an 8 ohm load in a bridge amplifier as they are designed for 4 ohms single ended (in a bridge each amplifier sees half the actual load impedance). For a 4 ohm load I'll need to halve the inductance value, which got me looking on ebay as usual. Plenty to choose from but the most promising where on the other side of the pond. So I decided to look for similar iron powder toroids as used in those inductors... No problem winding my own & I can go ott...

    I must be lucky as I stumbled accross a listing by an old work colleague who lives in the same city as me & who I'm still in touch with of the same type iron powder toroid but just a bit bigger So I get them at his BIN price & he'll drop them round at no extra cost... As an added extra he emailed me a list of components he has been put in charge of getting rid of in bulk from my old & now defunct employer, no reasonable offer refused


    It's good to keep in touch
    Bests, Mark



    "We must believe in free will. We have no choice" Isaac Bashevis Singer

+ Reply to Thread
Page 4 of 4 FirstFirst ... 234

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •