Tech Square

Posted by Granholm (A) on February 26, 2006 at 09:53:22

Hi,

I have thought of upgrading my PSU for a Class-D amp (Zap Pulse 2.3SE). I have the possibility to have a teroide custom made but dont know much more than they should have loads of VA rating ;o)

Any one that have any input as what would be an optimal design??

Main Current is 230V 50Hz supply current should be from 45 til 55V +/-.

I read something about an cobolt alloy core once some years ago? Would this till be a good idea or are there better materials for this aplication?

I have all the physical space I need (at least up to an 3KV teroide).

Current draw is not much on idle run as the Class-D amp only use about 20W (both mudules together) so I think the most important is to have as rigid supply as possible so that peak situations are supported without any voltage drop.

Also It might be a good idea to implement a screen? Or not??

Any imput would be much apreciated and any results that come out of this would be posted.

Have Fun

• Re: Any X former experts? - mettula 14:41:30 02/25/07 (0)
  In Reply to: Any X former experts? posted by Granholm on February 26, 2006 at 09:53:22:

  A 5 VA 240V to 24V step-down transformer operating at a frequency of 50Hz is being designed. Whatare the 3 types of materials which could be suitable for the construction of the core material. thanks
  
  

  ---
• Try Amveco #AA56252-038 - Russ57 10:28:50 03/02/06 (4)
  In Reply to: Any X former experts? posted by Granholm on February 26, 2006 at 09:53:22:

  That is a 2 by 38VAC, 625VA transformer. Dual 117VAC primary. If you contact them they will normally wind for whatever voltage you want for litte to no extra money. So getting 42 or 45 volt secondary should not be a problem.

  Yes I would get a screen and no I would not get cobalt. It almost always helps to have a transformer capable of 150 to 200% of actual current draw. I can offer no science for this but can say it certainly sounds better to me.

  Don't overlook good, fast, big caps with nice slew rate times. It has been a long time since I built solid state stuff but back then Siemens sikorel was a fantastic cap. Hard to find, not many sizes, and not cheap, but performance/specs much better than anything else. But time marches on and they might not be the best anymore.

  
  Russ
  
  

  ---

  • Might get in trouble as my main is 230V AC ;o) nt! - Granholm 12:32:09 03/02/06 (0)
    In Reply to: Try Amveco #AA56252-038 posted by Russ57 on March 2, 2006 at 10:28:50:

    nt!
    
    

    ---
  • Re: Try Amveco #AA56252-038 - Granholm 11:41:39 03/02/06 (2)
    In Reply to: Try Amveco #AA56252-038 posted by Russ57 on March 2, 2006 at 10:28:50:

    Hi Russ57,

    Was thinking og RIFA PEH 169 or 200 series. But yes the Siemens was very hot back then and I think they still are. But the Aerovox T-power I am using is not to bad eigther! Though they are only rated at 63 voltage wich "only" allow secundary AC to be 2x42VAC (59,xxDC).

    What I was wondering was if there would be even the sligthest benefit from adding even more VA (Actually only A)?? As I have space enough for an toroide up to 3000VA rating (standard build).

    What is your experience with cobolt?

    Have Fun.
    
    

    ---

    • Re: Try Amveco #AA56252-038 - Russ57 12:18:17 03/02/06 (1)
      In Reply to: Re: Try Amveco #AA56252-038 posted by Granholm on March 2, 2006 at 11:41:39:

      Yes, in my experience having a transformer capable of more current than you need makes a nice difference, but I see no point in going past 200%(I guess for you that is about almost 1,000VA tops?). One advantage is that the larger a transformer is the better it's regulation, the difference between loaded and unloaded voltage. So I think it makes for a stiffer supply. Cobalt's big advantage is lower core loss. That is not important to you unless you plan to own a power company and have a bunch of transformers:) Cobalt will have a bigger inrush current and that will already be a problem using "doughnut" transformer. So for the inrush reason I say don't bother with cobalt. Save the fancy core materials for audio coupling transformers such as those used in tube amps. Even then I'd suggest M4 and/or nickel.

      Actually an EI style transformer will not couple high frequency hash on the AC line as well as a "doughnut" type. But they tend to put out more electrical fields and are physically larger. Personally I like EI better.

      
      Russ
      
      

      ---

      • I have heard that before (EI is better) - Granholm 12:29:45 03/02/06 (0)
        In Reply to: Re: Try Amveco #AA56252-038 posted by Russ57 on March 2, 2006 at 12:18:17:

        But what if I ad endcaps?? Would this not prevent some of the stray fields? Or would it also deteriate the sound?

        Have Fun
        
        

        ---
• I'm not a whatever expert. - cheap-Jack 12:25:24 02/27/06 (11)
  In Reply to: Any X former experts? posted by Granholm on February 26, 2006 at 09:53:22:

  Hi.

  Assuming I understand what you said or what you really want to say,
  the total power power consumption of your class D power modules is
  20W.

  
  As class D switching amp assuming it is some typical pulse width modulation (PWM) design, the power stage should get a 90% efficiency.

  Factoring into the PSU efficiency, I suggest (correct my assumption if deemed neeeded), the AC power needed to power your power modules would be somewhat 25VA.

  A beefy PSU is always helpful. But for class D switching amp for audio use (assumed so since you did not mention specifiically), I don't think you need overkilling due to its relatively low power consumption.

  So to answer your question, you may want to custom order an power iron of 25-30VA if off-the-shelf standard product is not available.

  I know a brandname 2x500W class-D audio power amp only use basic CR filters & the heat sinks are very small, due to its 82% power efficiency.

  c-J

  
  
  

  ---

  • Might not have been all that clear ;o) - Granholm 12:42:04 02/27/06 (10)
    In Reply to: I'm not a whatever expert. posted by cheap-Jack on February 27, 2006 at 12:25:24:

    Sorry for not explaining well enough.

    The power used by the amp is in idle (no music signal what so ever). This is about 10W for each chanel (stereo amp = 20W). But it is capable of supplying 2x300W to the speakers if the supply voltage is about +/-65V.

    I Already have an 300VA teroide from the shelf (2x42V AC). But want to build this as extreem as possible. I don't mind spending a few hundred $$ or more if I can get just a marginal improvement. And I have heard from others with similar amps that increasign the VA of the X-former can improve sound.

    The manufacturer I have contact to can do almost any design but they are not into audio (or at least not over the top - totally lost in space kind of stuff).

    I have physical space to place a teroide as big as an of the shelf 3kVA in my current box and would not mind to have one made that big if this is the only design parameter that would ad to the quality. I dont think this is the case and therefore my question!

    Please ask again if I still not make any sence...

    Have Fun
    
    

    ---

    • You want to post the specs of the amp.. - cheap-Jack 13:00:38 02/27/06 (9)
      In Reply to: Might not have been all that clear ;o) posted by Granholm on February 27, 2006 at 12:42:04:

      Hi.

      so that I can know more about it before I can comment?

      c-J
      
      
      

      ---

      • Here thay are! - Granholm 13:13:26 02/27/06 (8)
        In Reply to: You want to post the specs of the amp.. posted by cheap-Jack on February 27, 2006 at 13:00:38:

        Hi cheap-Jack,

        Thanks for your time! The spec is at the bottom of this page: http://www.lcaudio.com/index.php?page=70

        I only use the amp modules and not the predetor PSU. In stead I use 2 T-Power 10.000UF 63V for each module at the moment. And the 300VA +/-42V AC Teroide.

        I am planning on upgrading the PSU and maybe later use the 700XE http://www.lcaudio.com/index.php?page=75 that in most regards are some what similar to the Zap Pulse 2.3SE.

        Have Fun
        
        

        ---

        • Are this the spec you need? - Granholm 10:16:22 02/28/06 (6)
          In Reply to: Here thay are! posted by Granholm on February 27, 2006 at 13:13:26:

          Output Power 8 Ohms: 200 Watts RMS
          Output Power 4 Ohms: 400 Watts RMS
          Output Power 2 Ohms: 750 Watts RMS
          Output Power 1 Ohms (+/-58V 20-20.000 Hz): 1000 Watts RMS
          Peak Current 0,6 Ohm 100 Amperes 1mS
          Distortion THD < 0,2% 8 Ohms Nom. Load
          Distortion THD 0,04% 1 Watt RMS
          Frequency Response 0 - 150.000 Hz +/- 3dB
          Input Impedance 17 kOhm +/- 1%
          Output Impedance 5,5 Milliohm (70 Hz)

          My current supply is one toroide (2x42V AC 300VA) After a pair of rectifier bridges follows to 4 T-Power electrolytic capacitors at 10.000uF each.

          If you need any further information please ask!

          Have Fun
          
          
          

          ---

          • Yes, I got it now. - cheap-Jack 11:15:56 02/28/06 (5)
            In Reply to: Are this the spec you need? posted by Granholm on February 28, 2006 at 10:16:22:

            Hi.

            So, what power O/P you want? What speaker impednace you are driving?

            With your 300VA power iron, taking 80% as overall efficiency, the max O/P power you can get would be around 120Wrms/channel theoretically assuming 2x42V were high enough for the power modules.

            See if this would help.

            c-J
            
            

            ---

            • Good! And some more details! - Granholm 12:26:10 02/28/06 (4)
              In Reply to: Yes, I got it now. posted by cheap-Jack on February 28, 2006 at 11:15:56:

              Hi,

              To answer your questions first.

              The P/O should be as much as I can get from the amp for maximum dynamic headroom. Though I might stay on my current capacitors that are rated at 63V wich would limit the max output to some degree.

              My speaker impedance is 8 ohm with some areas going down to 6.

              The efficiency is somewhat around 90-94% for the ZapPulse and it would run fine on voltages from 35 to 75V DV (+/-).

              I would not mind skipping my 63V capacitors for an 80V pair to go for the higher current but I am not sure if the sound would improve by doing this!

              Ny my ultimate goal is not getting the amp to sound louder but better! (More controle, dynamic headroom, less noice... ei more natural)

              Have Fun
              
              

              ---

              • More VA power, more O/P power. - cheap-Jack 07:59:15 03/01/06 (3)
                In Reply to: Good! And some more details! posted by Granholm on February 28, 2006 at 12:26:10:

                Hi.

                But your power iron is only rated for 300VA, so the best you may be able to get is 120Wrms/channel assuming the overall efficiency is 80%
                (PSU+class D power modules).

                Where you get 94% efficiency that high for your moduls. What switching topology was used there?

                Why you need so high power anyway? Certainly not for normal home use or your neighbours will surely complain you let alone your family.

                If you really want it for certain PA use, then you got to replace a
                higher VA power transformer with higher B+/B- voltage required by the module design. Say if you really want to get 2x200W/8R, then
                you surely need a power iron rated not less than 500VA with B+/B-
                to fit the modules. Otherwise, the power iron will get overloaded, causing B+/B- voltage drops & hence O/P power.

                The filter caps & bridge rectifiers got to be beefed up accordingly.

                c-J

                
                
                
                

                ---

                • Dead end? - Granholm 12:36:37 03/01/06 (2)
                  In Reply to: More VA power, more O/P power. posted by cheap-Jack on March 1, 2006 at 07:59:15:

                  Hi c-J

                  My original intend was to get a new power iron! I know I would not get full power with what I got!

                  I also do not need the power to be kicked out of my flat ;o)

                  But I want it for dynamic headroom! If you listen to uncompressed music like classical music there are much more dynamic information in the music that is realy putting strees on the amp. Therefore I want a solution with the current modules that leaves no limmit at all. No matter the cost.

                  I don't know the actual design but only 18-20W is disapated in a modules at full power (290W) and it can be decreased further by adding a seperate gate driver supply) You might find more information here: http://www.lcaudio.com/index.php?page=70

                  So back to my original question! What would be the optimal parameters/materials etc. for a state of the art torodial X-former?

                  Would I have any sonic benefit by oversizing the X-former to as much as 3kVA? Ore would there be other more important parameters (guage of wire, core material, shielding etc)

                  I know that the rest need to follow! I already have 2 rectifier bridges capable of 35A 400V and the current capacitors are not bad at all. But I might need new ones if I should go for higher voltage (if this would add any positive sonic results). And maybe even more capacity is needed? Currently I have 4x10.000uf. With a class-D amp that almost only draw power when music is pressent this should not be bad at all.

                  Have fun.
                  
                  

                  ---

                  • Not yet. (long) - cheap-Jack 13:14:09 03/01/06 (1)
                    In Reply to: Dead end? posted by Granholm on March 1, 2006 at 12:36:37:

                    Hi.

                    Browsing over the specs sheet you linked, you can boost up O/P power of the module to 200W/C & 350W/C by raising B+/B- to +/-65V & to +/-80V respectively. This will need a 500VA & 875VA power iron, EI or pancake type regardless.

                    But who needs such high power for home audio?

                    Dynamic headroom is not decided by O/P power alone. Too high power, too high damping factor delivered to speaker systems of high efficiecny only 'freeze' up the speakers. It can make things worse.

                    I play classical music, LP, CD & DVD-audio 95% to the time & my power amp is only 35W+35W & "dynamic headroom" is never a problem to me.

                    Why? Speaker impedance & efficienct matching & the driver pre-amp
                    never overdrive the power amp.

                    Headroom is the max. power O/P delivered by the amps before clipping. Gain should be just enough to give the enough power & sound pressure levels SPLs to your ears so that there will be amble of headroom reserved for dynamic climaxes.

                    Before you go further, tell me what pream you are driving the modules & what setting of the volume control you get for your
                    SPL you are happy with.

                    Higher the volume setting, say beyond 12:00 hrs, better will be the dynamic headroom available & more enjoyable will be the music.

                    This is my sonic experiences, shared with many other audiophiles.

                    c-J
                    
                    
                    

                    ---

                    • Not that long ;o) - Granholm 13:25:55 03/01/06 (0)
                      In Reply to: Not yet. (long) posted by cheap-Jack on March 1, 2006 at 13:14:09:

                      Hi,

                      Thanks for your reply! My preamp as an acttive design (have been using various passive as well - ELMA attenuators 24 step) The volume nob is a multiturn so there is no 12.00 setting. But ther is a damping scale in the display and I normaly have a damping of about 40db and down to 24db at some classical recordings whan I am home alone.

                      My initial request was also if I would get any benefit of beefing up the toroide and I was not sure my self this was the way to go. But as VA is not the only important factor I still look for information of what would make the best toroide? I can have one custom made and can choose my self how it should be made. Many owners of the ZapPulse have experienced improvement in sonic quality by using bigger toroides but to my knowledge bigger is not always better. Hence the question about best quality! If bigger is not the way to go then what?

                      By the way thanks for your patience ;o)

                      Have Fun
                      
                      

                      ---
        • Sorry, I missed the specs. Pls re-post. (nt) - cheap-Jack 07:32:53 02/28/06 (0)
          In Reply to: Here thay are! posted by Granholm on February 27, 2006 at 13:13:26:

          .
          
          

          ---