1. Please take 30 seconds to register your free account to remove most ads, post topics, make friends, earn reward points at our store, and more!  

MOSFET vs. Class D

Discussion in 'Amps and Cabs [BG]' started by terbay, May 5, 2005.

  1. 44me


    Jun 17, 2002
    Bedford, NH USA
    The home theater market is one that TI’s been focusing for a while, and it’s a good application for switching amplifiers. TI’s original products used class D modulation, but their newer products use a different modulation scheme that produces less distortion and requires a simpler, easier to design output filter. It’s quite impressive how much power they can get out of an integrated circuit output stage. These devices use DMOS output stages, which are essentially MOSFET’s built on an IC process. Efficiency is in the range of 85%, so 30W to the speaker will burn about 5W in the output stage. That’s a lot for an IC, but certainly manageable.

    - John
  2. slinkp


    Aug 29, 2003
    brooklyn, NY, USA
    Really? Hmm, maybe AI uses different power supplies in the different models? I got an email from Rick Jones that said:

  3. Jerrold Tiers

    Jerrold Tiers

    Nov 14, 2003
    St Louis
    May I assume that your distinction between Class-D (PWM) and naturally sampled refers to fixed clocked vs sample-driven clocking? Because both really are "PWM".

    In any case....among Ampeg

    Of course SVT-CL, V4BH and SVT-2PRO are tube pre and power.

    All current Ampeg heads that have non-tube poweramps are Mosfet output, unless I forgot one.....most are class-AB.

    SVT3PRO and SVT6PRO are hybrid power amps, tube driven mosfet.

    PortaBass heads and combos are a PWM variety, with "conventional" (linear) power supply.

    Preamps in heads and separate preamps vary, Some are tube, some are hybrid, and some are SS. "B" series and SVT-350H are SS preamp, SVP-CL is tube.

    BTW, the first mosfet amp was made by Yamaha, the model B2, if I remember right. It was a home hifi type amp that came out around 1978, using a special mosfet type.

    Hitachi subsequently made a range of very nice mosfets designed for analog usage, which many companies used in the 80s. They are all now essentially obsolete, although one British company had a source of similar parts for a while.

    Anyone using mosfets in an analog amplifier currently is probably using two particular types that have not yet been "improved", and as a result are still usable for analog.
  4. 44me


    Jun 17, 2002
    Bedford, NH USA
    Slinkp, I’m pretty sure the Clarus uses a conventional supply, and the higher powered Focus uses a switching supply.

    Jerold, you’re right – naturally sampled is self-oscillating, but is still PWM.

    - John
  5. Bob Lee (QSC)

    Bob Lee (QSC) In case you missed it, I work for QSC Audio! Commercial User

    Jul 3, 2001
    Costa Mesa, Calif.
    Technical Communications Developer, QSC Audio
    Class D can also be broken down into class AD (switching between +V and -V), which is the prevalent method of class D, and class BD, which switches between +V and 0, or -V and 0, depending on the instantaneous polarity of the signal. Class BD is still pretty rare but it is a promising technology.
  6. 44me


    Jun 17, 2002
    Bedford, NH USA
    Bob, I didn’t realize that a class name had been given to that scheme. We (where I work) call that approach mult-level. It’s pretty common for high power inverters and motor drives. I didn’t realize people were applying it to audio applications. I expect there will be a lot switching technology innovation over the next few years, with a lot of great, light weight gear introductions. For the record, I still play through tubes – go figure!

    - John
  7. barebones

    barebones Supporting Member

    Jan 3, 2005
    Denver, CO

    Thank you, John. Very helpful!

  8. brooklynbassguy

    brooklynbassguy Supporting Member

    Feb 8, 2004
    maplewood, nj, usa
    I just stumbled upon this old thread. If thunderfunk is the same design as the old amp 420, it uses mosfets.
    An engineer friend, and I did an experiment with a walter woods and a trace-eliot some years ago, using a sine wave generator and a meter to see the outputs. The trace held a firm 250 watts from 1k down to 40hz. The woods was about 250 at 1k and 25 at 40hz, and it's power was making alot of noise. This was done to explain what I had been hearing, but written specs didn't seem to agree
    What exactly is Class H?
  9. Interesting if this holds up in his newer models. I have been using various models of WW for years, and always have noticed that they seem to 'roll off' at a very high level (e.g., that there just isn't a lot of 50hz and below oooomphh versus an amp with a big linear power transformer). Some other WW users agree with me, others don't. Quite frankly, it's such a difference that I can't believe most others don't notice the issue. I guess if you are playing at low volumes or through smaller cabs that roll-off themselves you wouldn't notice it. However, wumping out of a 410 at volume, there is definitely a 'lack of quality' in the WW low end, versus most traditional amps. That being said, it might just be Walter's design versus a 'class D thing in general.

    I'm most surprised by Jim Bergantino's views on this. If I remember correctly, he (or others paraphrasing him) said that the WW is the only digital amp he likes.... maybe he's using different criteria than I am (since he obviously knows what he's talking about!)
  10. Passinwind

    Passinwind I know nothing. Commercial User

    Dec 3, 2003
    Columbia River Gorge, WA.
    Owner/Designer &Toaster Tech Passinwind Electronics
  11. From my understanding, the EBS amps are class H.
  12. boogiebass


    Aug 16, 2000
    KJung knows I disagree with him here and his assumption that it's only players who play at low volume with small cabs is, based on my experience, dead wrong regarding the Woods Ultra. However, we don't need to re-thrash that cat. ;)

    Here's a quote from Jim B. as referenced above:

    "Certainly, just like conventional amps, not all switching amps are created equal. I do have one favorite, as those who have spoken with me know. There's a reason why he has a patent on his design!

    Jim Bergantino

    Do a search if you want to read all of Jim's comments on switching vs. conventional amplifiers.

  13. I love my Walter too :) ... just found it interesting that the test quoted earlier exactly explained my (and some others) experience with the Walter. If you find the Walter perfect, that's a very good thing :) :)
  14. boogiebass


    Aug 16, 2000
    Right on. I wouldn't call the WWU "perfect," but it works for a lotta gigs! ;)

    And I believe the tests referred to above were earlier Woods amps, no?
  15. It seems that's the case... I was just wondering if the results held currently. However, just so nobody misinterprets my comments, I'm talking about listening to this stuff very, very critically. I would be very, very happy to play all the rest of the gigs in my life with my Walter (or a good Eden rig, or a good...etc. etc.). I just view this forum as a way to really get down to the nitty gritty. And, since the Walter's are WAY expensive, I tried to point out the good and possibly the bad so that people go into a possible purchase with their eyes wide open.
  16. boogiebass


    Aug 16, 2000
    Noted and appreciated.

    One of the things I've noticed with others (and myself) is that sonic impressions can become "lodged" in our minds sometimes. What might be happening (and might serve to explain our disagreement with the Woods here to some degree) is that your observations with older Woods amps has been somewhat transfered to the Ultra, de facto. I've done that myself, at times, e.g. SWR brightness, Ampeg growl, etc. These impressions, once seated in our minds, tend to sometimes shape our future perceptions, it seems.

    So just to be clear: my comments about no lack of low-end in a Walter Woods amp ONLY applies to the Walter Woods Ultra. I know from experience that some of the older Woods amps exhibit exactly the kind of low-end response issues you're reporting. But I believe Walter's current design is intended to address the low-end (among other things) as he realizes these amps are not just being used by low-vol, jazzers these days.

    The other thing is, I only use the WWU on some of my gigs. I also regularly gig with EA iAMP 350, 500 and 800's, as well as a Crest CA-6 powered rack system (mostly outdoors) and a Crown CE2000 powered rack system at times, too. So I never meant to imply that the WWU was the be-all, end-all of amps. Just a darn good one, IME! ;)
  17. Makes sense. I too use a big, analog power supply amp for louder gigs and the Walter for everything else. That's a perfect combination for me.... as you point out, the Walter does what it does very well, and actually does pretty much everything well! I haven't done the rack deal in a long, long time, but all these guys with the Milenia stuff and the 2000 watt power amps is starting to worry me and my back!
  18. brooklynbassguy

    brooklynbassguy Supporting Member

    Feb 8, 2004
    maplewood, nj, usa
    Let me be clear, that I make no assault WW amps I have an AI Clarus that gets alot of use on my smaller gigs in the city. A WW is certainly a great amp for those schlepping around in big cities. When I play larger venues and need the the "meat" I pull out the big guns and suck it up-my poor back :crying: ! Mind you my biggest gun nowadays is a rack with a QSC PLX 2402.
    One thing I have notice with high tech switching amps, and the QSC is on the fence with this, they are very sensitive to funky electrical situations. the older linear designs seem to be more rugged. That's only my experience...
  19. Class D, PWM, switching power supply,or as some marketing people like to (mistakenly) like to call it,digital amps need their AC cycle to be exactly 50, or in USA 60 Hertz. Otherwise you are in for trouble. Unfortunately most generators are not very good at this. So if you play a lot of gigs where you depend on generators for power i would suggest you either dont use a PWM amp, or use the absolutely best generator you can find, or use some kind of device that can stabilize the cycle.
  20. Bob Lee (QSC)

    Bob Lee (QSC) In case you missed it, I work for QSC Audio! Commercial User

    Jul 3, 2001
    Costa Mesa, Calif.
    Technical Communications Developer, QSC Audio
    That's not true in my experience. The Class D circuitry is getting DC from the power supply; why should the line frequency matter? What matters in Class D is that the supply rails be extremely clean (regulated and free of noise and ripple) and stable.

    Switch mode power supplies are a different matter than class D, but they too are tolerant of line frequency variations.

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.