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. Thor

    Thor Gold Supporting Member In Memoriam


  2. Since Bob obviously has better knowledge then me he must be right. May be i did not understand what i was taught. I was taught that 'digital amps' and generators do not mix well. And i remember the 50/60 hertz cycle being the reason but that might have been false.
  3. Jerrold Tiers

    Jerrold Tiers

    Nov 14, 2003
    St Louis
    Actually, many class-D amps don't have a problem with supply ripple. It depends entirely on their internal design........ *

    And I know of NO class-D that particularly cares about line frequency....

    * tech-time.... the supply ripple thing got started because theoretically the ripple could cause a "volt-second" difference between + and - supplies to the output stage, and that could let hum leak through.

    That is very true for certain types of class-D. But, it is very much NOT true for others, no more so than for an analog amp.
  4. 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
    If the amp has a bridged class-D output section, then yes, it could be designed so that the ripple on the + and - rails null out. For a single-ended output, though, ripple can get through because it directly modulates the PWM signal. This will cause sum-and-difference artifacts in the audio.
  5. It must be that this is what i was taught.... maybe most first generation class D amps were single-ended output? Or maybe just some, but enough to start this rumour?
  6. 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
    Well, it's still not related to line frequency.
  7. WalterBush


    Feb 27, 2005
    Yuma, Az
    Full disclosure, I'm a certified Fender technician working in a music store that carries Fender, Yamaha, and Ibanez products among others.

    You (or the person explaining to you) might have been getting the digital "brownout" phenomenon exhibited when computerized gear gets insufficient voltage confused with Class D amplifiers. Class D amps are called digital because they switch power on and off, not because they're necessarily computer-controlled. Lots of digital gear (reverbs, effects, computerized multiple-monitor-mix systems, digital snakes, etc) won't mix well with generators, because when they drop below a certain voltage, the digital logic chips inside simply can't function. This doesn't usually happen to your home computer because it has a well-regulated power supply, but most audio gear (power amps excluded) doesn't (this is one reason why Furman power regulators exist, BTW)

    Most generators are built to power construction equipment or household appliances, both of which will generally function adequately at less than optimal voltage, and so when a strain is placed on them, they will continue to deliver full current at the expense of voltage.

    What this means to musicians is that your digital gear could fail because the festival you're playing at has an underpowered generator, and the food vendors sharing it turned on a second deep frier to meat the lunchtime demand. Power amps shouldn't fail completely (although components inside can be strained badly), since by their very nature they've got regulated power supplies, but other digital equipment will fail, and attempt to reboot repeatedly.

    I first became aware of this phenomenon when I was using a computer to deliver 5 monitor mixes at a festival. I thought I was doing something wrong with the software, but after 45 minutes of no monitors and evil glares from the band, we used our analog board to deliver one mix and immediately our monitors worked. Of course, the band's digital equipment didn't function, no wireless, no digital 'verbs, nothing. At least we were off the hook, sort of ;)
    To make things even more embarassing, both myself and my assistant work as electricians during the day, we've installed large-scale backup generators before, and we should've been aware that this might happen, or at least figured out what was happening to our PA sooner.

    Sorry to derail the thread, it's just that this happened recently, and I have been carrying it around like a dead albatross for a while now. Just thought that might explain your misunderstanding, although once again, line frequency shouldn't affect any of this.

    And anyone, feel free to correct me if I got anything wrong. And buy a regulator if you do lots of festival shows :)
  8. I stand corrected..
    I guess its really a shame that new techiniques get a bad rep because people do not understand it and start presuming things that are incorrect.
  9. WalterBush


    Feb 27, 2005
    Yuma, Az
    Full disclosure, I'm a certified Fender technician working in a music store that carries Fender, Yamaha, and Ibanez products among others.
    It's depressing, really. The same thing happened to Fender amps when CBS took over. They only made the crummy ones for 8 months, but now "pre-CBS" is like gold in the instrument amp world, and subsequent years and designs are sometimes ignored. Kind of harkens to the posts below about first impressions coloring the way we percieve a company or amp later.
  10. I wouldn't call this new technology :confused: I've owned one version or another of the WW for 15 years or more, and other companies have had switching power supplies and/or digital amps for a long time. Like any technology, there are huge positives (power per size/weight, cool running, and in my case, completely reliable) and downsides (possibly some sound issues that are a trade-off with size and weight... but if you like that sound, it's no problem :bassist: ).
  11. fdeck

    fdeck Supporting Member Commercial User

    Mar 20, 2004
    Madison WI
    HPF Technology LLC
    Interesting info. I just received an AMP1 kit from 41 Hz Audio (www.41hz.com) and am planning on starting out with a conventional unregulated supply. This will give me a chance to measure the ripple.

    I suppose the question is whether there is enough negative feedback in the Tripath TA2022 switching power amp IC to ensure adequate power supply rejection. But in any event, I will be running the amp in bridge mode so it might not even be a problem.

    It's my impression that if you are smart enough to design a switching power supply (which I am not), then it's a cinch to make it regulated. The whole problem with regulating the supplies on a Class-AB amp does not apply to switching designs, namely the significant efficiency loss of a linear regulator.
  12. Thunderfunk


    Mar 27, 2004
    McHenry, IL

    I'm also with Bob on this although I can see both points. Digital implies a number stream that's processed as data, although RS-232 is serial (a 1-bit bus). Now in a PWM system you might say that it's 111110000000000000000001111111. Different pulse widths made up of a string of serial 1's and 0's, BUT it's transmitting power, not digital information.... although the digital stream, as well as the reconstructed analog signal IS information.... and turning your car's headlights on and off is also information... never mind. (see Marshall Mcluhan. He was paid $50,000 to tell GE that "Light is pure information.")

    "The Class-D Amplifier

    "(From the book Introduction to Electroacoustics and Audio Amplifier Design, Second Edition - Revised Printing, by W. Marshall Leach, Jr., published by Kendall/Hunt, @ 2001.)

    "A class-D amplifier is one in which the output transistors are operated as switches. When a transistor is off, the current through it is zero. When it is on, the voltage across it is small, ideally zero. In each case, the power dissipation is very low. This increases the efficiency, thus requiring less power from the power supply and smaller heat sinks for the amplifier. These are important advantages in portable battery-powered equipment. The "D" in class-D is sometimes said to stand for "digital." This is not correct because the operation of the class-D amplifier is based on analog principles. There is no digital coding of the signal. Before the advent of the class-D amplifier, the standard classes were class-A, class-AB, class-B, and class-C. The "D" is simply the next letter in the alphabet after "C." Indeed, the earliest work on class-D amplifiers involved vacuum tubes and can be traced to the early 1950s."

    Something that's been left out (I think) is that our transformers are large because they operate at 60Hz. If they were run through a switcher (chopper) at 100KHz, they'd get a lot smaller. That's why "digital" power supplies are so small and light (WW).

    Also, MOSFETs are high input impedance devices, "similar" to tubes, which is why (I think) some people say they sound like tubes. But their advantage is they are self limiting and nearly impossible to blow up.
  13. brooklynbassguy

    brooklynbassguy Supporting Member

    Feb 8, 2004
    maplewood, nj, usa
    So for laymans edification:what exactly is an acoustic image clarus? I've experienced weird hums in quasi brownout situations, ie nyc household current on a hot summer day, between 97-104 volts, that would disappear when voltage returned to normal-119-127. Also a certain nyc club would on somewhat rare ocasions spikes, or ground problems that would blow internal fuses on EBS Fafners, or kill Claruses? My QSC would only dim lights at badly wired house parties or blow home circuit breakers, but no humming beforehand.
  14. 44me


    Jun 17, 2002
    Bedford, NH USA
    Almost all switching power supplies are regulated, as most supply topologies give you pretty lousy characteristics unregulated. Regulation does not effect efficiency other than it alters the operating point. Line frequency generally has no effect either as long as the input filtering is sufficient for the lowest line frequency. There are topologies that don’t lend themselves to regulation (sometimes called DC transformers), but those tend to be used in applications that are fed from relatively fixed input voltages.

    Switching amplifiers can be more unforgiving of low input voltage than linears if the design requires a regulated input voltage. Once you run out of supply headroom the supply ripple goes way up and can cause noticible hum. An interesting characteristic of regulated switching supplies is that they present a negative impedance load to the AC line. What this means is that the current they draw goes up as the input line voltage goes down. This can cause fuses to blow with low input voltages that are sized appropriately for normal operation. Most designers will include under-voltage lockout circuitry to keep this from becoming a problem.

    - John
  15. Jerrold Tiers

    Jerrold Tiers

    Nov 14, 2003
    St Louis
    Not necessarily..... again the exact details of the system implementation affect whether there are measureable, hearable issues or not.

    Bridge amplifiers are not immune as a type, unless the bridge is perfectly balanced as to a fairly large number of part characteristics. You can count on the fact that it is not perfectly balanced, although with selection and care you can get close, if that is considered necessary. Bridge type units have another whole set of problems of their own, however.

    But, even with a "half-bridge", i.e. a single output vs ground, there need not be any problem. One of the finest audiophile amplifiers I have heard is in fact a half-bridge PWM design. Obviously they solved that hum problem......
    BTW, 44ME, you should look up the US patent for the QSC SMPS.
    It describes an unregulated SMPS of a particular type which is essentially a DC-DC transformer. The design is certainly used in the PLX, and may be also used in the Powerlight. (Bob Lee would be able to answer that.)

    The patent for the Woods SMPS/amplifier combination also describes a form of what is, IIRC, an unregulated supply. The patent covers the integration of the two parts in a particular way.
    As far as "digital" applied to a PWM amplifier.... it is not sensible to do that if you get fussy about definitions.

    A PWM amplifier is analog. Yep, analog.

    The parameter variance is not amplitude, but it IS time duration. So there is a transformation performed from the amplitude to the time domain.

    However there remains a 1 to 1 corresponcdence between the two, AND the time parameter can assume an unlimited number of values.

    A "digital" representation is something like PCM, where each small amplitude range has a discrete number applied to it. There are only a certain quantity of digitally represented values the signal can assume. And, the number of possible values is defined by the number of digital "bits" the system employs.

    It is conceivable that if the time variance were represented by a number, and generated by a clocked timing system, that the result would be a system as "digital" as PCM. Those systems exist, but have some problems. It is an active area of development still. Even so, there is a dual transfomation, from the number to a time, and from time to an amplitude. .

    We are WAAAAAAAYYY out in "geekland" now, folks......... having fun yet?
  16. B String

    B String Supporting Member

    Apr 11, 2002
    Los Angeles
    Ok, so how does all of this information relate to my Acoustic
    Image Focus? Is it a hunk o crap, or a technological marvel?
    Dats what I wanna know. Are there any down sides to this
  17. WalterBush


    Feb 27, 2005
    Yuma, Az
    Full disclosure, I'm a certified Fender technician working in a music store that carries Fender, Yamaha, and Ibanez products among others.

    Yes, yes I am, although I only understand about 1/2 to 3/4 what you're saying. It just gives me more to research. Hey, how's my Adcom amp work? It's a model they made for subwoofer usage, but it's a fantastic bass rig when paired with a Summit Audio TD100 and good speakers. Built like a tank, and the transformer and what I assume to be filter caps are ENORMOUS. I can't figure out how it works with such a small amount of electronic components inside. My dad said something about "the power supply IS the power amp" when he saw it, but that didn't mean much to me, he considered it significant. Anyone familiar with Adcom products?

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.