Class D or AB

[JimmyM]

Didn't have time to respond more fully because of travel yesterday, but calling a class D amp "digital" makes zero sense. First off, it's not even digital unless the sound is being produced digitally. TC and Line 6 are digital. Only the SMPS is digital and the rest of the amp is analog. Second, why not refer to them as class AB and class D? Everyone else in the industry does it. What's the fear? That people won't buy it if they know it's not digital? Seriously, that makes absolutely no sense.

It's a missed opportunity to help sort out the BS from the truth. I truly doubt they had ill intentions by doing that, but I can't begin to understand why they did it when it's not true.

 

[KJung]

Okay, to be simplistic and more accurate then. Weight savings comes from type of power supply, power efficiency is where the output topology class comes in. Too much info for musicians to use or understand? Not hard to guess I am the tech guy in the band I guess, but there are two others in the current project.

God help us all. All Markbass amps have SMPS. The difference in weight of the Class D 500 watt amps and the Class A/B 500 watt amps in their product line is a result of the amp topology.

Why in gods name would you be making a big deal (or ANY deal) about this? Unbelievable.

Hopefully, the OP had his question answere before this ridiculous discussion started.

Unsubscribed and out.

 

[chadds]

 

[Russell L]

Doesn't bother me in the least. Let's see, where's the on/off button on this thang...

 

[warnergt]

...calling a class D amp "digital" makes zero sense.

Jimmy, I almost always agree with you but I have to disagree here.
Class D amps are digital in the sense that the power is switched
on and off to achieve the output signal. They are either full on
or full off; there is no in between. It is on ('1') or off ('0').

That is why they are so cool and efficient. Unlike conventional
analog amplifiers, the switching devices (e.g. tubes or
transistors) dissipate no power. If you compute the power
dissipated by the devices, (watts = volts * amps), in all
cases, either volts is zero or amps is 0, so watts is always 0
(in reality, it's not 0 but close to it). All the power goes to
the load while the amp stays cool as a cucumber.

What makes this possible are some amazing transistors
(usually MOSFETS) that they have today that we didn't dream
of ten or twenty years ago. The amazing part is that they
have an "on" resistance that is close to 0 ohms (usually
measured in milliohms). They now have MOSFETS that have
less than 1 milliohm of "on" resistance. Maybe that means
nothing to most folks but, to us engineering types, that's insane!

 

[onosson]

Jimmy, I almost always agree with you but I have to disagree here.
Class D amps are digital in the sense that the power is switched
on and off to achieve the output signal. They are either full on
or full off; there is no in between. It is on ('1') or off ('0').

That is why they are so cool and efficient. Unlike conventional
analog amplifiers, the switching devices (e.g. tubes or
transistors) dissipate no power. If you compute the power
dissipated by the devices, (watts = volts * amps), in all
cases, either volts is zero or amps is 0, so watts is always 0
(in reality, it's not 0 but close to it). All the power goes to
the load while the amp stays cool as a cucumber.

Yeah, this is how I've always understood it. Digital does not have to mean "as opposed to analog".

I went from a 500-watt LMIII (class A/B) to a LMTube800 (class D) and my experience has been great with both.*

* My reasons for switching had to do with the two separate input gain controls, not the extra power or the solid-state/tube preamp mix control.

 

[Egbert89]

Jimmy, I almost always agree with you but I have to disagree here.
Class D amps are digital in the sense that the power is switched
on and off to achieve the output signal. They are either full on
or full off; there is no in between. It is on ('1') or off ('0').

That is why they are so cool and efficient. Unlike conventional
analog amplifiers, the switching devices (e.g. tubes or
transistors) dissipate no power. If you compute the power
dissipated by the devices, (watts = volts * amps), in all
cases, either volts is zero or amps is 0, so watts is always 0
(in reality, it's not 0 but close to it). All the power goes to
the load while the amp stays cool as a cucumber.

What makes this possible are some amazing transistors
(usually MOSFETS) that they have today that we didn't dream
of ten or twenty years ago. The amazing part is that they
have an "on" resistance that is close to 0 ohms (usually
measured in milliohms). They now have MOSFETS that have
less than 1 milliohm of "on" resistance. Maybe that means
nothing to most folks but, to us engineering types, that's insane!

+1

 

[MODELL]

Class doesn't matter at all.
They both sound the same.
The difference is in features of one amp over the other.

How does it matter to you?

I understand about the sound being the same, it's more about what is in print. I thought I had read here that they were AB, then I saw an advertisement in an old BP mag that says the Randy Jackson TTE500 head is Class D. That trumps something someone said a few posts up about 500 watt MB amps. It was "just curious" as I posted.

 

[KJung]

I understand about the sound being the same, it's more about what is in print. I thought I had read here that they were AB, then I saw an advertisement in an old BP mag that says the Randy Jackson TTE500 head is Class D. That trumps something someone said a few posts up about 500 watt MB amps. It was "just curious" as I posted.

Just FYI, without getting into the ridiculousness above.

The TTE500 head is class A/B (described as 'analog' in quotes in the specs). The new TTE800 is class D, again, described as 'digital' in quotes in the specs.

The Big Bang (replacing the F1 and F500) is the only 500 watt Markbass amp that is class D, as far as I know.

All these amps sound different, most likely not due to amp topology (i.e., the 800 watt class D Marbass amps sound quite different from the 500 watts class D Markbass, amps, which sound different from the various versions of the class A/B LMIII type models).

Again, all Markbass amps have SMPS (switch mode power supplies), regardless of the variation in amp modules.

Nice variation of power, tone and features in this product line.

 

[Bob Lee (QSC)]

That bums me out. Half the reason bass players don't know boo about equipment is that manufacturers do a piss-poor job of educating us.

+1 Well said, Jimmy!

 

[Bob Lee (QSC)]

Class D amps are digital in the sense that the power is switched
on and off to achieve the output signal. They are either full on
or full off; there is no in between. It is on ('1') or off ('0').

"On/off" (or more aptly, "positive/negative") switching does not make them digital. "Digital" implies some modicum of numeric, logical, or computational processing.

 

[Mark Reccord]

Jimmy, I almost always agree with you but I have to disagree here.
Class D amps are digital in the sense that the power is switched
on and off to achieve the output signal. They are either full on
or full off; there is no in between. It is on ('1') or off ('0').

That is why they are so cool and efficient. Unlike conventional
analog amplifiers, the switching devices (e.g. tubes or
transistors) dissipate no power. If you compute the power
dissipated by the devices, (watts = volts * amps), in all
cases, either volts is zero or amps is 0, so watts is always 0
(in reality, it's not 0 but close to it). All the power goes to
the load while the amp stays cool as a cucumber.

What makes this possible are some amazing transistors
(usually MOSFETS) that they have today that we didn't dream
of ten or twenty years ago. The amazing part is that they
have an "on" resistance that is close to 0 ohms (usually
measured in milliohms). They now have MOSFETS that have
less than 1 milliohm of "on" resistance. Maybe that means
nothing to most folks but, to us engineering types, that's insane!

I'd argue that pulse width modulation is an inherently analog process with some digital-like qualities, though the argument is somewhat semantic. In my world (eval board design for FPGAs), I don't consider anything that isn't sampled and quantized to be digital. I'd consider PWM to be mostly analog in that the analog signal is superimposed on the switched mode carrier and is never actually quantized (in the sampling theory sense of the word). The audio signal remains continuous all the way through the process.

In the words of Bruno Putzeys (class D guru, Hypex): "All power amps are analog but some are more analog than others."

 

[warnergt]

"On/off" (or more aptly, "positive/negative") switching does not make them digital. "Digital" implies some modicum of numeric, logical, or computational processing.

I disagree. Digital means that it has discrete levels (i.e. some
numerical values like '1' and '0') versus the infinite number of
levels found in analog design which is fully continuous.

Furthermore, there is digital processing being performed. The
output analog level is being compared to an input analog
reference level to determine when to switch on or off. While
these levels are analog, the computational result of this
comparison is digital; it is either '0' or '1'. This intermediate
digital result is necessary to obtain the correct analog output.

The output of the amplifier is analog but that output is created digitally.

 

[warnergt]

One more point: there is no denying that, in a class D amp,
there is a point in the signal path where the signal is
represented purely by 1's and 0's.

 

[seamonkey]

One more point: there is no denying that, in a class D amp,
there is a point in the signal path where the signal is
represented purely by 1's and 0's.

PWM is also "time". How long it stays a 1 or 0. That's not digital. Digital circuits do have to deal with time/clocks but that isn't part of the "digital"

And as others have pointed out, class-ab is adding two different waveforms together to reproduce the original. That's a real bastardization.

In any case, what comes out of a class-d amp is the exact same signal that comes out of a class-ab

They sound the same. (gain matching ...) (amp part, not pre-amp, or special feedback part)

I agree with the OP that Markbass isn't clear on this. Markbass probably should not have bothered with stating the amp class in the first place. Who really needs to know? I know by current draw to actual power out if the amp is efficient. Things like quiet fans or no-fans are features I look for. More efficient amps need less.

 

[dincz]

I disagree. Digital means that it has discrete levels (i.e. some
numerical values like '1' and '0') versus the infinite number of
levels found in analog design which is fully continuous

PWM in a class D amp is also continuously variable between pulse duty factors of 0 and 100%. The amplitude of the pulses is not a defining characteristic.

The output of the amplifier is analog but that output is created digitally.

A class D amp is no more digital than an electric light. There is no numeric encoding in either. Even referring to "0" and "1" is really not valid as they aren't logic states.

LED brake/tail lamps on cars use the same LED's for both functions by altering the on/off strobe periods They aren't digital either.

 

[will33]

Much enjoy the tech talk. Some of it is over my head but I like to think it "soaks in".

In the case of these amps, the term "digital" is an advertising buzzword as somebody put it. Pretty much dis-info and should be ignored.

 

[JimmyM]

How is a class D amp with an SMPS any more or less digital than a class AB amp with an SMPS?

 

[warnergt]

PWM in a class D amp is also continuously variable between pulse duty factors of 0 and 100%. The amplitude of the pulses is not a defining characteristic.

I agree. This "pulse duty factor" is another way of saying how
long we have a '1' versus how long we have a '0'. There are no
duty factors in an analog system. The level of the pulses ('0' or '1')
is a defining characteristic. These 0's and 1's are needed to
define the output signal.

 

[Bob Lee (QSC)]

I disagree. Digital means that it has discrete levels (i.e. some
numerical values like '1' and '0') versus the infinite number of
levels found in analog design which is fully continuous.

Furthermore, there is digital processing being performed. The
output analog level is being compared to an input analog
reference level to determine when to switch on or off. While
these levels are analog, the computational result of this
comparison is digital; it is either '0' or '1'. This intermediate
digital result is necessary to obtain the correct analog output.

The output of the amplifier is analog but that output is created digitally.

No, that's not digital. A class D amp doesn't switch between 0 and 1, but between a positive voltage and a negative one. The voltages are what's important, not any arbitrary numeric assignments to them. If you put the same pulse stream into a ±25V class D output section and into a ±50V one, you'll get twice as much output voltage (and 4× the power into the same load) with the higher-voltage one. That's not a characteristic of a digital system.

The modulation scheme you describe is analog. The output is not 0 or 1, but positive/negative. Instead of a continuously varying voltage as in the original audio signal, the pulse train has continuously varying time periods. That's still not digital.

You can do class D digitally, where a processor computes the positive and negative time periods, but that approach is so rare it's almost non-existent.