Dispelling Tube Amp Myths

Guys, I have tried to distill this information down to something the technically minded guitarist can understand. However, one may have to read through the post a couple of times.

There is a lot myth surrounding tube amps. That is mainly because myth is easier for the layman to understand than science. One of the main myths is tube watts are louder than solid-state watts. That is complete nonsense, a watt is a watt. A watt is equivalent to one joule of energy per second. There are couple of reasons for this myth. A big reason is that solid-state amp designers were not always honest about power ratings, preferring to state power in peak power terms instead of RMS power, which led to inflated power ratings.

That being said, the reason why tube amp often sounds louder than solid-state amps of the same RMS rating is because tube amps are poorly damped. Damping is the ratio between speaker impedance and amplifier output impedance. Damping factor determines how well the power stage of an amp can control speaker cone movement. Power tubes have high output impedances. In order to drive a speaker, they require an output transformer that steps the power tube output impedance down to the speaker nominal impedance. In the process, the output transformer converts a high voltage, low current signal to a lower voltage, higher current signal. It also increases that amp's damping factor somewhat. For example, A Fender Vibrolux has an output transformer with a primary impedance of 4,000 (4K) ohms with secondary impedance of 4 ohms. However, an output transformer knows nothing about impedances. It works using impedance ratios, that is, if we plug an 8-ohm speaker into the 4-ohm jack, the power tubes will see primary impedance of 8,000 ohms instead of 4,000 ohms because any change in the load attached to the output transformer's secondary winding (a.k.a. the wind that is attached to the speaker jack) is reflected back to the power tubes. This change is due the fact that output transformers actually have impedance ratios, not fixed impedances. The impedance ratio is the ratio between the primary and secondary impedances. In this case, the impedance ratio is 4000 / 4 = 1000. The impedance ratio is the square of the ratio between the number of turns of wire in the primary and the secondary windings, which is SQRT(1000) ~= 32, where SQRT is the square-root function. What that means in layman’s terms is that there is one turn of wire in the secondary winding for every thirty-two turns in the primary winding.
I remember we were trying out a guitarist for a rock band...and he brought his 100w Kustom solid state combo and he couldn't be heard over our drummer. He was on full tilt too. At the time I had a Mesa Triple rec that was turned up less that a 25%. Maybe that Kustom amp was underrated as you say. Still...i think you basically said tube amps often sound louder and "this is why"....yes...confirmed Tube amps generally are louder wattage ratings being equal. This has been my experience on several occasions.
 
I remember we were trying out a guitarist for a rock band...and he brought his 100w Kustom solid state combo and he couldn't be heard over our drummer. He was on full tilt too. At the time I had a Mesa Triple rec that was turned up less that a 25%. Maybe that Kustom amp was underrated as you say. Still...i think you basically said tube amps often sound louder and "this is why"....yes...confirmed Tube amps generally are louder wattage ratings being equal. This has been my experience on several occasions.
Tube amps regularly can exceed (sometimes considerably) their rated output. A properly designed solid state amp can exceed it’s rating as well, but tube amps can sound really good, maybe even their best, when pushed past their limits. Solid state gets ugly in a hurry if pushed past it’s limits. So many of them are designed to not let you go past them.

Also, when discussing “solid state” in the past that generally meant “class a/b” designed power stages. Now, class D is the most prominent SS type power stage in use, and it is VERY VERY different, as we’ve all seen. Tiny little lightweight amps rated for a lot of power…. Class D clipping is nasty, and most designs have limiters built right in for this reason. But there’s another big difference. Many are extremely light in power supply, and while the design calls for less than a class A/B amp, IMO, many of these designs are light on bottom end extension and control, because of the old adage… “there’s no substitute for horsepower. I was looking into taking a couple affordable designs, putting a few better quality caps in the signal path, and better and bigger caps and power source, in hope of increasing the fidelity of the amps. But most them them are SMD designs and are tough to do much with.
 
Many are extremely light in power supply, and while the design calls for less than a class A/B amp, IMO, many of these designs are light on bottom end extension and control, because of the old adage…
Most Class D amps use switch-mode power supplies. Compare a modern Class D bass amp with switch-mode power supply to an old tube bass amp with a linear power supply and the tube amp will not come remotely close to the Class D amp when it comes to keeping the bottom end tight. A switch-mode power supply is much more efficient than a linear power supply and it does not need a large traditional power transformer, which is why most Class D amps are so light when compared to their Class AB counterparts. A switch-mode power supply can be used with a class AB amp as well. The reality is that all a power amp is is a modulated power supply.
 
The Science on this post is so cool and interesting and I am nerdy enough and techie enough to get it.
BUT
There are often times the human condition defies science.
Technically amps can easily cover a wider frequency range but the user will often find them woofy or shrill , their will be artifacts.
When I was building pedals I spent lots of time looking at waveforms and comparing that to what I liked sound wise , then how that waveform effected the input portion of an amp , then ultimately the power section. ( not just guitar amps , I tried a wide range of amps , preamps and Mixers )
I found it much more difficult to design a pedal ( Gain or Compression or Boost ) that sounded "right" on a solid state or higher end audio circuit.
I watched those circuits and amps struggle with an analog square wave or any single clipping , boost , basically turning to noise any signal it couldn't process , ( like a video signal acts )
It amazes me the happy accidents that often created the tones we chase and love today.
Its so interesting that a Klon or a Dumble , an original Marshall Plexi etc should be no problem to recreate and folks get close some might say better , some not all that certain is that its never 100% the same, Like trying to see what the Mona Lisa looked like new.
Thanks Em7 for another fun post :)

I find the 5th harmonic to be the worse on a distorted guitar signal. If something is not done to roll this harmonic off, a guitar signal takes on a metallic tone. Tube input stages tend to have more miller capacitance than solid-state input stages due to the fact they have input impedances. Miller capacitance results in high frequency roll off, which is important when we are working with a clipped signal. The highest fundamental frequency on a 24-fret PRS is 1318.510 Hz. The fifth harmonic of that frequency is 1318.510 * 5 = 6592.55, which is high enough to be rolled off by the speaker. The frequency of the high E string when plucked open is 329.628 Hz. The 5th harmonic of 329.628 Hz is 1648.14 Hz, which will not be rolled off by the speaker. If you look at the schematic for some of the Mesa Boogie amps, you will find that Randall cut bass before entering a stage that was designed to clip while he cut treble after exiting that stage. Bass distortion not desired with guitar and neither is higher-order treble.

Here is the preamp schematic for Mesa Mark III:


1JLiJC2.jpg


If one views the "Lead Circuit," one encounters capacitor that is wired across the plate resistor. What this cap does is shunt frequencies above the RC constant to the power supply. There is a low-pass filter before the stage. The cutting of bass before a stage followed by the cutting of treble after leaving a stage is responsible for the midrange honk found on these amps.
 
A switch-mode power supply is much more efficient than a linear power supply and it does not need a large traditional power transformer,
Exactly, but MOST of the cheaper ones are over rated and under powered. You know what I mean... things like tiny little amps that are rated for 100 watts, but it's at 10% THD, and at clipping it's not as loud as an 18 watt tube guitar amp.

I know there are now high end Class D amps. I was reading about an "affordable" high end home stereo power amp last week, and after two pages of the guy bragging about it's imaging capability, flat frequency response, deep, extended and tight bass, super transparent and smooth high end, and tons of clean power reserves, in the summary it showed the "affordable price" of over $11K. :) The main point being that the high end has also no embraced the class D designs (which at first, they certainly did not).

But most of the designs you see that seem too good to be true (100 watts for $79 and those type things) don't have the "horsepower" to control a big woofer. But at least, the talented designers aren't writing them off as toys now, like they did at first. Hopefully this means more affordable, small, light and powerful amps will be designed. I'm thinking about snagging both a Power Station and something like a Seymour Duncan PS170 to use with a modeler and a real guitar cab, to see how something like that compares to a real tube amp. I could even through one of my Adcom power amps in the mix as well, just for fun.
 
Inflating power figures for solid-state amps has been a problem since day one. It always pays to find out what integrated power amp IC they are using and pull the datasheets. The only rating that counts with solid-state is the RMS power rating of the chip and as has been mentioned, that limit needs to be more than one needs. Solid-state designs cannot be pushed to their rails like a tube amp. A tube will never go all the way to its power rail, a transistor will peg it. For those who do not know, that is where the term "soft clipping" comes about. A tube clips softer than a transistor because it cannot go all the way to its power rail, which results in a more rounded off clipped signal, which produces fewer higher-order harmonics.

What is interesting about the Quilter amps is the way Pat incorporated current feedback. A lot of people are crediting him with the invention of current feedback to lower damping factor, but Marshall implemented it in their ValveState power amp several decades ago. What I believe that Pat did was discover a way to implement it in a Class D power amp (the ValveState power amps are Class AB).

Here is the power amp topology from Pat Quilter's patent application:
wooVcEY.jpg
 
Last edited:
Inflating power figures for solid-state amps has been a problem since day one.
Absolutely! And, with what you address afterwards about clipping, it becomes clearer why people love pushed tube amps, and hate pushed solid state. So, SS designers often tried to go for more clean headroom, because that clipping was nasty if you pushed SS to hard. So then you further fuel the "tubes are louder than SS" because you can push tubes past it's limit and you probably want to keep SS well under it, or it won't sound as good.

A lot of people are crediting him with the invention of current feedback to lower damping factor
Have you tried any of his guitar amps? I haven't. I know some love them, some don't. And as I mentioned above, like other SS designs, many that love them use them mainly as a clean+pedals platform, but some do like the overdrive as well. I'd be interested to hear how they sound, as well as how they handle the load of various guitar speaker/cab configurations.
 
Absolutely! And, with what you address afterwards about clipping, it becomes clearer why people love pushed tube amps, and hate pushed solid state. So, SS designers often tried to go for more clean headroom, because that clipping was nasty if you pushed SS to hard. So then you further fuel the "tubes are louder than SS" because you can push tubes past it's limit and you probably want to keep SS well under it, or it won't sound as good.


Have you tried any of his guitar amps? I haven't. I know some love them, some don't. And as I mentioned above, like other SS designs, many that love them use them mainly as a clean+pedals platform, but some do like the overdrive as well. I'd be interested to hear how they sound, as well as how they handle the load of various guitar speaker/cab configurations.
His amps are the best of the best with respect to solid-state. I owned a MicroPro Mach 2-8 for over a year. I liked the amp, but the digital reverb circuit was too noisy for me. The noise appears to be specific to the Mach 2. There are two things I do not like about Quilter amps. The first is that they are bright. Fender blackface and to a greater extent silverface amps can be too bright for me. Quilters can be like silverfaces on steroids. However, the deal killer thus far has been mid-gain tones. Pat Quilter has not cracked that one, but I am 100% positive that he will. He is working for the love of amp design. He does not draw a salary from Quilter Labs. Mid-gain tones would be easy to incorporate using derivations on overdrive pedal circuits. Pat is a Western Swing player, so distortion tones are not his bag.
 
Great info. Thanks for sharing!

Interesting about the gain tones. I can see where that may be a harder nut to crack for someone who just doesn't play those tones. Sounds like he needs a rocker on staff. :) And yes, some pedals can nail the gain tones pretty well, so there is a path there. We need to introduce him to David Barber. That guy knows how to make some GREAT sounding low and mid gain circuits!

That said, for now, a Quilter and a Barber Direct Drive sounds like it could be a really nice combination.
 
What is interesting about the Quilter amps is the way Pat incorporated current feedback. A lot of people are crediting him with the invention of current feedback to lower damping factor, but Marshall implemented it in their ValveState power amp several decades ago

Paul Rivera used an approach to lower damping factor on the Class A/B amps he designed for Yamaha in the early 80s. Not sure whether that was before or after Marshall. I have one, a G100-112, I bought used circa 1984. Gigged with it in college. It sat unused for many years. Bought some tube amps along the way. Dusted off the G100 a year or so ago. With its eq circuits, I can dial in the exact tone that I want. I really like it. I alternate between that and a Tungsten Creama Wheat, which I like just as much. Lately, it’s been all G100 all the time, in part because of a fear of the current tube shortage. I think I could be content with the G100 if I had to pick just one. It’s really that good.
 
The tube shortage is no more valid than the toilet paper shortage last year
Just read your expanded comments on this topic in the other thread about whether folks are considering switching to solid state. Completely understand your point and agree. A bunch of hoarders drained the supply.
 
Touché. I should’ve said current tube unavailability.

You accept that the big Chinese plant burned in 2019 and has been reopened, yes? Is it debatable that the supply is markedly less than it was before that? Not trying to be a jerk. Just trying to clarify what you are saying.
Yes, absolutely didn’t think you were being a jerk at all! I just don’t want someone to sell great tube amps due to the misperception that tubes are going the way of the dodo bird. I’ve seen comments already from guys at TGP that did this. Social media has a habit of creating drama and panic and creating it’s own problems.
 
Touché. I should’ve said current tube unavailability.

You accept that the big Chinese plant burned in 2019 and has been reopened, yes? Is it debatable that the supply is markedly less than it was before that? Not trying to be a jerk. Just trying to clarify what you are saying.

Don't panic. Shuguang is indeed in ashes but Psvane very much operational - TAD Redbase are made by Psvane and they have whole range of 12a*7 tubes
 
Absolutely! And, with what you address afterwards about clipping, it becomes clearer why people love pushed tube amps, and hate pushed solid state. So, SS designers often tried to go for more clean headroom, because that clipping was nasty if you pushed SS to hard. So then you further fuel the "tubes are louder than SS" because you can push tubes past it's limit and you probably want to keep SS well under it, or it won't sound as good.


Have you tried any of his guitar amps? I haven't. I know some love them, some don't. And as I mentioned above, like other SS designs, many that love them use them mainly as a clean+pedals platform, but some do like the overdrive as well. I'd be interested to hear how they sound, as well as how they handle the load of various guitar speaker/cab configurations.
Tubes amps are louder than solid-state amps partly because they dissipate more than their rated wattage when pushed. That is absolutely true a 50W Marshall can disspate 75W when it pushed causing it to switch from class AB to class B. However, a 50% increase in power is barely audible. It is not even a 3dB increase. It is the constant current operation of a tube power amp that makes it sound louder than a constant voltage power solid-state amp. You can put a 50W Marshall up to a 80W constant voltage solid-state amp and the Marshall will sound louder when playing aggressively. That is because power increases with respect to speaker impedance and speaker impedance increases as we move up the fretboard. That is a large part of the reason why a tube amp cuts through the mix while soloing whereas a non-constant current solid-state amp struggles to cut through the mix.

With that being said, different approaches are being used by solid-state designers to mimic what guitarists like about tube amps. If you read the entire patent applications. Pat has chosen to take a interesting approach with his amp. He has an even harmonic generator and a soft-clipping cell, neither of which are novel. However, I have never encountered a zero crossing processor circuit in a solid-state amp. That circuit simulates the notch at the zero crossing that forms when power tubes in a class AB amp are driven into class B (it is the result of each half of the push-pull circuit no longer operating for more 180 degrees).
 
Back
Top