I'm saying it's abrupt because, even though there are about "four stages of amplification" about all of them are overdriven to very high degree and you get square-ish wave output as a result.
IME, most of these high gain amps employ cascaded gain stages - not for gradual slight clipping in each stage as you often hear claimed - but to get dynamic harmonic shifts from interstage DC offset shifting caused by asymmetric clipping.
The effect would not even be very pronounced with just "slight" clipping, that's why, for example, the "cold cathode" stage of SLO clips almost one halfwave off the signal. Anything but gradual and slight.
Lot of these "high gain" designs were made to imitate Marshall sound, and Marshall powerstage clipping is not actually very gradual or "soft". Its brickwall limiting and dynamic crossover distortion would likely scare off the folk who are a accustomed to seeing mere drawings of how tubes supposedly clip.
About the only high gain amp I know that does this "gradual slight clipping" thing to extent is the weird Carvin Quad X preamp with its insane 7 to 11 gain stages. And even that one was admittedly designed for dynamic DC offset shifting, lot's of it.
IME, if you want to build a traditional high gain amp then you need that hard clipping for its harmonics, and you need lots of it for its compression. "Slight" just doesn't work.
I'm not saying it doesn't go on in e.g. AC30 or 5E3 but different amps are for different effects. You don't choose a AC30 for over the top soaring grindcore, or a 5150 to play smooth jazz. But all of those are "tube amp tones" and it is indeed a very broad category.
IME, most of these high gain amps employ cascaded gain stages - not for gradual slight clipping in each stage as you often hear claimed - but to get dynamic harmonic shifts from interstage DC offset shifting caused by asymmetric clipping.
The effect would not even be very pronounced with just "slight" clipping, that's why, for example, the "cold cathode" stage of SLO clips almost one halfwave off the signal. Anything but gradual and slight.
Lot of these "high gain" designs were made to imitate Marshall sound, and Marshall powerstage clipping is not actually very gradual or "soft". Its brickwall limiting and dynamic crossover distortion would likely scare off the folk who are a accustomed to seeing mere drawings of how tubes supposedly clip.
About the only high gain amp I know that does this "gradual slight clipping" thing to extent is the weird Carvin Quad X preamp with its insane 7 to 11 gain stages. And even that one was admittedly designed for dynamic DC offset shifting, lot's of it.
IME, if you want to build a traditional high gain amp then you need that hard clipping for its harmonics, and you need lots of it for its compression. "Slight" just doesn't work.
I'm not saying it doesn't go on in e.g. AC30 or 5E3 but different amps are for different effects. You don't choose a AC30 for over the top soaring grindcore, or a 5150 to play smooth jazz. But all of those are "tube amp tones" and it is indeed a very broad category.
Last edited:
I also build the AC30, but I'm absolutely sure that the EL84 tubes are guilty for its signature more than anything else.It is massively different than anything else.If Fender is clean , AC30 is sparkling.If the only audio amplifiers used by humans were the guitar amplifiers, the AC 30 would have been the most praised High end amplifier of all times , something like an Accuphase while a Fender would have been the JC or PL of that market.
The Peavey Classic 30 has basically the same tube setup as Vox AC30, yet it is a very different amp. I put more weight on overall circuit design, which makes the AC30 what it is. It would be a totally different amp with fixed bias, cathodyne phase splitter, "closed loop" and equipped with a Fender-ish preamp and distortion channel... You know, like the Classic 30.
...And AC30 is a pretty good example of significant variety of amps because the original one - designed for accordions and for "multipurpose" - was actually a fixed bias amp with dual EL34's and a single speaker. After that there have been dual channel EF86 equipped and three-channel 12AX7 equipped versions, those also with different voicing versions for "bass" and "guitar", then the addition of top boost channel, another gain stage, different speaker setups including Celestion, Goodman or Fane speakers, tube or solid-state rectified versions, getting rid of the wonderful pitch shift vibrato and et cetera. So when someone mentions "signature tone" of AC30 I have to wonder what that actually means since different pedigrees can be quite different amps from another. It's like saying signature "Bassman" tone or signature "JC-120" tone, and each person likely has a different idea what that actually means.
...And AC30 is a pretty good example of significant variety of amps because the original one - designed for accordions and for "multipurpose" - was actually a fixed bias amp with dual EL34's and a single speaker. After that there have been dual channel EF86 equipped and three-channel 12AX7 equipped versions, those also with different voicing versions for "bass" and "guitar", then the addition of top boost channel, another gain stage, different speaker setups including Celestion, Goodman or Fane speakers, tube or solid-state rectified versions, getting rid of the wonderful pitch shift vibrato and et cetera. So when someone mentions "signature tone" of AC30 I have to wonder what that actually means since different pedigrees can be quite different amps from another. It's like saying signature "Bassman" tone or signature "JC-120" tone, and each person likely has a different idea what that actually means.
There is another point aside from clipping, is at low signal amplitudes there still remains some inherent non-linearity of the tube stages, particularly triodes. Maybe not glaringly obvious but adding some percentage of low order harmonics, which to me at least is where a lot (not all) of SS amps have their ''tell''.
Apart from cascaded gain stages, the classic 4 hole jumped amp (High Brilliant/Low Normal) ends up being a parallel mixable blend of non-linear waveshaping and EQ shaping.. which is something I haven't run across in solid state versions, maybe some examples are out there.
I wished more people here read your book https://www.thatraymond.com/downloads/solidstate_guitar_amplifiers_teemu_kyttala_v1.0.pdf
It is an eye-opener on tube amp mysticism
Thanks. 2nd revision has been under work for more than ten years. Seems to be one of those projects that grows more out of control the more I put effort to it though. Currently 700+ pages and going.
Anyway, there are so many fascinating additions that I hope I will get time to finish that thing before it is totally outdated (I have a dayjob and other responsibilities, can't sit just hours and hours writing everyday). Analog solid state guitar amps with class AB outputs are likely history soon.
Anyway, it's been an extremely educating journey to write that stuff along with huge amount of research it has taken. It has totally changed several of my points of views about guitar amps, related technology, and et cetera. There were so many things I had "learned from the Internet" that I had to re-evaluate once I had took the time to research and test them myself. Tube amps, for instance, seem to have this magical aura of perfection attributed to them, which quite often features aspects that won't hold closer scrutiny, and many still seem to evaluate solid-state amps as if the technology was still the same as in the late 1960's or simply synonymous to cheap quality, so again by aspects that do not hold closer scrutiny.
There are wonderful examples of this even in this very thread.
Anyway, there are so many fascinating additions that I hope I will get time to finish that thing before it is totally outdated (I have a dayjob and other responsibilities, can't sit just hours and hours writing everyday). Analog solid state guitar amps with class AB outputs are likely history soon.
Anyway, it's been an extremely educating journey to write that stuff along with huge amount of research it has taken. It has totally changed several of my points of views about guitar amps, related technology, and et cetera. There were so many things I had "learned from the Internet" that I had to re-evaluate once I had took the time to research and test them myself. Tube amps, for instance, seem to have this magical aura of perfection attributed to them, which quite often features aspects that won't hold closer scrutiny, and many still seem to evaluate solid-state amps as if the technology was still the same as in the late 1960's or simply synonymous to cheap quality, so again by aspects that do not hold closer scrutiny.
There are wonderful examples of this even in this very thread.
It's regretful, really.. I maintain some hope that musicians are aware of and could appreciate that good analog solid state designs do exist. It's still a good topic of discussion for DIYers as to what elements make for a good sounding guitar amp. Whether it's DSP based, analog solid state or vacuum tube, there are examples of each that are considered "good" and also some that are "bad" ..which is hugely subjective.. and may at times be based on a pre-judgement of one technology over another. I'm in the middle of reading through 3 technical books at the moment, but I'm hoping to get to yours soon, Teemu. Best of luck with the revision work!
Idea: What would happen if a limiter was placed right after pick up or the guitar?
Ive always perceived it as; An electric guitar sounds like sh.. if not very gently plugged at low volume.
All the (right) imperfections of pedals, amplification, speaker and drivers makes it sound good. A nice imperfect tubeamp does away with some of the nastyness, but if you plug in some good headphones or hifi speakers to speaker out, it would still sound bad and still with earsplitting initial spikes, even its less than through a hifi amp.
I really like Gnobuddys observations as allways. I believe he could design a successful comercial amp with all the details he is conserned with in the whole chain of sound from guitar to listener.
Cheers!
Ive always perceived it as; An electric guitar sounds like sh.. if not very gently plugged at low volume.
All the (right) imperfections of pedals, amplification, speaker and drivers makes it sound good. A nice imperfect tubeamp does away with some of the nastyness, but if you plug in some good headphones or hifi speakers to speaker out, it would still sound bad and still with earsplitting initial spikes, even its less than through a hifi amp.
I really like Gnobuddys observations as allways. I believe he could design a successful comercial amp with all the details he is conserned with in the whole chain of sound from guitar to listener.
Cheers!
There's great info in Gnobuddy's starting posts about transient and asymmetric nature of guitar signals.
And then it gets off the mark. Why does not he observe what this magical tube amp actually does to the signal when it slams the very same transient signal, amplified by factor of 60x, to a grid that clips at the threshold of 1 to 2 volts. I can tell it's not that nice and soft.
Grid conduction is pretty much akin to transfer characteristic of a generic diode. There is slight gradual gain compression in plate clipping, but only with triodes which have gain dependent on plate voltage. Revered power tube clipping is usually pentodes and their transfer characteristics do not differ much from generic bipolar junction transistors.
We should acknowledge this before assuming magical properties to tubes. They are likewise equally subject to clipping a lot with transient peaks. They do not have magical traits to circumvent that.
...And then in the very next post we are shown very nice soft clipping, which is actually produced by these nasty solid state circuits.
I do like this thread, in its long run it has featured many interesting ideas, but sometimes some claims have to be taken with a grain of salt. My sincere thumbs up for Gnobuddy but I still can't agree on everything with him.
And then it gets off the mark. Why does not he observe what this magical tube amp actually does to the signal when it slams the very same transient signal, amplified by factor of 60x, to a grid that clips at the threshold of 1 to 2 volts. I can tell it's not that nice and soft.
Grid conduction is pretty much akin to transfer characteristic of a generic diode. There is slight gradual gain compression in plate clipping, but only with triodes which have gain dependent on plate voltage. Revered power tube clipping is usually pentodes and their transfer characteristics do not differ much from generic bipolar junction transistors.
We should acknowledge this before assuming magical properties to tubes. They are likewise equally subject to clipping a lot with transient peaks. They do not have magical traits to circumvent that.
...And then in the very next post we are shown very nice soft clipping, which is actually produced by these nasty solid state circuits.
I do like this thread, in its long run it has featured many interesting ideas, but sometimes some claims have to be taken with a grain of salt. My sincere thumbs up for Gnobuddy but I still can't agree on everything with him.
I'm trying to read how you mean this - is the emphasis on "analog" (as in an analog-only path being replaced by digital modeling), or the Class AB being replaced by Class D, both of those, or something else?
Both. When I look at new (solid state) guitar amplifiers the trend is digital signal processing and class-D amplification. The "new" tech is finally starting to sink in. IMO, old analog circuits and class-AB topologies can't order much of competition. They may remain in use for nostalgic reasons but the newer tech is in many regards superior.
That was a point made earlier in the thread. DSP has only recently reached the price/performance level to actually produce very convincing amp models in affordable musical amp products. And there was a lot of effort put in to programming multiple wave-shapers and EQ shaping to emulate both preamp and power amp overdrive. The Katana series was a real eye opener for a lot of players, apart from that it was spending quite a lot of $$ for Axe FX, Kemper etc.. but most of the big commercial amp manufacturers have had to up their game on the DSP. I'll comment on one thing, if the Flamma FS06 DSP preamp is that good (as it sounds in the demo videos), then for about $100 you have a front end preamp that could do about the same as a Fender Master Tone DSP based amp which is way more money, just run it through a decent power amp and speaker.
I have to disagree that class ab will disappear. You can buy something like a Sessionette or much cheaper Roland 405?
Then you have ONE great amp, and thats all you need for making the best possible guitar sounds.
Tubeamps will prevail cause dsp can never simulate all the different feedback interactions between their speakers and the strings of the guitar.
On the other hand guitaramps will MOSTLY disappear cause DSP now sounds so great directly into a computer, moniter- it PA-system.
Cheers!
Then you have ONE great amp, and thats all you need for making the best possible guitar sounds.
Tubeamps will prevail cause dsp can never simulate all the different feedback interactions between their speakers and the strings of the guitar.
On the other hand guitaramps will MOSTLY disappear cause DSP now sounds so great directly into a computer, moniter- it PA-system.
Cheers!
Yes, real soon after those silly vacuum tube amps disappear.
Your Search Results for “tube amps”
(3631 Items)Obviously it's not going to disappear completely, but I have a hunch that in large manufacturers will lean towards class-D in future. Already in quite many applications class-D amps have displaced class-AB amps.
DSP will similarly displace analog signal processing. This is happening in large scale too.
This is repeated often but I don't agree with it. Today many modeling units can actually do a better job in featuring those interactions than some genuine tube amps. And not only that, units like Axe FX can fine tune even minuatiae interactions and parameters with unparalleled tweakability.
Tube amps won't disappear either but the reason they still exist is not their superiority or that they perform some unmatchable tricks.