You are slightly wrong about the effect of the negative feedback - or lack of it. Sure negative feedback driving into a resistor will give you more bass and treble, but not feeding a pair of lightweight coned 12" speakers! The output impedance of the Vox AC30s I was simulating* is not hifi - less than a tenth of an ohm. The output impedance of an AC 30 is 80 ohms!!!! And to my mind this absolutely dominates the AC30 sound. With the open backed cabinet, the loudspeaker/cabinet assembly has a bass resonance of around 80-100 hertz, around a low E-G. This resonance is entirely un-damped by the amplifier. With the original loudspeaker this gives a lovely deep boom . But that's not the end of it. A loudspeaker coil is not just a resistor. It has inductance. And coupled with that high output impedance that means that there is more voltage available and present at high frequencies. So there is a massive treble boost effectively applied to the speakers up in the sort of 'telecaster' style pickup range - the tinkly chinky end of things. Not only is this effect present at low power, its present under overdrive conditions, too. - there may be 30watts at 1khz into an 8 ohm load but there is an effective far far higher voltage available at the loudspeaker top end and at the bass resonance. In order to simulate this I ended up using a 100W transistor amp with a pre clipper and deliberately raised output impedance, which was pretty good except I didnt have a pair of ultra light flappy coned loudspeakers in the cabinet, but a pair of solid high power selection G12 50s and they didn't sound quite the same. An AC30 will absolutely bring out the loudspeakers natural sound. We used to joke 'the furthest distance from hifi you can get' . The Marshal 100 and the Fender Twin reverb I had then on test, both had decent feedback and had output impedances of I think less than an ohm. Both overloaded pretty similarly, and the main difference in output sound was that the Marshall had a weak power transformer that sagged under full load so the initial 'attack' of a power chord was louder than it sustained at ! The Fender was the nearest to hifi that there was. Albeit with the usual bass cut in all guitar amps. Frankly I don't think the tube type made much difference, it had a bit less feedback than the Marshall so it overloaded more softly. What it did have, were some particularly decent speakers in it. British amps generally came with Celestions and the Marshall stack typically had a sealed cabinet, and that's responsible for a lot of its 'sound'. The Fenders were all open backed combos. You have to remember that the speaker and cabinet is the 'soundbox' of an electric guitar. Lightweight cones were louder, but prone to resonances as the different bits of cone flapped at different rates! If you want to get a bit of a feel for the 'sound' of a particular cabinet, unplug it from the amp and just put a battery across it. That is like tapping the soundbox of an acoustic guitar!. The final trick the Marshall had up its sleeve is something that probably has been copied by all the overdrive pedals out there - asymmetric clipping in the preamp. There is - from memory, and its a long time ago (1975?) - an under biased half of an ECC 83 in there right before the tone controls. Ought to have a 1k cathode bias, has a 10k. Something like that. And not much of a bypass capacitor either. That little gem of bad - or good - design meant in effect two things. Firstly when it clipped you didn't get odd harmonics generated, you got even ones. The musical instruments that have sound generators clipped on one side of the waveform, but not the other are the reeded wood winds. Clarinet, oboe, saxophone. A Marshall preamp would do the same, and like reed instruments the fact that it was clipped on one side didn't mean it was clipped on the other - it could go on getting louder even though one half of the waveform was fully smashed into the HT rail. And even that wasn't the end of it. I always suspected that the valve there, under overload, biased itself more off, increasing the distortion. So the distortion at the start of a chord or note was a bit less than the sustain distortion. That coupled with the cheap underpowered mains transformer allowed the Marshall to have a lot more attack power than it did in fact sustain. Pushed into some pretty heavily designed Celestions in a sealed cabinet, the final sound was a lot heavier than the Fender. Ideal for muddy Humbuckers. Whereas the true quality of single coils and P90s goes beautifully with a classic fender amplifier - until you overdrive anyway.! As an electronic engineer from a hifi background the Fenders were far more 'hifi'. But what the British amps had was something borne out of WWII or even before. Neville Shute, in his autobiography 'Slide Rule', states it this way: "An engineer is someone who can do for five bob what any damned fool can do for a quid" (5 bob being one quarter of a pound sterling). The Vox especially, and to a lesser extent the Marshall epitomised their age - more loud, less money spent. The Vox used no feedback, saving an extra amplification stage. It used the lightest coned loudspeakers available because they had more decibels per watt! That gave it a heavy boomy bass and a sparkly treble. The Marshall cut back on the mains transformer - and used more heavier coned loudspeakers in a sealed box. giving it a heavier deep bass sound and an unusual attack. The more affluent Americans could afford to buy Fenders. Better built and with extremely nice loudspeakers for clean country sounds. I will finish with one overall observation. What makes one amp different from another is not usually the tone controls. Until you get into narrow band EQ like a graphic equalisers or a wah wah pedal you aren't really doing more than changing the balance., What really makes a difference is the loudspeakers, especially if you have a high driving impedance. Which the AC30 does. When you get into overdrive, however other elements enter into the equation, like how the overdrive happens. Fender class B amps overload very cleanly and because of lower feedback quite smoothly. Marhsalls with more feedback overload harder. People claim its the pentode/tetrode sound but I think that one valve when overloaded sounds much like another with similar feedback levels. The AC 30 is as far as I know unique in having no feedback - or is it? Another thought is what the output impedance and overload is like on a single ended class A amp - like a Champ. Many people used to swear by them for overdrive. Well that's two years of research in one rather long post. Hope it is food for thought. * spent two years designing transistor guitar amps to sound like tubes - got pretty close!
Solid state amps are basically voltage to voltage and gain is set by V = 1 + R1 / R2, but that doesnt sound good. The amps I am building are voltage to current and the gain is V = 1 + Z / Rsmall. The amp now has to measure the speaker current via Rsmall and modulate whatever voltage needed onto the speaker. For a stupid amp, voltage frequency response is linear and current frequency response looks like the inverse of the impedance graph of the loudspeaker. For a good amp it has to be the other way round, i.e. frequency response looks like the impedance graph of the speaker, frequency-response looks linear currentwise. The Trick ist that my amps use all of their computational power to maintain linear current, while at the Same time PSRR, temperature drift and linearity are computed (I see amps as sorts of analog computer). Bad amps only maintain PSRR, "low distortion", Offset and temperature.
.@@alexandergraf8855 Yes, that's how to do it. I have a Fender Ultimate Chorus that uses that trick. It allows the natural resonances of the loudspeaker cone to become unrestrained - the speaker/cabinet is like the soundbox of an acoustic guitar - and if you manage to soften the clip and keep the impedance high at clip also you van exactly simulate an AC 30. I used on one model a 150W amplifier that was deliberately clipped in the preamp to deliver only 30W. - the power stage never clipped at all. That sounded very good. The chief problem with simulating the AC 30 properly is getting hold of those particular ultra light cone speakers. I was always wondering, but never got round to it, whether simulating a soundbox using digital reverberations techniques would works to simulate - at least for studio work - famous cabinets, like the AC30 or a Marshall 4x12 ...
HI kley nice presentation just want to point out that the ef 86 valve has a gain of 200 twice the 12ax7. I built ac 15 from that schematic and it's pretty well round sound witout the bright cap and added a master volume with ef 86 valve and you get kind of hard working compressed sound at love volume really great amp 😀
fantastic Video, thanks a lot. I especially appreciate that you included the EF86 since that is what i have in addition to the Top-Boost circuit on my Bad Cat Black Cat.
AC30 is a Class A push pull, fairly rare beast, distinguished from AB in how its biased. Also: Class A PP power stages eliminate any harmonic distortion they create thru phase cancellation in the output transformer, its a very clean power stage. So neg feedback isn't going to do anything really, the power stage does the same thing itself. The preamp gain distortion remains, this amp wouldnt be itself with neg feedback. The EF86 is triode strapped so it's different from pentode gain stages like the earliest single ended Tweeds or the Bell & Howell projector amps like the 385. I believe the Gibson GA30 has a pentode input. The EF86 is distinct either way but without direct B+ on the screen the break up is less ragged and has less odd order distortion.
This is just my opinion but it’s somewhat informed-I’ve owned two vox amps(3 if you count a 30 watt valvetronix) in my time. One was a 2012 Ac15 CC1 and the other is an Ac-4 which I built from a Roberts reel to reel amplifier that has the perfect setup tube and transformer wise for a 1960’s Ac-4. Honestly ,the tone is pretty drastically different. You KNOW I’m going to say the amp that I built is better. But it IS BETTER,lol! The. 60’s Ac 4 and Ac15 have an Ef86 pentode for the preamp tube and boy does it make a difference. That “chime” associated with Vox amps? That seems to be one of the main ingredients. The 2012 Ac15 gave it a good try. It’s a good amp! But it sounds to me like an Ac15 if Fender had built it. You’d be hard pressed at times to tell the difference between a modern Ac15 and a Blues junior or even a Princeton reverb reissue. The tremolo and reverb in the new Ac’s sound pretty much identical to Fender amps to me. The tremolo controls in the 60’s schematic ac4 is EXTREMELY interactive, and immediate for lack of a better description. I DID add a depth control to the usual speed control only of the classic ac4. It acts almost like a mild boost or gain knob when the tremolo is off. Really nifty! But the main thing is the tone. It’s SO distinct compared to modern vox amps. It’s almost night and day. The “chime in the new amps can get pretty harsh pretty easily. It’s possible to get it under control but it’s just not the same as the classic vintage AC designs. I’ll tell ya this, the real Ac4 blows the Champ away in my opinion. At least mine does! A bright cap definitely helps with the ac4,tho. If you wanna try turning an Akai /Roberts M8 reel to reel amp into an Ac 4 (actually mine is more like an ac-6) ,I’d DEFINITELY recommend it! It’s a GREAT project. Those amps have all the bones to make a great classic Ac-4! Just lemme know if you’d like info about it or a sound demo! Keep up the good work! I dig this channel! You’re a great player too! Much better than I am!
So, i try to understand this: In the Vox-Cut-Control, you have now drop in Gain, just a filtering of Treble trough phase-canellation. In a Presence Control (like a Marshall) with negative feedback you have an attenuation of Mids/Bass wich results in a perceived treblier sound, but also decreases Gain a bit? Is that right? How does it affect the Tone and interactiveness of the circuits that one happens bewtween the PI and Power-Section and the other feeds in before the PI? Thanks a bunch, your videos are great.
NEGATIVE FEEDBACK - you throw away amplifier gain for extended frequency response and less harmonic distortion. CATHODE BIAS - more compression since the bias is not consistent. Whereas NEGATIVE BIAS is fixed.
I,m interested in how accurate the ac 30 from Weber Amps is. It's purported to be a copy of a 1965 head. I've seen that the board layout from Hoffman is problematic. But their are 20 bullshit posts on forums for every post that is based on actual experience.
Hi Kley. I have 3 VOX AC30s, 63 64 67. Two are pre top boost and one is a top boost. I have been trying to fin eligible schematics without luck, could you help please. Great uploads by the way 😁
After looking at the circuit I wonder why hasn't anyone built a clone of the VOX AC30 albeit with a 56-watt cathode biased power-amp section using the 7868 tubes. The 7868 was a true pentode equivalent of the 6GM5 beam tetrode, the 6GM5 itself was a noval base variant of the 7591A beam tetrode. On paper the 7868 is like a EL84 on steroids.
@@jaycee30865 Are we talking about the same thing? Cathode bias = a large cathode resistor. Fixed bias = cathode straight to ground and a negative voltage on the grid.
I love these amp circuit analysis vids. They feel like a mix of science and magic. I've learnt a lot from them.
You are slightly wrong about the effect of the negative feedback - or lack of it. Sure negative feedback driving into a resistor will give you more bass and treble, but not feeding a pair of lightweight coned 12" speakers!
The output impedance of the Vox AC30s I was simulating* is not hifi - less than a tenth of an ohm. The output impedance of an AC 30 is 80 ohms!!!! And to my mind this absolutely dominates the AC30 sound.
With the open backed cabinet, the loudspeaker/cabinet assembly has a bass resonance of around 80-100 hertz, around a low E-G. This resonance is entirely un-damped by the amplifier. With the original loudspeaker this gives a lovely deep boom . But that's not the end of it. A loudspeaker coil is not just a resistor. It has inductance. And coupled with that high output impedance that means that there is more voltage available and present at high frequencies. So there is a massive treble boost effectively applied to the speakers up in the sort of 'telecaster' style pickup range - the tinkly chinky end of things. Not only is this effect present at low power, its present under overdrive conditions, too. - there may be 30watts at 1khz into an 8 ohm load but there is an effective far far higher voltage available at the loudspeaker top end and at the bass resonance.
In order to simulate this I ended up using a 100W transistor amp with a pre clipper and deliberately raised output impedance, which was pretty good except I didnt have a pair of ultra light flappy coned loudspeakers in the cabinet, but a pair of solid high power selection G12 50s and they didn't sound quite the same.
An AC30 will absolutely bring out the loudspeakers natural sound. We used to joke 'the furthest distance from hifi you can get' .
The Marshal 100 and the Fender Twin reverb I had then on test, both had decent feedback and had output impedances of I think less than an ohm. Both overloaded pretty similarly, and the main difference in output sound was that the Marshall had a weak power transformer that sagged under full load so the initial 'attack' of a power chord was louder than it sustained at !
The Fender was the nearest to hifi that there was. Albeit with the usual bass cut in all guitar amps. Frankly I don't think the tube type made much difference, it had a bit less feedback than the Marshall so it overloaded more softly. What it did have, were some particularly decent speakers in it. British amps generally came with Celestions and the Marshall stack typically had a sealed cabinet, and that's responsible for a lot of its 'sound'. The Fenders were all open backed combos. You have to remember that the speaker and cabinet is the 'soundbox' of an electric guitar. Lightweight cones were louder, but prone to resonances as the different bits of cone flapped at different rates! If you want to get a bit of a feel for the 'sound' of a particular cabinet, unplug it from the amp and just put a battery across it. That is like tapping the soundbox of an acoustic guitar!.
The final trick the Marshall had up its sleeve is something that probably has been copied by all the overdrive pedals out there - asymmetric clipping in the preamp. There is - from memory, and its a long time ago (1975?) - an under biased half of an ECC 83 in there right before the tone controls. Ought to have a 1k cathode bias, has a 10k. Something like that. And not much of a bypass capacitor either.
That little gem of bad - or good - design meant in effect two things.
Firstly when it clipped you didn't get odd harmonics generated, you got even ones. The musical instruments that have sound generators clipped on one side of the waveform, but not the other are the reeded wood winds. Clarinet, oboe, saxophone. A Marshall preamp would do the same, and like reed instruments the fact that it was clipped on one side didn't mean it was clipped on the other - it could go on getting louder even though one half of the waveform was fully smashed into the HT rail.
And even that wasn't the end of it. I always suspected that the valve there, under overload, biased itself more off, increasing the distortion. So the distortion at the start of a chord or note was a bit less than the sustain distortion. That coupled with the cheap underpowered mains transformer allowed the Marshall to have a lot more attack power than it did in fact sustain. Pushed into some pretty heavily designed Celestions in a sealed cabinet, the final sound was a lot heavier than the Fender. Ideal for muddy Humbuckers. Whereas the true quality of single coils and P90s goes beautifully with a classic fender amplifier - until you overdrive anyway.!
As an electronic engineer from a hifi background the Fenders were far more 'hifi'. But what the British amps had was something borne out of WWII or even before. Neville Shute, in his autobiography 'Slide Rule', states it this way:
"An engineer is someone who can do for five bob what any damned fool can do for a quid"
(5 bob being one quarter of a pound sterling).
The Vox especially, and to a lesser extent the Marshall epitomised their age - more loud, less money spent. The Vox used no feedback, saving an extra amplification stage. It used the lightest coned loudspeakers available because they had more decibels per watt! That gave it a heavy boomy bass and a sparkly treble.
The Marshall cut back on the mains transformer - and used more heavier coned loudspeakers in a sealed box. giving it a heavier deep bass sound and an unusual attack.
The more affluent Americans could afford to buy Fenders. Better built and with extremely nice loudspeakers for clean country sounds.
I will finish with one overall observation. What makes one amp different from another is not usually the tone controls. Until you get into narrow band EQ like a graphic equalisers or a wah wah pedal you aren't really doing more than changing the balance., What really makes a difference is the loudspeakers, especially if you have a high driving impedance. Which the AC30 does.
When you get into overdrive, however other elements enter into the equation, like how the overdrive happens. Fender class B amps overload very cleanly and because of lower feedback quite smoothly. Marhsalls with more feedback overload harder. People claim its the pentode/tetrode sound but I think that one valve when overloaded sounds much like another with similar feedback levels. The AC 30 is as far as I know unique in having no feedback - or is it?
Another thought is what the output impedance and overload is like on a single ended class A amp - like a Champ. Many people used to swear by them for overdrive.
Well that's two years of research in one rather long post. Hope it is food for thought.
* spent two years designing transistor guitar amps to sound like tubes - got pretty close!
Solid state amps are basically voltage to voltage and gain is set by V = 1 + R1 / R2, but that doesnt sound good. The amps I am building are voltage to current and the gain is V = 1 + Z / Rsmall. The amp now has to measure the speaker current via Rsmall and modulate whatever voltage needed onto the speaker.
For a stupid amp, voltage frequency response is linear and current frequency response looks like the inverse of the impedance graph of the loudspeaker. For a good amp it has to be the other way round, i.e. frequency response looks like the impedance graph of the speaker, frequency-response looks linear currentwise.
The Trick ist that my amps use all of their computational power to maintain linear current, while at the Same time PSRR, temperature drift and linearity are computed (I see amps as sorts of analog computer). Bad amps only maintain PSRR, "low distortion", Offset and temperature.
.@@alexandergraf8855 Yes, that's how to do it. I have a Fender Ultimate Chorus that uses that trick. It allows the natural resonances of the loudspeaker cone to become unrestrained - the speaker/cabinet is like the soundbox of an acoustic guitar - and if you manage to soften the clip and keep the impedance high at clip also you van exactly simulate an AC 30. I used on one model a 150W amplifier that was deliberately clipped in the preamp to deliver only 30W. - the power stage never clipped at all. That sounded very good. The chief problem with simulating the AC 30 properly is getting hold of those particular ultra light cone speakers. I was always wondering, but never got round to it, whether simulating a soundbox using digital reverberations techniques would works to simulate - at least for studio work - famous cabinets, like the AC30 or a Marshall 4x12 ...
+1000 for pulling up the schematic in GIMP.
HI kley nice presentation just want to point out that the ef 86 valve has a gain of 200 twice the 12ax7. I built ac 15 from that schematic and it's pretty well round sound witout the bright cap and added a master volume with ef 86 valve and you get kind of hard working compressed sound at love volume really great amp 😀
Good point! I have built one amp with a pentode in v1 and enjoy it quite a lot. A 6au6 I believe. Definitely changes the feel and breakup curve.
Love the video Kley! I'm a small step behind you on the amp schematic understanding but this all made total sense and helped me a lot! Keep em coming!
Thanks! I've done a few others and plan on doing some more!
Great job explaining this - just found your channel 🙂
Thanks! I'm a big fan of Wampler pedals!
The magic is in the normal channel imo
Great video! Thanks
fantastic Video, thanks a lot. I especially appreciate that you included the EF86 since that is what i have in addition to the Top-Boost circuit on my Bad Cat Black Cat.
AC30 is a Class A push pull, fairly rare beast, distinguished from AB in how its biased. Also: Class A PP power stages eliminate any harmonic distortion they create thru phase cancellation in the output transformer, its a very clean power stage. So neg feedback isn't going to do anything really, the power stage does the same thing itself. The preamp gain distortion remains, this amp wouldnt be itself with neg feedback.
The EF86 is triode strapped so it's different from pentode gain stages like the earliest single ended Tweeds or the Bell & Howell projector amps like the 385. I believe the Gibson GA30 has a pentode input. The EF86 is distinct either way but without direct B+ on the screen the break up is less ragged and has less odd order distortion.
Where did you get the schematic from, tho ones I find for Vox are terrible, your are very well laid out
This is just my opinion but it’s somewhat informed-I’ve owned two vox amps(3 if you count a 30 watt valvetronix) in my time. One was a 2012 Ac15 CC1 and the other is an Ac-4 which I built from a Roberts reel to reel amplifier that has the perfect setup tube and transformer wise for a 1960’s Ac-4. Honestly ,the tone is pretty drastically different. You KNOW I’m going to say the amp that I built is better. But it IS BETTER,lol! The. 60’s Ac 4 and Ac15 have an Ef86 pentode for the preamp tube and boy does it make a difference. That “chime” associated with Vox amps? That seems to be one of the main ingredients. The 2012 Ac15 gave it a good try. It’s a good amp! But it sounds to me like an Ac15 if Fender had built it. You’d be hard pressed at times to tell the difference between a modern Ac15 and a Blues junior or even a Princeton reverb reissue. The tremolo and reverb in the new Ac’s sound pretty much identical to Fender amps to me. The tremolo controls in the 60’s schematic ac4 is EXTREMELY interactive, and immediate for lack of a better description. I DID add a depth control to the usual speed control only of the classic ac4. It acts almost like a mild boost or gain knob when the tremolo is off. Really nifty! But the main thing is the tone. It’s SO distinct compared to modern vox amps. It’s almost night and day. The “chime in the new amps can get pretty harsh pretty easily. It’s possible to get it under control but it’s just not the same as the classic vintage AC designs. I’ll tell ya this, the real Ac4 blows the Champ away in my opinion. At least mine does! A bright cap definitely helps with the ac4,tho. If you wanna try turning an Akai /Roberts M8 reel to reel amp into an Ac 4 (actually mine is more like an ac-6) ,I’d DEFINITELY recommend it! It’s a GREAT project. Those amps have all the bones to make a great classic Ac-4! Just lemme know if you’d like info about it or a sound demo!
Keep up the good work! I dig this channel! You’re a great player too! Much better than I am!
Thanks for that. Interesting that you can cut treble in two places, pre or post distortion. Are the values of the pots the same?
Great videos.
So, i try to understand this: In the Vox-Cut-Control, you have now drop in Gain, just a filtering of Treble trough phase-canellation. In a Presence Control (like a Marshall) with negative feedback you have an attenuation of Mids/Bass wich results in a perceived treblier sound, but also decreases Gain a bit? Is that right? How does it affect the Tone and interactiveness of the circuits that one happens bewtween the PI and Power-Section and the other feeds in before the PI? Thanks a bunch, your videos are great.
my Vox ac30 cc2 was replace the VR1 and VR2 potentiometer (A500K) into 1mega ohm by technician ..what is the pros and cons about it?....tnx
NEGATIVE FEEDBACK - you throw away amplifier gain for extended frequency response and less harmonic distortion. CATHODE BIAS - more compression since the bias is not consistent. Whereas NEGATIVE BIAS is fixed.
I,m interested in how accurate the ac 30 from Weber Amps is. It's purported to be a copy of a 1965 head. I've seen that the board layout from Hoffman is problematic. But their are 20 bullshit posts on forums for every post that is based on actual experience.
Hi Kley.
I have 3 VOX AC30s, 63 64 67. Two are pre top boost and one is a top boost. I have been trying to fin eligible schematics without luck, could you help please.
Great uploads by the way 😁
After looking at the circuit I wonder why hasn't anyone built a clone of the VOX AC30 albeit with a 56-watt cathode biased power-amp section using the 7868 tubes. The 7868 was a true pentode equivalent of the 6GM5 beam tetrode, the 6GM5 itself was a noval base variant of the 7591A beam tetrode. On paper the 7868 is like a EL84 on steroids.
I think that would be very doable. Need a little beefier set of transformers and to do a little trial and error for biasing, but very doable.
Next time don’t film when robbers are doing your home.
Thank you.
JK, but it sounded like that in some parts. Great video. Cheers.
Highest power cathode bias you are aware of? Have you ever heard of a fender twin from the 1970s?
I have not! I know of twin reverb using fixed bias. What circuit is that?
Silverface era twin is cathode biased. As is bassman and I think basically everything.
@@jaycee30865 Are we talking about the same thing? Cathode bias = a large cathode resistor. Fixed bias = cathode straight to ground and a negative voltage on the grid.
Kley De Jong ab763 vs ab568.
@@jaycee30865 I will check it out, thanks!