A list of the modules, as currently envisioned:
* input section: two input jacks, input level (pull bright), 6-pos HPF switch.
* active EQ: treble, mid freq, mid level, bass, LED. (2x)
* type M gain stage: gain (pull colder), bass, mid, treble, level, LED.
* type F+ gain stage: gain (pull 2-stage), bass (pull alt voice), mid (pull raw), treble (pull bright), level, LED. (2x)
* reverb: 6-pos source switch, L mix, R mix, LED.
* bypass switches: a panel of 6 switches which replicate the footswitch functions.
* master section: NFB in/out, resonance, presence, master volume, mute switch, preamp switch, power amp switch, two power LEDs.
In addition, I have designed two other "horizontal" modules, which can go into the same slots as the gain modules; however, for my own amp, I won't use these:
* 5-band graphic EQ: LRC resonant circuits with tube opamp, center frequencies based on the Mesa Boogie Mark IV.
* dual semi-parametric EQ: similar to the variable mid controls from the four-knob active EQs, possibly with wider frequency range.
Monday, February 11, 2019
reverb and effects
This amp will have two spring reverb tanks, to produce a true stereo image. Even with identical tanks, there will be subtle differences between the left and right signals, leading to a stereo effect. But it will be possible for users to install different tanks for left and right, e.g., tanks with different numbers or lengths of springs, or different decay times, to obtain a more dramatic stereo effect. The left and right return signals can each be separately mixed from full-dry to full-wet, enabling unusual stereo settings such as full reverb from one speaker and full dry tone from the other.
The spring reverbs are bypassed if external effects are plugged into the return jacks; the reverb and effects loop share the same bypass relay, and it's not possible to have both spring reverb and external stereo effects active at the same time (although one could plug a mono effect into one channel, while keeping the spring reverb in the other channel, for some interesting sonic variations).
For an unusual capability that no other amp provides (that I know of), the reverb and effects loop will have a six-position switch to select the "send" source. In the rightmost position, the send will come from the end of the signal chain as normally expected, i.e., from the output of the second EQ. But turning the control to the left will select progressively earlier points in the chain: the output of each gain stage, the first EQ, or the output of the first input stage itself. Thus, depending on the settings of the EQs and gain stages, and depending on the position of this source selector knob, the reverb or external effects can process a much cleaner early signal, with later distortion and EQ colouration only applying to the "dry" signal. There are not a lot of examples of this type of sound in the recorded canon, but I suspect it may prove musically useful.
In addition to the spring reverbs and the main (final) stereo effects loop, there are two earlier points where mono effects can be inserted: each of the two active EQ modules provides an effects loop, switched by the same bypass relay as the EQ. Plugging an effect into the return jack overrides the output of the EQ, but two send jacks are provided, pre- and post- EQ, so the external effect can either replace the EQ, or can cascade after it.
The motivation for providing these earlier mono effects loops in addition to the main stereo loop, is that just like in the case of EQs, there can be a radical difference in the tonality produced by an effect when placed in front of the distortion stages, as opposed to afterwards. Of course one could plug the effects into the chain ahead of the amp entirely, but then the guitar sees the load of the effect circuit, whatever that may be; with the arrangement here, even the earliest effect position still has tube buffering, so the guitar always sees the same input circuit (which is a one megohm load to ground, with 34 kohms series into a tube grid: i.e., the traditional Fender-style guitar input).
None of the effects loops has its own dedicated buffering; instead, I simply exploit good buffered "send" locations where they already exist in the chain. Thus, the external effects units themselves are responsible for not loading their inputs too much, and for driving their outputs the "right amount" to re-insert into the signal chain. And probably the most inconvenient part, the effects units must be able to handle "raw" tube-circuitry signals, which are nominally supposed to be at line level, but which can potentially climb much higher, into the realm of several dozens of volts peak-to-peak. Tubes don't especially mind this kind of input overload, but solid state circuits certainly will! (Effect sends from tube circuits often include some type of pad or divider, to reduce the signal level, and then of course added gain on the return inputs to compensate.) Probably it would be more accurate to call these "patch points", rather than "effects loops", but I stick with the more widely understood terminology.
The spring reverbs are bypassed if external effects are plugged into the return jacks; the reverb and effects loop share the same bypass relay, and it's not possible to have both spring reverb and external stereo effects active at the same time (although one could plug a mono effect into one channel, while keeping the spring reverb in the other channel, for some interesting sonic variations).
For an unusual capability that no other amp provides (that I know of), the reverb and effects loop will have a six-position switch to select the "send" source. In the rightmost position, the send will come from the end of the signal chain as normally expected, i.e., from the output of the second EQ. But turning the control to the left will select progressively earlier points in the chain: the output of each gain stage, the first EQ, or the output of the first input stage itself. Thus, depending on the settings of the EQs and gain stages, and depending on the position of this source selector knob, the reverb or external effects can process a much cleaner early signal, with later distortion and EQ colouration only applying to the "dry" signal. There are not a lot of examples of this type of sound in the recorded canon, but I suspect it may prove musically useful.
In addition to the spring reverbs and the main (final) stereo effects loop, there are two earlier points where mono effects can be inserted: each of the two active EQ modules provides an effects loop, switched by the same bypass relay as the EQ. Plugging an effect into the return jack overrides the output of the EQ, but two send jacks are provided, pre- and post- EQ, so the external effect can either replace the EQ, or can cascade after it.
The motivation for providing these earlier mono effects loops in addition to the main stereo loop, is that just like in the case of EQs, there can be a radical difference in the tonality produced by an effect when placed in front of the distortion stages, as opposed to afterwards. Of course one could plug the effects into the chain ahead of the amp entirely, but then the guitar sees the load of the effect circuit, whatever that may be; with the arrangement here, even the earliest effect position still has tube buffering, so the guitar always sees the same input circuit (which is a one megohm load to ground, with 34 kohms series into a tube grid: i.e., the traditional Fender-style guitar input).
None of the effects loops has its own dedicated buffering; instead, I simply exploit good buffered "send" locations where they already exist in the chain. Thus, the external effects units themselves are responsible for not loading their inputs too much, and for driving their outputs the "right amount" to re-insert into the signal chain. And probably the most inconvenient part, the effects units must be able to handle "raw" tube-circuitry signals, which are nominally supposed to be at line level, but which can potentially climb much higher, into the realm of several dozens of volts peak-to-peak. Tubes don't especially mind this kind of input overload, but solid state circuits certainly will! (Effect sends from tube circuits often include some type of pad or divider, to reduce the signal level, and then of course added gain on the return inputs to compensate.) Probably it would be more accurate to call these "patch points", rather than "effects loops", but I stick with the more widely understood terminology.
modular layout, gain stages, footswitches
In terms of physical layout, this amp will be modular. Each significant function block will be built as a separate module. Each module comprises a rectangular region of the front panel, containing the associated knobs and/or switches. The module also contains a circuit board (not necessarily a PCB, however; wiring will probably be done point-to-point, using so-called turret boards); and on the back panel are the associated tubes. Most preamp modules have one or two 12AX7 tubes; some may have up to four. The modules are housed in segments of extruded aluminum "U"-channel, which forms the front and back panels as well as the "floor". Thus, quite a bit of EMI shielding is provided to the electronics, even though I plan to build the outer case of wood. The modular design is to facilitate incremental development, construction, and testing, as well as to encourage others to develop their own modules. Each module will come with "flying leads" (wires), which are soldered to connection points on the amp chassis; this choice favours reliable connections over the convenience of rapidly swapping modules in the field.
The heart of this (or any) guitar amp are the gain stages. This is where the bulk of any desired distortion is produced; even with so-called "clean" guitar tones, the mild distortion produced by tube gain stages is a critical element of the sound. A Fender amp can produce a nice "clean" tone for a guitar; plug that same guitar directly into a hi-fi stereo amp or mixing board, and you'll hear a very thin and disappointing tone, lacking in the interest and "sparkle" of the Fender amp tone. An electric guitar is really only part of a musical instrument; the formants and non-linearities imposed upon the signal by the amp or other electronics, become crucial components of the complete musical instrument whose sound we call "electric guitar".
In my amp, rather than having a single high-gain section to produce the distortion, I have elected to provide several lower-gain sections, which can be cascaded to produce just about any desired level of distortion. Each section has gain, tone, and level controls, allowing subtle control over the character of the distortion as it is built up. Alternatively, although this is technically a single-channel amp, different gain sections can be set up for different tones and can be switched in and out by the footswitches, affording some of the flexibility of a multi-channel amp, while also providing the ability to run some or all of the sections in series, simultaneously, for extreme levels of gain.
Initially, I plan to build two types of gain stage: the "Type M", which copies a Marshall 2203 preamp, and the "Type F-Plus", which is a Fender AB763 preamp with an additional switchable gain stage (i.e., either one tube with two triode sections, or two tubes with four sections). The amp will have space for three gain modules, which I'll probably populate as F+ -> M -> F+.
The footswitch board will have six buttons, each a regular push-on/push-off metal footswitch, with an associated status LED. The LEDs will be replicated on the amp front panel, with each LED colour-coded and located with its associated preamp module. Each footswitch activates one of the six bypass relays: two EQs, three gain stages, and the final stereo effects loop. For convenience in cabling, the footswitch unit will connect to the amp using a standard 8-conductor "CAT-5" cable, of any desired length (six switched lines plus power and ground).
The heart of this (or any) guitar amp are the gain stages. This is where the bulk of any desired distortion is produced; even with so-called "clean" guitar tones, the mild distortion produced by tube gain stages is a critical element of the sound. A Fender amp can produce a nice "clean" tone for a guitar; plug that same guitar directly into a hi-fi stereo amp or mixing board, and you'll hear a very thin and disappointing tone, lacking in the interest and "sparkle" of the Fender amp tone. An electric guitar is really only part of a musical instrument; the formants and non-linearities imposed upon the signal by the amp or other electronics, become crucial components of the complete musical instrument whose sound we call "electric guitar".
In my amp, rather than having a single high-gain section to produce the distortion, I have elected to provide several lower-gain sections, which can be cascaded to produce just about any desired level of distortion. Each section has gain, tone, and level controls, allowing subtle control over the character of the distortion as it is built up. Alternatively, although this is technically a single-channel amp, different gain sections can be set up for different tones and can be switched in and out by the footswitches, affording some of the flexibility of a multi-channel amp, while also providing the ability to run some or all of the sections in series, simultaneously, for extreme levels of gain.
Initially, I plan to build two types of gain stage: the "Type M", which copies a Marshall 2203 preamp, and the "Type F-Plus", which is a Fender AB763 preamp with an additional switchable gain stage (i.e., either one tube with two triode sections, or two tubes with four sections). The amp will have space for three gain modules, which I'll probably populate as F+ -> M -> F+.
The footswitch board will have six buttons, each a regular push-on/push-off metal footswitch, with an associated status LED. The LEDs will be replicated on the amp front panel, with each LED colour-coded and located with its associated preamp module. Each footswitch activates one of the six bypass relays: two EQs, three gain stages, and the final stereo effects loop. For convenience in cabling, the footswitch unit will connect to the amp using a standard 8-conductor "CAT-5" cable, of any desired length (six switched lines plus power and ground).
the preamp -- pre and post EQ
The preamp of my new guitar amp will have lots of knobs and lots of tubes. However, through bypass relays, it will be possible to cut the signal path down to a one-tube preamp with not much more than a level control, akin to early amplifiers like the Fender "tweed" models.
One key factor in shaping overdriven guitar tones, to me, is the capability to shape the spectrum of the tone (i.e., EQ) both before and after the distortion-producing gain stage(s). The spectrum of the tone going into the gain stage has a big effect on the character or "texture" of the distortion; changes such as turning up the bass will not necessarily produce more bass at the output (indeed, the result could well be more treble and high harmonics), but the nature of the distortion and the playing-feel will be greatly altered. Changing the EQ after the gain stage, on the other hand, will have the expected results on the final sound: turning up the bass will result in more bass being heard, etc..
Different guitar amps have historically placed their tone controls at different points in the signal path (e.g.: Fender: close to the input; Marshall: close to the output), and this is one big reason that different amps sound different from each other. Many guitarists, perhaps unwittingly in some cases, have discovered the importance of being able to EQ the tone in at least two places in the chain; but since most amps don't directly address this, they use EQ pedals or other frequency-selective boost pedals, ahead of either pedal- or amp- produced distortion, and then they use the amp tone controls to shape the final sound. Many of the popular distortion pedals distinguish themselves through the tonality they impart with built-in pre-distortion EQ. For maximum flexibility, my amp will provide two full sets of active EQ controls (bypassable), before and after the gain stages.
The usual tone-stacks used in Fender and Marshall amps provide bass, mid, and treble controls, but due to the design of the passive R-C circuits, it is not actually possible to generate a mid-frequency "hump", just more or less relative proportion of "mid scoop". A mid hump, aka formant, is a fundamental aspect of the "voice" of most musical instruments, including guitar. Most guitar-plus-amp systems exhibit a formant, often chiefly due to the frequency response anomalies of the speaker cabinet. But the only way to change such a formant is to swap cabinets. Twiddling knobs on most amps won't make much difference to the formant. Again, some guitarists employ frequency-selective pedals to produce or alter the formant shape of their systems. EQ pedals are an obvious approach, but also the common trick of using a wah pedal kept at a fixed setting, has been used by many guitarists over the years to produce a dramatic formant in their tone.
Because of the importance of formants, I am electing to use a 4-knob active EQ, in both the "pre" and "post" positions of the signal chain. This EQ provides fixed bass and treble controls, based on the "Baxandall" circuit, which enables both boost and cut of the fixed-frequency "shelves". For midrange, a third boost/cut level knob is provided, along with a frequency knob (initially, I'm looking at a 10:1 range, from 200 Hz to 2 kHz, though this may change with experimentation). There is no control to change the "Q"; i.e., this is a "semi-parametric" mid.
One key factor in shaping overdriven guitar tones, to me, is the capability to shape the spectrum of the tone (i.e., EQ) both before and after the distortion-producing gain stage(s). The spectrum of the tone going into the gain stage has a big effect on the character or "texture" of the distortion; changes such as turning up the bass will not necessarily produce more bass at the output (indeed, the result could well be more treble and high harmonics), but the nature of the distortion and the playing-feel will be greatly altered. Changing the EQ after the gain stage, on the other hand, will have the expected results on the final sound: turning up the bass will result in more bass being heard, etc..
Different guitar amps have historically placed their tone controls at different points in the signal path (e.g.: Fender: close to the input; Marshall: close to the output), and this is one big reason that different amps sound different from each other. Many guitarists, perhaps unwittingly in some cases, have discovered the importance of being able to EQ the tone in at least two places in the chain; but since most amps don't directly address this, they use EQ pedals or other frequency-selective boost pedals, ahead of either pedal- or amp- produced distortion, and then they use the amp tone controls to shape the final sound. Many of the popular distortion pedals distinguish themselves through the tonality they impart with built-in pre-distortion EQ. For maximum flexibility, my amp will provide two full sets of active EQ controls (bypassable), before and after the gain stages.
The usual tone-stacks used in Fender and Marshall amps provide bass, mid, and treble controls, but due to the design of the passive R-C circuits, it is not actually possible to generate a mid-frequency "hump", just more or less relative proportion of "mid scoop". A mid hump, aka formant, is a fundamental aspect of the "voice" of most musical instruments, including guitar. Most guitar-plus-amp systems exhibit a formant, often chiefly due to the frequency response anomalies of the speaker cabinet. But the only way to change such a formant is to swap cabinets. Twiddling knobs on most amps won't make much difference to the formant. Again, some guitarists employ frequency-selective pedals to produce or alter the formant shape of their systems. EQ pedals are an obvious approach, but also the common trick of using a wah pedal kept at a fixed setting, has been used by many guitarists over the years to produce a dramatic formant in their tone.
Because of the importance of formants, I am electing to use a 4-knob active EQ, in both the "pre" and "post" positions of the signal chain. This EQ provides fixed bass and treble controls, based on the "Baxandall" circuit, which enables both boost and cut of the fixed-frequency "shelves". For midrange, a third boost/cut level knob is provided, along with a frequency knob (initially, I'm looking at a 10:1 range, from 200 Hz to 2 kHz, though this may change with experimentation). There is no control to change the "Q"; i.e., this is a "semi-parametric" mid.
the outlines of a new guitar amp
So, on to the basic parameters of what I'm planning to build. My primary concern is the preamp, because that is where most of the tone-shaping takes place, at least in the way I use guitar amps. So what I'm designing will be amenable to a preamp-only realization (e.g., a rack mounted unit), or to the typical guitar arrangement of preamp and power amps in a box (the "head"), with separate cabinets for the speaker(s). But for my own use, I want a small and portable unit, with limited power -- but unlimited flexibility. So, my remarkably-complicated preamp will be wedded to a very small and simple power amp and speaker setup, in a single case: a "combo" amp.
My ideal amp will be small in size and power, but it will be stereo. I find that stereo is crucial for reverb and certain delay effects. Unlike the wide stereo image which is the final result of a mix in recording or live sound, stereo as applied to a single instrument such as guitar, serves a quite different purpose; as such, it is not usually important or even desirable to have a wide distance between the stereo speakers. The stereo effect creates a doubling or chorus-like ambiance, similar to the double unison strings on harpsichords or the upper strings of a 12-string guitar. The two side-by-side speakers in a "twin combo" guitar amp are ideally suited for guitar-stereo; it's strange to me that there are not many examples of this (if any?) available on the market. (*) But then, that's why we are here! (A number of twin-combo amps would be good candidates to adapt to stereo: especially those with four output tubes, which could be divided into two push-pull pairs, by adding a second output transformer and a few other components. But here, we are designing from scratch.)
I am considering several different combinations of power amps and speakers. For sonic reasons as well as size, I'm interested in speakers smaller than the traditional 12". Probably for the combo amp, I'll try a pair of 8" speakers, similar to what's in the Fender Champ -- but with a closed box. For the separate-head version, a twin-10" stereo-wired cabinet might be the thing.
For the power amps, I am considering three main possibilities. If I end up manufacturing and selling these, I will likely offer all three options, and perhaps others. To wit:
* 2 x 6V6 stereo (single-ended amps, a la Fender Champ).
* 4 x 6V6 stereo (push-pull, a la Fender Princeton and others).
* 4 x EL84 stereo.
In all cases, the power tubes will be cathode-biased, with selectable negative feedback; tube rectifiers will be used, for their voltage "sag" and slow power-up characteristics. (If it wasn't completely obvious, the aim here is anything but "high fidelity".)
Given the number of preamp tubes I am planning to use (13 in the current design), it is hard to find a single power transformer which can supply all the required 6.3vac heater current for preamp and power amps, while also supplying the right high-voltage (many transformers I've seen which have enough heater current, are seemingly made for non-audio applications, and have inconveniently-high HV values, often 500vac or higher); and to employ tube rectifiers, the transformer must also provide one or more separate heater windings for the rectifiers, at 5vac or 6.3vac, since these must "float" at the high DC voltage level of the rectifier output. One obvious solution would be to include a second, beefy filament transformer, to feed all the hungry preamp heaters. However, given the need for two transformers anyway, I've decided to split up the power supplies more completely, so that there can be a separate power switch for the preamp, and one for the two power amps. (The main complication here is that now two (or more) rectifier tubes are needed. In fact, certain configurations might require three: one of the power amp designs I'm looking at (from the Eico HF-81 stereo amp) uses a pair of EZ81s for its quartet of EL84 output tubes, instead of a more regular choice like a single GZ34.)
The system will be set up so that either the preamp or the power amps can be operated independently. This has certain subtle implications. E.g., the low-voltage 5vdc used for the bypass relays, LEDs, and the footswitch, is controlled by the preamp power switch. When only the power amps are turned on, there's no 5vdc. The issue is that the left and right inputs to the power amps are the effects-return jacks, when the preamp is operating; hence, the inputs are only enabled when the stereo effects loop is switched in by its corresponding bypass relay. In order that the effects loop is "in" rather than byassed when preamp power is off, this bypass relay is wired "backwards" relative to the others: it is by default switched "in" when off. Sending power to the relay puts it into the "bypass" state. (Even though the relay is "in" by default, the LED will only light up if the preamp is turned on.)
Instead of the usual "standby" switch, I plan to fit a "mute" switch, which simply grounds the power amp inputs. (Chatter on the Internet has convinced me that standby switches lack benefits and are possibly harmful to tube lifespans. A mute switch addresses the usual use of a standby switch, with fewer downsides. One industry example I'm aware of: Matamp.)
I have found that negative feedback in the power amp section makes a significant difference in the tone, but there is not a single "best" configuration; different amounts of NFB sound good for different playing styles, speaker cabinets, etc.; so, a switch will be provided to turn off NFB entirely, and "presence" and "resonance" controls will be provided, to shape the spectrum of the NFB when it is turned on.
So, to summarize the controls of the power section, or what I'm calling the "master" section:
* preamp on/off.
* power amp on/off.
* mute.
* master volume.
* NFB on/off.
* resonance (LF boost).
* presence (HF boost).
All of these controls are active when the power amp is on, independent of the preamp. Other controls which I'll speak about in upcoming posts, are only active when the preamp is on.
(*) Note: a quick check of the Internet suffices to show how wrong I was: there are actually quite a few stereo, all-tube, guitar amplifiers available today. The first three I happened across were Magnatone, Orange, and Blackstar: all three are making nice, low-wattage, tube stereo amps for guitar, with the right provisions for stereo effects, etc.. So, it should be easy these days to "audition" the sound of stereo effects as applied to guitar, through closely-spaced speakers, if you're not sure whether you want to undertake the extra effort and cost of building a stereo amp. Indeed, one of these existing amps may fully satisfy you, and if you've got the bucks, then I guess you're all set.
But nobody else has my preamp! Hee, hee.
My ideal amp will be small in size and power, but it will be stereo. I find that stereo is crucial for reverb and certain delay effects. Unlike the wide stereo image which is the final result of a mix in recording or live sound, stereo as applied to a single instrument such as guitar, serves a quite different purpose; as such, it is not usually important or even desirable to have a wide distance between the stereo speakers. The stereo effect creates a doubling or chorus-like ambiance, similar to the double unison strings on harpsichords or the upper strings of a 12-string guitar. The two side-by-side speakers in a "twin combo" guitar amp are ideally suited for guitar-stereo; it's strange to me that there are not many examples of this (if any?) available on the market. (*) But then, that's why we are here! (A number of twin-combo amps would be good candidates to adapt to stereo: especially those with four output tubes, which could be divided into two push-pull pairs, by adding a second output transformer and a few other components. But here, we are designing from scratch.)
I am considering several different combinations of power amps and speakers. For sonic reasons as well as size, I'm interested in speakers smaller than the traditional 12". Probably for the combo amp, I'll try a pair of 8" speakers, similar to what's in the Fender Champ -- but with a closed box. For the separate-head version, a twin-10" stereo-wired cabinet might be the thing.
For the power amps, I am considering three main possibilities. If I end up manufacturing and selling these, I will likely offer all three options, and perhaps others. To wit:
* 2 x 6V6 stereo (single-ended amps, a la Fender Champ).
* 4 x 6V6 stereo (push-pull, a la Fender Princeton and others).
* 4 x EL84 stereo.
In all cases, the power tubes will be cathode-biased, with selectable negative feedback; tube rectifiers will be used, for their voltage "sag" and slow power-up characteristics. (If it wasn't completely obvious, the aim here is anything but "high fidelity".)
Given the number of preamp tubes I am planning to use (13 in the current design), it is hard to find a single power transformer which can supply all the required 6.3vac heater current for preamp and power amps, while also supplying the right high-voltage (many transformers I've seen which have enough heater current, are seemingly made for non-audio applications, and have inconveniently-high HV values, often 500vac or higher); and to employ tube rectifiers, the transformer must also provide one or more separate heater windings for the rectifiers, at 5vac or 6.3vac, since these must "float" at the high DC voltage level of the rectifier output. One obvious solution would be to include a second, beefy filament transformer, to feed all the hungry preamp heaters. However, given the need for two transformers anyway, I've decided to split up the power supplies more completely, so that there can be a separate power switch for the preamp, and one for the two power amps. (The main complication here is that now two (or more) rectifier tubes are needed. In fact, certain configurations might require three: one of the power amp designs I'm looking at (from the Eico HF-81 stereo amp) uses a pair of EZ81s for its quartet of EL84 output tubes, instead of a more regular choice like a single GZ34.)
The system will be set up so that either the preamp or the power amps can be operated independently. This has certain subtle implications. E.g., the low-voltage 5vdc used for the bypass relays, LEDs, and the footswitch, is controlled by the preamp power switch. When only the power amps are turned on, there's no 5vdc. The issue is that the left and right inputs to the power amps are the effects-return jacks, when the preamp is operating; hence, the inputs are only enabled when the stereo effects loop is switched in by its corresponding bypass relay. In order that the effects loop is "in" rather than byassed when preamp power is off, this bypass relay is wired "backwards" relative to the others: it is by default switched "in" when off. Sending power to the relay puts it into the "bypass" state. (Even though the relay is "in" by default, the LED will only light up if the preamp is turned on.)
Instead of the usual "standby" switch, I plan to fit a "mute" switch, which simply grounds the power amp inputs. (Chatter on the Internet has convinced me that standby switches lack benefits and are possibly harmful to tube lifespans. A mute switch addresses the usual use of a standby switch, with fewer downsides. One industry example I'm aware of: Matamp.)
I have found that negative feedback in the power amp section makes a significant difference in the tone, but there is not a single "best" configuration; different amounts of NFB sound good for different playing styles, speaker cabinets, etc.; so, a switch will be provided to turn off NFB entirely, and "presence" and "resonance" controls will be provided, to shape the spectrum of the NFB when it is turned on.
So, to summarize the controls of the power section, or what I'm calling the "master" section:
* preamp on/off.
* power amp on/off.
* mute.
* master volume.
* NFB on/off.
* resonance (LF boost).
* presence (HF boost).
All of these controls are active when the power amp is on, independent of the preamp. Other controls which I'll speak about in upcoming posts, are only active when the preamp is on.
(*) Note: a quick check of the Internet suffices to show how wrong I was: there are actually quite a few stereo, all-tube, guitar amplifiers available today. The first three I happened across were Magnatone, Orange, and Blackstar: all three are making nice, low-wattage, tube stereo amps for guitar, with the right provisions for stereo effects, etc.. So, it should be easy these days to "audition" the sound of stereo effects as applied to guitar, through closely-spaced speakers, if you're not sure whether you want to undertake the extra effort and cost of building a stereo amp. Indeed, one of these existing amps may fully satisfy you, and if you've got the bucks, then I guess you're all set.
But nobody else has my preamp! Hee, hee.
designing my ideal guitar amp
If you take a glance across my blog topics on this site, it may become clear that I am building a small collection of unusual musical instruments and equipment. This will all be used to realize my musical vision, through live performance, and particularly through recordings. Guitar has always been my "main" instrument, and electric guitar will figure prominently in the new sonic mix I am seeking to create. Up until recently, however, my designing and building efforts have been aimed elsewhere, particularly towards finishing my Pandalon. Consequently, I've been making do with less-than-ideal guitar sounds, produced with the help of DSP gadgets and whatever old tube amps I have laying around. It has always been my intention to remedy this situation, and at last the time seems to be right.
In the past, I have obtained good results by assembling guitar systems out of various rebuilt and re-purposed vintage tube gear. My most recent system (as of about 2000!) involved a homebrewed 2 x 12AX7 preamp, feeding DSP digital reverb/effects, into a 4 x EL84 hi-fi amp (Eico HF-32) and 12" closed-box speaker. Tone was not entirely unlike the Vox AC-30, although cleaner from the power amp given negative feedback and the closed speaker -- and dirtier overall thanks to my preamp, because I love distorted and overdriven tones.
As I started to sketch out the outlines of a new guitar system, I soon realized that I want to design and build from scratch this time. Given that decision, I realized that whatever I come up with, may be of interest to other builder-musicians. Hence, this blog! Enjoy...
Please note, some of the circuits I will present are my own design entirely (I'm proud to say), but others are either copied verbatim, or derived or adapted, from well-known circuits available publically on the Internet, from companies such as Fender, Marshall, and others. In many cases, these circuits were probably originally patented. The patents may well have expired by now; and in any case, I consider reverse-engineering or copying circuits for personal use, to be "fair use", and the same would apply to any reader of this blog who wishes to replicate my work for themselves. However: if anyone (possibly including me in the future) wants to build amps based on this work, to sell to the public, then I simply warn that a closer look should be taken at possible legal problems with any of the copied circuits. Caveat Venditor!
In the past, I have obtained good results by assembling guitar systems out of various rebuilt and re-purposed vintage tube gear. My most recent system (as of about 2000!) involved a homebrewed 2 x 12AX7 preamp, feeding DSP digital reverb/effects, into a 4 x EL84 hi-fi amp (Eico HF-32) and 12" closed-box speaker. Tone was not entirely unlike the Vox AC-30, although cleaner from the power amp given negative feedback and the closed speaker -- and dirtier overall thanks to my preamp, because I love distorted and overdriven tones.
As I started to sketch out the outlines of a new guitar system, I soon realized that I want to design and build from scratch this time. Given that decision, I realized that whatever I come up with, may be of interest to other builder-musicians. Hence, this blog! Enjoy...
Please note, some of the circuits I will present are my own design entirely (I'm proud to say), but others are either copied verbatim, or derived or adapted, from well-known circuits available publically on the Internet, from companies such as Fender, Marshall, and others. In many cases, these circuits were probably originally patented. The patents may well have expired by now; and in any case, I consider reverse-engineering or copying circuits for personal use, to be "fair use", and the same would apply to any reader of this blog who wishes to replicate my work for themselves. However: if anyone (possibly including me in the future) wants to build amps based on this work, to sell to the public, then I simply warn that a closer look should be taken at possible legal problems with any of the copied circuits. Caveat Venditor!
Subscribe to:
Posts (Atom)