I was watching a demo of the Matchless Chieftain on YouTube and I really liked the sound. I can't quite describe it, but it just sounds powerful. I pulled up the schematic for it and I noticed it's unusual in a few ways.
The first is that the preamp and reverb driver are 12AX7s with both sides of the tube wired in parallel (share plate resistors, cathode resistors and bypass caps, when used). Anyone know why that is?
Another thing that's unusual is that the tone stack is split between the first and second stage. The bass and mid tone controls follow the first stage and the treble (and volume) follow the second stage.
I'm just curious about it from a design perspective. Can anyone kind of give me an overview of the whys of the paralleled tubes and the split tone stack?
The first is that the preamp and reverb driver are 12AX7s with both sides of the tube wired in parallel (share plate resistors, cathode resistors and bypass caps, when used). Anyone know why that is?
Another thing that's unusual is that the tone stack is split between the first and second stage. The bass and mid tone controls follow the first stage and the treble (and volume) follow the second stage.
I'm just curious about it from a design perspective. Can anyone kind of give me an overview of the whys of the paralleled tubes and the split tone stack?
Last edited:
The reason the reverb driver is paralleled is to give a higher current output. This provides a stronger signal current to shake the springs of the reverb tank harder, generating more effect. That 12AX7 is really used as a low output power amp and the transformer it drives is equivalent to an output tranny.
As for the split tone stack, I suppose Mark Sampson was going for a different sound than the typical TMB or Baxandall stack. Different responses tonally.
As for the split tone stack, I suppose Mark Sampson was going for a different sound than the typical TMB or Baxandall stack. Different responses tonally.
Splitting sections of the tone stack either side of a stage prevents interaction between them. In the usual TMB stack there is a lot of interaction between the three controls. Placing the treble control after a stage would prevent it interacting with the bass and mid controls.
Thanks guys. That makes a lot of sense. That's an interesting approach to the tone stack interaction issue.
The parallel wiring of 1st stage is a bit like jumpering 2 input on a Marshall. IMO you will see lowered 1st stage output impedance, will get combined 1st stage harmonics. Driving the tone stack like that is clever, and as discussed separated the treble.
It's also interesting that they chose to not reduce the cathode resistor at the input. Normally when a tube is paralleled, the current doubles, so to keep the bias the same, you would cut the resistor in half, 820 ohms being the closest standard value. They also didn't bypass the cathode on this stage, lowering gain and distortion but adding headroom because of cathode degeneration. They make up for this by adding bypass cap on the next stage, increasing gain here.
Also, the signal to noise ratio will theoretically be cut by 3dB, making the input quieter when paralleling the triodes. This helps to keep the overall amp noise quieter because the first stage noise is lower.
Also, the signal to noise ratio will theoretically be cut by 3dB, making the input quieter when paralleling the triodes. This helps to keep the overall amp noise quieter because the first stage noise is lower.
I just posted this answer to a very similar question in MEF:
Most Guitar amps are based on Fender, including Marshall and Vox; this Matchless looks unusual because it's based on a Gibson, in fact it's basically a knockoff of one of them whose model name today I don't remember by heart.
Rather than going through 200 schematics to find it, here's one which illustrates the basic Gibson ideas:
Some typical Gibson traits:
1) a midrange killing T net .
In this Gibson it's fixed; in the model I mention it has a series pot between it and ground, to make it adjustable.
2) a Bass cut only control, in both a very small cap in series with a resistor to ground (which may be a volume pot) bypassed by a pot which controls how much it cuts.
3) this one is easy: a treble cut only cap to ground, with a series pot to control how much.
Then, how can such crude excuses for a "tone control" exist in such a good sounding amp?
Because Matchless starts on the right foot: a guitar tube amp is a *speaker driver* which sets the main sound; if properly chosen it has lots of punchy bass and biting treble even run flat, so much so that cutting the excess is fine.
FWIW Matchless always used killer speaker combinations.
The amp head by itself , plugged into a generic run of the mill speaker, sounds bland, lots of amp home makers here felt disappointed and asked me about it (there was a rash of cloners after a couple local guitar gods started using them) .
My answer was: "that's the way it sounds, plug it at least into a Marshall 4x12" and you'll see"
Here's a free sample: Strat into a Matchless combo:
https://www.youtube.com/watch?v=7enztRmXJOA
Similar to early AC30 (pre Top Boost) versions: tone control was only a crude Cut one, plus it had massive fixed Bass cut , but amp was incredibly clear and jangly, and "the loudest 30W ever seen" because it drove killer Celestion Blue speakers.
Matchless combined both design ideas into their own.
Most Guitar amps are based on Fender, including Marshall and Vox; this Matchless looks unusual because it's based on a Gibson, in fact it's basically a knockoff of one of them whose model name today I don't remember by heart.
Rather than going through 200 schematics to find it, here's one which illustrates the basic Gibson ideas:
An externally hosted image should be here but it was not working when we last tested it.
Some typical Gibson traits:
1) a midrange killing T net .
In this Gibson it's fixed; in the model I mention it has a series pot between it and ground, to make it adjustable.
2) a Bass cut only control, in both a very small cap in series with a resistor to ground (which may be a volume pot) bypassed by a pot which controls how much it cuts.
3) this one is easy: a treble cut only cap to ground, with a series pot to control how much.
Then, how can such crude excuses for a "tone control" exist in such a good sounding amp?
Because Matchless starts on the right foot: a guitar tube amp is a *speaker driver* which sets the main sound; if properly chosen it has lots of punchy bass and biting treble even run flat, so much so that cutting the excess is fine.
FWIW Matchless always used killer speaker combinations.
The amp head by itself , plugged into a generic run of the mill speaker, sounds bland, lots of amp home makers here felt disappointed and asked me about it (there was a rash of cloners after a couple local guitar gods started using them) .
My answer was: "that's the way it sounds, plug it at least into a Marshall 4x12" and you'll see"
Here's a free sample: Strat into a Matchless combo:
https://www.youtube.com/watch?v=7enztRmXJOA
Similar to early AC30 (pre Top Boost) versions: tone control was only a crude Cut one, plus it had massive fixed Bass cut , but amp was incredibly clear and jangly, and "the loudest 30W ever seen" because it drove killer Celestion Blue speakers.
Matchless combined both design ideas into their own.
That T net you're talking about, is that the RC network just before the 2nd stage with the 2-220k and a 470k resistors, 330 and 470pF caps and .0068 to gnd? I was going to buy one of these once but didn't understand what the heck all that was for. If I did get it, I was going to experiment and put a jumper around the whole thing to see what it did. If not, do you know what this network is for?
Also interesting is the labeling of it as "isolation preamp." Looks to me as a normal gain stage where all signal of the channel will pass through it. Just what does it isolate?
Also interesting is the labeling of it as "isolation preamp." Looks to me as a normal gain stage where all signal of the channel will pass through it. Just what does it isolate?
The previous stages / channels.
Think of it this way: Let's first assume a scenario where there aren't those two "isolation" gain stages but a simple resistive mixing where each of those T-filter networks, the reverb output etc. would simply terminate to a potentiometer (or resistor) of suitable value. Now all those branches would effectively look parallel to each other and that "mixing" resistor. The signal would pass through one T notch filter per channel, then terminate to an impedance consisting of another t-notch filter (in "reverse"), volume controls and such, and all other "mixing" resistors in parallel.
Well... figuring out response of such a mess gets way too unpredictable ...especially if any of those impedances happens to be variable as well.
Now, with those "isolation" gain stages there each of those branches terminates to high input impedance of the grid and no signal "bleeding" to other channels takes place because the gain stage isolates all of them from each other.
Now with "fixed" loading and channel circuits "isolated" the response of each channel circuit is much more predictable.
Last edited:
Some of it, yes. At least minimises it by great deal vs. typical "FMV" tonestack.
I would assume the "bass" control circuit - loaded by a 220K resistor in parallel with varying impedance of the "mid" control circuit - and "mid" control circuit - driven by source impedance largely defined by varying impedance of the "bass" control" - will effectively have interaction with each other. For obvious reasons.
The treble circuit, on the other hand, is buffered by that gain stage in between so it will probably just interact with the volume control.
However, what this arrangement does as well is that it effectively places midrange and bass equalisation before distortion produced by gain stage made out of V2 tube: Now the user can "thin" out lower frequencies before clipping and effectively minimise nasty intermodulation distortion of lower frequencies. It would sound pretty much like farting and ruin all note separation otherwise.
Similarly treble equalisation takes place post V2 distortion: Now all those high order harmonics created by distortion can be conveniently attenuated at the higher frequencies. Those are the frequencies that would sound like obnoxious "fizz".
Pre and post distortion EQs will introduce identical frequency response but they will have major impact on harmonics created by distortion in concerned gain stages. In essence pre and post distortion EQs will sound and react differently from one another. If you look at it; bass reduction pre distortion and treble reduction post distortion is a voicing trick you'll find from pretty much every guitar amp that actually sounds good, or even decvent, overdriven.
Many Crate and Fender amps, for example, actually divide the tone controls to more effective places like this: Bass controls before distorting stages and treble controls after. Location of mid control is usually more a "preference" issue. Certain Mesa/Boogie amps are accused "hard to dial in" because ALL tone controls are located before distorting stages and this weakens their effectiveness: Treble control won't affect harmonics created by distortion (and distortion may easily re-create all those frequencies the control dialled out). Similarly bass control won't affect the final tone and excessive bass boost settings merely turn overdrive into heinous farting.
Last edited:
They are "notch" filters; IOW a "band reject" filter for specific frequency range. Q value and frequency pretty much depend on overall circuit and component values.
Gibson uses these filters for both "mid-range notch" and for another notch that takes place below effective bandwidth of the guitar. (e.g. at 50 Hz) At that point the filter practically turns to hi-pass -type because frequencies below the notch f are not produced by the amp to begin with.
Personally I like "Post" everything EQ, usual say Cathode follower/direct coupled- gobs of distortion just because of it's design...there can be plenty of tailoring/Voicing Post EQ. and more interesting harmonics.
Vs say like some Mesa's or non-cathode Follower amps...Pre EQing/Tone controls are not as effective and do come off as being hard to dial in....with CF design it's easy to assume the Fizz/Fuzz will be hardest to fight..but again can very well be tamed in with Post EQing/Snubbers/bypass etc/General RC voicing to shift Responses to more desired sound etc.
Non-Cathode Follower/Pre Eqing to me is not as Fine of control and actual range/useable amount of Control imo...and again very noticeable with treble a lot can be rolled off/tamed but at times it's like too much/taking too much away. it's like trying to balance a bit too much..and further comes off as Dull/not quite as intresting Harmonics/overtones/Voicings imo.
Last edited:
- Status
- Not open for further replies.