Question about tone stack

[Alan0354]

I have been experimenting with controlling the tone of the amp. I started out trying to just have normal tube stage ( say Fender 100K plate resistor, 1.5K cathode with 20uF bypass cap). I just experimented using simple high pass by small caps to couple to get the treble, or simple lowpass by using a pot in series with a cap to ground, adjusting the pot to attenuate the highs. Just very simple things to see the effect.

It did not take long for me to realize the simple filter like this does not sound too good. They tend to be very mid heavy, muddy with not enough bass and treble. You can't get the clear sound like Fender or Vox from simple high pass or low pass filters. Then I look at the Fender tone stack and some others, seems like the circuit is more taking away mid or low mid frequency rather than simple high pass and low pass circuits. More like a notch type of characteristic where it attenuates the mid band and let more of the bass and treble through. Of cause, they have pots to adjust also, but the general idea is to cut certain amount of mid frequency.

I remember I played with some tweet amps like old Gibsons with very simple tone control, they also sound very muddy until they break up and give a very sweet compressed sound.

That leads me to think maybe the tube stages are not the ones that govern the sound, the sound is mostly from the tone stack. All else are secondary.

Is that true?

[Dave H]

The simple high and low pass filters work best with amplifiers which don’t have power amp negative feedback like Vox, Marshall 18W, 5E3 etc. Not having negative feedback enhances the bass and treble with a loudspeaker load. I particularly like the tone control on the Watkins Dominator/Marshall 18W normal channel. It combines both high pass and low pass on one control and is tuned just right (for me). If the power amp has feedback the standard mid cut tone stack works better.

[Mick Bailey]

I've carried out extensive trials with tone stacks, based around a separate experimental breadboard mounted on top of the amp with three connections to where the original tone stack connected - in/out/ground. Every time I tried something different I simulated it in PSpice to see what effect it would have. It's not an exaggeration when I say I've tested hundreds of combinations and designs. I've also experimented with NFB tone stacks too. My aim was to achieve a design with low insertion loss and reduced interaction between BMT controls, but keep the component count reasonably low. I did get pretty close in the end.

My conclusion is that for a given amp topology the tone stack has the greatest influence on the sound. Important factors are insertion loss and how the mids are handled. Too much insertion loss means reduced drive. Positioning in the circuit also has a large effect. Also, many simple treble controls influence the upper mids too much, so you turn down the treble and the amp begins to sound a little wooly.

For a while I ran an amp with a slightly modified Roy Bean tone stack. Some really good sounds out of that but way too difficult to use live and a lot of insertion loss. Massive spectrum and gave the amp modelling-type of control with you-name-it sounds. The same amp now has a modified DR Z Route 66 control and it sounds nothing like it does with a FMV tonestack. I like it, but the absence of a mid control means It will probably get changed.

What did come out of my trials is that what you hear and what PSpice shows have a lot of separation.

[Dave H]

...What did come out of my trials is that what you hear and what PSpice shows have a lot of separation.

Did you simulate the whole amp with a speaker load in PSpice or just the preamp?

[Chuck H]

That leads me to think maybe the tube stages are not the ones that govern the sound, the sound is mostly from the tone stack. All else are secondary.

That's too much generalization. Every part of an amp makes the tone. The tone stack is a feature panel that allows some manipulation, or not. The "not" part can also be a feature for some amps as Mick noted.

In general, the typical Fender, Marshall, Vox stack does cut mids and work best when an amp is run clean to only slightly clipped. The Marshall values cut the least mids and the most bass. This could be why the Marshall emerged as the choice for hard rock and metal since less bass and less mid scoop tends to sound better when an amp is overdriven. A popular setting for overdriven Marshalls is 10 on all the knobs! But in truth there is usually some attenuation used on the bass and presence controls.

An important note here is that once an amp is clipping hard the tone controls don't do much to change the actual frequency balance because:
1) There is typically enough gain to elevate even attenuated frequencies to where the controls can't reduce it much. And
2) The hard clipped signal is mostly overtones and harmonics that are generated after the tone stack.

In this case the tone stack doesn't so much work as tone controls, but rather "distortion character" controls. A good illustration is that there are many posts here about flabby bass from overdriven Fender BF type amps. These posters are usually instructed to stop trying to get more bass from the amp by turning up the bass knob! When an amp is clipping hard the way to adjust the bass is to turn it up to where the bass starts to sound too loose and then back it down a little. At this point the amp is indeed making all the bass it can without clipping it more than is useful or sounds best.

Some early amps sound VERY good clipped because they DO NOT use the typical mid cutting tone stack. As you noted, they can sound wooly or too bass and low mid heavy. But remember I said every part of the amp makes the tone? Well it's actually every part of the system, including the guitar. Some of these vintage amps were small combo types. Not capable of much bass generally. They also used speakers that were typically rated closer to the amps power rating. Now consider that the guitars popular at the time didn't use hot, high output humbuckers, but more often single coil pickups or humbuckers much lower in bass like PAF's or FilterTrons. So... If you plug a Telecaster (with a hot single coil in the bridge) into one of these vintage amps you end up with the input signal having just about the right frequency balance to set up for good overdrive. Now the amp would be pushed into heavy clipping. And even if there is still some flab on the bass there is also the likelihood that the speaker and enclosure itself will attenuate some of the flabby sounding stuff due to inefficiency. This is the real world scenario of how it would have been when players started to overdrive their amps. This is why trying to overdrive these vintage circuits into the highest possible gain with modern speakers and using modern guitars usually produces unsatisfying results.

As mentioned, tone stack placement also plays into it. Basically, the earlier in the circuit the tonestack is, the more effective it will be at modifying dynamics and the less effective it will be for actual frequency control as the amp begins to clip. This, again could be part of why the BF Fender amps emerged as the choice for cleaner amps while the Marshall amps are preferred for heavier rock.

Some amps are charming because of the restrictions they place on their usefulness as dictated by their design. Because they wouldn't sound like they do for certain things if they weren't designed that way! If you get my meaning. Many more modern Do-All type designs are useful for stage work where a large array of less than perfect tones may be preferable to an amp that can only do one or two good tones, but these amps are often considered more or less equally mediocre for all things in studio recording or for a band that specializes in a narrower genre.

So I'll repeat. The tone is in the complete system and how it's implemented.

Probably the best do all tonestack is a high pass and a low pass filter combined with a mid cut/neutral/boost type control. The mid knob included in the tonestack never has enough adjustment. But even at this point the specific frequency knees will force an amp into charateristics that work best with certain guitars and speakers for certain things!?!

That's as complete a tutorial about tonestacks as I'm up for right now. If after reading it you still don't get it could be because it has a lot to do with how the subject has grown along with changing styles under the control of guitarists

[Alan0354]

Thanks everyone for the educational opinions. I want to specify that I am talking about clean sound. The clipping sound is a totally different animal all together.

[Alan0354]

I've carried out extensive trials with tone stacks, based around a separate experimental breadboard mounted on top of the amp with three connections to where the original tone stack connected - in/out/ground. Every time I tried something different I simulated it in PSpice to see what effect it would have. It's not an exaggeration when I say I've tested hundreds of combinations and designs. I've also experimented with NFB tone stacks too. My aim was to achieve a design with low insertion loss and reduced interaction between BMT controls, but keep the component count reasonably low. I did get pretty close in the end.

My conclusion is that for a given amp topology the tone stack has the greatest influence on the sound. Important factors are insertion loss and how the mids are handled. Too much insertion loss means reduced drive. Positioning in the circuit also has a large effect. Also, many simple treble controls influence the upper mids too much, so you turn down the treble and the amp begins to sound a little wooly.

For a while I ran an amp with a slightly modified Roy Bean tone stack. Some really good sounds out of that but way too difficult to use live and a lot of insertion loss. Massive spectrum and gave the amp modelling-type of control with you-name-it sounds. The same amp now has a modified DR Z Route 66 control and it sounds nothing like it does with a FMV tonestack. I like it, but the absence of a mid control means It will probably get changed.

What did come out of my trials is that what you hear and what PSpice shows have a lot of separation.

That is exactly my motivation of playing with the tone stack. I feel the Fender type of tone stack has too much insertion loss. I am trying to use as fewer stages as possible and the tone stack loss is in the way. Is BMT bass middle and treble?

What is "amp topology"? Do you mean push pull, class AB, single end, NFB, no NFB?

[big_teee]

You can change the darkness and brightness in tone by changing the bypass cap size.
If too dark you can change the Cap size between the PI and Power Tubes, also.
The resistor and cap before the treble pot is also a place to experiment.
20uf bypass caps are larger than I like, somewhere between the .68uf marshall, and the 20uf fender.
Also on tone stack insertion loss, the CF was used earlier by fender and marshall, to place the tone stack after the CF.
The CF doesn't load down the previous stage.
http://www.valvewizard.co.uk/accf.html
T

[eschertron]

That is exactly my motivation of playing with the tone stack. I feel the Fender type of tone stack has too much insertion loss. I am trying to use as fewer stages as possible and the tone stack loss is in the way. Is BMT bass middle and treble?

What is "amp topology"? Do you mean push pull, class AB, single end, NFB, no NFB?

If you want only two preamp stages before the power section (like a champ, for instance) then you're absolutely right. You'll never worry about compression or wave shaping with the input stage, just gain. The second stage will be the only one to see a high enough input level to allow compression/clipping. I've used the "tone stack lift" mod to get more grind out of a champ. Basically bypassing the tone stack entirely, which I gather is not really what you want.

Just my observation, but preamp design can benefit from the *right amount* of insertion loss between stages. It's not all about maximizing gain or clipping. This is assuming you'll have more than one preamp tube. The stages, as they cascade, can add a little more 2HD to the signal if they are set up so the interstage loss complements the tube's gain figures. This involves calculating the input signal (vpp) desired at each tube's grid and then coming up with a tonestack (or other interstage loss) design that produces that voltage.

In a nutshell, topology relates to that. There's a kind of standard "Fender low-gain" topology that is typified by the champ. Placement of vol control, tonestack, all in relationship to the two preamp stages. A lot of other amp makers lifted that design and used it. The Marshall (3 gain stages and a CF) preamp (yes, lifted from the Bassman, and modified) is another topology that has been copied and standardized. Because it works. The things in between the stages are as important as the gain stages themselves... so the sound comes as much from the tonestack (and other interstage circuitry) as from the gain stages themselves. I agree with you.

[Mick Bailey]

That is exactly my motivation of playing with the tone stack. I feel the Fender type of tone stack has too much insertion loss. I am trying to use as fewer stages as possible and the tone stack loss is in the way. Is BMT bass middle and treble?

What is "amp topology"? Do you mean push pull, class AB, single end, NFB, no NFB?

Yes, BMT - Bass Middle Treble.

By topology I mean all of the components, nodes and connections - including the operating class - that make up the amp. I said that to eliminate changes to additional factors such as coupling caps, parallel input stages, bypass caps etc. All of these will affect the sound but practical trials and over a year of experiment have shown me that the greatest tonal change to an amp can be achieved through manipulating the tone stack. I'm not saying that the tone is generated by the stack, but what potential is there within a given amp can be unleashed through careful design.

[Mick Bailey]

Did you simulate the whole amp with a speaker load in PSpice or just the preamp?

The entire amp. The trouble is that the Pspice models for many components are imperfect or too simplified. You don't get the speaker characteristics that you get with a live speaker in a real amp, and the distortion characteristics of the sum of the components.

Edit: Just remembered I simulated the entire amp in the full Spice version - I had too many nodes for PSpice.

[Alan0354]

If you want only two preamp stages before the power section (like a champ, for instance) then you're absolutely right. You'll never worry about compression or wave shaping with the input stage, just gain. The second stage will be the only one to see a high enough input level to allow compression/clipping. I've used the "tone stack lift" mod to get more grind out of a champ. Basically bypassing the tone stack entirely, which I gather is not really what you want.

Just my observation, but preamp design can benefit from the *right amount* of insertion loss between stages. It's not all about maximizing gain or clipping. This is assuming you'll have more than one preamp tube. The stages, as they cascade, can add a little more 2HD to the signal if they are set up so the interstage loss complements the tube's gain figures. This involves calculating the input signal (vpp) desired at each tube's grid and then coming up with a tonestack (or other interstage loss) design that produces that voltage.

In a nutshell, topology relates to that. There's a kind of standard "Fender low-gain" topology that is typified by the champ. Placement of vol control, tonestack, all in relationship to the two preamp stages. A lot of other amp makers lifted that design and used it. The Marshall (3 gain stages and a CF) preamp (yes, lifted from the Bassman, and modified) is another topology that has been copied and standardized. Because it works. The things in between the stages are as important as the gain stages themselves... so the sound comes as much from the tonestack (and other interstage circuitry) as from the gain stages themselves. I agree with you.

The reason I want a low loss tone stack is because I want fewer stages. It is my believe that the move stages, the more electronics you add in the signal path, you loss tone. That's the reason I try very hard to have a low insertion loss tone stack to keep the number of stages down. Like the old Gibson or Tweeds, they mostly have two stages before the power section. But at the same time, their tone stack is mostly simple high pass or low pass type that the clean sound tends to be muffled with not enough highs or big bass like the Fender BF and SF. But they do break up nicely, just the clean sound cannot compare to Fender or Vox.

[Alan0354]

For a while I ran an amp with a slightly modified Roy Bean tone stack. Some really good sounds out of that but way too difficult to use live and a lot of insertion loss. Massive spectrum and gave the amp modelling-type of control with you-name-it sounds. The same amp now has a modified DR Z Route 66 control and it sounds nothing like it does with a FMV tonestack. I like it, but the absence of a mid control means It will probably get changed.

What did come out of my trials is that what you hear and what PSpice shows have a lot of separation.

Do you have the Dr. Z Route 66 schematic?

[Mick Bailey]

Here it is;

http://i72.photobucket.com/albums/i1...6CORRECTED.gif

[Alan0354]

Here it is;

http://i72.photobucket.com/albums/i1...6CORRECTED.gif

Thanks, I really appreciate it.