tone control in negative feedback line

[mediatalk]

i try to avoid any tone shaping components in the audio signal path. i like to add a passive filter section into the negative feedback line. to alter the frequency response via phase cancelation seems to be a very musical sounding one. i do it all the time with my recording gear.

i heard this was done before.

i need a filter with a 3db/oct shelf boost at 90hz, 2.3db band-cut at 330hz with a very wide band so that the response curve gets flat again at around 2.4khz. the outcome should be a decrease in the bass response below 90hz and a low-mid boost centered at 330hz.

my amp has a 2 and 4 ohm OT and the NF is connected to the 4 ohm tab. could i avoid the business of an make-up amplifier for the NF signal, because of the attenuation caused by the passive filter i like to add, by adding a second OT connected in reverse to the speaker output (4 ohm and ground of 2nd winding of the existing OT to 4 ohm and ground of 2nd winding of extra OT) and use the 8 ohm tab of the extra OT as the double of the NF signal? of course i will secure the leads of the primary winding of the extra OT.

nick.

[Rob Mercure]

Your tone shaping network in your negative feedback loop effectively turns your entire amplifier into an active filter - just about the most 'non-passive" tone control setup (hmmmm...."agressive" tone controls?). And all components shape tone to some degree.

Rob

[jrfrond]

One of the limitations of this is assuming that you even WANT to use negative feedback. In my own designs and mods, it's a no-no, because I just think the feel is better without it.

Rob was entirely correct about these type of controls being active. In many tone control designs, you will see active tone controls designed into local negative feedback loops of opamps, transistors, etc.

I can understand why you want to keep the signal path clean. I feel the same way in many instances. However, you CAN design local NF controls into the cathodes of the preamp tubes. One way to do a bass control would be to bypass the cathode of a preamp stage with a small value cap (e.g. .22uF) and use a pot to gradually dial-in a 22uF cap in parallel across smaller cap. This is just one idea. There are MANY ways to skin the tone control cat!

[mediatalk]

jrfrond, my amp as a toggle NF switch. 2, 4 ohm and off. in some clean settings i put it into off. most high gain lead settings sound better with the NF on 4 ohm. introducing a out of phase signal to my recordings with its own eq settings has always been very rewarding and is the main reason why i'm interested in this is for my amp. i tried to do it via the fx loop (dry/eq-ed out off phase mix) with sonic success, but the main draw-back is a loss in direct response. to much electronics in the signal path.
can any one confirm the extra OT trick? the passive filter i added into the NF line attenuated the NF to a degree of being almost non-existent. i even thought to quadruple the NF via a 16 ohm tab in the extra OT.
local NF, i agree. i have tried this mod you mentioned on my tape echo sitting in the fx loop with great success.
on my amp i don't change anything in the preamp section because this thing sounds ridiculously amazing. the bypass switch for the entire tone stack is always engaged. the difference in directness and clarity is day and night. i have tweaked and optimized my amp with various tube types and some tube miss-matching to my liking. but in the end i always feel the need of some eq-ing with the settings i mentioned in my first post. eq-ing in front of amp is forbidden in my house. my tone control on my lead guitar is by-passable. i added a delay-line to it, also known as a prs sweet-switch. again, tone shaping in front of the amp happens only via phase cancelation through the pickup settings (string-, bridge-neck-, pickup-pickup distances and humbucker rewiring options etc.) and cable types and lenghts. as you can see i'm pretty obsessed with phase tricks. its for me the most musical sounding eq-ing.
since i heard that "tone control in NF line" is possible and has been realized in commercial products, i'm determined to pursue it. i must hear it and then decide. has anyone some references about this?

nick.

[daz]

You might wanna quickly impliment a one knob tone like the one on some of the 18 watters. When i built my 18 i had that and it works pretty well and probably is less destructive to the tone than a TMB. It would take a few minutes to try it. I know exactly where you're coming from tho.

[HTH]

both presence and resonance controls are tone-shaping devices in the NFB loop - I have both in a 2203 I've got here. Not sure how to introduce any kind of midrange control in the NFB loop though.

[Rob Mercure]

Mediatalk,

I suggest you find a primer on electronic amplifiers - may suggest "Amplifiers" by "Lewis York" (or something else simple and thorough) and learn to differientate between a passive and active electronic component. Again, in a nutshell, an active filter uses some sort of amplifying element to offset gain lost by frequency selective elements - you amplifier is "active" so despite your frequency selective elements being "passive" and overall you have an active filter system - sorta like a crippled parametric equalizer. Take a few evenings off and do a thorough reading of your background material while avoiding the internet, television, and anything having to do with sales of HiFi equipment (most hifi folks woudn't know an "active" rattlesnake if it were chewing on their thumb!). I think the change in your knowledge base would be: ridiculously amazing

[Steve Conner]

Why does everyone hate on negative feedback all the time?! I think it just gets a bad press because of its name. We should re-brand it to inverse feedback, or something.

Anyway, be careful, messing with the feedback loop can make it unstable, which could cook your O.T. from the resulting high-voltage high-frequency oscillations. Whenever I build an amp, I use a scope to check for this kind of thing the first time I power it up, and any time I alter the feedback loop.

I think a make-up amplifier would be more convenient than the size and weight of an extra O.T, if it turns out that you really need more loop gain to get the response you want. The stray capacitance and inductance of that extra O.T. would probably eat into your loop phase margin too, bringing you closer to instability.

A bridged-T network in the NFB line would be one way to make a midrange control. Gibson amps used to use them for a midrange cut. Putting it in the NFB line reverses its action, so you'd get a midrange boost.

And like Rob says stay away from the internet! ;-)

[jrfrond]

Why does everyone hate on negative feedback all the time?! I think it just gets a bad press because of its name. We should re-brand it to inverse feedback, or something.

Negative feedback has it's place. The nomenclature, at least for this old warhorse who understands it, has nothing to do with it. I just happen to like guitar amps with little to no power amp negative feedback. Is it for everyone, and will it work in all designs? No, certainly not. Can you mod the feedback loop to become task-specific? Of course! The good ol' Presence and Resonance controls are fine examples.

I agree with you about monitoring design mods with a scope to check for ultrasonic oscillation. The way I see it, you aren't a REALLY serious builder or tech unless you have a scope. It is an invaluable design and troubleshooting tool.

[mediatalk]

thanks steve, so basically you are confirming that the extra OT trick is doable. the extra OT is only temporarily. if this project bears fruit, i invest in a custom build 2, 4, 8, 16 ohm OT. current OT is 2, 4 ohm only.
gibson realized a midrange boost via the NF loop? thats great news. do you know which model it was.
yeah, the internet. i haven't played for weeks after i stumbled over this site and especially this applet.
http://www.till.com/articles/PickupR...emo/index.html
but it turned out to be very rewarding.

nick.

[Steve Conner]

Sure, the extra OT trick is doable, in the same way that smoking a cigar with your pants soaked in gasoline is doable. It might work, but it's pointless and could go badly wrong.

If you have an extra OT with a higher tap that's suitable for your amp's power level, just swap it out. If your extra OT is smaller, your trick won't work, as it'll saturate before your amp reaches full power.

Gibson used the bridged-T circuit in the forward path, not the NFB loop, to realize a midrange cut. See the schematic of a Gibson Scout GA-17RVT:

http://www.harpamps.com/gibson/pg_0135.jpg