So, NFB Is Good, But......?

[martin manning]

Ever notice that anybody who is really great at anything has mastered both the theory and the practice, aka the art of it? Those people are exceptional, but that's reason enough to pursue both if you want to improve.

[The Ballzz]

FWIW,
It's true that my intention is to "ABSOLUTELY" climb up and stand on the shoulders of giants! Most, if not all, of the work has already been done to create the best circuits for the intended tones/sounds/responses of my planned project. I see little benefit in trying to reinvent wheels that already roll quite nicely.

My project surrounds using tried and proven circuitry in a new & unique configuration and package to facilitate convenient/proper use of nearly any type of effects commonly used by guitarists, in a properly impedance & level matched manner. The concept involves much more than simply designing a highly flexible effects loop, although some aspects of many different, existing effects loop designs will certainly be a big part of the picture.

The main requirements of the "tweaks" to existing loop designs surround the ability to switch between serial & parallel operation and the flexibility to properly match levels and impedances.

The first phase of this endeavor is to finalize/build a nicely working amplifier with two completely separate preamps:

A) A careful clone of Marshall JMP-2203 preamp

&

B) A careful clone of 1965 Super Reverb (reverb/tremolo channel) preamp, without the reverb or tremolo
C) Both of these preamps configured to be switchable and/or blendable into a 15-20-ish watt power section. I'm still torn as to whether to use EL84 or 6V6 for this. I would actually prefer to use 6AQ5, but due to their lower voltage requirements, it may complicate getting enough voltage to the the preamp. Will likely be cathode biased.

Once this is accomplished, the rest of the project will involve mostly switching, effects loop and routing configurations.

The intended name is the "EFFECTS HOG" as it will greedily gobble up any effect pedal, box or rack mount unit in the most optimal manner. Many of the switching/routing functions & options will not be intended or designed to be switchable on the fly. The user will need to have a pretty good idea/understanding of what they want/need before choosing their configuration, although it will be laid out so that in standard, right out of the box configuration, simply plugging a guitar straight into it will provide an amazingly flexible and great sounding amp!

It will also include a flexible power supply to facilitate "phantom" powering of outboard foot pedals/switches/etc. to avoid running ac power to a pedal board.

It is my considered observation that while many great strides have been made in effects loops in recent years, it is still a somewhat jerry rigged mish/mosh of solutions.

Still Dreamin'
Gene

[R.G.]

FWIW,
It's true that my intention is to "ABSOLUTELY" climb up and stand on the shoulders of giants! Most, if not all, of the work has already been done to create the best circuits for the intended tones/sounds/responses of my planned project. I see little benefit in trying to reinvent wheels that already roll quite nicely.

Good on ya! Go for it!! Learn as you go!

[The Ballzz]

Please forgive me for resurrecting my own "NECRO-THREAD" but there are so many great observations, that I feel it not prudent to go through them again. I'll endeavor to give a good example of why/where/etc, I'm asking these questions. I've got another successful build under my belt and have continued studying and at least HAVE Merlin's second edition preamp book that I have not yet had the chance to read! Here's the amplifier example:

A) Nearly bone stock JMP 2203 preamp.
B) Dual, 6AQ5, cathode biased, PP power section.

I have laid this out with several different, standard "Presence" configurations to get the most unbelievable "CLEAR" and articulate sounds, even while driven into distortion and mayhem. The problem is, that while this fantastic sound/tone is achievable as a single guitar, by itself, when placed in a band situation, it becomes almost inaudible until it is at almost ear bleeding volume levels. Now, when I remove the "Presence" loop and simply ground the LTP, it becomes supremely "hearable" at reasonable volume levels, but simply lacks the clarity and "sparkle" that it had with the "Presence", even though the sound is still pretty good. Quite perplexing!

I have a scope that I've not learned enough about yet and need a proper frequency generator, but.......... I'm a little bit scared to allow myself into the realm of stacked pizza boxes/beer cans/packs, while driving my wife out the door to her Mom's with weird noises from my shop, for weeks on end! I'm like a dog with a bone and once I get hold of it, I won't let go until it's been properly chewed! If ya know what I mean?

Oh, and FWIW, my recent, successful Tweed 5E3 Deluxe build has been a real eye opener!


TIA,
Gene[/i]

[tubeswell]

A significant limiting factor when considering global NFB is the way your OT is wound. The more interleavings your OT windings have, the greater the phase shift will be at HF. Too little NFB and there's no effect. Too much NFB and you'll get HF oscillation. Best way to find out - if you don't understand open-loop versus closed-loop gain and phase-shift (or how your OT is wound) - is to experiment.

[pdf64]

A significant limiting factor when considering global NFB is the way your OT is wound. The more interleavings your OT windings have, the greater the phase shift will be at HF. Too little NFB and there's no effect. Too much NFB and you'll get HF oscillation. Best way to find out - if you don't understand open-loop versus closed-loop gain and phase-shift (or how your OT is wound) - is to experiment.

My understanding is that a well interleaved OT will have a more linear phase/frequency response than one with poor / no interleaving.
And so an amp with a well interleaved OT should have a higher margin of stability than one with a poorly interleaved OT, and so maintain stability with a level of global NFB than might otherwise be the case, all else being equal.

[strelok]

@OP Norman Crowhursts 3rd book has a nice coverage of Negative feedback. Its starts out fairly basic and works its way into Nyquist stability and associated diagrams. It's presented in a nice plain language format, thats easy to understand. At least in comparison to most engineering texts on the subject.

http://www.tubebooks.org/Books/crowhurst_basic_3.pdf

[The Ballzz]

@OP Norman Crowhursts 3rd book has a nice coverage of Negative feedback. Its starts out fairly basic and works its way into Nyquist stability and associated diagrams. It's presented in a nice plain language format, thats easy to understand. At least in comparison to most engineering texts on the subject.

http://www.tubebooks.org/Books/crowhurst_basic_3.pdf

Holy Crowhurst!
Thank you so much for that link! Seems to be just where I was hoping to start. It looks like good primer to help me understand Merlin's more detailed stuff!

Thanks Again,
Gene

[V2]

Gene -

1. Home amp settings often don't cut through the mix in a band setting. Have you tried cranking the mids in a live setting to cut through? Maybe you need to alter your tone stack values too (e.g., replace 25k mid pot with 50k).

2. I think the Reason Bambino uses the same cathode-biased 6AQ5 power section. Perhaps get hold of one of those for testing...

3. Have you looked at Kevin O'Connor's FX Loop chapter in the first volume of his TUT series of books? Maybe something in there is useful.

[SilverFox]

The mention of process control and open-closed loop process' brought back fond memories for me of the early 90's and comparisons to the mid 80's just prior to the "Revolution". I worked in the field for a few years and had some opportunities to build a several systems, (Wonder Ware). I don't claim to have the knowledge you seem to hold pdf but it was one of those fields that came naturally to me and fit well with my computer and electronics experience. I'm mentioning all this just prior to getting to my on-topic point. Anyway, along came the computer and Windows and what happened to the print industry with the introduction of "Desk Top Publishing", happened to the process engineering industry with the introduction of, "Drag and Drop" object oriented database programming and every PC nerd was a process engineer. Now the secretary can be the engineer. Ah well progress.

Regarding the FB loop: I would think that, due to the frequency limitations of guitar the complexity associated with the presence control and FB loop is limited. I don't believe there is much room for improvement but hey, "School Them Dice". I'm currently working on refining the tone just prior to the phase inverter via Active Filters. I may indeed open a topic on OP Amp filter design as I'm one of the "Unschooled" in math and could definitely use some assistance if this isn't very basic. Speaking of process control- What about a "Feed Forward" approach to presence? Or is that what Active Filtering is?

Thanks for the memories,

silverfox.

[Structo]

Those Crowhurst text books are great.

There is so much reference material on tubes it's overwhelming.

But, it's nice when a contemporary author (Merlin Belcowe) explains it in
a way that is better understood than some of the vintage authors.

We are all at different stages in our education of electronics.

Everybody learns in a different way, be it by reading, by doing, by seeing, by hearing, how to do it, etc.

Find the best way that you personally can learn the information.

[strelok]

Yeah I really like the Crowhurst books. However it is really hard to beat Merlin's website. Especially when the focus is more tuned to guitar amps. I really need to get his book. I think I've read everything on his website at least half a dozen times at this point.

[tubeswell]

A significant limiting factor when considering global NFB is the way your OT is wound. The more interleavings your OT windings have, the greater the phase shift will be at HF. Too little NFB and there's no effect. Too much NFB and you'll get HF oscillation. Best way to find out - if you don't understand open-loop versus closed-loop gain and phase-shift (or how your OT is wound) - is to experiment.

My understanding is that a well interleaved OT will have a more linear phase/frequency response than one with poor / no interleaving.
And so an amp with a well interleaved OT should have a higher margin of stability than one with a poorly interleaved OT, and so maintain stability with a level of global NFB than might otherwise be the case, all else being equal.

That's the aim of interleaving. Perhaps a better way to state this is that there is probably an optimum amount of interleaving? It seems to me that more interleaving + a bigger OT winding = more LC reactance, hence my thinking about phase shift between primary and secondary windings. I stand to be corrected (although I have found HF instability in some of my amps with interleaved OTs, but it could be caused by other factors.)

[pdf64]

there is probably an optimum amount of interleaving? It seems to me that more interleaving + a bigger OT winding = more LC reactance, hence my thinking about phase shift between primary and secondary windings

Yes, I imagine that past a certain degree of interleaving, the drawbacks may outweigh the benefits.
However, I think that the interleaving would be a quite a high degree before parasitic LC overcame the benefit of improved coupling.
This info sheet indicates that increased sectioning (assume this means interleaving) leakage inductance reduces http://www.lundahl.se/wp-content/upload ... tp_typ.pdf