so how do you decide...?

[rathmann]

For folks who build "original" designs, meaning anything not striving to be a specific clone, how do you make you topological decisions? Specifically I'm looking for things like NFB or no NFB, cathode/fixed bias, LTP or cathodyne phase inverter.

I know the technical distinctions here. I just want to know if folks here labor over the decisions, or if you've just decided what your favorites are and roll with em until the end of time.

I'm so non-commital that I even built a project a few years back that allowed remote switching between all of these things. I still can't decide what to do for my next project...

I have some transformers from a mid 90s Matchless and am thinking of doing something AC30-ish but with 6BM8s and a brownface trem using the triodes of the 6BM8s as the business part of the oscillator.

So what is it? Coin tosses? Vintage circuit loyalism? parts count? How do you decide?

[DIY_Guy]

With my amps, I add in the things I like, and remove the things I have no use for.

When it comes to specific preamp parts, I simply breadboard ideas until it sounds the way I like it! I don't think anything I've built is similar to another amp out there. Well except for my SLO I cloned, but even that was changed from the original design.

[kg]

you just have to try it out, and run with what you think its best.

it's all subjective anyway. for example, if it makes you FEEL better to run an LTP splitter, then (in my experience) it'll make you PLAY better.

[GregS]

So what is it? Coin tosses? Vintage circuit loyalism? parts count? How do you decide?

For me the decision is based on a few things and also depends on what the intended use will be. For an amp that I will use in the band I play with, the design/features have to meet those requirements. If it's something to play around with by myself at home then there are more options.

Also, if I've already done a particular type of circuit (e.g. cathode biased, no NFB) then I like to try some other type of design for variety and to increase my knowledge.

Greg

[J M Fahey]

"Loyalism", "I always did it the same way", etc., are not "real" design in my book.
Can be valid shortcuts, though.
I think the logical step is to first decide what I want as a final product, such as "tight/loose easily_distortable/clean_punchy 15W amp .... " etc., then have a rough idea of what I have available (do I *have* to use a given set of iron / certain tubes / a certain chassis / whatever ?).
That constrains the horizon somewhat.
Then work backwards: needed voltage/current swing --> search related datasheets --> this tells me the voltage needed to drive them --> I choose a LTP for highest voltage available or cathodine for an easier to saturate one --> etc.
You can also decide between using NFB (tighter, cleaner amp) or not (looser creamier amp).
This way you are really designing on a blank sheet of paper with your own specs.
Just my 2 cents; others may use different yet equally valid methods.

[Wilder Amplification]

Basically, if something's been done before, I don't do it and make sure I do the opposite.

[overtone]

That is a great question. Especially as it is not limited to amps.

I write out my goal, how ever mad,
so that I can review it,
measure the faltering progress,
and to know when to call it done.

I try to go for goal quality over just using up parts that are handy.
Each part has its day, you decide when, according to the goal.

Usually that first idea is strongest. Always refer back to it.

Best,
tony

[Bruce / Mission Amps]

Original? Ha ha ha...
The honest truth is that most modern guitar amps are derivatives of every vacuum tube amp that has come before... ALL of us are standing on the shoulders of the amp giants from yesteryear, 40's, 50's 60's 70's ... and if an amp builder doesn't agree with that... then I'm sorry, but I think that one is simply delusional.
Newer builders, with limited resources and experience, should start with an easily clone-able circuit and experiment with values and gain stages to see why others have gone a specific route.
There isn't really an easy path here and it will take some time for you to build, play, tweak, tear apart, rebuild, tweak again... and ultimately, decide for yourself what is best for you. However that will be "today", because as you progress, you'll find other circuits and other's ideas will effect you differently and you need to hear them first hand before you can really decide whether or not you want some NFB, none, a LTP PI, a paraphase driver, or a cathodyne driver... etc etc.
Regardless, use a large enough chassis so you don't get frustrated with layout constrictions. Use decent quality parts, especially with respect to the transformers and start with a basic amp design that you already think you like.

[defaced]

For folks who build "original" designs, meaning anything not striving to be a specific clone, how do you make you topological decisions? Specifically I'm looking for things like NFB or no NFB, cathode/fixed bias, LTP or cathodyne phase inverter.

Simple:
Draft design
Prototype
Break it and make it what you want
Final design
Final layout
Build
Turn it up
Enjoy

For my design goals on my most recent amp, the answers to those questions were: NFB, fixed bias, cathodyne. Some was derived experimentally, others just based on common knowledge that some setups sound a certain way.

Everything I build gets prototyped and messed with until I like it, then I build the "real" version. This way I can verify the circuit and the layout. For the most part, I work from existing designs or circuit snippets I've found and tweak/finalize as needed. But for new stuff, like the SS FX loop I'm working on now, It's trial and error until I get what I want. IMO, designing really starts when you wire something up and try it, the schematic paper step before wiring is just to get your thoughts together.

[MWJB]

What Bruce says...a preamp is a bunch of triodes strung together, a grid, a cathode a plate - pretty well anything you try here has probably been done before. A cathodyne/LTPI/paraphase can easily all be made to fit in the same space, just leave a couple of spare eyelets...try a presence? Start with an SPST switching in/out a cap to see if one or the other works, or you feel something in between. NFB, another SPST in/out? Basically when developing you want to be able to switch between whatever modes you want to compare...also be critical...listen for any characteristics, or traits that you don't like, then aim to eliminate them...just for a noise to come out at the end of the day is NOT the mark of success.

If you need to get into blind/double blind tests then there probably isn't enough difference between the modes you are trying to test.

ATo reiterate what Bruce says again...leave your self room to move, with a chassis that will accomodate rearranging things whilst you pay attention to good layout (you need to be sure that the circuit works or not, not that you have bodged it into a space it can't handle causing avoidable oscillations for instance). Before you pick up the first component...draw your layout, keep analysing it for a few days looking for improvements, I typically end up redrawing things half a dozen times before turn on the iron.

Even if you do build vintage clones, most amps were revised during their production run...so to build the one that sounds best to your ears still requires critical listening tests. Most in demand reproduction builders (I HATE the term clone) have analysed multiple models of the "same" amp before finalising their design.

I literally try any tube that will fit in any socket, within reasonable operating perameters, because you never really "know" what the whole shebang will sound like until you actually try it. It takes patience, don't be in a rush...even a 3 tube SE amp might take a couple of months of permutations. If you aim to start with the cheapest 12AX7s & power tubes (not that cheap is always bad, far from it...a tube only makes a sound in a circuit, peculiarities of that circuit can sometimes buck "conventional wisdom") that will meet your features & power output, then you probably already work for a large scale manufacturer! ;-)

Jon wrote: "Basically, if something's been done before, I don't do it and make sure I do the opposite." Pah! You actually have to DO something?!...One day, Grasshopper, you will just have to visualise the circuit to know if it sounds good & all your designs will appear in your dreams! ;-)

[Amp Kat]

Good stuff fellows and I have to lol sometimes when I see some builders that come in here and say their amps are totally original and nothing like them. Maybe now theres nothing like them but there was at one time because every thing that can be accomplished with tubes has already been done. It may not be on the market but it's somewhere out there and in written notation but usually in complex Engineering formular lingo that most amp builders don't understand and for what they are doing don't need too.

Also agree with above that it really depends on what you plan to do with the amp as to how you want to design it like :
Using it for clean tones only
clean tones and overdrive mixed
distortion only
combination of all above
power tube distortion
effects in the front or through the loop

One thing that many amps do is use the 1st tube for both channels and this puts you in a bad situation because you are confined to that frequency and changing the components to make it either cleaner are dirtier can effect everything down stream. So if you want a two channel amp you may opt for each channel to use independant 1st gain stages as it is so critical in the overall tone setting. It's actually quite endless the opportunities especially in the preamp section but most say keep it simple so have a good time with it.

[tubeswell]

If I knew how to decide exactly how I would design the perfect amp in advance I wouldn't spend so much time tweaking by builds 'afterwards' to see whether I can get them more perfect or not. ;-) Tweaking and re-building them is half the fun for me. (You can read that last statement more than one way ;-) )