Lego Block Amp-in-a-Box

[Vivek]

It seems all Amp-in-a-box follow the same general theory

Input buffer
filter
distort
filter
distort
filter
distort
filter
output buffer

with tone controls thrown in somewhere


Given this Lego type connectivity, it seems possible to propose a building block approach to an Amp in a Box design, where a myriad of different Amps can be made, by using same PCB, and builder makes a few easy design choices when choosing which Lego blocks he would like to try out.

These need not be simulations of actual existent Amps, yet they would be totally valid Amp designs, with perfectly usable sounds.

Or they could be simulations of actual Amps, if someone just writes down a bunch of design parameters that attain a particular sound and feel of the target amp.

So we could post "Patches" for hardware just like MFX users post online patches for "Comfortably Numb"

It should be possible to say something similar to

"Hello Richard, Have you tried this great Amp in a Box which I call "Persian Lime" :

{Start Amp version 1.5}
PCB = Amp_5(18)
Buffer1 = Buffer_C (1)
Clipper1 = Clipper_T (300, 5000, 1.5)
Tone = Tone_F (400,1200,6000)
Clipper2 = Clipper_Y (700,4000,3)
Clipper3 = Clipper_W (500,6000,6)
Buffer2= Buffer_L (3)
{End Amp}
"

and above would totally specify a complete build of an Amp in a Box hardware that follows all known rules and knowledge re impedances, signal levels, filters etc with no issues while linking the Lego blocks together.


and someone could invent a new block and say

"Guys, I have just released Clipper_U1 which uses FET instead of Opamps. It follows the standards for gain, input impedance, output impedance etc etc for Lego Block Amps and you can use it anywhere in the Lego chain. Enjoy"

[Vivek]

All schematics below are, in effect, the same Amp in a Box, but using different colored Lego Blocks in different positions.

They are all derivatives from the same concept.

Yet that all sound different

It is possible to build any variant on the same PCB

It is possible to invent new amps using same architecture but different design parameters

[niektb]

Isn't this more or less the same concept as in the topic you created earlier? https://www.diystompboxes.com/smfforum/index.php?topic=127648.msg1225130#msg1225130

[Vivek]

Yes, it builds upon the earlier concept of Unit Processes

But I wish to investigate it further, specifically for AIAB applications

and actually having some deliverables this time


Like designs for :

2-3 different designs of input buffers

Some modular clippers where it is very easy to calculate the gain, LPF, HPF
The clippers will differ in the transfer curve

some existing tone controls but values chosen so that impedances are somewhat standardized

A PCB that is flexible enough to build some of these ideas ( I am miserable at this, l need help)



Some ideas to kickstart would be

INPUT BUFFER
DC blocked at input and output
gain 1
frequency response 50Hz to 10,000 Hz
Input impedance more than 600K
Output impedance less than 300 Ohms
Supply voltage 7 to 30V

CLIPPER TYPE 1
gain either set by resistors, or variable pot, from 1 to 500
input impedance more than 100K
output impedance less than 300 ohms
LPF and HPF with easy online calculator

Transfer function: rail saturation

CLIPPER TYPE 2
same as above but Transfer function: Asymmetric

etc

[teemuk]

What you describe basically IS how many commercial products are built:
One printed circuit board. Component values are altered or components are not fitted. In many cases components of entire circuit blocks are omitted if a feature is not needed.

This is cheaper than making a different PCB for all different designs.

Proper layouts support several extra circuit features. i.e. easier to design support for emitter bypass and Miller caps, and just omit fitting them if not needed. Easier to design support for inverting, non-inverting, and differential opamp stage, with all imaginable features than just one of them and then have to alter the entire board in case you find out you actually needed an inverting stage instead of non-inverting and it would have been nice to have clipping diodes and HF bypass cap there too. Etc.
I've seen boards that can transfer to various different designs depending on what parts are fitted. Vox VT+ series, for example, has some serious "easter eggs" that were simply never fitted to final design. Have you ever looked at modern Roland amp schematics?

Then add a wee bit of modularity: preamp with all the goodies on one board, power amps in various power ranges on another, and supply on third. Combine in various formats and you have an entire serie of products.

And let's not even go to subject how transferring signal processing from a complex analog circuit to a DSP can offer soooo wide range of possibilities. Now we need just a few common analog stages for input and output, user interface and codecs and rest can be nearly anything.

[Steben]

It's intriguing after the buffer thread you come up with a unity buffer in a gain circuit. 

[Jbassfunk775]

Well that's certainly a novel idea, and it looks like you've given this a decent amount thought. I also think the fact that while early and largely conceptual. You've already begun laying a framework that's organized and process oriented. I'm going to be honest in saying that I'm entirely clear on where the idea meets the tangible. It's important to have a specific concrete end result always present in your mind. At this point particularly, the physical real world development needs to move from the the page/diagram to the breadboard ASAP!

I do see a number of pitfalls in the analog hardware version of software free-share communities. Namely that the software sharing gets tons of "user original" submissions because it requires a zero prerequisite level of specialized applied skills, experience based knowledge/understanding of electronic design principles, active/passive component functionality. I could literally go on and on.. And while we have communities like this one available the individual experience/knowledge of members runs the gamut from absolute zero up to professional working engineers in fields that venture wayy past the scope of instrument audio equipment. That's compounded by the often narrow areas of interest; ie one guy is a diehard vintage fuzz collector and builder and isn't interested in anything but biasing germanium transistors. Myself I'm big into parametric/semi-parametric rackmount eq's   compressors, and tube or discrete fet transistor preamps (which is actually what drew me to your post).

All that said I hope you stick with it and your ideas come to fruition!! I'd love to see what happens and come up with!