Fractal used for real amp design!

nicolasrivera

Fractal Fanatic
Wanted to share something i find pretty amazing, an amp designer using Fractal to virtualy design a analog tube amp for production. I'm talking about the Carol Ann Triptik 2.
here is what the designer said in a TGP thread:

When I designed the Triptik 2 model, this was done in reverse. It was proto-typed as a model in Axe long before it was a physical prototype. I saw value in testing the concept first, kind of used the Axe as a first level design / audio test tool. It worked tremendously well and certainly shortened the design timeline quite dramatically and with much less expense. While this toolset is not available through the user interface, Cliff is a close friend of mine and made some suggestions to the original Triptik model that I took on board and drew up a schematic. He then modelled it in the Axe using the original Triptik has a foundation.

https://www.thegearpage.net/board/index.php?posts/23945706/
 
Last edited:
I can't wait to try the Triptik 2. One of the first real amps in years that I'm thinking about buying.

The Tucana and Triptik are my favorite amps in the Axe.
Brit 800 #34 is next in line.
 
Buuuuut...holy shit that thread in general. There's a lot of stupid in the world.
Dude. I totally read that with the voice of Rick Sanchez in my head. (The word "But" was burped.)

anigif_enhanced-buzz-26804-1395548370-7.gif
 
Cool anecdote. Husky (John Suhr) in that thread was also dishing some interesting stuff.

Buuuuut...holy shit that thread in general. There's a lot of stupid in the world.

I lurk on TGP and read it pretty frequently, but not registered. That thread is a good example of why I stay on the "off-topic" forums. The idiocy just about drowns out the smart people (Huskey/John Suhr, Alan Phillips, Cliff C., etc.)
 
Alan is one of the few designers actually trying to do new things rather than rehashing old designs and claiming that certain brands of capacitors are what gives their amps "mojo".
So are we to infer from this info that the Triptik model now in our boxes is indeed the Triptik 2? (as I attempt to play it as I type and it sounds spectacular in 8.01..)
 
Reading through the responses on that one page was a physically painful experience, and theres dozens more pages...sheesh. More OT, I never knew that about the triptik and honestly it's a wonder why more manufacturers don't take advantage of the axe in this way. I think at least friedman amps has a relationship with CC on this kind of level where they provided the schematics for their models and know the kind of publicity the get for it, oh well
 
In the power generation business I think all circuits are digitally modeled before they ever run any conduit or wire. They use PSpice I think and they use another tool called Bentley but I think it is more of a tool for BIM which is like digital modeling for an entire building design. I am not an EE myself, just develop training for a lot of them.
 
I love it when hardware in the loop (HIL) testing comes back around to actually improve the models once the product is out. If Carol Ann or Friedman use this as a HIL platform, hopefully they reciprocate and collaborate on the final model. If so, that's really cool symbiosis!
 
You can.....
Change the tone stack...change tubes....change all kinds of internal component parameters.

Did you have something else in mind?
Yup - front end of one model, power amp of another, etc etc

One of the coolest models in the Axe was designed by accidentally doing exactly that.

A tube swap is not in the same ballgame as actually meshing amps :)
 
Electrical engineers working in analog are used to modeling their designs in SPICE since the 1970's, and I believe that SPICE has vacuum tube amp models, so why isn't that going on in the amplifier design industry?
Does SPICE lets you plug a guitar and load an IR?
 
it's an interesting perspective for the future, rigorous "physical" modeling. I wouldn't say outright it's impossible (a modern graphics card would have qualified as a supercomputer not too long ago) but it's clearly an uphill battle with dynamic range / accuracy and step size, probably ending up in the MHz range.

edit: thinking about this for some more, the "interesting" effects like for example speaker impedance aren't easily modeled in (time domain transient) SPICE so I'm not sure if there is any point in the first place, other than marketing.
 
Last edited:
Back
Top Bottom