Official Luthiers Forum!

There has been a whole new fad in pro audio where individuals are taking an impulse response of a particular famous room (essentially sampling the properties of the room then using a computer to recreate that room in the studio.) This has also been done with guitar amps with companies such as line 6. One of the new companies is Fractal audio with the axe fx II. You have the ability to take an impulse response from any source then play it back through and acoustic or electric guitar. A bubby of mine and I are going to play around with this this weekend, I plan to put several homemade piezo pickups inside several acoustic guitars underneath at key locations to try and capture the full dynamic range of the sample. My question is should I use just white noise or use a sine signal and sweep the frequencies that are applicable to an acoustic guitar as the impulse response in taken. Trust me I am not looking to take a cheap acoustic and make it sound like a Somogyi. I am just looking for some new different tones to play with

As Always thanks
Richard

If you tap the guitar top, it's the equivalent of a white noise input and will excite all of the frequencies of the top and their proper relationship to one another. Seems like the easy way to go to me.
I recall Joseph Curtin, a violin maker, did something very similar on violins and presented it a GAL convention a while back. He impacted the bridge to get his response. It was interesting stuff. Good luck.

_________________
Jim Watts
http://jameswattsguitars.com

personally, i would focus way more on learning to actually play an instrument, as opposed to trying to digitally recreate some nondescript room where some famous musician may have once recorded something. the venue never made or broke any talent, i guarantee you.

I know nothing of the technology, but...

I'd certainly do white noise to control any environmental influences. Maybe some times I want and sometimes not, like feedback.

Then I'd do the sign sweep to define the range of frequencies, and also amplitudes. Also, probably would want to see if duration to sound matters, like the tap of a hammer vs hum.

I'd also try to drive the instrument from another instrument and or voice, which should be simple after the other stuff.

I don't know if this is what you're looking for, but Rick Turner and Seymour Duncan have worked together in this area.

http://www.d-tar.com/mama_bear.shtml

Pat

_________________
formerly known around here as burbank
_________________

http://www.patfosterguitars.com

what are you trying to accomplish with this, if i may so ask? insert the IRs into some convolution engine?

_________________
member of the guild of professional dilettantes

I'm not sure what this statement means or if it has anything to do with my question. If you are referring to me "learning to play", I play very well. I was teaching guitar at a local music store at the age of 16, toured briefly before I went to college and have been in countless session in the studio over the years. This was 25 years ago.

Hi Miguel, The purpose of this is just experimental. I'm not looking to apply this to an acoustic guitar. You can't make $100 Chinese made dreadnought sound like a Pre War Martin D-28. The purpose of the Axe-Fx is to model any amp, cabinet, mic etc and have all of those combinations at your disposal as a player. See http://www.fractalaudio.com/ There have been many attempts to do this in the past, but this guy has really nailed the technology. The artist who are using this unit are world famous players.

To my knowledge no one has used an acoustic top as a source. I don't expect a Strat to sound like and acoustic guitar, but i'm just curious what we will get out of it. Since an acoustic top is a very complex resonating structure it can serve a source for in the impulse response. I'm just asking opinions on ways to excite the top.

Richard

By definition an 'impulse response' is a tap or equivalent, so far as I know. It will contain equal amounts of energy at all frequencies up to a limit that is set by the contact time between the 'hammer' and the 'anvil'. As long as the hammer is in contact the object is not vibrating freely, so the high frequencies are damped. If the hammer stays in contact for 1/1000 of a second, there will be essentially no energy above 1000 Hz, and the drop off starts around 500 Hz.

What you want, then is a small, light hammer that will bounce off quickly. I use a hard plastic or wood bead, 5/8" or 3/4" diameter, and arrange for it to hit the saddle between the 4th and 5th strings. Ideally the hammer should also be hung from a string, rather than having a rigid arm that could introduce it's own resonant frequencies. This makes a very sharp sound when it hits, due to the high frequency content, but there is very little energy involved, and I doubt it would mark anything but the most fragile cedar soundboard. Having it hit the saddle is insurance.

you said:


i said:


you responded:


my point is this: when i see guitar players(and it is always electric guitar players, and yes, i am one as well) seeking The Holy Grail, in the form of: amplifiers, effects, "mustard capacitors", carbon composite resistors, Drake transformers, Mullard tubes, vintage guitars, and now, digitally simulated recording venues, all to chase some luck some famous musician had long ago....the cork-sniffery just makes me groan a bit.
let's just say hypothetically you manage get your delay/reverb/EQ set up to make a recording sound JUST LIKE Sun Records Room X thet Elvis recorded "I Wanna Play House" in in 1955 or whatever.....then what? what is the implication....? it wasn't the venue and it wasn't the equipment that was responsible for the historical/legendary status of the recordings, it was the TALENT and CREATIVITY and hard work, and you can't digitally synthesize that...

I agree totally, there are no short cuts. I also agree that there are those who PLAY guitar and those who want to sound like (insert your favorite guitar player) and are constantly searching for the "THE TONE"

But this is quite a bit off subject from my original question which was how would you capture the signature of an acoustic guitar body/top (without string) ? In other words how would you artificially excite that top so that you capture the response of the top from lets say 60hz to 10000K All guitars will have a unique signature. So here is what I did. I made 5 piezo pickups and used double stick tape to attach the pickups in different areas inside the top. I ran the pickups into a mixer and sent them into the Axe-fx. I put one of my full range monitors as close to the top as possible. I took 2 samples one was a sine wave sweep for 60hz to 8K. The other was just white noise. These were sent to the Axe-fx where it took the signals from the pickups and created a response file. It essentially turns this file into a speaker, which basically is what an acoustic guitar top is. Now you can take any guitar, electric or acoustic play it through this unit and the "speaker" that is being simulated is this guitar top that was captured. Like I said this is just an experiment. I plan to do several guitars while I have everything set up. This very well may be a big waste of time, but I like to try and think outside the box sometimes

Sounds fascinating, I can't wait to hear the results!


Ken Jones
Mountain Song Guitars

_________________
Mountain Song Guitars www.mountainsongguitars.com

Interesting experiment. I play acoustic guitar in a small amateur orchestra and I often wish I could swap out my steel string OM for an arch top or nylon string for certain songs, but there just isn't time. I've tried a few effects pedals but they just don't provide a believable sound. I'll be interested to see how this works out.

_________________
Steve Smith
"Music is what feelings sound like"