Anti-buffer

[blueherdim7]

Guys-

I am wondering if such a thing exists or can be made.  I have a buffer at the beginning of my chain that goes into my switcher.  Everything is great except for my BM which sounds awful with a buffered signal.  Is it possilble to make something that will unbuffer the signal for my BM? 

Cheers.

Jon

[R.G.]

An anti-buffer is an impedance. You can use a resistor of some value between 8K and 100K to suit your taste.

There are some people who like to try faking a guitar pickup's impedance (as opposed to just resistance) by putting a resistor in series with an inductor. The common wah inductor works, as does the primary of the small audio transformers sold by Mouser and possibly Radio Shack. This will be a tune to taste solution.

Someone will pop in here in a minute with info about Jack Orman's having written an article on this.

[earthtonesaudio]

Pop!
http://www.muzique.com/lab/pickups.htm

[blueherdim7]

Thank you guys...that is exactly what I am looking for!

Jon

[blueherdim7]

Is there any reason why I could not combine a pickup simulator with a jfet buffer (output buffer) to rebuffer my signal after is passes through the fuzz so that my downline effects and amp will still receive a buffered signal?

Cheers.

Jon

[R.G.]

If your fuzz has a low output impedance the rest of the line doesn't need a specific buffer. There are a very few effects with large output impedances that profit from buffering, but not many. You likely don't need it. Won't hurt much though.

[blueherdim7]

I am refering to a 1983 BMP and a Kay Fuzz Clone built from the schematic and pcb from ustomp.  This may seem a very noob-like question, but rest assured, I am a noob...how would I measure the impedence at the outputs?

Jon

[earthtonesaudio]

The quick and dirty way to measure output impedance is to hook up a variable resistor to ground on the output, and gradually turn it down until you notice a drop in output level.  Then you measure the resistor and it should be pretty close to the output impedance, or just slightly lower.

The accurate way involves math.

[blueherdim7]

Thanks Alex!  I can give that try.  I am not one for public math, but can anyone tell me how to calculate the output impedence?

Cheers.

Jon

[d95err]

Is the interaction with the guitar pickup really that important for a good fuzz sound? Isn't the main problem that the buffered signal does not loose the high end it normally does due to the low input impedance of the fuzz? If so, wouldn't a simple lowpass filter at the input of the fuzz solve the problem?

[jayp5150]

Is the interaction with the guitar pickup really that important for a good fuzz sound? Isn't the main problem that the buffered signal does not loose the high end it normally does due to the low input impedance of the fuzz? If so, wouldn't a simple lowpass filter at the input of the fuzz solve the problem?

I just know in practice, anytime I ran my fuzz after my morley wah (buffered bypass), it severely affected the sound of the fuzz.

I'm not sure a lowpass would do anything, as the affect the buffer had was the complete absence of low end from the fuzz. I had a variable capacitor setup on the front end of the fuzz, and even with it at it's fattest (3.3uf, I think), it was nasally, and thin. I built a bypass box for the wah, and it was night and day.

I need to try out the impedance matching thing, though, because I like my wah, and I like fuzz... it sucks to separate them.

[blueherdim7]

I agree with you Jay, a buffered signal into my fuzzes sounds nothing like a direct signal from the guitar.  I plan to make a box that will unbuffer the signal into my fuzz and then rebuffer that signal out to the rest of my chain to the amp.  I don't want to mod my vintage BMP.

Jon

[ashcat_lt]

The problem is, as mentioned in that link above, that the pickup and cable create low-pass filter.  One thing that article leaves out is that the input impedance has a major impact on the overall slope of that filter.  At 1M, it is fairly resonant (as pictured in that same link).  At lower input Z, the resonant peak is flattened.  Eventually, with very low in-Zs the corner frequency starts to slide downward.

My issue with this idea is that the "pickup simulator" is really just another filter.  Once you've plugged your guitar into the buffered input, the signal has already been affected by that highly resonant filter.  Then you're going to run it through this second filter.  It will not be the same thing as plugging the guitar directly into the lower Z input.  That second filter can't undo the resonance from the first.  I made some pictures to prove this point, but can't get them uploaded to my server right now.   

I think it will help, and be better than nothing, but it won't be anywhere near the same.

Edit ___________

Okay, got it up there.  So, I've based my experiment on these graphs, which show the filter created when plugging a "typical" guitar into various input impedances.
First 100K - 1M

and then 10k - 100K

I ran white noise through filters set up to approximate those curves.  The blue represents the 1M input at your buffer.  Green is the 40K input of the big muff (that's a value I got from googling).  The red is the result of running these in series.  This assumes that the pickup simulator filter matches the pickup/cable filter pretty closely in terms of the L and C values. 

[blueherdim7]

Ashcat-

Thank you for your reply!!  The generosity and knowledge of this forum is trememdous!  If I read your post correctly, specifically the last graph...the top line is the blue line?  The blue and green are difficult to distinguish.  If that is correct, then your chart seems to show why a BMP sounds so bad with a buffered signal.  What I would like to do is make the red line appear more like the green one.  I agree that once buffered, the signal cannot be truly "unbuffered" back to its original state.  I just want to try to bring it closer and more "pleasant" without having to run my fuzzes in a parallel set up to give them unbuffered inputs.  Also, what software are you running to create those.  It looks like something I need to get. 

Cheers.

Jon

[ashcat_lt]

I don't really know anything about this stuff, just passing on information I've picked up elsewhere.  The first 2 graphs were generated by my friend JohnH, using 5spice to model the circuit.  They originally appeared in this thread.  It has very little to do with what we're talking about here, but it does show the circuit and component values he used.

The last graph I made.  You got it correct, it's the topmost (the one with the big bump) which I'm calling blue.  First I made a short file full of white noise in SoundForge.  Imported that file into Sonar 4 where I used the channel EQ to create a filter that looked like John's graph.  The result was exported into 3 seperate files and opened back in SoundForge where I took snapshots in its Spectrum Analyzer.  Then I took a screenshot and exported to .jpg from Corel PhotoPaint.

I'm not sure that there is a passive way to "unbuffer" the signal, or more correctly to smooth out that sharp knee in the red line.  I've never tried this, and I'm not sure how close you'd need to get to the green line to actually sound acceptable.  Either way, I'm certainly not the one to design the necessary filter.  I believe, however, that this forum is a good place to find somebody who could.

[gez]

Scroll down:

http://www.home-wrecker.com/wahmods.html

'Good enough'...