Buffers and excess highs

[shredgd]

Hi,

I built an AMZ Super Buffer (using just one IC, a 4558) and a JFET Buffer (schematic at GGG) because my true bypass pedals chain has become a bit too long for my guitar signal and I'm experiencing a loss in treble in my guitar signal.

I found they both suffer from the same problem: the tone I get with each in my pedalboard is even richer in highs (presence, I'd better say) than the tone I get from a single 5m cable from my guitar straight to the amp!!
This might seem good for some, but for me it means my tone losing warmth and beauty.

Comparing the two type of buffers, I must say I prefer the JFET one, the Super Buffer being even a tad richer in highs than the former.
Curiously, I have a Boss SD-1 (which I don't use because it is not true-bypass), and it gives me a better bypass to my ears (again in terms of presence) than both the JFET and the Super Buffer, though still too treble-oriented and with a warmth loss.

Sometime ago I read a thread in which the phenomenon of the extra-highs was explained: those frequencies are normally lost by the cable, even a single cable, from the original pickup sound, so they are not added but restored in the sound, it said.

However, I love the sound I get from a single 5m cable straight to my amp and I'm looking for a really transparent buffer to keep that sound while using my 8 true-bypass pedals and two 5m cables.

I'd rather accept a little loss in highs, compared to the single cable setup, than a gain in highs.
In other words, I realize I'm losing too much highs with all those pedals and cables, and I want to restore some or all of them, but nothing more.

Does this kind of buffer exist?

Thank you,

Giulio

[Gilles C]

Just as a reference, check what they do in this schematic to simulate a cable...

http://www.jensentransformers.com/as/as004.pdf

[brett]

Hi
These are interesting observations. 
Why are 1M gate to ground resistors used very frequently in JFET buffer/input stages?  Probably coz it's a nice round number.  I doubt whether the size of this resistor has received serious attention. 

For a while now, I've been noticing that input impedance of 200 to 500k is a good level in a range of effects and amps.  The old effects books of Penfold use 220k resistors and 250k pots in many circuits. 
A good, simple option would be a JFET buffer with a 330k resistor or 500k trimpot to dial up the tone you want .
I've been using JFETs with 470k for a while, but only because one day a long while ago I thought that 1M seemed too high and I reached for a lower value that I had a lot of.
cheers

[amz-fx]

Why are 1M gate to ground resistors used very frequently in JFET buffer/input stages?  Probably coz it's a nice round number.  I doubt whether the size of this resistor has received serious attention. 

The input resistor has been given some attention...  for example, in my  Rat mods and Fat Gnat designs, I use 100k input impedance.  There are reasons you might use higher or lower impedances according to what you are trying to accomplish: http://www.muzique.com/lab/imp.htm

1M has historically been chosen because it is similar to the input impedance of a tube amp...  as noted by  Pete Cornish. However, like his observations about true bypass in that article, his estimate of 1M/20pF for a tube amp input are wrong for the typical 12AX7 type of input stage (maybe closer for a pentode input but those are much less common).

If you want to trim a few of the highs on the AMZ Super Buffer, you can make R6 a lower value. Start with 220k and see how you like that. If 220k is not exactly right, go to 100k if you need to trim more highs or 470k if you want to let a few more through.  You should be able to easily zero in on a value that sounds good to you.

regards, Jack

[sfr]

I don't want to sound like a jerk or nothing - but when I started using a buffer and having this problem, my solution was to use the tone control on my guitar.   That was actually my favorite thing about the buffer - my tone control actually worked!  W/O it, it seemed like anything less than "10" was a muddy mess.

[Meanderthal]

 It sounds like you've gotten used to the sound of cable loss, and like it , but not too much. That would imply that you're really looking for a rather inefficient buffer, so the Super Buffer would be way too effective for your needs. A buffer does not add any highs, it only prevents them from being lost.
Why not just use that Boss? It already has a buffer when bypassed, and through comparison you've found it to be the least effective of the 3 you mentioned.
Now with the Boss in there you need more cable loss to dull it down a little more... How 'bout a really huge Hendrix style coiled cord? They're known to be pretty bad that way... Or the Monster Cable high capacitance cables ...

[Pushtone]

You could try adding in a circuit that would ADD LOADING.

Variable input impedance (from 10k to 1meg) on a control knob would allow you to dial in just the right amount of HF loss to suit.

http://www.diystompboxes.com/smfforum/index.php?topic=39559.0

http://www.diystompboxes.com/smfforum/index.php?topic=39543.0

[shredgd]

Thanks for all your opinions/suggestions.

I did some more experiments today...

The 1000p cap from signal to ground before the buffer to simulate the capacitance of a standard cable, as suggested by the .pdf indicated by Gilles C, is too much, but I didn't have any smaller cap around, so I can't say it's a wrong direction to explore.

Apart from that, I realized I hadn't included any pull-down resistor at the input of my JFET Buffer, because I used the original schematic of the "Basic Buffers" page at AMZ. Adding a 1M resistor smoothed out things a bit, even though this means the input impedance of the buffer is now about 330k, not 500k anymore.

However, this fact (and reading the pages you linked in your replies) led me to think high input impedance is not so desirable!
I mean, I understand its importance in keeping the signal as strong as possible, mostly in the high frequencies, which are the first to get lost, but in my opinion the target tone we have to achieve with a buffer is the tone you get without it and with a single cable to the amp! Amps are designed to be used with a guitar and a cable, I suppose, so feeding them with an excess treble (ok, I should say "normally lost in the cable") alters its original sound and response!

Regarding the Boss pedal, I don't know if I prefer its bypass to the JFET Buffer anymore... They sound different, that's for sure, and they ARE very different, because the Boss incorporates two buffers in bypass mode, and there are resistors in series with the guitar signal. Some times during my experiments it seemed to me it sounded better, some times it didn't. It colors the tone more than the JFET Buffer, and that was predictable comparing the two schematic, that's why I'm leaning more toward the JFET Buffer, now.
To answer to Meanderthal, I couldn't use the Boss permanently in my pedalboard even if I loved its bypass, because I need a switchable buffer to get rid of it if I'm using my diy fuzz face, which sounds horrible with a buffer in front of it (like any fuzz).

I think I have to experiment a little more with resistors (in series and to ground) and caps (to ground) in front of the JFET Buffer and give another try to the Super Buffer reducing the input Z resistor like Jack suggested, but the Boss schematic takes me to another question: what do series resistors do after a buffer? (Resistors to ground in this position are simply part of a high pass filter together with a cap, isn't it?)

One last question for Jack: I was looking at the Bipolar Buffer in your basic buffers page and wondered if one could raise those 220k resistors to raise input Z (and how much).

[amz-fx]

One last question for Jack: I was looking at the Bipolar Buffer in your basic buffers page and wondered if one could raise those 220k resistors to raise input Z (and how much).

You can raise R1 and R2 to 330k each, or even 470k each to see how you like the tone.

regards, Jack

[bwanasonic]

Don't know what your recording conditions are, but don't neglect the beauty and simplicity of a small tube amp, some effects, and a microphone. I have been doing extensive recording in the past year or so, all at reasonable volumes in a residential setting. The first clip is a good contrast between the direct recorded sounds of the other guitar player ( Tim Casey), and my more *filtered*, mutated*, and atmospheric tone. You can also get pretty raucous sounds without unduly disturbing your neighbors. The Sansamp GT2 has performed well for me in direct recordings, but this usually involved a good bit of post-production compression and EQ twiddling, and never felt near as good in *real-time* as playing a *real* amp.

Kerry M

*mostly thanks to the Frostwave Blue Ringer.

[shredgd]

Some more experiments today...

I built the Bipolar Buffer using 470k resistors: this is the best sounding of the three buffers in terms of overall tone (this explains my tendency to like the Boss bypass..) but the input impedence has to be raised some more, because it loses some highs. I will soon mod it with a 10k/10k voltage divider to bias the base of the transistor via a 560k resistor or so.

This made me realize that what I perceive with the other buffers is not properly an excess in highs (probably those are simply the frequencies best managed by those buffers), but rather a lack of something in the other frequencies.
If I had to describe it with a single word I would say that the tone I get with a single cable to the amp is more "alive" (or "natural"). And my preliminar impression with the Bipolar Buffer is that it keeps this lifeness (except for a certain lack in presence, which I hope to revert by raising the input Z as I wrote above and/or by using the parallel 0.1u film cap trick at the end of the circuit as in Jack Orman's Super Buffer).

I know I'm probably getting boring with all these observation, but mine is a common disease among guitar players: being VERY picky about tone!

Giulio

[Meanderthal]

 Yeah, I'm that way about bass... but buffers aren't my problem. With me I get all wierd about the midrange(bottom end is more than covered by my SVT cab). It's at the point where other people swear they can't hear stuff that bugs me(That sounds GREAT, what are you talking about?), so I'm pretty sure it's just me, but damn it, when it bugs me it just does...

[shredgd]

Hi,

I finally reached a decision. I simply can't use a buffer, IMO they alter the real sound and dynamics of a guitar.

I shortened all the connection cables between my pedals, now I still have a little treble (presence) loss, but at least the overall tone and response is the same. It might even help, as I usually find guitar amplifiers a bit treble-oriented. Also, I won't have to remember to turn the buffer off when I turn on my fuzz face!

[JonFrum]

I don't want to sound like a jerk or nothing - but when I started using a buffer and having this problem, my solution was to use the tone control on my guitar.   That was actually my favorite thing about the buffer - my tone control actually worked!  W/O it, it seemed like anything less than "10" was a muddy mess.

The same thought did enter my mind.    Then there is that treble pot on the amp.

[shredgd]

Just an update to my thoughts and observations...

I came to the conclusion that I really need a buffer to avoid my sound from becoming dull because of all the passages through my pedals and cables.
The bipolar buffer is the best IMO. One at the beginning of my pedals chain (my case) restores most of the highs without changing the character and response of my guitar.

Giulio

[bioroids]

I don't understand something:

If you have a buffer that has the same impedance as the amp, and you use the same cable,
then you should have the same highs loss, between the guitar and the buffer, or the guitar and the amp.

The buffer prevents extra highs to be lost AFTER it, on the way to the amp, so adding more effects won't dull your sound.

So why adding a buffer would make those extra highs "reappear"?
I've experienced myself the extra highs obtained by using a simple buffer with 1M input impedance between the guitar and the amp. I always wondered why.

Is it because the amp impedance is not as high as it is always assumed (the classic 1M)?

Regards

Miguel

[shredgd]

Hi Miguel,

if you look at this thread

http://www.diystompboxes.com/smfforum/index.php?topic=30987.0

you can see we are not the only two wondering why this curious (from a non-electronic engeineer point of view, at least) phenomenon occurs. I still didn't find a real answer. One thing is sure: our buffers and our amps "see" our guitar signal differently, otherwise we wouldn't experience this phenomenon. Input impedance differences, like you say in your last sentence, probably play a role at this point.

Giulio

[Mark Hammer]

It's the wire.

Read the article titled "When is a cable not a cable" here: http://hammer.ampage.org/files/Device1-5.PDF
It explains just about all.