Stupidly Awesome Buffer [documentation]

[jeepe]

[quote="Manfred"]Hi Jeepe, you meant in that way?

exactly !!! thank you so very much !!!

= = =
you know what?
I was, after some 5-10 times of going at it and getting stuck, right now coming back to Supidly Awesome Buffer, cause
our fellow-forumer, Iforgot posted at tagboardeffects, under IvIark's buffer post http://tagboardeffects.blogspot.com/201 ... ffers.html that he had been using that vero for some time and he was
never really satisfied with it, and so he was about to try Stupidly Awesome Buffer, and mentioned that he even made a vero of it, and posted this link... as of Aug 2015...
AND here I was, again, after so many times

I was always tempted by this circuit, but since I have NOT made any buffer yet, I was also tempted by simplicity... (of Jack Orman's)
but then, thanks to Iforgot, I would, again and again, get back here, to the sophisticated, and beautifully performing Stupidly Awesome Buffer
but since I need two (to mix a bass guitar with a stereo "AUX" in, in a practice headphone amp, CMoy, you know), I'd have to draw the vero, and it's not only becoming complex, but also annoying, cause I know that 1 or 2 sections needn't to be duplicated... (around the +IN and the VCC+ legs)...
but how will I know that it works well?
or how will I know why it doesn't if it doesn't?


so, as I came here, again, I just saw this schematic, and right away I knew what that was
and then I realize that it's posted there only lately, oh, and even because I asked
THANKS so much!!!

also for the explanatory indication of modification (!! )

and I hope that anybody will find it who's into building a buffer....

AND THEN a strip layout? (???)
with PDFs?? and PCB style photo illustrations???

you're simply generous
and extremely elegant

BUT, I'm trying to view this NOT from my personal aspect only, but from that of the community of DIY-musicians, who can not design circuits, but only build them...

and from Soulsonic's aspect, too... (hi! it is a privilege to have you in this part of the conversation ! )
his work is not extended, both as a DUAL BUFFER, and in time, as a Buffer which is an active project even in the 2020s

anyhow, thanks to the both of you,
as well as thanks to Iforgot, who gave it a try seeking a better one than that simple Jack Orman "building block"

[jeepe]

I've built several different splitters over the years, and I would usually use 3 buffers: One like this circuit at the input and then splitting into two or more, one for each output. Having the buffer at the input lets you split to many buffered outputs without any additional loading on the input. I also did a few that had isolation transformers on one output to break hum loops that would frequently happen when using multiple amps.

thank you for sharing the sophisticated splitting concept!
sounds absolutely logical even to me!
and I'm happy that anyone wanting to build a very high quality DUAL BUFFER / splitter (I'm repeating these terms for google's bots) can find it right here, from the OP

thank you!

I think I posted a schematic for this somewhere here already?

I couldn't find it... but this could be my fault (I couldn't find a "search _this_user_'s posts" link)

and of course, if you posted it again, it would be great, just for completeness' sake

(I am always taken aback by the fragile ephemerality of BASS and guitar culture in DIY circles... I mean, many of the websites don't even have SSL, picture links are broken, not to mention that parts have died out in the mean time, so it looks like it is not gonna make it to the 2030s, for example,
but luckily, freestompboxes.org is updated to today's standards, and to have those "drawings" of yours here, could be great)

[AndrewGallios]

hi

i want to ask about the R6-1k output resistor

does it have to be 100 ohm for low output impedance or the 1k its fine?

[bmxguitarsbmx]

hi

i want to ask about the R6-1k output resistor

does it have to be 100 ohm for low output impedance or the 1k its fine?

Either is probably fine.

You could totally overthink this and calculate the output resistor for the power supply and the maximum output current of the op-amp. But, since this buffer doesn't have input or output protection, I wouldn't be bothered. (Input protection from being driven while buffer is not powered. Output protection from the discharge of C3 when powered off)

[george giblet]

does it have to be 100 ohm for low output impedance or the 1k its fine?

The resistor is there to stop oscillation with a capacitive load, 1k will work 100% of the time.

[AndrewGallios]

Hi I have a question if someone knows

What values in the parts should be changed to get a little bit of gain in this buffer? Just a touch more from unity gain

[Cybercow]

Should be as simple as adding some resistance between the output of the opamp and the negative input of that same opamp. Try a 20K or 25K trimpot there and see what you get sound wise as the trimpot is turned with a signal going thru it.

[mauman]

With a non-inverting op amp, you'll also need a resistor R10 to the reference voltage (junction of R3/R4). And a cap in line (C6) will roll off the gain to zero <edit: unity> at DC, which is good practice. For a gain of 2 (+6 dB), make R9 = R10, and as you increase R9, the gain goes up. For a gain of 1.5 (+3.5 dB), make R9 half the value of R10, and as you decrease R9, the gain goes down. I'd suggest R10 = 10k, and C6 = 4.7 uF or 10 uF. Make R9 = 4.7k for a little over +3 dB, or 10k for +6 dB.

Buffer with gain

[Cybercow]

What mauman states is a better proposition for increasing the gain on this buffer circuit. Providing a reference to pin #2 helps stabilize the circuit by keeping the audio signal at or near a zero reference. Without it, as I postulated in a previous comment, the opamp can end up with a DC offset and the audio signal will be off-balance from a zero reference point.

[m4268588]

I don't understand necessity of C6.
Connecting C6 to Vref instead of GND results in low noise. And you are using the FET input device OP-Amp.
Is C6 inevitable?

[mauman]

I don't understand necessity of C6.
Connecting C6 to Vref instead of GND results in low noise. And you are using the FET input device OP-Amp.
Is C6 inevitable?

It will work without C6, but this blocking capacitor rolls off the gain to unity at DC. The value is large enough so it doesn't affect audio frequencies. The goal is to reduce the effect of any input offset voltage in the op amp, which can be more than 10 mV in a TL071.

[m4268588]

I got it

[Lukasen]

I have quite a lot of experience with transistor followers aka buffers, so I will list some of them here. I will only be happy if they are an inspiration to others (originally I had a really long article written, but my browser crashed while I was writing).

The first of them is a bipolar NPN transistor buffer. I use a version with 1M in the collector for a small reduction in input impedance. I recommend using polypropylene input and output capacitors. The output section 2u2//10k is due to the minimal load change on output of the input impedance of the buffer (switching between Zin amp or Zin fuzz after buffer).



It's not a problem to do the same with a PNP transistor (actually I don't know why I didn't see that when it might be an advantage). IMHO it sounded similar to an NPN transistor buffer (in the simulation it showed a slightly higher headroom before clipping compared to an NPN buffer).



I also tried JFETs. Borbely's source follower scheme with feedback appealed to me the most, but I was bothered by the Idss variations (I built five prototypes and each one sounded different, madness), the "soft" sound (compared to bipolar transistors) and the slight "coloring" of the sound (I used J201) .



I finally got back to bipolar transistors and their "brown" sound thanks to an ESP Products article on voltage followers. I was intrigued by Sziklai's (complementary) circuit for low output harmonic distortion, and after a series of many tests, I had the schematic ready. I liked it so much that I built it into a narrow Hammond 1550A with two outputs (one for the tuner).



Compared to the others, it has a fairly high input impedance and sounds better to me, more "acoustic" and undistorted. It sounds to me like he's pulling the body resonance out of the instrument, but I don't know if I'm writing it right (sorry for the wrong translation).

A few words at the end of the post, I tried MOSFETs, but I didn't like them much (lots of noise: try fuzz after Zvex Hard-on Booster (or similar with BS170), you really don't want that).
I also tried buffers with operational amplifiers, but it made sense to me only with types that are close to rail-to-rail (for using with single supply) and "do not sound artificial", i.e. I finally used only LM4562 and OPA2132 types (later I found them too exclusive for DIY). With these types, a buffer at the end of the pedalboard really makes sense because it can handle a really long cable and its capacity. You can find more in this article by Rane, Practical Line-Driving Current Requirements: https://www.ranecommercial.com/kb_artic ... ticle=2113.

[lykwydchykyn]

Hi, thanks for the schematics. What do the Zeners do in the Sziklai buffer?

[Lukasen]

Hi, thanks for the schematics. What do the Zeners do in the Sziklai buffer?

Hi, Zener diodes are on the input only to protect against ESD and higher input voltage (more diy solution). A more suitable solution for ESD would be to use a transil at a similar voltage (definitely below the supply voltage). But some diodes have a higher capacitance, so they reduce the heights even before the buffer, so watch out for them .

[Ichabod_Crane]

And I ask about this last schematic, please: none pulldown input resistor? What is the input impedance?

[Lukasen]

I have a rather strict opinion about pull-down resistors. It's probably the worst DIY "solution" (except when you really want to set a certain input impedance with it or there's no other way). Not only do you not have to remove "popping" noise, but at the same time you change the input impedance and thus the frequency of the input high-pass filter at the input of the circuit. At the same time, it is sufficient to properly connect the 3PDT switch so that when the 3PDT is switched off, the input of the circuit, including the capacitor, is discharged to the ground through a low resistor (perhaps 47 Ohm).

So back to the question: I didn't measure the input impedance (but a pull-down resistor would definitely reduce it to unusable). But I estimate it more than 1 megohms and less than 2.2 megohms (it will definitely not be bigger, see the impedance of the source). It will probably also vary across the frequency range...
The higher input impedance was one of the reasons I used a 5n6 rather than a 10nF input capacitor (lots of low frequencies even for bass).

[Ichabod_Crane]

I noticed that in your schematic the pulldown resistor is absent.

Do you have a drawn of your alternative "true bypass" wiring without pulldown resistor?
It's a bit off topic, so just you want to do it.

[bmxguitarsbmx]

Personally, I think 1M input impedance is good. Tubescreamers are 470k/510k depending. Most tube amps are 1M. My buddies who wire "direct to jack" and install a switch to mute are delusional, IMO, as it makes the pickup have way to much transient and the pickups get a "filter" like sound due to excess peaking at resonance. To each their own, of course, and I'm sure a lot of pickups out there sound great into as high an input impedance as possible.

[Lukasen]

I noticed that in your schematic the pulldown resistor is absent.

Do you have a drawn of your alternative "true bypass" wiring without pulldown resistor?
It's a bit off topic, so just you want to do it.

Yes, here it is (unfortunately, I don't know where I took it from anymore...).
I'll admit that I don't like the crossing of the wires, but otherwise I think it's the most suitable connection (except for the relay, of course).
(I actually have other utilities as well, but I don't have the ability to post to that part of the discussion forum yet.)