Analogman - Buffer

General documentation, gut shot, schematic links, ongoing circuit tracing, deep thoughts ... all about boutique stompboxes.
User avatar
sinner
Old Solderhand
Information
Posts: 4713
Joined: 06 Nov 2008, 17:16
Location: ...no more
Has thanked: 1031 times
Been thanked: 907 times

Post by sinner »

mxrmxr wrote:Yeah you've got me all figured out haven't you Govmnt_Lacky. Well done :applause:

To the moderators: Please remove me from the forum. Leechers like myself offer nothing to the forum and are not welcome here.

Thank you,
mxrmxr
Dude don't be silly :hug:

User avatar
Govmnt_Lacky
Information

Post by Govmnt_Lacky »

mxrmxr wrote:Yeah you've got me all figured out haven't you Govmnt_Lacky. Well done :applause:

To the moderators: Please remove me from the forum. Leechers like myself offer nothing to the forum and are not welcome here.

Thank you,
mxrmxr
I am not pretending to "know" you at all. I am only stating what is fact. Someone had the gumption to ask you politely to share some of the information you might have (whether you have it or not is not known since you never responded) and you never replied. I doubt you did not see the request as I am sure you are keen on viewing threads that you posted in. I am just hoping that you would set things straight and respond.

And please don't pull the "POOR ME" routine and bail! Sack up and just respond bro :horsey:

If you are so deeply offended, please see that I have been around these forums for several years and get quite tired of seeing people reap information and not give back.

User avatar
ryanuk
Solder Soldier
Information
Posts: 200
Joined: 29 May 2008, 11:18
Has thanked: 26 times
Been thanked: 26 times

Post by ryanuk »

An old thread, but I was curious about this as well, after seeking a video on YouTube about using this buffer in front of an Ernie Ball Vol Pedal.
Downloaded some photos, which showed the traces.

I managed to come up with this schematic. Cant see one or two of the values. Also not sure about the 2nd op amp stage and the parallel cap/resistor.

Seems to be a Tubescreamer/boss discrete buffer, into a opamp gain stage (not unity?), and then another IC stage.
Unsure of the logic.

Any thoughts/comments?
Attachments
1.pdf
(162.55 KiB) Downloaded 139 times

User avatar
ryanuk
Solder Soldier
Information
Posts: 200
Joined: 29 May 2008, 11:18
Has thanked: 26 times
Been thanked: 26 times

Post by ryanuk »

Bumping this...
Noticed my scrappy schematic has been downloaded a few times.

Does anyone have any other pics of the board so I can confirm the values and traces and get a proper schematic together?
For completeness more than anything else. :thumbsup

Ryanuk

User avatar
grrrunge
Diode Debunker
Information
Posts: 778
Joined: 20 Mar 2010, 10:51
Completed builds: Loads!
Location: Copenhagen, Denmark
Has thanked: 48 times
Been thanked: 192 times
Contact:

Post by grrrunge »

I did some googling. Let's get cracking :)
Attachments
sfx12vpflipjd2xd3c2d.webp
PXL_20210429_162122094.jpg
PXL_20210429_162054117.jpg
mvrhne8ha8harucoij0i.jpg
mjeylubar7fzlxegmhd0.webp
buffersmlinsidesml.jpg
buffersmlinsidesml.jpg (66.18 KiB) Viewed 640 times
bufferminiinsidesml.jpg
bufferminiinsidesml.jpg (73.36 KiB) Viewed 640 times
bufferboardinminisml.jpg
bufferboardinminisml.jpg (102.21 KiB) Viewed 640 times
bufferboard.jpg
bufferboard.jpg (57.6 KiB) Viewed 640 times
A true believer in the magic of Sherwood Forest Pedal Pirates
---
New base of operations: http://www.knucklehead.dk

User avatar
CheapPedalCollector
Resistor Ronker
Information
Posts: 283
Joined: 19 Oct 2020, 15:15
Has thanked: 334 times
Been thanked: 99 times

Post by CheapPedalCollector »

I don't understand anyone making a buffer with a 4558 instead of a 4556. That's pretty odd.

User avatar
grrrunge
Diode Debunker
Information
Posts: 778
Joined: 20 Mar 2010, 10:51
Completed builds: Loads!
Location: Copenhagen, Denmark
Has thanked: 48 times
Been thanked: 192 times
Contact:

Post by grrrunge »

How come? If you don't need 70mA current drive capability, they should behave practically identical in the audio spectrum. What am i missing here?
A true believer in the magic of Sherwood Forest Pedal Pirates
---
New base of operations: http://www.knucklehead.dk

User avatar
CheapPedalCollector
Resistor Ronker
Information
Posts: 283
Joined: 19 Oct 2020, 15:15
Has thanked: 334 times
Been thanked: 99 times

Post by CheapPedalCollector »

You need the current to drive long cable runs with a very low output impedance if you don't want to damage the part or have other issues. TL072 doesn't have enough current either and has a lot of distortion and is noisy. RG Keen used several 5532's in parallel for his design which is also works pretty well but more expensive. 2 4556 stages in parallel is less expensive and should work excellent with something else before it that's high impedance input, I'd have to check spec sheets for something that's JFET input, has low THD and low noise as I can't remember which op-amps fit that criteria off the top of my head.

It depends on what you're after I guess. I wouldn't use a 4558 for anything myself.

User avatar
grrrunge
Diode Debunker
Information
Posts: 778
Joined: 20 Mar 2010, 10:51
Completed builds: Loads!
Location: Copenhagen, Denmark
Has thanked: 48 times
Been thanked: 192 times
Contact:

Post by grrrunge »

I must say i disagree with you on that.

If you have a typical input impedance around 500k ohms and a maximum signal amplitude of 9 volts you get 9V/500kOhm=18uA. A typical op amp with 20-ish mA drive current will drive loads down to 450 ohms before giving in.

The feedback reduces the output impedance by a factor of loop gain, effectively giving you an output impedance in microohms. Even for long cable runs that would push the corner frequency of your output impedance and cable RC filter into the far MHz range.

Distortion is reduced by a factor of loop gain, resulting in Distortion in the ppm range.

TL072, albeit not being low noise by modern standards, is still the low noise option of the TL0xx series, and is widely used in lots of high end audio equipment. No need to bash an old workhorse ;)

Anyway... Let's get back on track.
A true believer in the magic of Sherwood Forest Pedal Pirates
---
New base of operations: http://www.knucklehead.dk

User avatar
Dirk_Hendrik
Old Solderhand
Information
Posts: 4152
Joined: 03 Jul 2007, 08:44
Location: Old Amsterdam
Has thanked: 215 times
Been thanked: 854 times
Contact:

Post by Dirk_Hendrik »

grrrunge wrote: 24 Sep 2022, 21:49 No need to bash an old workhorse ;)
Let me second that.
Yes, while there may be more modern designs with better specs I have problems with discarding a TL07x for with the statement "it has a lot of distortion", especially when referring to a 5532 in the same TL07xx design may be old (approx 1975) the 5532 isn't that much better (1979) that's some 42 years as well. First of all the amount of distortion of an opamp is also depending on the application and secondly a TL07x just does not have "a lot of distortion", It just has a very little more than some more modern and usually far more expensive counterparts.
Sorry. Plain out of planes.

http://www.dirk-hendrik.com

User avatar
grrrunge
Diode Debunker
Information
Posts: 778
Joined: 20 Mar 2010, 10:51
Completed builds: Loads!
Location: Copenhagen, Denmark
Has thanked: 48 times
Been thanked: 192 times
Contact:

Post by grrrunge »

Here's the mirrored bottom side with reference designators added. All the text became somewhat small. I guess I wass zoomed in quite a bit :mrgreen:

The red reference designators for C51 and C52 indicate that they are guesses, but based on the logic behind the rest of the numbering, they should be good.
Attachments
bottom-refdes.jpg
A true believer in the magic of Sherwood Forest Pedal Pirates
---
New base of operations: http://www.knucklehead.dk

User avatar
grrrunge
Diode Debunker
Information
Posts: 778
Joined: 20 Mar 2010, 10:51
Completed builds: Loads!
Location: Copenhagen, Denmark
Has thanked: 48 times
Been thanked: 192 times
Contact:

Post by grrrunge »

Here's what I've got.
ryanuk wrote: 14 Sep 2022, 15:32Downloaded some photos, which showed the traces.
Would you mind sharing your photos. It looks like we've got some different values here and there :)
Attachments
R62 R63.png
R62 R63.png (182.83 KiB) Viewed 444 times
R58 R59.png
R58 R59.png (243.32 KiB) Viewed 444 times
R52 R53 R54 R56.png
R51.png
R51.png (144.51 KiB) Viewed 444 times
am buffer.png
am buffer.png (9.13 KiB) Viewed 444 times
A true believer in the magic of Sherwood Forest Pedal Pirates
---
New base of operations: http://www.knucklehead.dk

User avatar
ryanuk
Solder Soldier
Information
Posts: 200
Joined: 29 May 2008, 11:18
Has thanked: 26 times
Been thanked: 26 times

Post by ryanuk »

@grrrunge

Many thanks! Some progress on the project. The photos I found and used to trace my schematic were the same ones you found.

I guessed some of the values for the buffer. Assume this would be a bipolar transistor, but could be a FET?

Re R55, I wasn't sure whether this should feedback to the C52 or whether should connect to the large ground trace on the topside? If as shown on your schematic, I cant understand what R54/R55 and C53 are doing?

User avatar
grrrunge
Diode Debunker
Information
Posts: 778
Joined: 20 Mar 2010, 10:51
Completed builds: Loads!
Location: Copenhagen, Denmark
Has thanked: 48 times
Been thanked: 192 times
Contact:

Post by grrrunge »

ryanuk wrote: 26 Sep 2022, 12:24Many thanks! Some progress on the project. The photos I found and used to trace my schematic were the same ones you found.
Fair enough. I thought you had some gut shots from a youtube video as well :)
ryanuk wrote: 26 Sep 2022, 12:24I guessed some of the values for the buffer. Assume this would be a bipolar transistor, but could be a FET?
The Analogman website describes the buffer as a tubescreamer input buffer with an op-amp based output buffer. This points towards a BJT based input buffer.
ryanuk wrote: 26 Sep 2022, 12:24Re R55, I wasn't sure whether this should feedback to the C52 or whether should connect to the large ground trace on the topside? If as shown on your schematic, I cant understand what R54/R55 and C53 are doing?
For low frequencies, the gain of the inverting amplifier IC2B is given by -R56/R54
At higher frequencies C53 acts as a short, so you have R54 and R55 in parallel, and your gain becomes -R56/(R54||R55). Essentially this gives you a well defined behavior for a treble booster.

All in all it does seem like a far stretch to market this as a buffer for the tone purists. After all it's an input buffer followed by two very frequency altering amplifier stages. The output is - however - very low impedance :)
This does in turn explain why i thought the clips i heard of this thing on youtube had a somewhat raspy character to it, whereas other buffers just reduce cable loss to a negligible amount.
A true believer in the magic of Sherwood Forest Pedal Pirates
---
New base of operations: http://www.knucklehead.dk

User avatar
grrrunge
Diode Debunker
Information
Posts: 778
Joined: 20 Mar 2010, 10:51
Completed builds: Loads!
Location: Copenhagen, Denmark
Has thanked: 48 times
Been thanked: 192 times
Contact:

Post by grrrunge »

I gave it a spin in LTspice to guesstimate the remaining values. Here goes:
Attachments
am buffer.png
am buffer.png (11.06 KiB) Viewed 367 times
SIM.PNG
A true believer in the magic of Sherwood Forest Pedal Pirates
---
New base of operations: http://www.knucklehead.dk

User avatar
grrrunge
Diode Debunker
Information
Posts: 778
Joined: 20 Mar 2010, 10:51
Completed builds: Loads!
Location: Copenhagen, Denmark
Has thanked: 48 times
Been thanked: 192 times
Contact:

Post by grrrunge »

@ryanuk: If you check out the gain/phase plot of amp 1 above, you can clearly see the frequency dependent gain introduced by R55 and C53, which then gets damped again at the output by R59 and C55.
A true believer in the magic of Sherwood Forest Pedal Pirates
---
New base of operations: http://www.knucklehead.dk

User avatar
ryanuk
Solder Soldier
Information
Posts: 200
Joined: 29 May 2008, 11:18
Has thanked: 26 times
Been thanked: 26 times

Post by ryanuk »

grrrunge wrote: 26 Sep 2022, 14:13 Fair enough. I thought you had some gut shots from a youtube video as well :)
I misled you there. I was inspired to reverse this buffer, as I had seen a video on youtube where Analogman demonstrates the buffer in front of an Ernie Ball VP. No close ups of the board though.
grrrunge wrote: 26 Sep 2022, 14:13 At higher frequencies C53 acts as a short, so you have R54 and R55 in parallel, and your gain becomes -R56/(R54||R55). Essentially this gives you a well defined behavior for a treble booster.
Thanks for the explanation - I had it in my mind it was some sort of a tone shaping circuit. Kinda reminded me of a Boss FA1 in typology. Thank you for your patience in explaining. Im still learning.

I guess we can now replicate the PCB. I'll have a bash when I get opportunity.

User avatar
Dirk_Hendrik
Old Solderhand
Information
Posts: 4152
Joined: 03 Jul 2007, 08:44
Location: Old Amsterdam
Has thanked: 215 times
Been thanked: 854 times
Contact:

Post by Dirk_Hendrik »

Great work Grrrunge!
I am somewhat amazed.
In many cases we have had to read about Boss pedals (yes, I know this thread is not about Boss and I'll get to that) and about their notoriously bad buffers, associated with tone suck. Nasty thing about that is that Boss does use a variety of different circuits for their buffers and therefore the "bad Boss buffers" generalization cannot be. However, I very often had the feeling this complaint was far more common for the time delay based pedals (Delay, chorus, flange) than for others. MOst of those have a high boost filter before the time delay and a high cut filter after the delay. Both filter (frequency) curves are each others opposite and therefore the effect is theoretically "inaudible". (the phase response of the signal is totally ff-ed and that does have influence on the sound). Together that's the pre and de-emphasis filtering.

What amazes me here is that this "buffer" is in essence a transistor buffer followed by a pre/de-emphasis filter section. That's more or less the buffering that is generally disliked for its tone suck character., as well that this "buffer" definitely will have a very sonic character. Not something that a buffer should be associated with. And that in turn does make me wonder (for the so-manieth time again) what exactly is going in in Analog Mikes head that makes him go exactly the wrong direction.
Sorry. Plain out of planes.

http://www.dirk-hendrik.com

User avatar
grrrunge
Diode Debunker
Information
Posts: 778
Joined: 20 Mar 2010, 10:51
Completed builds: Loads!
Location: Copenhagen, Denmark
Has thanked: 48 times
Been thanked: 192 times
Contact:

Post by grrrunge »

Admittedly, the design decisions has me puzzled as well. Thats what caught my interest in the first place
The 16dB boost at high frequencies will most certainly add harmonic content to signal transients, when the op-amp is driven into clipping. This might in turn be interpreted at “warmer“ sounding - although that's another discussion. In terms of transparency, this is not at all sonically transparent circuit design.
A true believer in the magic of Sherwood Forest Pedal Pirates
---
New base of operations: http://www.knucklehead.dk

Post Reply