Bob Cordell's Super Gain Clone PCB (LM3886) and a stripped-down version: Compact3886

Member
Joined 2018
Paid Member
Here is the pencil and paper method of gain calculation. It gives answers that are about 5% below yours. More importantly, pencil+paper matches the gain number Bob Cordell wrote in his book: page 686 of the 2nd edition, page 541 of the 1st edition.

_

Thanks for that.

The schematic shows R6 as 47k and R7 as 470. You used different values for the calculation. Any reason for this that I should be aware of before the build.
 
Member
Joined 2011
Paid Member
Referring to Bob's official schematic in post #9 of this thread, gain is equal to (R11 / R9).

796420d1574297970-bob-cordells-super-gain-clone-pcb-lm3886-stripped-version-compact3886-sgc_from_book-jpg
 
Member
Joined 2011
Paid Member
Each of the three DIP-8 integrated circuits has its own dedicated pair of supply bypass capacitors. On the printed circuit board, these caps are immediately adjacent to the supply pins of the DIP, for tight coupling and effective bypassing. However, on the schematic, these caps have been moved out of the way, to make the signal paths easier to read (less cluttered).

"U4" (the input receiver chip) has C10 and C19

"U5" (the Klever Klipper chip) has C16 and C21

"U8" (the DC servo chip) has C15 and C20

You can verify this when you've got a PCB in your hands.

As far as I'm aware, nobody has built a set of four Super Gainclone boards, two with C10=C15=C16=C19=C20=C21 = 100nF, and two more boards with different capacitance values in those positions, then listened to the resulting pair of stereo amplifiers. So "any audible differences" is not a hypothesis that's been tested, at least not so far.

Remember that it's your D.I.Y. project and you can build it any way you like. It seems like a fairly low risk experiment to try different values of bypass capacitor(s) and discover whether you enjoy the results. It might be a lot of fun!
 
Member
Joined 2011
Paid Member
In November I sold (at my cost) the excess boards I hadn't used myself, from the initial prototyping run at the PCB fab. There weren't a lot of PCBs; all were gone after only eleven orders.

Then in December I launched a Group Buy, for much greater quantities. This time there were many more buyers, and everyone got a nice volume discount. The orders have been placed, the money has been collected, and now we're waiting for the GB boards to ship from China to California.
 
value of C23, mute pin capacitor

l notice that the datasheet application circuit uses 100 uf instead of 10 uf in the supergainclone circuit .

i am learning electronic and in this circuit if l use 100uf, there will be a switch off thump on powering off because the power supply to the DIP op amp collapsed earlier .

Am l correct with the reasoning?
 
Member
Joined 2011
Paid Member
Bob Cordell's design is shown in post #9. It has ten microfarads on the LM3886 MUTE pin. The schematic for this Super Gain Clone PCB (in post #1) also has ten microfarads on the LM3886 MUTE pin. I recommend that you connect ten microfarads to the MUTE pin.

Despite this recommendation, if you still want to experiment, please go right ahead! It's okay to deviate from the recommended values, as long as you remember you are taking a risk. In snow skiing it's called "off piste" -- you have chosen to ski where there is no marked trail -- and if you fall down & get hurt, nobody will find your dead frozen body for several days. This is D.I.Y. after all, emphasis on the Y. Do what You want to do and have some fun! Play with whatever "knobs" you like.

There are other, non-Bob-Cordell designs for LM3886 boards "out there" which install a Zener diode in the mute circuit, supposedly for extra margin of safety. (Bob's Super Gain Clone design has no Zener diode in the mute circuit). An early, early builder of this PCB has deviated from the official schematic and installed a Zener diode (not on the schematic!) in her/his Mute circuit of her/his board. I understand that s/he is happy with the result.
 
Member
Joined 2011
Paid Member
Please have a look at the "silk screen" lettering on the top side of the PCB. Its purpose is to help you know which layout footprint corresponds to a certain schematic component.

You can also treat the question as a recreational puzzle: figure out what goes where by gathering all the clues and then following them to a conclusion. For example ...

The schematic says that one of R18's legs is connected to LM3886 pin 3, and its other leg is connected to LM3886 pin 9.

Therefore we expect to find R18 somewhere close the LM3886. And indeed it is close! R18 is the very first component below the LM3886. See plot below.

_
 

Attachments

  • sch.png
    sch.png
    26.2 KB · Views: 1,200
  • position_of_resistor.png
    position_of_resistor.png
    8.4 KB · Views: 1,157